Stics Soil Crop Model Ed1 v1
Stics Soil Crop Model Ed1 v1
Stics Soil Crop Model Ed1 v1
Conceptual framework,
equations and uses
Éditions Quæ
Collection Update Sciences & Technologies
This book was funded with support from the French National Fund for Open Science
(project AAPFNSO2019OPEN-STICS-13999) and from CIRAD.
Éditions Quæ
RD 10, 78026 Versailles Cedex
France
www.quae.com www.quae-open.com
Dedication...................................................................................................................................... 11
Preface. . ........................................................................................................................................... 13
Nicolas Beaudoin, Patrice Lecharpentier and Dominique Ripoche-Wachter
The first book..................................................................................................................................... 13
A new book based on an innovative approach.................................................................................. 14
Project funding.. ................................................................................................................................. 16
English revision.................................................................................................................................. 16
Chapter 1. Introduction.............................................................................................................. 17
Dominique Ripoche-Wachter, Nicolas Beaudoin and Eric Justes
1.1 History. . ........................................................................................................................................ 17
1.2 Purpose. . ....................................................................................................................................... 19
1.3 The STICS open book.................................................................................................................. 20
1.4 Human-machine interfaces.......................................................................................................... 22
1.5 Tools for STICS users. . ................................................................................................................. 24
Chapter 3. Development............................................................................................................. 47
Nadine Brisson, Iñaki García de Cortázar Atauri, Marie Launay and Dominique Ripoche-Wachter
3.1 The importance of phenology on crop development. . ................................................................. 47
3.2 Simulated events.......................................................................................................................... 48
3.3 Main development processes....................................................................................................... 50
3.4 Emergence and initiation of crop development and growth....................................................... 57
3
STICS soil-crop model
Chapter 10. Transfers in soil: water, nitrate and heat fluxes............................................. 201
Joël Léonard, Bruno Mary, Guillaume Jego, Nicolas Beaudoin and Nadine Brisson
10.1 Water and nitrate fluxes........................................................................................................... 201
10.2 Soil temperature....................................................................................................................... 208
4
Contents
10.3 Modifications of surface conditions that influence water and heat transfer........................... 210
Chapter 12. Carbon and nitrogen transformations in soil and balances....................... 233
Bruno Mary, Fabien Ferchaud, Hugues Clivot and Joël Léonard
12.1 Organic and mineral C-N pools............................................................................................... 233
12.2 Decomposition and mineralisation of organic matter............................................................. 235
12.3 Nitrification.............................................................................................................................. 243
12.4 Denitrification.......................................................................................................................... 248
12.5 N2O emissions.......................................................................................................................... 252
12.6 Ammonia volatilisation............................................................................................................ 256
12.7 Carbon and nitrogen balances................................................................................................. 259
Chapter 15. Tools for smart use of the standard STICS model version.. ........................ 323
Dominique Ripoche, Christine Le Bas and Nadine Brisson
15.1 Driving use options.................................................................................................................. 323
15.2 Strategy use options................................................................................................................. 327
15.3 Model formalism options......................................................................................................... 331
5
STICS soil-crop model
References.................................................................................................................................... 405
Chapter 17. Definition of symbols.......................................................................................... 437
17.1 Definition of parameters.......................................................................................................... 437
17.2 Definition of output variables.................................................................................................. 472
17.3 Internal variables definition..................................................................................................... 504
6
Foreword
Peter Thorburn
7
STICS soil-crop model
Some of these applications have indirect or direct links with government policy.
Modelling in this context raises new challenges for model development and applica-
tion coming from the increased scrutiny to which the results will be subjected (Moore
et al., 2014). In agriculture, models started as tools of scientific enquiry. For example,
CT de Wit’s interest in modelling was sparked by the desire to know the potential
yield of a crop (Keating and Thorburn, 2018). The scrutiny of such modelling was
likely limited to scientific peers who likely understood and accepted the strengths and
weaknesses of modelling [although the was not always the case; e.g. Passioura (1996)].
As models developed and modellers started using them to inform farmers how to
improve their management, scrutiny expanded to include farmer stakeholders as well
as scientific peers. However, in many farmer interactions the model (or simulation
output) acted as a “boundary object” facilitating discussions between the modeller and
farmer (Jakku and Thorburn, 2010). The modeller explaining to the farmer the simu-
lation results and their meaning built trust in the farmer of the modeller (provided the
explanations made sense to the farmer!). This was/is essentially a social process and
the technicalities of the model application itself were not necessarily scrutinised – if
the farmer trusted the modeller, she trusted the model. Further, the farmer was free to
change farm management, or not, as a result of these interactions, and solely bore the
consequences of any changes (whether positive or negative).
In public policy applications, the link between the modeller and (government) stake-
holder is likely to be much less personal than between modellers and farmers or
scientific peers. Further, the application of the policy will often create “winners” and
“losers”. It is natural for the “losers” to want to scrutinise the technical basis behind
the policy impacting them. A recent example of this is the examination of model-
ling behind water quality policy for agricultural lands in New Zealand (Johnson et al.,
2021). The “losers”, and other stakeholders, will likely ask questions about the quality
of the science in the model, whether that is accurately implemented in the code (and
for the specific version of the model used in the analysis) and whether the model
was competently run. Publication in the peer reviewed literature is often the means
of assuring the quality of the science. As a community, however, cropping systems
modellers have less established methods of quality assurance for implementation and
running models than some other communities. For example, just for calibration of
phenology, an important but limited part of running a crop model, there is a huge
diversity of approaches used by different modellers (Seidel et al., 2018) and there
will be benefits from having some consistency in the approach (Wallach et al., 2021).
Conversely, ensuring accurate implementation of science in model code, i.e. having
good software development practices, has received less attention across cropping
systems models (Holzworth et al., 2015). It is therefore significant that this issue has
been discussed in the 2nd edition of the “STICS red book”.
What does the future hold for cropping systems models? That question has been
addressed in a number of recent papers (Jones et al., 2017; Keating and Thorburn, 2018;
Silva and Giller, 2020) and their conclusions do not need repeating here. However,
those authors agree that application of cropping systems models will be an important
methodology in meeting the coming challenges faced by food and agricultural systems
and the models will need to be further improved and developed. That development
will necessitate increasing efforts in collecting data to underpin those developments.
8
Foreword
Data availability has always been both a limitation and driver of model development:
in many respects cropping systems models have been created to overcome the scarcity
of data. As we enter the age of “big data”, data will increasingly be available from remote
and proximal sensors. That raises the questions of how those data will aid model devel-
opment and/or application, and how will they affect the relevance of cropping systems
modelling? An example of the first question is the potential use of multi-year high
resolution data on crop growth and development to inversely parameterise models,
e.g. soil water (He and Wang, 2019) or phenology (Araya et al., 2016) parameters,
aiding subsequent application. The implications of the second question are less clear.
With rich data, possibly less biophysical detail is needed in a model if it is designed for
use in conjunction with those data (e.g. Donohue et al., 2018). Even further, there may
be no role for a biophysical model at all. However, this “struggle” between models and
data for prediction and understanding is not new. An example is the prediction of the
optimum rates of nitrogen fertiliser in the mid-west corn-belt of USA. Large datasets
have been gathered and developed into a tool for forecasting the Maximum Return
to Nitrogen (Sawyer et al., 2006; http://cnrc.agron.iastate.edu). However, recently
developed approaches based on cropping systems modelling are showing promise in
increasing accuracy of those forecasts (Puntel et al., 2018). And, unlike purely data-
driven approaches, the cause of a result from a cropping systems model can be tracked
down and understood, enlightening the modeller and their stakeholders. Thus, it is
unlikely there will be a single “winner” in the “struggle” between models and data for
prediction. What is clear however, is that modelling systems (structure and software)
will need to evolve to be easily applied with these new sources of data. The STICS
model is well advanced down that development road and thus will remain relevant for
a long time. I forecast there will be a 3rd edition of the “STICS Red Book” in the future!
9
Dedication
This book is dedicated to Nadine Brisson… Nadine Brisson was an ever-enthusiastic
captain of the STICS ship and kept her sights set far ahead. With her intelligence,
energy, insightfulness, determination and presence, she was always able to bring
people on board with her who would remain ever faithful. Her vessel covered great
distances, towards the shores of all the continents, sometimes facing storms along the
way but continuing cheerfully and steadfastly onwards, committed to fulfilling her
pledge: to give the scientific community a tool to help tackle food security, climate
change, agroecological transition and other major challenges. She would have been so
pleased with this new edition, which is available at zero cost (free numeric version),
in keeping with the values of Open Science that she promoted before the movement
had even emerged. Her life ended much too soon, but it was full of professional and
personal adventures. She became fast friends with nearly everyone she met, and this is
how we will forever remember her.
To Nadine
Source: LyricFind
Parolier : Georges Brassens (1964)
Extrait des paroles de Les Copains d’abord © Universal Music Publishing Group
11
Preface
Nicolas Beaudoin, Patrice Lecharpentier
and Dominique Ripoche-Wachter
The first edition of this book was written primarily by Nadine Brisson and was quite
original in that synthesised scientific knowledge about cropping systems. The book
covered the STICS model formalisms in an exhaustive way. But, more than ten years
on, it was in need of a comprehensive update following the profound changes to the
capabilities of the STICS model.
The following authors contributed to the original formalisations according to their
affiliations:
– INRA (now INRAE): R. Antonioletti, N. Beaudoin, P. Bertuzzi, T. Boulard,
N. Brisson, S. Buis, P. Burger, F. Bussière, Y.M. Cabidoche, P. Cellier, P. Debaeke,
13
STICS soil-crop model
F.
Devienne-Barret, C. Durr, M. Duru, B. Gabrielle, I. García de Cortázar Atauri,
C. Gary, F. Gastal, J.P. Gaudillère, S. Génermont, M. Guérif, G. Helloux, C. Hénault,
B. Itier, M.H. Jeuffroy, E. Justes, M. Launay, S. Lebonvallet, G. Lemaire, B. Mary,
T. Morvan, B. N icolardot, B. Nicoullaud, H. Ozier-Lafontaine, L. Pagès, S. Recous,
G. Richard, R. Roche, J. Roger-Estrade, F. Ruget, C. Salon, B. Seguin, J. Sierra,
H. Sinoquet, R. Tournebize, C. Valancogne, A.S. Voisin
– ESA-Angers: Y. Crozat
– ARVALIS Institut du végétal: P. Gate
– CEMAGREF (now INRAE): B. Rebière, J. Tournebize, D. Zimmer
– CIRAD: F. Maraux
1. https://www.zotero.org/
14
Preface
Figure 0.2. The workflow for the new book showing the dynamic interaction between the three
activities of the STICS project team
The book was produced using essentially R language (R Core Team, 2020) and a
specific format for reproducible document writing, R Markdown (https://rmarkdown.
rstudio.com/, Allaire et al., 2021). The chosen configuration was based on the ‘book-
down’ package, which is designed for book or other document content formatting
(https://bookdown.org/yihui/bookdown, ; Xie, 2016, 2021).
These packages are distributed under the GPLv3 licence and allow users to generate
different output formats such as HTML, PDF, DOC, EPUB (Figure 0.3). The R Mark-
down format makes it easy to integrate lines of R code, for example, in order to generate
illustrations (tables, graphs, etc.). Mathematical equations and literature references
can be automatically formatted using specific syntaxes.
15
STICS soil-crop model
For a subset of these types of insertions and formatting, operations need additional
R functions which have been developed separately from the packages used for this
book (for equation implementation and plots), as well as data from simulations or
observations, or which have been formatted in a specific way.
The book production is managed as a development project in the form of an RStudio
project (RStudio Team, 2021, https://www.rstudio.com/). All necessary files (code,
text and data) are managed using a version control system (Subversion, https://subver-
sion.apache.org/) shared among all contributors (essentially STICS team members at
the moment). This means that every author is able to maintain the content of the
book project: they can save changes, get changes made by other authors, and finally
produce the book or parts of it in different file formats. This package permits dynamic
interaction between STICS book editions, scientific design activities and software
maintenance (Figure 0.2).
Project funding
This project was funded with the support of:
- the French National Fund for Open Science (FNSO),
- CIRAD, and
- the Agroecosystems Division of INRAE.
English revision
Teri Jones-Villeneuve (teri@jonesvilleneuve.com)
Throughout the revision process for the book, Teri offered helpful feedback by
pointing out areas where additional clarity was needed and suggesting improvements.
The entire STICS project team thanks her for her professionalism and are very grateful
for this contribution to the book.
16
Chapter 1
Introduction
Dominique Ripoche-Wachter, Nicolas Beaudoin
and Eric Justes
Reviewed by: Guillaume Jego and Patrice Lecharpentier
This introduction will outline the purpose and brief history of the STICS model, along
with the general concepts. We will also discuss the collective dynamics of its develop-
ment, evaluation and governance. Finally, we will introduce the different chapters of
this book which describes the algorithms in the model. A list of user network services
can be found at the end.
Some sections of this chapter come from the translation of the book chapter
« Modélisation du fonctionnement des agro-écosystèmes: l’épopée STICS (Beaudoin
et al. 2019), with the permission of QUAE Edition ».
1.1 History
The model design started with a first meeting held in France in 1996, where the foun-
dations of the key model principles were laid out. The aim was to create “a single
model for different crops that integrated both specialist and generalist knowledge in
order to be general, robust, simple, operational and flexible.” (Beaudoin et al., 2019)
The specifications of STICS were co-developed by researchers and engineers from
various disciplines in the fields of agronomy (sensu lato) and modelling, who recom-
mended four main characteristics:
– A balance between the different compartments and interacting processes in the
soil-plant-atmosphere system, in order to produce a generic soil-crop model, appli-
cable to a wide variety of themes and contexts.
– Genericity of the plant functioning description, based on general ecophysiological
concepts, guiding to the definition of a single model. The same basis applies to the
simulation of soil functioning processes.
– Simple and uncomplicated input data, with easily accessible parameters that are not
very sensitive to change of scale, which facilitates the model’s operational use in real
agricultural situations.
– Robustness of formalisms and their parameterisation, which ensures realism in a
wide range of agro-environmental conditions, and including management practices.
17
STICS soil-crop model
An additional characteristic scalability has emerged over time. The original strategy
was to work together to design a dynamic, functional, one-dimensional model, with a
strong agro-environmental aim. This soil-crop model can perform simulations either
at a plot scale (the model’s aim) or at a macro-regional scale, but only in conjunction
with other models or tools (e.g. software platforms).
STICS was developed at INRAE (formed following the merger of INRA and IRSTEA)
in collaboration with other research and educational institutes such as the French
agricultural research and cooperation organization CIRAD, the Graduate School-
Ecole des Mines de Paris and the Laboratory for Sciences of Climate and Environment
– LSCE, along with French professional institutes (ARVALIS, Terres Inovia, CTIFL,
ITV, ITB, Agrotransfert), and few other partners.
The STICS model is open source (Licence CeCiILL C, v2.1, the French equivalent
of a lesser general public license, or LGPL). Its code has been filed with the Agency
for the Protection of Programs (APP), the European organisation for the protec-
tion of authors and publishers of digital creations, under the reference number:
IDDN.FR.001.360007.000.S.C.2021.000.10000.
In the early STICS development stages, many well-known models were available
(CERES: Ritchie and Otter (1985), ARCWHEAT: Weir et al. (1984), EPIC: Williams
et al. (1989), SUCROS: van Keulen and Seligman (1987), among others) that were
18
Introduction
developed from the pioneering work of de Wit (1978) or Duncan (1971). However,
new models appeared regularly in the literature (SSM: Amir and Sinclair (1991),
CROPSIM: Hunt and Pararajasingham (1995), WANGRO: Rao Kanneganti and Fick
(1991), GRAMI: Maas (1993), SHOOTGRO: McMaster et al. (1991); Teittinen et al.
(1994), etc.). As Sinclair and Seligman (1996) explained, this is because there cannot be
a single universal model can exist in the field of agricultural science and it is necessary
to adapt system definitions, simulated processes and formalisations must be adapted
to specific environments or to new problems (technical, genetic, environmental, etc.).
It is also understandable that each modelling team wants to develop its own model in
order to control the code and its development. These authors emphasize the heuristic
potential of modelling, which was a crucial element in the STICS development. The
work on STICS was firstly published in 1998 by Brisson et al. (1998b).
These timelines shown in Figures 1.1 and 1.2 describe the different versions that have
been developed and the changes in formalisms and simulated processes.
1.2 Purpose
The aims of the STICS model (the acronym stands for Scientific, Technical and Inter-
disciplinary simulator of soil-Crop System functioning) are similar to those of many
existing crop models (Whisler et al., 1986). However, STICS also takes into account
cropping system diversity, including soil functioning. STICS is a soil-crop model, and
not just a crop model, which runs at a daily time-step and using input variables related
to climate, soil and the cropping system and its management.
According to Daniel Wallach et al. (2018), STICS can be defined as a deterministic
process-based model. It can continuously simulate successions of various crop species
and fallow periods. Additionally, it can be considered as a cropping system model
because it accounts for a wide range of dynamic interactions between the different
modelled compartments: crop, soil, climate and agricultural management techniques
and functions in continuous by simulating the management of crops and fallow
periods over long periods of time.
The output variables describe crop yield in terms of quantity and quality (with criteria
associated with C and N) as well as environmental impacts linked to soil-C changes
and CO2 emissions, water drainage, nitrate N leaching and gaseous N emissions. The
elementary simulated object is a local cropping situation for which a physical medium
(soil and weather) and a crop management schedule are set through simulation input
parameters. The main simulated processes are crop development, growth and yield
production, as well as carbon, water, nitrogen and energy balances, and soil functioning.
From a conceptual point of view, STICS includes many original features compared
with other well-known crop models, such as simulation of crop temperature, a snow
module, simulation of various techniques and management options, bi-specific inter-
cropping, and many others. While most of the basic features are based on classical
formalisms or have been adapted from existing models. Nevertheless, several strong
points of the STICS model should be noted (Brisson et al., 2003; Brisson et al., 1998b):
– Crop genericity: adaptability to various type of crops covering a wide range of
botanical families (wheat, maize, soybean, sorghum, flax, grassland, tomato, sugar beet,
sunflower, vineyard, pea, rapeseed, banana, sugarcane, carrot and lettuce, among others)
19
STICS soil-crop model
STICS can simulate annual and perennial crops as well as bi-specific intercropping
systems (§ 2.2.3).
20
Introduction
1.3.3 The chapters related to the mortality and recycling of crop residues
Mortality is described in each chapter relating to the growth of each organ: for the
leaf (§ 4.1.2.2), for the roots (§ 5.5.4 and 5.4). The residues that remain on the ground
decompose according to mechanisms described in § 12.2 and 7.8. The existence of this
recycling depends on the cultivation techniques is described in § 13.5.3.
21
STICS soil-crop model
22
Introduction
All the parameters that are involved in simulated processes and attached to a USM are
described in the following chapters. A full list is also included in the appendix section
(§ 17.1). The simulation output variables of a simulation describe the different simu-
lated processes on a daily time step, such as crop variables (phenological stages, yield,
biomass, etc.) and soil and environmental variables (water, N, C balances, etc.). These
variables are also listed in the appendix section (§ 17.2).
1. http://www6.paca.inrae.fr/stics_eng
2. https://www6.inrae.fr/record_eng/
3. https://galaxyproject.org
23
STICS soil-crop model
Figure 1.6. Phases of modelling steps in the model design and validation process. Adapted
from Schlesinger and Bellocchi (Bellocchi et al., 2011)
24
Introduction
While the aim of this book is not to discuss model performance in great detail, the
tools and the methods used in the evaluation process are described in § 16. Perfor-
mance is addressed briefly for illustrative purposes in § 14.5.5.1.
4. https://subversion.apache.org
25
STICS soil-crop model
– Ozone absorption and its effects on plant growth that could induce stress and
reduce biomass and yield production;
– Cryptogamic diseases interactions with plants: using a generic module to be coupled
to STICS. The implemented formalism has been already published (Caubel et al., 2017;
Caubel et al., 2014, 2012).
Other branches are also planned to improve the model robustness and accuracy for
specific situations, such as:
– Cumulative effects of organic amendments on soil C-N storage (Levavasseur et al.,
2021);
– Perennial crops functioning (Strullu et al., 2020; Strullu et al., 2015), regarding
diversity of grassland types and managements techniques (including animal returns at
grazing) and their ability to store C in soil (Graux et al., 2020).
These efforts illustrate that STICS stands for a past, present and also future soil-
crop and cropping system model that is engaged in a dynamic process of continuous
improvement.
5. https://www.jenkins.io
26
Introduction
Figure 1.8. Diagram of the test bench of the model mobilising IDE-STICS.
1.5.3.1 SticsRPacks
The SticsRPacks suite project was initiated at the end of 2018 to develop tools for
driving the STICS model using R language6. This collaborative project (INRAE/
CIRAD) is open source and hosted on the GitHub platform7.
The purpose of these tools is to:
– perform operations that are not provided in JavaSTICS, such as generate/modify
parameters files, produce various graphs, handle statistical processing, etc.;
– automate these operations using scripts;
– reduce computation time to perform simulations.
More information about these tools is given in § 15.5.2.
6. https://www.r-project.org
7. https://github.com/SticsRPacks
27
STICS soil-crop model
– defining and implementing functionalities for data requests for a given test;
– allowing scalability of model skills between STICS versions or several soil-crop
models, thanks to a formalised thesaurus (following Porter et al. (2014) and the
European AnaEE project).
Integrating data into IDE-STICS will require time, agronomic skills and reciprocal
interest between modellers and agronomists. A prototype of data integration process
has been already implemented and tested by experimental agronomists but is not yet
available for users.
8. https://www6.paca.inrae.fr/stics_eng/About-us/Project-Stics-Team
28
Introduction
colleagues, such as Bruno Mary, who co-created the STICS model with Nadine;
Nicolas Beaudoin, coordinator of the STICS-ProTeam when Nadine left us; and
Françoise Ruget and Patrick Bertuzzi, who recently joined the contingent of emeritus
members. We are ever grateful to all our retired colleagues who have helped make
STICS what it is today!
29
Chapter 2
Overall description of the modelled system
Nicolas Beaudoin, Dominique Ripoche-Wachter,
Marie Launay, Eric Justes and Nadine Brisson
Reviewed by: Françoise Ruget
31
32
STICS soil-crop model
Figure 2.1. Schematic representation of a crop model, adapted from Bonhomme and Ruget (1991).
* Bioagressors can be accounted for thanks to STICS coupling.
Overall description of the modelled system
(and N2 fixed from atmosphere for legumes). Depending on plant type, crop devel-
opment is driven either by i) a thermal index (degree-days), ii) a photo-thermal
index which also takes photoperiod into account or iii) a vernalo-photo-thermal
index which also takes vernalisation into account. The development module drives
the kinetics of the leaf area index (lai) and the roots, which capture abiotic resources,
and defines the harvested organ filling phase. Any water or nitrogen stresses will
reduce leaf growth and biomass accumulation, and the degree of reduced growth
is based on stress indices calculated by water and nitrogen balance modules. Other
abiotic stresses, such as thermal (frost or high temperatures), trophic or waterlog-
ging stresses are also taken into account, as limiting factors of growth and yield
formation. The response function to environmental constraints can also integrate
enzymatic activities, such as nitrate uptake.
Crop organs are defined as plant compartments to witch generic ecological func-
tions are assigned. Here, the distinction between the structural and temporary pool
of carbohydrates is a key point. Only the structural aspect is assigned to the leaves,
stems, fruit, perennial reserves and roots. Conversely, the temporary reserves are not
located into the crop organs, in order to support the model genericity for all crops,
including annual cereals and perennial forage crops. The storage organs and harvested
organs can be grains, fruits, tubers or even stems. The new STICS version 10 now
considers the C and N fluxes to and from temporary and perennial storage organs.
The model simulates the water and nitrogen elements balances in soil, to assess
plant availability and environmental losses (§ 2.1.3). Their descriptions are based on
the classic compartmental approach, which defines different pools in the system,
their evolution and their relationships (input and output functions; see for instance
Figure 12.5 from (Nicolardot et al., 2001). The functions encompass physical,
physic-chemical and bio-chemical processes. The compartment size depends on
the process and ranges from centimetric (elementary layer) for solute content and
transfers to decimetric (pedologic layer) for defining soil properties.
The processes are described in a way which can be qualified of functional and process-
based or analogic, which aims to favour a simple integration of all the processes
interacting in the soil-crop system, at the crop cycle scale as well as the crop rotation
scale. The most emblematic illustrations of this conceptual choice are:
– The interception efficiency of the total solar radiation uses Beer’s law, as an expo-
nential response of the shadowing effect to leaf area index (lai), in the case of a
homogeneous canopy.
– The conversion of photo synthetically active radiation (PAR) in aerial biomass is
based on the concept of radiation use efficiency (RUE), assuming a specific poten-
tial linear response of biomass accumulation, at a stable radiation level when there
is no carbohydrate remobilisation (Monteith, 1972); the simulated RUE depends on
radiation level, water stress and N stress, CO2 concentration and the crop stage.
– The nitrogen uptake and accumulation in the plant is based on the concept of N
dilution curve in the aerial biomass of the canopy, in the case of vegetative growth, for
either an isolated plant or dense canopy (Lemaire and Gastal, 1997). This allows the
user to calculate the crucial concept of nitrogen nutrition index (NNI) which drives
the N status of the crop in the vegetative phase.
– Fruit filling includes the concept of dynamic harvest index, as proposed by Spaeth
and Sinclair (1985), for crop species with determinate development.
33
STICS soil-crop model
– The downward soil water storage, when daily rainfall exceeds the actual evapotran-
spiration amount, uses the tipping bucket concept, at the elementary layer infiltration
scale, based on the hypothesis of infinite hydric conductivity.
– Nitrate leaching is simulated using a mixing cell model, with similar results to those
of the resolution of the convective-dispersive equations (Darcy’s law and Richard’s
law), except for soils with low hydric conductivity, with easier soil parameterisation
and a much lower time of calculation (Van Der Ploeg et al., 1995).
– Soil C and N mineralisation are simulated using a compartmental approach, with
only three pools of organic matter, which resembles the AMG model (Clivot et al.,
2019); this compartmental approach allows an independent parameterisation linked
to a residue typology (Nicolardot et al., 2001). This approach is based on the crucial
concept of C/N ratio of residues which drives their decomposition and humification
rates, and determines the C and N mineralisation rate of the residues.
These processes are implemented by default in the standard STICS version; if their
basic hypothesis is not met, an alternative process, involving a more complex concep-
tual approach, is described and workable. For instance, for a heterogeneous canopy
(e.g. row crops or intercropping systems), the simple ‘big leaf ’ model associated with
the Beer’s law can be replaced by a more complex energy balance based on crop archi-
tecture and a resistive approach. This alternative process allows determine the sunlit
and shaded leaf layers of the bispecific intercrop and then the light capture differen-
tiation between these two layers and furthermore, using a spatial discretisation along
the crop inter-row (§ 9.3).
The effect of crop management on the dynamics of the soil-crop-microclimate systems
is also given particular attention (§ 13). The reason is that crop specificities influence
both ecophysiology and crop management (e.g. accounting for the various forms of
forage cutting, fertilizer composition, plastic or crop residue mulching, etc.).
Finally, STICS model is either functional or process-based at daily scale; it is mech-
anistic at the crop cycle scale, since it deals with the main interactions between the
system components. It can design the emergent properties at the crop cycle and crop
rotation scales.
34
Overall description of the modelled system
The biological cycles include those pertaining to the crop(s), which may cover one
or more cycles a year, and those related to the soil microbial biomass which depends
on the decomposing residues and soil organic matter. In established perennial crops,
a new crop cycle resumes according to either the imposed or calculated date of crop
regrowth. In consequence, the simulated RUE encompasses both aerial biomass and
perennial reserves building as opposed fine root system.
The biogeochemical cycles correspond to those for carbon and nitrogen. Each part
of their cycles occuring in the soil-crop-atmosphere system is modelled based on
several kinetic principles driven by the system forcing variables and according to the
law of conservation of mass. The simulated C balance at the soil-crop level is almost as
complete as compared to their description in literature (Hyvönen et al., 2007): all net
C fluxes are simulated except for CH4 emission and dissolved organic or inorganic C
in water drainage; however, only the difference between the gross primary production
and its respective gross respiration is simulated. The N fluxes are complete, allowing
to simulate positive or negative net N balance in function of the cropping system
management and pedoclimatic conditions (Autret et al., 2020).
35
STICS soil-crop model
From a practical point of view, the actual validity domain covers all the situations
for which the model has been shown to produce acceptable results. This empirical
concept is based on the model-use bibliometric (§ 14). Soil-crop model performance
depends on how realistic is the representations are as well as the reliability of the code,
and the quality of the parameters (Loague and Green, 1991). Users who feed in the
data and choose the simulation options and some parameters, can have a significant
impact on the validity domain (Confalonieri et al., 2016; Wallach et al., 2021).
The STICS application domain can also be extended in space and time, thanks to the
model’s intrinsic ability for either successive simulations or bi-specific intercropping
systems (see the next two sections) or by coupling the model with GIS (§ 14.3).
36
Figure 2.2. Simplified representation of the transferred variables between two successive units of simulation (USM); (*) organic C&N contents are
37
Overall description of the modelled system
simulated outputs while their initiale values belong to the soil characterisation.
STICS soil-crop model
38
Overall description of the modelled system
The soil environment is also assumed to be the same for both crops (i.e. the horizontal
differentiation within the soil profile is disregarded in favour of the vertical differen-
tiation. The assumption is made that the interactions between the two root systems
result from the influence of the soil on each crop root profile, based on its penetrability
and water dynamics.
This theory is applied within the STICS code via multiple calls to the elementary
subroutines and re-calculation of the state variables as a function of the considered
sub-system. Specific modules or options were added to account for the ecophysiolog-
ical features of these complex systems. These modules cover radiation interception and
the energy budget that drives water requirements and microclimate, and root system
dynamics which are influenced by soil status over the various layers of the whole soil
profile. Shoot growth was slightly modified to account for the understorey shaded
crop growing under limiting radiation. Those modules and options are described
in the relevant thematic chapters of this book. Reciprocally, the involved formalism
options can be applied for sole crop simulations, like the energy budget for row crops.
Figure 2.3. Simplified diagram of the model: on the right the system with its three sub-systems (D:
dominant canopy; U: understorey canopy divided into a shaded part (SU) and a sunlit part (LU));
in the centre, the number of calls to each module devoted to a particular part of the system; on the
left, the modules (grouped according to the way they are named in the code). * Corresponds to the
modules modified for the adaptation to intercropping (from Brisson et al., 2004).
39
STICS soil-crop model
A first set of three modules deals with the ecophysiology of above-ground plant parts
(phenology, shoot growth, root growth, yield formation). A second set of four modules
deals with how the soil responds in interaction with underground plant parts (root
growth, water balance, nitrogen balance, soil transfers). The crop management module
deals with the interactions between the applied techniques and the soil-crop system.
The microclimate module simulates the combined effects of climate and water balance
on the temperature and air humidity within the canopy.
Within each module, there are options that can be used to extend the scope of STICS
application to various soil-crop systems. These options relate aspects of ecophysiology
and crop management, such as:
– competition between vegetative organs and storage organs for assimilates (hereafter
referred to as trophic competition);
– the canopy geometry when simulating radiation interception;
– description of the root density profile;
– use of a resistive approach to estimate the evaporative demand by plants;
– mowing of forage crops;
– plant residues or plastic mulching under vegetation.
Another of the model’s strong point is its conceptual modularity: sub-programs are
identified for each group of ecophysiological processes, such as N or water balance,
crop growth, changes in soil C-N stocks, etc.
Figure 2.4. Simplified diagram of the order in STICS between processes within the daily loop.
40
Overall description of the modelled system
At each step, another order between the elementary functions is defined. Users will
find it is well worth their time to specify the order of priority of these elementary func-
tions, especially within the following items:
– Crop growth: shoot growth, leaf senescence, yield elaboration, C and N assimilate
allocation;
– C and N transformation: fertiliser, volatilisation, mineralisation, nitrification,
denitrification;
– Water requirements: soil evaporation, crop transpiration;
– N nutrition of the crop(s): symbiotic fixation for legumes, N uptake (minimum of N
demand and N supply);
– Stress indices calculation: water, nitrogen and abiotic factors (frost, anoxia);
– N partitioning in the crop(s): to the grain; between leaves, stems and reserves.
41
STICS soil-crop model
42
Overall description of the modelled system
43
STICS soil-crop model
The same principle applies for crop management, as illustrated in Figure 2.6. Indeed,
many different technical options may be activated or not, according to the crop onset,
fertilisation type, irrigation management, canopy control or harvest decision rule. For
intercropping, some techniques must be identical for both crops, according to site
specific practical considerations, such as tillage or harvest criteria decision.
With less modalities, similar illustrations could be applied to soil or climate station.
Finally, from a practical point of view, there is an asymmetry between the plant file, for
which the parameterisation is already set by STICS team for some crops and other input
files, which characterise the management, soil, climate and initial conditions. Filling
these input files requires a minimal data collection and expertise by the user (chapter 16).
Figure 2.5. Examples of parameterisation strategies as the result of activating a set options in
each specific plante file. Acronym: PET = potential evapotranspiration.
44
Overall description of the modelled system
Figure 2.6. Examples activated option pathways to parameterise the crop management.
Acronym: lai = leaf area index.
45
STICS soil-crop model
Figure 2.7. List of parameterised plant species as of 2021. The species given in bold have been
fully parameterised while those in regular font are at the prototype one.
46
Chapter 3
Development
Nadine Brisson, Iñaki García de Cortázar Atauri, Marie
Launay and Dominique Ripoche-Wachter
Reviewed by: Nicolas Beaudoin and Bruno Mary
47
STICS soil-crop model
48
Development
As in most crop models, the development stages simulated by STICS can differ from
the stages defined in classical agronomic scales. The development stages in STICS are
growth stages rather than organogenetic stages (Brisson and Delécolle, 1992). The stages
actually correspond to changes in the trophic or morphological strategy of the crop that
influence the evolution of leaf area index or grain filling (see examples in Figure 3.1).
Figure 3.1. Illustration of the stages of interest for various types of crops such as wheat (annual determinate), tomato (annual
indeterminate), grapevine (perennial indeterminate) and forage crop (perennial determinate interrupted by cuts symbolised by
scissors). The flowering stage IFLO is mostly confounded with the IDRP stage (in bold).
49
STICS soil-crop model
Using generic terms to name the various stages means that different species can
be simulated, exhibiting either determinate growth (vegetative and reproductive
growth occur successively) or indeterminate growth (vegetative and reproductive
growth occur simultaneously, at least partly). The stage (named iamfs below)
equates to the beginning of stem elongation and is generally not far from the end of
leaf initiation: it is the ‘1 cm ear’ stage for wheat and graminaceous forage crops, just
slightly later than the double-ridge stage for most varieties, whereas it is the floral
induction for corn. For indeterminate crops like tomato and grapevine, it is more
difficult to find an equivalent in organogenesis, so this stage is instead regarded as
the stage when the plant reaches a specific number of leaves (3, 4 or 5). The stage
must be regarded as a growth stage since it is the end of leaf onset, which can
occur before or after the stage. Nevertheless, some stages can also provide
information about other developmental stages for some crops. This is the case, for
example, with the stage, which is a proxy of the veraison stage (beginning of
ripening) for grapevine. Moreover, the beginning of grain filling ( ) is always
preceded by a key stage for the onset of the number of harvested organs (grains or
fruits) which can be for determinate crops and (named inous below)
for indeterminate crops. At physiological maturity ( ) (named imats below)
the harvested organs stop growing in dry matter terms and the - period
depends on the required quality for the final product.
50
Development
induction (Blum, 1996; Seghieri et al., 1995). Nitrogen nutrition conditions can also
have an effect on the progress of the cycle (Girard, 1997), as well as light conditions
through plant density (cryptochrome).
In STICS, crop temperature (udevcult) drives development. It may be slowed by
sub-optimal photoperiod conditions (rfpi ), by non-compliance with vernalisation
requirements (rfvi ) or by the effect of water or nitrogen stress (𝐬𝐭𝐫𝐞𝐬𝐬𝐝𝐞𝐯P and
turfac or innlai ). Thus, each day, the phasic course of the crop (upvt) is given
according to the Eq. (3.1):
As far as the emergence period is concerned, a specific calculation is made using the
conditions prevailing in the soil (see § 3.4) as for the root lifespan (𝐝𝐞𝐛𝐬𝐞𝐧𝐫𝐚𝐜P).
Leaf lifespan is expressed in exponential type time (also called Q10 time) for reasons
explained in § 4.1.2.
Most phasic courses between two successive stages are regarded as variety-specific
(Table 3.2), as are the parameters indicating the sensitivity to the photoperiod and
vernalisation requirements.
Table 3.2. Table summarising the various parameters of developmental duration and the
driving variables used to calculate those durations. The tcult variable is the crop tempera-
ture and 𝗍𝗌𝗈𝗅 is the soil temperature at the root front level.
Positive thermal response 1
developmental
nitrogen stress
requierement1
Slowing water
photoperiod1
Parameter of
stress effect1
Action of
2tcult/10
duration
Slowing
effect1
Cold
tcult
tsol
stpltgerP x x
stdordebourV x x x x
stlevamfV x x x x x
stlevdrpV x x x x x
stdrpmatV x x x x x
stdrpnouV x
stdrpdesV x
stflodrpV x x x x
dureefruitV x
durviefV x
phyllothermeP x
debsenracP x
1 If appropriate, this option is activated according to the plant sensitivity to the relevant factor.
51
STICS soil-crop model
52
Development
Figure 3.3. Photoperiodic limiting factor for phasic development ( ) when varying the
sensitivity to photoperiod a) with response type of wheat or the photoperiodic response type
b) with for both species, for wheat ([ , ] = [8.20]) and for
soybean ([ , ] = [18.15]).
53
STICS soil-crop model
Figure 3.4. Annual variation of the photoperiod for two northern latitudes and the consequence
on the corresponding limiting factor for phasic development ( ) calculated for wheat crop.
54
Development
3.3.4.1 Vernalisation
Vernalisation requirements are defined by the genotype-dependent number of
vernalising days (𝐣𝐯𝐜P), and the vernalising value of a given day (𝗃𝗏𝗂) depends on crop
temperature (Figure 3.5). Vernalising days are counted from germination (igers) for
annual crops, because an active metabolism is required to initiate the process, or
from the 𝐣𝐮𝐥𝐯𝐞𝐫𝐧𝐚𝐥P day for perennial crops. A minimum number of vernalising days,
𝐣𝐯𝐜𝐦𝐢𝐧𝐢P is required (Eq. (3.4)). The progress in crop vernalisation, rfvi, gradually
increases until it reaches the value of 1.
Figure 3.5. Vernalising value of a given day ( ) as a function of the mean crop temperature ( )
for wheat ([ , ] = [6.5, 10]) and for rapeseed ([ , ] = [6.5, 20]).
55
STICS soil-crop model
Figure 3.6. Sensitivity to the parameter (assumptions of 10°C and 20°C) on the
calculation of the period of vernalisation (rfvi) and its consequences on leaf growth ( ) for a
ryegrass catch crop sown in late summer.
3.3.4.2 Dormancy
This section deals with the perennial dormancy and not with the dormancy break of
seeds of annual crop grains such as wheat, barley or pea that can lead to germination
of the grain on the plants before harvest which are described in § 3.4.1.
The aim is to calculate the day of dormancy break, which makes it possible to change
the rfvi variable from 0 to 1, bearing in mind that it is always possible to impose this
date and ignore the following dormancy calculations (using the parameter 𝐢𝐟𝐢𝐧𝐝𝐨𝐫𝐦P).
For perennial plants, the active onset of vegetative development generally occurs after
a period of winter rest (if this is not the case, chilling requirements are set to 0). The
dormancy duration is calculated by meeting the chilling requirements. If the simula-
tion is initialised at the 𝐢𝐝𝐨𝐫j stage, the model then assumes that this is the onset of
dormancy (idebdorms) and that the chilling requirements are not met.
In 1965 Bidabe proposed a formula to calculate dormancy and post-dormancy dura-
tions for apple trees, based on the Q10 notion which corresponds to exponential-type
responses to temperature. This is a well-known formula used for fruit trees for both
vegetative or reproductive buds. In STICS, we only use that which concerns the
dormancy period, since the post-dormancy period (i.e. from ifindorms). The daily
responses are accumulated (𝖼𝗎, Eq. (3.5)) until the current day (I) from a starting date
(𝐢𝐝𝐞𝐛𝐝𝐨𝐫𝐦P), generally taken to be during the autumn or the summer (García de
Cortázar Atauri, 2006), which shows that the initial date has a little effect on the calcu-
lation (August 1st for grapevine). The genetic-dependent parameter for the amount
of chilling requirement is 𝐣𝐯𝐜P.
56
Development
The Bidabe’s formula (Bidabe, 1965) provides good results for grapevine (García de
Cortázar Atauri et al., 2009a).
57
STICS soil-crop model
In STICS, the emergence phase is broken down into three sub-phases: seed imbibi-
tion (moistening), followed by germination, and lastly, shoot elongation. The physical
soil conditions influence not only the duration of emergence but also the number of
emerged plants, especially in dry conditions or when there is a surface crust.
3.4.1.1 Moistening
Seed moistening can be regarded as a passive process starting at a species-dependent
water potential prevailing in the seedbed (𝐩𝐨𝐭𝐠𝐞𝐫𝐦𝐢P in MPa). The relationship
from Clapp and Hornberger (1978), parameterised by the characteristic soil water
contents of field capacity and wilting point, was used to convert 𝐩𝐨𝐭𝐠𝐞𝐫𝐦𝐢P into water
content (see § 10.1.4). Once the seed is moistened, it has a limited number of days
(𝗇𝖻𝗃𝗀𝗋𝖺𝗎𝗍𝗈) during which it uses endosperm reserves for plantlet growth (Eq. (3.6)).
This number has a species-dependent component (𝐧𝐛𝐣𝐠𝐞𝐫𝐥𝐢𝐦P) but also a thermal
one, since it is thought that at low temperature (i.e. the average soil temperature
in the seedbed, 𝖲𝖡, from the beginning of moistening, 𝖨𝖬𝖡), respiration processes
and the consumption of reserves are slower (the minimum at high temperature is
𝐩𝐫𝐨𝐩𝐣𝐠𝐞𝐫𝐦𝐢𝐧P × 𝐧𝐛𝐣𝐠𝐞𝐫𝐥𝐢𝐦P). When the temperature is lower than the germina-
tion base temperature, 𝐭𝐠𝐦𝐢𝐧P, then the day number is maximal (𝐧𝐛𝐣𝐠𝐞𝐫𝐥𝐢𝐦P). Above
𝐭𝐝𝐦𝐚𝐱P, the seed uses up its reserves in the least time, parameterised by default to
20% of the maximum (𝐩𝐫𝐨𝐩𝐣𝐠𝐞𝐫𝐦𝐢𝐧P=0.2).
Figure 3.7. Evolution of the number of days of autotrophy as a function of temperature for two
sets of cardinal temperatures.
58
Development
3.4.1.2 Germination
Germination is achieved when the growing degree-days from planting in the seedbed
(𝗌𝗈𝗆𝗀𝖾𝗋) reaches a given threshold (𝐬𝐭𝐩𝐥𝐭𝐠𝐞𝐫P), with a condition as to the dryness of
the soil (Eq. (3.7)).
t
𝗍𝗌𝗈𝗅 is the soil temperature and 𝐭𝐠𝐦𝐢𝐧P is the base temperature for germination. Soil
moisture in the seedbed (𝖲𝖡 = depth of sowing ± 1 cm) influences germination through
the 𝗁𝗎𝗆𝗂𝗋𝖺𝖼 variable (Eq. (3.8)).
where 𝗁𝗎𝗆𝗌𝗈𝗅, 𝗁𝗇 and 𝗁𝗑 are the actual water content, the wilting point and the field
capacity in the seedbed, respectively, and 𝐬𝐞𝐧𝐬𝐫𝐬𝐞𝐜P is a plant parameter which can
be given a value between 0 and 1. If 𝐬𝐞𝐧𝐬𝐫𝐬𝐞𝐜P , the effect of soil dryness on all the
functions of root growth is only effective for water contents below the wilting point
(Figure 3.8); conversely, if 𝐬𝐞𝐧𝐬𝐫𝐬𝐞𝐜P , this effect is the highest.
Figure 3.8. Evolution of the variable as a function of the parameter and the
values of seedbed water contents at field capacity ( ) and at wilting point ( ).
59
STICS soil-crop model
If the seedbed dries out, it may delay germination significantly. This does not impair
grain viability as long as the grain has not already imbibed water. If, however, the
soil water content has been high enough to allow grain moistening, grain viability
is reduced. To account for this effect, we relied on work by Bradford (2002, 1990)
showing that too long a time for germination after moistening reduces the germina-
tion rate if the number of days of moistening (𝗇𝖻𝗃𝗁𝗎𝗆𝖾𝖼) is higher than a plant- and
temperature-dependent threshold duration (𝗇𝖻𝗃𝗀𝗋𝖺𝗎𝗍𝗈, Eq. (3.6)). It is assumed that
germination occurs (igers being the germination day) but at a reduced plant density
(ratio between density of germinated plants, densite(igers), to sowing density,
𝐝𝐞𝐧𝐬𝐢𝐭𝐞𝐬𝐞𝐦T) proportional to the thermal time deficit (Eq. (3.9)). An illustration of
the chronology of germination in various soil conditions is given in Figure 3.9.
Figure 3.9. Chronology of germination represented for two different soil conditions: a) soil
wetting and b) soil drying. The first arrow indicates the moistening date (soil above )
and the second arrow the germination date. In the first case the required thermal time for
germination ( = 50 degree days) is not reached by six days ( ) of moistening,
which causes a decrease in density (78%).
60
Development
The parameterisation of Eq. (3.10) can be significantly different in actual soil conditions
when compared to laboratory conditions (finely sieved soil) because the presence of
clods or compacted earth slows down the shoot’s vertical upward growth. Emergence
occurs when elongation (𝖾𝗅𝗈𝗇𝗀) is greater than sowing depth (𝐩𝐫𝐨𝐟𝐬𝐞𝐦T) as shown in
Figure 3.10.
Water status (𝗁𝗎𝗆𝗂𝗋𝖺𝖼) is calculated as described in Eq. (3.8) by using the average soil
moistures between the seedbed and the root front zrac (layer denoted HB). The variable
𝖼𝗋𝗎𝗌𝗍 stands for soil crusting conditions and will be explained in the following section.
In Eq. (3.10), 𝐞𝐥𝐦𝐚𝐱P, 𝐛𝐞𝐥𝐨𝐧𝐠P and 𝐜𝐞𝐥𝐨𝐧𝐠P are species-dependent parameters.
As for germination, if the duration between germination (igers) and emergence (ilevs)
is too long (𝐧𝐥𝐞𝐯𝐥𝐢𝐦1P and 𝐧𝐥𝐞𝐯𝐥𝐢𝐦2P parameters in Figure 3.11), there may be emer-
gence deficiencies represented by the variable 𝖼𝗈𝖾𝖿𝗅𝖾𝗏 (i.e. the ratio of the emerged to
the germinated density (Eq. (3.11)).
61
STICS soil-crop model
The effect of frost on young plantlets can be simulated and causes an additional reduc-
tion in population density. The plantlet stage (𝑖𝑙𝑒𝑡𝑠) is assumed to end at a defined
number of leaves (𝐧𝐛𝐟𝐠𝐞𝐥𝐥𝐞𝐯P), calculated from the plastochrone (𝐩𝐡𝐲𝐥𝐥𝐨𝐭𝐡𝐞𝐫𝐦𝐞P).
The frost damage function for emergence (𝖿𝗀𝖾𝗅𝗅𝖾𝗏) is calculated in the same way as
other frost functions (§ 4.4.4.1) with thresholds of specific sensitivity for the plantlet
stage (𝐭𝐠𝐞𝐥𝐥𝐞𝐯10P and 𝐭𝐠𝐞𝐥𝐥𝐞𝐯90P) and reduces the plant density in a multiplicative
way (Eq. (3.12)).
62
Development
post-sowing rainfall may destroy soil fragments, and then drought renders this layer
almost impenetrable for the plantlets, since the resistance to emergence depends on
the weather through its evaporative demand and on the force exerted by the plantlet.
The formalisation of these processes in STICS relies partly on the work of Dürr et al.
(2001). Surface crusting is assumed to occur only after sowing once a certain amount
of rainfall (soil-dependent parameter 𝐩𝐥𝐮𝐢𝐞𝐛𝐚𝐭S) has occurred. The crust is assumed to
be dry when the natural mulch depth (𝗑𝗆𝗎𝗅𝖼𝗁: variable calculated from the soil evapo-
ration formulations: see § 11.2.3)) is greater than the threshold parameter 𝐦𝐮𝐥𝐜𝐡𝐛𝐚𝐭S,
in which case 𝗑𝗆𝗎𝗅𝖼𝗁 is considered as the thickness of the crusted layer.
The subsequent delay in emergence can, just like the water deficit in the seedbed,
reduce plant density. Yet not all the plantlets will be affected because of the hetero
geneity of the crust and the differences in individual plantlet vigour. In STICS it is
assumed that the ease of crust penetration is accounted for by a plant-dependent
parameter (𝐯𝐢𝐠𝐮𝐞𝐮𝐫𝐛𝐚𝐭P). The delay in emergence is formalised by stopping the accu-
mulation of thermal time in Eq. (3.10) when the shoot reaches the base of the crust
(𝖼𝗋𝗎𝗌𝗍=0.0 calculated in Eq. (3.13).
t
The density reduction law is specific to the crusting phenomenon (𝖼𝗈𝖾𝖿𝗅𝖾𝗏𝖻) but
analogous to the other constraint law (𝖼𝗈𝖾𝖿𝗅𝖾𝗏 depicted in Figure 3.12), with a
minimum threshold corresponding to the 𝐯𝐢𝐠𝐮𝐞𝐮𝐫𝐛𝐚𝐭P parameter: if 𝐯𝐢𝐠𝐮𝐞𝐮𝐫𝐛𝐚𝐭P
is greater than 0, which means that when the soil is crusted a proportion of plants
succeed in emerging, the COEFEVB function is less effective than the water content
and temperature-dependent 𝖼𝗈𝖾𝖿𝗅𝖾𝗏 function. The combination of both relation-
ships is made dynamically by calculating the daily derivatives of both laws: if 𝖼𝗋𝗎𝗌𝗍=0,
which means a crust obstacle occurs the current day, the density reduction is calcu-
lated according to the law; otherwise it is the COEFLEVB law that prevails
(Figure 3.12).
Thus, as soon as significant rainfall occurs, the shoot continues to growth normally.
Table 3.3 shows the sensitivity of the formalisations described above to the effect of
soil crusting by varying the three required parameters.
63
STICS soil-crop model
Figure 3.12. Combination of the two laws ( depending on non-optimal water content
and temperature conditions and depending on the crust layer) affecting the emerged
density as a function of the occurrence of the soil crust factor =0.0, which means a crust
obstacle occurs, and the plantlet vigour ( ). The parameters and
are defined in Eq. (3.11).
Table 3.3. Sensitivity analysis of the soil crusting parameters on the emergence variables
(example of a maize crop in western France).
Sensitivity to
No SC High SC Low SC High SC
crusting (SC)
Plantlet vigour (PV) – High PV Low PV Low PV
pluiebatS (mm) 50 3 9 3
mulchbatS (cm) 0.5 0.5 1.5 0.5
vigueurbatP – 0.8 0.15 0.15
Sowing - emergence
12 27 24 27
duration (days)
Emerged density
relative to sown 77 64 31 19
density (%)
Note: pluiebat and mulchbat are soil parameters, vigueurbat is a plant parameter.
64
Development
Two cardinal temperatures limit the function of the linear response: Tmindeb and
Tmaxdeb.
t
idebdorm
Finally, if the model is initialised at the 𝐢𝐥𝐞𝐯T stage, the model assumes that the chilling
requirements are met (note that this does not apply to annual crops). When the model
is run for several years, the phasic and trophic status of the plant is saved from one
year to the next (see § 14.2).
65
Chapter 4
Shoot growth
Loïc Strullu, Alain Mollier, Jean-Louis Durand
and Nadine Brisson
Reviewed by: Nicolas Beaudoin and Bruno Mary
As all crop models do, STICS characterises the plant subsystem by its shoot biomass
and leaf area index. Once calculated, the shoot biomass is partitioned into the various
organs and feedback occurs between this partitioning and shoot growth for indeter-
minate plants. In STICS, indeterminate denotes species for which there is significant
trophic competition between vegetative and harvested organs. This definition differs
from the botanical one, and species like rapeseed or pea are considered as determi-
nate in STICS. This is because the assumption of independence between vegetative
and reproductive growth is acceptable, though these two developmental scales can
overlap. Meanwhile, species like sugarbeet are regarded as indeterminate because the
growing tuber greatly influences shoot growth. The harvested organs (grains, fruits
or tuber) are the only ones characterized in terms of number (see § 8). The present
chapter touches on various interrelated processes that are covered in other chapters.
See figure 4.1 to see how the paragraphs of this chapter relate to other chapters.
67
STICS soil-crop model
cited several studies that showed that simulating leaf area index directly on a canopy
scale produces results that are just as good as a leaf-to-leaf model. Baret (1986), Milroy
and Goyne (1995), and Chapman et al. (1993) worked on a canopy scale and suggested
dividing the evolution of leaf area index into two curves. The first curve represents
growth (always a logistic curve) and the other senescence (logistic or exponential).
Figure 4.1. Main functional links between the paragraphs of this chapter and other chapters.
Several authors have proposed to make a direct link between the evolution of leaf area
index and crop development (Dale et al., 1980; Dwyer and Stewart, 1986; Hammer and
Muchow, 1994; Nelder, 1961; Teittinen et al., 1994). Jamieson et al. (1995) designed
their model with four stages of evolution for leaf area index.
In STICS, leaf area growth is driven by phasic development, temperature and stresses.
An empirical plant density-dependent function represents interplant competition.
For indeterminate plants, trophic competition is taken into account through a trophic
stress index, while for determinate plants a maximal expansion rate threshold is calcu-
lated to avoid unrealistic leaf expansion. In the first version of STICS (Brisson et al.,
1998b), net leaf growth was directly simulated, without splitting the evolution of the lai
into gross growth and senescence, leading to a crude representation of leaf area index.
However, when thinking in terms of efficiency of radiation interception, it appears that
there is a plateau and the impact on radiation interceptions of high values is negli-
gible (Allen and Richardson, 1968; Cowan, 1968; Otegui et al., 1995; Varlet-Grancher
and Bonhomme, 1979). However, simulation of senescence is necessary to have a good
representations of C and N fluxes linked to leaf fall. Both options of simulations are
always available in the model.
68
Shoot growth
Figure 4.2. Leaf growth rate as a function of phasic development with the parameterisation
corresponding to a wheat crop as given in Singels and de Jager (1991) ( ,
and ) for two values of the parameter .
69
STICS soil-crop model
The thermal function relies on crop temperature and cardinal temperatures (𝐭𝐜𝐦𝐢𝐧P
and 𝐭𝐜𝐦𝐚𝐱P), which differ from the temperatures used for the phasic development.
The extreme threshold 𝐭𝐜𝐱𝐬𝐭𝐨𝐩P is the same as for development.
The density function (𝖽𝖾𝗅𝗍𝖺𝗂_𝖽𝖾𝗇𝗌), is active when the leaf area index threshold
(𝐥𝐚𝐢𝐜𝐨𝐦𝐩P) is reached and if the plant density (in plant m–2 calculated as explained
in § 3.4 and possibly decreased by early frost, see § 4.4.4) is greater than the 𝐛𝐝𝐞𝐧𝐬P
threshold, below which the plant leaf area is assumed independent of density (Eq. (4.4)).
Beyond this threshold, leaf area per plant decreases exponentially. The 𝐚𝐝𝐞𝐧𝐬V param-
eter represents the ability of a plant to withstand increasing densities. It depends on
plant species and may depend on the variety (Figure 4.3).
For branching or tillering plants, 𝐚𝐝𝐞𝐧𝐬V represents the plant branching or tillering
ability (e.g. for wheat or pea). For single-stem plants, 𝐚𝐝𝐞𝐧𝐬V represents competition
between plant leaves within a given stand (e.g. for maize or sunflower).
Figure 4.3. Density function versus plant density for various species (wheat, maize,
pea and sunflower).
70
Shoot growth
In the case of intercropping, the model calculates an equivalent plant density for the
understorey crop (densiteequiv), which accounts for the presence of the dominant
crop. If 𝖽𝖾𝗇𝗌𝗂𝗍𝖾_𝖽 and 𝖽𝖾𝗇𝗌𝗂𝗍𝖾_𝗎 are the planting densities of the dominant and the
understorey crops respectively and 𝐛𝐝𝐞𝐧𝐬𝐝P and 𝐛𝐝𝐞𝐧𝐬𝐮P are the threshold densities
for inter-plant competition, the equivalent density is calculated as in Eq. (4.5):
Figure 4.4. Illustration of the calculation of the equivalent plant density for the understorey
crop.
Water and nitrogen can be limiting factors for growth, with stress indexes whose
values vary between 0 and 1 (see § 4.4). Water (turfac) and nitrogen indexes (innlai) are
assumed to interact, thereby justifying the use of the most severe of the two stresses.
Meanwhile at the whole plant level, in the rare occasions where water deficit and water
logging (exolai) occur in different horizons of the root zone, the water-logging stress
index is assumed to act independently (Eq. (4.6)).
71
STICS soil-crop model
Figure 4.5 illustrates the evolution of the growth rate of a wheat crop receiving
reduced radiation (20% of incoming radiations), which can happen under a tree canopy
compared to a crop in the open field.
Figure 4.5. growth rate dynamics ( ) of a durum wheat crop in southern France with
100 % and 20 % of the incoming radiation (RG) without any stress and the evolution of values
during the growing phase.
72
Shoot growth
As a consequence, the leaf area index can decrease markedly during the growth phase
if the crop experiences severe stresses during the harvested organ filling phase, as
shown in Figure 4.6 for sugarbeet.
4.1.2 Senescence
In STICS, shoot senescence only concerns leaf dry matter and leaf area index. For
crops that are harvested more than once (e.g. temporary or artificial grasslands), shoot
senescence also affects the aerial biomass remaining after harvest. While senescence
was implicit in the first versions of the model (Brisson et al., 1998b), it is now explicit,
with a clear distinction between natural senescence due to ageing of leaves, and senes-
cence accelerated by stresses (water, nitrogen, frost). The concept of leaf lifespan, used
for example by Maas (1993), is applied to the green leaf area and biomass produced.
The leaf area and a fraction of the leaf biomass produced on a given day (see § 7) is
therefore lost through senescence once the lifespan has elapsed (Duru et al., 1995).
This fraction corresponds to the 𝐫𝐚𝐭𝐢𝐨𝐬𝐞𝐧P parameter (0-1), and its complement to 1
represents the fraction remobilised by the plant during senescence.
73
STICS soil-crop model
Until the iamfs stage, the maximum lifespan, calculated for the day when the leaves
are emitted ( ) is 𝖽𝗎𝗋𝗏𝗂𝖾𝖨; from iamfs to ilaxs, the maximal lifespan increases between
𝖽𝗎𝗋𝗏𝗂𝖾𝖨 and 𝐝𝐮𝐫𝐯𝐢𝐞𝐅P as a function of the leaf development variable ulai:
Water or nitrogen stress may shorten the current lifespan if the stress on day t is more
intense than the previous stresses encountered since time (Eq. (4.10)). Two specific stress
indices for senescence are introduced: senfac and innsenes (see § 4.4). Frost (fstressgel,
which can be either 𝖿𝗀𝖾𝗅𝗃𝗎𝗏 or 𝖿𝗀𝖾𝗅𝗏𝖾𝗀: see § 4.4.4) may also reduce or even cancel lifespan.
In case of high availability of nitrogen (inn >1), the foliage lifespan is increased from the
iamfs stage up to a maximum given by the 𝐝𝐮𝐫𝐯𝐢𝐞𝐬𝐮𝐩𝐦𝐚𝐱P parameter:
Leaf lifespan is not expressed in degree days (like phasic development), because doing
so has the disadvantage of stopping any progression as soon as the temperature drops
below the base temperature (𝐭𝐝𝐦𝐢𝐧P). To remedy this problem, the senescence course
Figure 4.7. Comparison between phasic development courses expressed in degree days and in
Q10 units for two values.
74
Shoot growth
between and t (𝗌𝗈𝗆𝗌𝖾𝗇) is expressed by cumulative Q10 units (with Q10=2), i.e. an
exponential-type function:
The senescence course is affected by the same cardinal temperatures as phasic deve
lopment and can be slowed down by stresses (see § 3.3.1. The leaf lifespan parameter
(𝐝𝐮𝐫𝐯𝐢𝐞𝐅P) expressed in Q10 units represents about 20% of the same lifespan expressed
in degree days (Figure 4.7).
where pfeuilverte is the proportion of leaf mass to total biomass produced on the
current day, and the parameter 𝐫𝐚𝐭𝐢𝐨𝐬𝐞𝐧P represents the fraction of leaf biomass
which becomes senescent since part of the dead leaf biomass is remobilised and does
not completely disappear.
The cumulative senescent foliage area is laisen. In forage crops (e.g. grasslands or
alfalfa) with residual dry matter from the previous regrowth cycle (𝐦𝐬𝐫𝐞𝐬𝐢𝐝𝐮𝐞𝐥T), the
senescence of residual dry matter (𝖽𝖾𝗅𝗍𝖺𝗆𝗌𝗋𝖾𝗌𝖾𝗇) starts from cutting. The senescent
rate is calculated as follows:
Leaves falling onto the soil during crop growth are another source of organic residue.
The falling rate is calculated with the parameter 𝐚𝐛𝐬𝐜𝐢𝐬𝐬𝐢𝐨𝐧P which is the propor-
tion of senescent leaves falling down. This phenomenon can be significant for some
crops, such as rapeseed in winter after frost events. The decomposition of the fallen
leaves at soil surface is simulated by the decomposition module (category 2, young
plant residues). The C/N ratio of leaves when they fall off (𝖢𝗌𝗎𝗋𝖭𝗋𝖾𝗌𝗂𝖽) is calculated
based on the nitrogen nutrition index of the whole crop using the plant parameter
𝐩𝐚𝐫𝐚𝐳𝐨𝐟𝐦𝐨𝐫𝐭𝐞P, as proposed by Dorsainvil (2002):
75
STICS soil-crop model
The assumption is that the photosynthetic function of storage organs lasts from the
beginning of grain/fruit filling (idrps) to the beginning of dehydration (idebdess)
stages.
76
Shoot growth
The parameter 𝐥𝐚𝐢𝐩𝐥𝐚𝐧𝐭𝐮𝐥𝐞P is the plant ground cover at planting if the crop is trans-
planted rather than sown. Figure 4.8 shows the simulated evolution of ground cover
for a lettuce crop with two planting densities.
Figure 4.8. Ground cover dynamics for a lettuce crop comparing two plant densities.
Parameters: trecouvmaxP = 0.072, infrecouP = 0.85, pentrecouvP = 4.5, adensV = –0.4, bdensP = 5,
laicompP = 0.14.
Water and nitrogen shortage and waterlogging stresses are applied to the rate of ground
cover growth, calculated as the derivation of Eq. (4.16). The method of combining
stresses is the same as for the leaf area index: 𝖽𝖾𝗅𝗍𝖺𝗂stress described in Eq. (4.6).
77
STICS soil-crop model
The variable 𝗍𝗎𝗋𝗌𝗅𝖺 is the mean water stress turfac experienced since emergence, and
𝐬𝐥𝐚𝐦𝐚𝐱P and 𝐬𝐥𝐚𝐦𝐢𝐧P are two parameters which define the limits of variation in
specific leaf area sla between a satisfactory water level and a state of extreme stress.
where ebmax is the maximum radiation use efficiency, raint is the intercepted PAR
(photosynthetically active radiation) and the parameter 𝐜𝐨𝐞𝐟𝐛G corresponds to the
radiation saturating effect. This effect is the result, even buffered, of the saturation
occurring within a short time step on the individual leaf scale and is easily observed
when daily calculations are made with instantaneous formulae of canopy photosyn-
thesis (Boote and Jones, 1987); such calculations lead to a value of . Note that
some variables are relative to previous day because of the consecutive nature of the
calculations and modules. Summing up the variable dltams throughout time gives
the biomass of non-perennial organs for annual crops or both perennial and non-
perennial organs for perennial crops, depending on simulation options (see § 7).
78
Shoot growth
79
STICS soil-crop model
where the parameter 𝐚𝐥𝐩𝐡𝐚𝐂𝐎2P represents the sensitivity of the crop growth to the
CO2 concentration. It is calculated so that the curve passes through the point (600,
𝐚𝐥𝐩𝐡𝐚𝐂𝐎2P). It mainly varies with the plant metabolism (C3/C4), being around 1.1
for C4 crops and 1.2 for C3 crops (Peart et al., 1989; Ruget et al., 1996; Stockle et al.,
1992). The effect of CO2 on stomatal resistance will be covered in the paragraph on
water requirements (see § 9.3).
Figure 4.10. Relative radiation use efficiency versus atmospheric CO2 concentration for two
crop species. The parameter is 1.20 for wheat and 1.06 for maize.
80
Shoot growth
81
STICS soil-crop model
Each stress index has its specific threshold (tetstomate, teturg and 𝗍𝖾𝗍𝗌𝖾𝗇) and a
comon parameter (𝐬𝐰𝐟𝐚𝐜𝐦𝐢𝐧P) (Figure 4.11).
The calculation of the tetstomate and teturg thresholds is explained in the chapter on
transpiration (see § 11). 𝗍𝖾𝗍𝗌𝖾𝗇 is proportional to teturg thanks to the 𝐫𝐚𝐩𝐬𝐞𝐧𝐭𝐮𝐫𝐠P
parameter (Eq. (4.23)).
82
Shoot growth
The hierarchy between the three stress indices is generally that which is indicated in
Figure 4.11, with 𝐫𝐚𝐩𝐬𝐞𝐧𝐭𝐮𝐫𝐠P > 1. The functions of these three stress indices are summa-
rised in Table 4.1. The germination and epicotyl growth phases can also be affected by
water shortage in response to soil moisture in the seed bed (𝐡𝐮𝐦𝐢𝐫𝐚𝐜G index).
Table 4.1. Impact of water stress on physiological functions through the various water
stress indices.
83
STICS soil-crop model
When the option of daily partitioning is chosen, perennial and non-perennial organs
are distinguished as their N demand (see § 7 and § 6.1), and the is calculated only
for non-perennial organs.
Figure 4.12. Critical nitrogen dilution curve (NC) and calculation as the ratio between
and .
All nitrogen stress indices accept 𝐈𝐍𝐍𝐦𝐢𝐧P or 𝐈𝐍𝐍𝐢𝐦𝐢𝐧P as the floor value for the
and the inni options, respectively. By definition, the s index corresponds to
the inn between 𝐈𝐍𝐍𝐦𝐢𝐧P and 1. The innlai and innsenes indices (Figure 4.14) are
defined by point [1, 1] and by points 𝐈𝐍𝐍𝐦𝐢𝐧P, 𝐢𝐧𝐧𝐭𝐮𝐫𝐠𝐦𝐢𝐧P and 𝐈𝐍𝐍𝐦𝐢𝐧P, 𝐢𝐧𝐧𝐬𝐞𝐧P,
respectively.
Setting the parameters in this way means the effect of nitrogen deficiency on photo-
synthesis can be differentiated from that for leaf expansion. In practice, it seems that
these two functions react very similarly and 𝐢𝐧𝐧𝐭𝐮𝐫𝐠𝐦𝐢𝐧P is similar to 𝐈𝐍𝐍𝐦𝐢𝐧P,
while 𝐢𝐧𝐧𝐬𝐞𝐧P is greater, indicating that the plants accelerate their senescence later
than their growth decrease, just as for water stress. A commonly accepted value for
𝐈𝐍𝐍𝐦𝐢𝐧P is 0.3 and 𝐈𝐍𝐍𝐈𝐦𝐢𝐧P is 0.0. The functions of these three stress indices are
summarised in Table 4.2.
84
Development
Figure 4.13. Comparison between and inni for a grapevine crop with nitrogen reserve at
the beginning of the cycle.
85
STICS soil-crop model
Table 4.2. Impact of nitrogen stress on physiological functions through the various
nitrogen stress indices.
The stress variable splai is the ratio of the trophic sources to the sinks, sourcepuits (see
§ 7 for an explanation on how to calculate it). The splai and spfruit options are defined
by the 𝐬𝐩𝐥𝐚𝐢𝐦𝐢𝐧P, 𝐬𝐩𝐥𝐚𝐢𝐦𝐚𝐱P, 𝐬𝐩𝐟𝐫𝐦𝐢𝐧P and 𝐬𝐩𝐟𝐫𝐦𝐚𝐱P parameters (Figure 4.15).
The various trophic stress indices cannot be considered as equivalent to biomass
allocation coefficients because they are not all applied to biomass. Accordingly, the
relative position of the functions sourcepuits and splai does not indicate any priority
between fruit and leaves: the priority needs to be calculated in terms of biomass and
depends largely on the relative sink strengths of the organs.
86
Development
4.4.4.1 Frost
The stress variable is the minimum crop temperature, tcultmin (see § 9). The frost
stress indices correspond to frost damage (1 for no frost and 0 for lethal frost). The
response to frost as well as the damage varies as a function of the developmental stage
(Table 4.4).
Table 4.4. Impact of frost on physiological functions through the frost stress indices.
Each response is defined by four parameters (Figure 4.16). Two of them are inde-
pendent of the developmental stage: 𝐭𝐝𝐞𝐛𝐠𝐞𝐥P (temperature at the beginning of frost
action) and 𝐭𝐥𝐞𝐭𝐚𝐥𝐞P (lethal temperature); the two others are relative to frost damage:
temperature inducing 10% or 90% frost damage. For the plantlet phase, the parameters
are 𝐭𝐠𝐞𝐥𝐥𝐞𝐯10P and 𝐭𝐠𝐞𝐥𝐥𝐞𝐯90P which act on plant density through the index 𝖿𝗀𝖾𝗅𝗅𝖾𝗏;
for the juvenile phase (up to iamfs stage), the parameters 𝐭𝐠𝐞𝐥𝐣𝐮𝐯10P and 𝐭𝐠𝐞𝐥𝐣𝐮𝐯90P
act on foliage (acceleration of senescence) through the index 𝖿𝗀𝖾𝗅𝗃𝗎𝗏. After the iamfs
stage, the parameters 𝐭𝐠𝐞𝐥𝐯𝐞𝐠10P and 𝐭𝐠𝐞𝐥𝐯𝐞𝐠90P are also active on foliage through
the index 𝖿𝗀𝖾𝗅𝗏𝖾𝗀. For frost affecting flowers and fruits, the parameters 𝐭𝐠𝐞𝐥𝐟𝐥𝐨10P and
𝐭𝐠𝐞𝐥𝐟𝐥𝐨90P define the dynamics of the fgelflo index.
87
STICS soil-crop model
Figure 4.16. Frost stress indices (fgellev, fgeljuv, fgelveg, ) as a function of minimal crop
temperature (Tcultmin). Parameters: tdebgelP = –1; tgellev10P = –6; tgellev90P = –8; tgeljuv10P = –10,
tgeljuv90P = –12, tgelveg10P = –8, tgelveg90P = –10, tgelflo10P = –2, tgelflo90P = –5, tlethalP = –13.
Table 4.5. Temperature stress factor or driver for each physiological function.
Function and
Physiological function Temperature Role
thermal stress index
Daily average soil
Emergence pilot
temperature
Daily average crop
Aboveground development pilot
temperature
Daily average crop
Vernalisation and dormancy stress
temperature
Daily average crop
Leaf growth and senescence pilot
temperature
Daily average crop
Root growth and senescence pilot
or soil temperature
88
Development
Function and
Physiological function Temperature Role
thermal stress index
Daily average crop
Radiation use efficiency (decrease) stress ftemp
temperature
Minimum crop
Filling at low temperatures (stop) stress ftempremp
temperature
Maximum crop
Filling at high temperatures (stop) stress ftempremp
temperature
The temperature stress factor acting on the radiation use efficiency (RUE), ftemp, is
calculated as:
The smooth shape of the RUE versus crop temperature (Figure 4.17) is quite classical
(Ritchie and Otter, 1985) and comes from the combined responses of photosynthesis
and respiration to temperature. Yet the cardinal temperature values are highly
dependent on the time step used: in our case this is daily average crop temperatures.
As far as fruit filling is concerned, the response in the model is yes/no.
Figure 4.17. Thermal stress indices ( and ftempremp) as a function of temperature using
cardinal temperatures (TeminP = 2; TeoptP = 10; TeoptbisP = 20; TemaxP = 30; TminrempP = 5;
TmaxrempP = 27).
89
STICS soil-crop model
4.4.5 Waterlogging
Under waterlogged conditions, the model calculates the variable exofac (Eq. (4.26))
which represents the proportion of root length which is under anoxic conditions, in
saturated layers:
The variable (𝖺𝗇𝗈𝗑) is equal to 1 if the layer z is anoxic and 0 in the opposite case.
Three anoxic stress indices are calculated (Eq. (4.27)): izrac, exolai and exobiom. They
are relative to root growth, lai growth and RUE, respectively. They are based on the
experimental work by Rebière (1996), reviewed by N. Brisson et al. (2002).
The root stress index izrac limits root growth at an efficient depth and density (see § 5).
These relationships, applied to a wheat crop which is assumed to have a high sensitivity
to water logging, are illustrated in figure 4.18.
90
Development
If the species (or variety) has developped adaptative mechanisms such as aerenchyma,
the effects of excess water will be less pronounced and this is simulated by reducing the
parameter 𝐬𝐞𝐧𝐬𝐚𝐧𝐨𝐱P. If 𝐬𝐞𝐧𝐬𝐚𝐧𝐨𝐱P = 1, the sensitivity is maximal, if 𝐬𝐞𝐧𝐬𝐚𝐧𝐨𝐱P = 0,
the plant is indifferent to excess water (for example, rice).
91
STICS soil-crop model
Table 4.7. How stresses are combined in the model for each physiological function.
Combination of stresses
Physiological function
(*only for indeterminate crops)
Emergence duration Water deficiency x Crusting
Plant density establishment (Water deficiency x Crusting) x Frost
Development min(Water deficiency, Nitrogen deficiency)
min(Water deficiency, Nitrogen deficiency)
Leaf growth
x Water logging x Trophic*
Senescence min(Water deficiency, Nitrogen deficiency, Frost)
Root growth Water deficiency x Nitrogen deficiency x Frost
Water deficiency x Nitrogen deficiency x Temperature
Radiation use efficiency
x Water logging
Number of fruits Nitrogen deficiency x Frost x Trophic*
Fruit growth Temperature x Trophic*
Transpiration MIN(Water deficiency, Water logging)
92
Chapter 5
Root growth
Nadine Brisson, Nicolas Beaudoin, Alain Mollier,
Florent Chlebowski, Marie Launay and Loïc Strullu
Reviewed by: Gaetan Louarn and Bruno Mary
5.1 Introduction
Apart from anchoring plants in soil, the root system is responsible for specific func-
tions, such as the uptake of water and mineral elements (N, P, K…), N2 symbiotic
fixation (in legumes), rhizodeposition and possible generic functions like carbohy-
drate or N storage, which crop models may or not take into account and in different
ways. The development of N-fixing nodules and their activity are less dependent on
the root system (which plays a supporting role) than on the shoot dynamics and the
physicochemical conditions of the surrounding soil (Burger, 2001).
The plant’s ability to trap mineral elements relies on root system efficiency. Efficiency is
not just related to the actual root length profile or root biomass – it is highly dependent
on the mobility of the element of interest within the soil. For water and nitrate ions,
the minimum root length density for unrestricted uptake is 0.5 cm cm–3 according to
Bonachela (1996), equating to an average soil-root distance of 0.8 cm, which falls within
the range of 0.5 cm to 1.0 cm proposed by Aura (1996). According to other authors
(Kage and Ehlers, 1996; Robertson et al., 1993) the minimum root length density can be
lower. This means that the efficient root profile is different from the actual root system,
especially in the subsurface layer where roots can more than adequately handle nitrate
and water uptake, although they are needed for the uptake of less mobile ions. This effi-
ciency must be dynamically estimated in order to correctly evaluate the supply/demand
ratio. The effect of the soil (constraints to penetration, sensitivity to anoxia, etc.) on
the shape of the root system (Nicoullaud et al., 1994) must also be accounted for.
While all these elements are taken into account in architectural root growth modelling
approaches (Drouet and Pagès, 2003), this is seldom the case in crop models, where
roots are not individualised but simply layered in the soil. In crop models, because the
soil is considered in only one dimension, growth in depth must be treated separately
from growth in density. The progression rate of the root front is generally based on
degree-days (Giauffret and Derieux, 1991; Hunt and Pararajasingham, 1995) and the
root density assumption mostly relies on an exponential decrease of roots with depth
(de Willigen et al., 2002; Gerwitz and Page, 1974; Heinen et al., 2003). Although we can
93
STICS soil-crop model
94
Root growth
𝐜𝐨𝐝𝐞𝐝𝐢𝐬𝐫𝐚𝐜P and 𝐜𝐨𝐝𝐞𝐦𝐨𝐧𝐨𝐜𝐨𝐭P), which can interact with the technical option
𝖼𝗈𝖽𝖾𝗂𝗇𝗌𝗍𝖺𝗅, to describe either seedling or established crop behaviour. The daily incre-
ment of root front, deltaz, is then determined only using parameters driving deltaz
response to physical soil conditions, and the specifical 𝐜𝐫𝐨𝐢𝐫𝐚𝐜P parameter. Finally,
the deltaz stop is specifically determined by the combination of a soil parameter
(𝐨𝐛𝐬𝐭𝐚𝐫𝐚𝐜S), a specific crop parameter (𝐬𝐭𝐨𝐩𝐫𝐚𝐜P), which can be either 𝐢𝐥𝐚𝐱T, 𝐢𝐟𝐥𝐨T or
𝐢𝐦𝐚𝐭T, and two options (𝐜𝐨𝐝𝐞_𝐚𝐜𝐭𝐢_𝐫𝐞𝐬𝐞𝐫𝐯𝐞P, 𝐜𝐨𝐝𝐞𝐩𝐞𝐫𝐞𝐧𝐧𝐞P).
95
96
STICS soil-crop model
Figure 5.1. Flowchart of options controlling vertical root growth, leaving aside the options for thermal time calculation.
Root growth
The 𝗁𝗎𝗆𝗂𝗋𝖺𝖼 variable, which reflects the influence of water content on germination
and root growth calculated in Eq. (3.8) during emergence, becomes a bilinear variable
after emergence Eq. (5.5). This variable depends on two parameters, i.e. root sensi-
tivity to drought (𝐬𝐞𝐧𝐬𝐫𝐬𝐞𝐜P), and the wilting point (𝗁𝗇):
Under waterlogged conditions, the proportion of root length which is under anoxic
conditions, in saturated layers is 𝖾𝗑𝗈𝖿𝖺𝖼 (Eq. (4.26)). The 𝗂𝗓𝗋𝖺𝖼 root stress index limits
root growth at an efficient depth and density (Eq. (4.27) and Figure 4.18).
The efda variable constitutes a constraint to penetration in compacted soils, or in rare
cases a slowing of root penetration linked to a lack of soil cohesiveness. The formalisation
proposed by Jones et al. (1991) and validated by Rebière (1996), was adapted for STICS
(Figure 5.2). Root penetration is not constrained between the bulk density thresholds
𝐝𝐚𝐜𝐨𝐡𝐞𝐬G and 𝐝𝐚𝐬𝐞𝐮𝐢𝐥𝐛𝐚𝐬G. Above a bulk density threshold 𝐝𝐚𝐬𝐞𝐮𝐢𝐥𝐡𝐚𝐮𝐭G the effect
of bulk density (𝖽𝖺) on root penetration is equal to the parameter 𝐜𝐨𝐧𝐭𝐫𝐝𝐚𝐦𝐚𝐱P. The
parameter 𝐜𝐨𝐧𝐭𝐫𝐝𝐚𝐦𝐚𝐱P is assumed constant and corresponds to the plant’s sensi-
tivity to the penetration constraint. The 𝐝𝐚𝐬𝐞𝐮𝐢𝐥𝐛𝐚𝐬G and 𝐝𝐚𝐬𝐞𝐮𝐢𝐥𝐡𝐚𝐮𝐭G values are
1.4 and 2.0 respectively. The 𝐝𝐚𝐜𝐨𝐡𝐞𝐬G value is poorly understood, and we have only
97
STICS soil-crop model
provided an order of magnitude. The bulk density 𝖽𝖺 is the effective value, and takes
into account fine earth and pebbles. Applying this formalism in presence of pebbles
needs caution because their role on root penetration can be dual.
Figure 5.3. Flow chart of options controlling root length production and root biomass recycling.
98
Root growth
Users can choose the true density option to simulate both root emission and root
senescence versus time and depth. This approach is more relevant for simulating
intercrops or low-density crops, for which root density is never optimal. This
approach can also take into consideration the effects of constraints imposed by
the soil on root distribution. The true density option contains itself six formalism
options pertaining to root length density expansion and spatial distribution, which
are each presented in separate diagrams.
Secondly, dead root biomass at harvest can always be estimated, as shown in Figure 5.3.
The profiles of living roots and dead root biomass can be obtained dynamically.
Figure 5.4. Reference root length density profile for rapeseed described by the efficient root
density as a function of the root system depth and according to the root front depth .
99
STICS soil-crop model
100
Root growth
Figure 5.5. Flow chart of options controlling root length expansion and vertical distribution
with True density option selected.
101
STICS soil-crop model
5.3.3.2 Root length expansion and its possible shoot growth dependency
Two options are available to calculate root length expansion. In the ‘self-governing
root length expansion’ option, to ensure the robustness of the model predictions of
water and N balances, we chose to simulate the root length growth directly, without
passing dealing with root biomass. The root biomass to shoot biomass ratio varies
depending on the stresses suffered by the plant (Brisson et al., 1998b).
Conversely, with the ‘trophic-link root length expansion’ option, root length expansion
is driven by shoot growth, so that a reliable C soil-crop balance can also be predicted
(although more parameters are required).
The logistic curve describing the root length expansion rate 𝗋𝗅𝗃_𝖽𝖾𝗏 (Eq. (5.8)) depends
on the maximum root growth parameter 𝐝𝐫𝐚𝐜𝐥𝐨𝐧𝐠P and the normalised root develop-
ment unit urac, ranging from 1 to 3 (such as ulai, the calculation of which is described
in § 4.1.1). The logistic curve is thermally driven, even when the plant has vernalisation
or photoperiod requirements. The plant parameters 𝐩𝐞𝐧𝐭𝐥𝐚𝐢𝐦𝐚𝐱P and 𝐯𝐥𝐚𝐢𝐦𝐚𝐱P are
already used to calculate leaf growth rate (Eq. (4.2)).
The thermal function 𝗋𝗅𝗃_𝖳 relies on crop temperature and cardinal temperatures
(𝐭𝐜𝐦𝐢𝐧P and 𝐭𝐜𝐦𝐚𝐱P) which are the same values as for leaf area growth calculation
(Eq. (4.3)). The inter-plant competition function 𝗋𝗅𝗃_𝖽𝖾𝗇𝗌 is the same as that calculated
for leaf area growth 𝖽𝖾𝗅𝗍𝖺𝗂_𝖽𝖾𝗇𝗌 (Eq. (4.4)).
With this formalism, unlike this designed for the leaf area index, water and nitrogen
deficiencies in the plant do not play any role in root length, which promotes root
growth relative to aboveground growth in the event of stress. In contrast, anoxia acts
via the waterlogging stress index 𝗋𝗅𝗃_𝗌𝗍𝗋𝖾𝗌𝗌, derived from the izrac indicator, which is
calculated in Eq. (4.27) and used in Eq. (5.9). In view of the difference which may exist
between true density and efficient root density (as much as tenfold), the raw application
102
Root growth
of izrac could have no effect on efficient root density, which would contradict exper-
imental results (Rebière, 1996). So when izrac is less than 1 (i.e. under waterlogging
stress conditions), it is multiplied by the ratio between efficient cumlracz to total rltot
root length before it is applied to the rlj variable (Eq. (5.9)).
At the root front, the root density is imposed and estimated using the parameter
𝐥𝐯𝐟𝐫𝐨𝐧𝐭P, and the growth in root length depends directly on the root front growth rate
deltaz (Eq. (5.10)).
Figure 5.6. Example of the root length growth as a function of the root development unit
compared to the underground/total biomass partitioning coefficient and the daily
production of shoot biomass ; is the daily aboveground/underground ratio of
biomass partitioning; is the daily growth of plant biomass.
103
STICS soil-crop model
Finally, the trophic effect can be combined with the effects of soil water and NO3 limi-
tation. The 𝐜𝐨𝐝𝐞_𝐈𝐍𝐍_𝐫𝐨𝐨𝐭G is designed to ignore the effect of stress on root growth.
It allows users to simulate a preferential allocation of biomass to the roots in case of
soil water or nitrogen stress.
The number of days of heterotrophy is therefore equal to twice the duration of the germi-
nation phase (Deleens et al., 1984). This calculation is used to allocate half of the seed’s
104
Root growth
reserves on the day of emergence, and the remainder until a period equivalent to the time
of germination. This quantity of daily biomass from the seeds is calculated as follows:
We therefore assign over nhet days from germination, a root length from seeds having
emerged (densite) and their specific weight 𝐩𝐠𝐫𝐚𝐢𝐧𝐦𝐚𝐱𝐢P converted into length
𝐥𝐨𝐧𝐠𝐬𝐩𝐞𝐫𝐚𝐜P. The new variable resulting from part of the seed reserves (𝐚𝐥𝐥𝐨𝐩𝐞𝐫𝐢𝐫𝐚𝐜P)
thus makes it possible to define the root length allowed by the heterotrophy phase on
a daily basis. The default value can be selected for this new parameter 𝐚𝐥𝐥𝐨𝐩𝐞𝐫𝐢𝐫𝐚𝐜P,
which is equal to 0.25 since 50% of the seed reserves are used for respiration and 50%
of the remaining share are allocated to coleoptile growth. The variable 𝗇𝖾𝗐𝗋𝖺𝖼 is there-
fore added to rlj during nhet days from germination. The 𝐜𝐨𝐝𝐞𝐝𝐢𝐬𝐫𝐚𝐜P option can
thus be activated to simulate the root activity in early stages.
Figure 5.8. Example of the daily root growth of winter wheat according to a) scenario
options; b) parameter value.
105
STICS soil-crop model
Figure 5.9. Simulated root length density profile and cumulated proportion of root
length density (Cum.RL inset plot); a) = 2 i.e. driven by initial root length and
b) = 1 i.e driven by vertical distribution parameters.
106
Root growth
– finally, it is possible to end up with a root profile where most of the roots are in the
deep horizons, which contradicts the distribution of roots observed in many species
(Fan et al., 2016).
5.3.3.3.1.2 Dynamical distribution of increment of length versus depth according to verti-
cal distribution parameters
When 𝐜𝐨𝐝𝐞𝐝𝐢𝐬𝐫𝐚𝐜P option is set to the option of , the background of the STICS
standard profile formalism is directly applied to the new root generation, as an expo-
nential function of the depth. The function 𝖽𝗂𝗌 drives the root distribution function,
involving the 𝐥𝐯𝐟𝐫𝐨𝐧𝐭P parameter and a new 𝐤𝐝𝐢𝐬𝐫𝐚𝐜P parameter, as follows:
We therefore have a decreasing function of the depth , (in cm), whose the shape is
determined by 𝐤𝐝𝐢𝐬𝐫𝐚𝐜P and whose asymptote is equal to the parameter 𝐥𝐯𝐟𝐫𝐨𝐧𝐭P. A
default value of 𝐤𝐝𝐢𝐬𝐫𝐚𝐜P can be taken equal to ; this is the optimised value
in order to obtain a root profile variability similar to those obtained from the meta-
analysis by Fan et al. (2016).
If the crop is a monocotyledon, root length distribution starts from the surface and
not from 𝐩𝐫𝐨𝐟𝐬𝐞𝐦T.
where 𝗁𝗎𝗆𝗂𝗋𝖺𝖼 (Eq. (3.8)) defines the effect of soil dryness and, accounts for the plant’s
sensitivity to this effect. The variable efda defines the effect of soil compaction through
bulk density (§ 5.2.2 and Figure 5.2). The anoxia index of each soil layer anoxmoy is
assigned the value of 1 if the horizon has reached saturation; it is associated with the
sensitivity of the plant to waterlogging 𝐬𝐞𝐧𝐬𝐚𝐧𝐨𝐱P.
The variable efNrac_mean defines the effect of mineral nitrogen, which contributes
to root distribution in the layers with high mineral nitrogen content. The variable
depends on the specific parameters 𝐦𝐢𝐧𝐚𝐳𝐨𝐫𝐚𝐜P, 𝐦𝐚𝐱𝐚𝐳𝐨𝐫𝐚𝐜P and 𝐦𝐢𝐧𝐞𝐟𝐧𝐫𝐚P,
which characterise the sensitivity of plant root growth to the mineral nitrogen
content in the soil (Eq. (5.17); Figure 5.10). This last constraint is optional and can be
deactivated in the model.
107
STICS soil-crop model
5.3.3.4 Senescence
5.3.3.4.1 Introduction
The concept of senescence is only present in the true density option. The root senes-
cence formalism plays a direct role with regard to the actual root profile, and possibly
an indirect role regarding root distribution, when the 𝐜𝐨𝐝𝐞𝐝𝐢𝐬𝐫𝐚𝐜P option is deacti-
vated (§ 5.3.3.3). It also directly drives root biomass turn over (see following § 5.4). The
main option which drives senescence is 𝐜𝐨𝐝𝐞_𝐝𝐢𝐟𝐟_𝐫𝐨𝐨𝐭P, which allows users to choose
between two types of roots (setting ) or a single mean root type (setting )(Figure 5.11).
If 𝐜𝐨𝐝𝐞_𝐝𝐢𝐟𝐟_𝐫𝐨𝐨𝐭P equals , the role of the root diameter on root lifespan follows the
background of the ArchiSimple model (Pagès et al., 2014). Modalities of 𝐜𝐨𝐝𝐞_𝐝𝐢𝐟𝐟_𝐫𝐨𝐨𝐭P
can interact with options concerning either forage crops or perennial grasses.
108
Root growth
Figure 5.11. Flow chart of options controlling root mortality with true density option selected.
The specific root length is calculated for fine (𝗅𝗈𝗇𝗀𝗌𝗉𝖾𝗋𝖺𝖼𝖿) and coarse roots
(𝗅𝗈𝗇𝗀𝗌𝗉𝖾𝗋𝖺𝖼𝗀):
109
STICS soil-crop model
The daily root length emission per elementary layer (𝖽𝗋𝗅𝗂𝗓) is divided between fine and
coarse roots in a proportion close to the initial value:
The living biomass can be calculated for each type of root. These calculations are
performed for each soil layer.
110
Root growth
the existence of fine roots; the final higher value with this option is influenced by
the additional root parameters which it needed. Conversely, choosing the LAX stage
for 𝐬𝐭𝐨𝐩𝐫𝐚𝐜P reduces the final value. The 𝐜𝐨𝐝𝐞𝐝𝐢𝐬𝐫𝐚𝐜P option has low influence on
mortality, as expected.
Figure 5.12. Example of the root length kinetics (cm root/cm2 soil surface) versus time
according to either scenario options or parameter value: a) length density of living
; b) length density of senescent roots .
111
STICS soil-crop model
When annual crops are harvested or when perennial crops are destroyed, the death of
the whole living root system results in additional root residues. The biomass of the dead
roots (𝗆𝗌𝗋𝖺𝖼𝗆𝗈𝗋𝗍) at soil depth z on day t is also calculated using either the mean root
length when a single kind of root is simulated ((5.24)), or the root length of fine and coarse
roots if the 𝐜𝐨𝐝𝐞_𝐝𝐢𝐟𝐟_𝐫𝐨𝐨𝐭P option is activated (𝗋𝗅𝖿 and 𝗋𝗅𝗀, respectively): Eq (5.25).
The daily inputs of C due to dying roots (𝖽𝗅𝗍𝖺𝖰𝖢𝗋𝖺𝖼𝗆𝗈𝗋𝗍) are the product of
𝖽𝗅𝗍𝖺𝗆𝗌𝗋𝖺𝖼𝗆𝗈𝗋𝗍 and the root C content (assumed to be equal to 380 g C kg–1). The
daily inputs of N due to dying roots (𝖽𝗅𝗍𝖺𝖰𝖭𝗋𝖺𝖼𝗆𝗈𝗋𝗍) are the ratio of 𝖽𝗅𝗍𝖺𝖰𝖢𝗋𝖺𝖼𝗆𝗈𝗋𝗍
and the mean C/N ratio of the roots (CsurNrac) depending on the amount of N in the
living root (QNrac). The cumulative C and N inputs over the whole root profile are
then calculated as indicated previously:
Part of these C and N inputs occur in the biologically active layer and thus undergo
decomposition. The remaining fraction enters the deep soil layer (below 𝐩𝐫𝐨𝐟𝐡𝐮𝐦S)
and is summed up in the variables Crprof and Nrprof. This organic matter accumu-
lates because this deep layer has no simulated biological activity.
The Figure 5.13 illustrates the daily prediction of root length and root biomass
is given for two types of roots in a winter wheat crop (from the aforementioned
example in northern France). The respective contribution of fine and coarse roots
strongly varies in terms of length and biomass; obviously, this contribution depends
on the values of their respective parameters.
112
Root growth
Figure 5.13. Illustration of root growth and root senescence versus time with different types
of roots ( activated; suffixes on root types: f = fine; g = coarse): a) in total root
length and cumulative length of senescent roots ; b) in living root biomass
and dead root biomass .
– When the standard profile option is selected (𝐜𝐨𝐝𝐞𝐫𝐚𝐜𝐢𝐧𝐞P = 1), roots are assumed
to accumulate in soil until plant death (usually at harvest). The biomass of dead roots
(msracmort) is assumed to be a linear function of the aboveground biomass at harvest
(masecnp):
where 𝐩𝐫𝐨𝐩𝐫𝐚𝐜G is a allocation coefficient between below- and above- ground organs
(0.20) and 𝐲0𝐦𝐬𝐫𝐚𝐜G is the minimum root production when aboveground growth is
very limited (0.7 t ha–1). The two coefficients are assumed to be similar for all plants.
– When the true density option is selected (𝐜𝐨𝐝𝐞𝐫𝐚𝐜𝐢𝐧𝐞P = 2), root growth and senes-
cence are simulated continuously, while the biomass of died root is recycled back into
soil only at harvest if the option 𝐜𝐨𝐝𝐞_𝐫𝐨𝐨𝐭𝐝𝐞𝐩𝐨𝐬𝐢𝐭𝐢𝐨𝐧P is not activated, as already
shown in Eq (5.23).
For both options, standard profile or true density, the amount of carbon in the dead
root material (QCracmort) is the product of the dead root biomass and its mean C
content which is assumed to be 380 g C kg–1.
As there is no absorption of N by roots when the option 𝐜𝐨𝐝𝐞_𝐫𝐨𝐨𝐭𝐝𝐞𝐩𝐨𝐬𝐢𝐭𝐢𝐨𝐧P is not
activated, we assumed that the N content in roots is equal to the N content in the vegeta-
tive organs of the crop. The amount of nitrogen in the dead root material (QNracmort)
is equal to the product of the dead root biomass and N content of vegetative organs.
These inputs are assumed to be homogeneously distributed in the biologically active
layer [0, 𝐩𝐫𝐨𝐟𝐡𝐮𝐦S]. Please note that in this case, the N balance does not loop.
113
STICS soil-crop model
For water and nitrogen absorption, an efficient root length density 𝗅𝗋𝖺𝖼𝗓 is calculated
by applying the threshold 𝐥𝐯𝐨𝐩𝐭G to the total root length density, 𝗋𝗅. By default, the
value used for the optimum root density threshold 𝐥𝐯𝐨𝐩𝐭G is 0.5 cm cm–3 soil (Brisson
et al., 1998b). In this way, it is possible to represent a root system for various species
exhibiting fasciculate or pivotal type root systems (see Figure 5.14).
Figure 5.14. Root length density profile, with the standard profile option, for a root front depth
of 150 cm, described by the efficient root density as a function of depth , for rapeseed,
maize and durum wheat.
114
Root growth
Figure 5.15. Root length density profile as a function of the root depth , at tillering and at
grain filling stages, simulated with the standard profile option ( option deactivated)
and the true density option.
115
STICS soil-crop model
ii) to a lesser extent, the thermal time differs because of the crop temperatures if the
growth of the root front is calculated with the crop temperature option. This inter
action mainly depends on the N supply (Corre-Hellou et al., 2007).
Figure 5.16. Comparison of root profiles in sole crop and intercrops at stage
(23 June).
116
Root growth
Figure 5.17. Dynamics of total root length for fine and coarse roots ( and , respec-
tively) compared to those of in a deep loamy soil in organic system for Medicago sativa L.
with the STICS model v10.
Figure 5.18. Dynamics of fine and coarse root mass in different soil layers for living root
biomass ( ) and dead root biomass ( ) for fine and coarse roots (suffixes and
respectively) for Medicago sativa L. with the STICS model v10.
117
Chapter 6
Nitrogen acquisition by plants
Bruno Mary, Fabien Ferchaud, Loïc Strullu
and Nadine Brisson
Reviewed by: Nicolas Beaudoin
119
STICS soil-crop model
Finally, during the reproductive phase, the N demand of the storage organs has to be
taken into account.
Nmax depends on two factors: the density of the canopy and the presence of storage
organs; the first factor defines the parameters of the curves Nmax = f(W) (according
to Lemaire and Gastal (1997) or Justes et al. (1997)) and the second one defines W.
In the case of isolated plants (𝐜𝐨𝐝𝐞𝐩𝐥𝐢𝐬𝐨𝐥𝐞𝐍P = 2), the critical dilution curve (Ni) is
described by two successive power functions:
The parameters 𝐚𝐝𝐢𝐥P and 𝐛𝐝𝐢𝐥P are common to both options. The additional parame-
ters used for the isolated plant option (adili and bdili) are obtained using the following
assumptions:
– the N contained in young plantlets is only used for N metabolism, not in structures.
The concentration of this metabolic-N (𝐍𝐦𝐞𝐭𝐚P) is a function of species metabolism:
6.47% for C3 crops (e.g. wheat) and 4.80% for C4 crops (e.g. maize) (Justes et al., 1997;
Lemaire and Gastal, 1997);
– N dilution starts when the plantlet biomass exceeds a biomass threshold called
𝐦𝐚𝐬𝐞𝐜𝐦𝐞𝐭𝐚P (0.04 t ha–1; Justes et al. (1997));
120
Nitrogen acquisition by plants
– the difference between the maximum and the critical N concentrations, when
biomass is equal to 𝐦𝐚𝐬𝐞𝐜𝐍𝐦𝐚𝐱P, is 𝐍𝐫𝐞𝐬𝐞𝐫𝐯𝐞P;
– in dense canopy, when biomass is greater than 𝐦𝐚𝐬𝐞𝐜𝐍𝐦𝐚𝐱P, the curvature of the
maximal curve is the same than that of the critical curve: 𝐛𝐝𝐢𝐥𝐦𝐚𝐱P = 𝐛𝐝𝐢𝐥P.
Using these assumptions, the missing parameters of Eq. (6.4) and (6.5) can be calcu-
lated as follows:
Figure 6.1. Maximal (Nmax) and critical (NC) dilution curves for wheat (dense canopy) and
vineyard (isolated plants).
121
STICS soil-crop model
(W) used to calculate the N demand from the maximal dilution curve depends on the
vegetative biomass (masecveg), the daily increase of grain biomass (dltags) and the
crop N status (𝖺𝖻𝗌𝗈𝖽𝗋𝗉):
The relationship between inns and 𝖺𝖻𝗌𝗈𝖽𝗋𝗉 is shown in Figure 6.2 for winter wheat
(𝐢𝐧𝐧𝐠𝐫𝐚𝐢𝐧1P = 0.26 and 𝐢𝐧𝐧𝐠𝐫𝐚𝐢𝐧2P = 1.60).
122
Nitrogen acquisition by plants
where 𝖢𝖢𝗋𝖺𝖼 and 𝖢𝖢𝗉𝖾𝗋 represent the carbon concentration in root and perennial
organs (380 and 440 g C kg–1, respectively).
The convection flow in each elementary soil layer (𝖼𝗈𝗇𝗏, in kg N ha–1 day–1) is the
product of the water flow (i.e. the transpiration flow 𝖾𝗉𝗓, in mm day–1, see § 11.5.3)
and the mean NO3 concentration (𝖼𝗈𝗇𝖼𝖭, in kg N ha–1 mm–1 water). The exchange-
able NH4 is not included, since it is assumed to be immobile. There is no nitrate
transport by convection if the transpiration is nil (due to absence of roots or severe
water stress):
The diffusion flow in each elementary soil layer (𝖽𝗂𝖿𝖿, in kg N ha–1 day–1) is the product of
the effective diffusion coefficient of mineral N (𝖽𝗂𝖿𝖭𝖾, in cm2 day–1) and the horizontal
gradient of mineral N concentration in the soil (in kg N ha–1 mm–1 water cm–1 soil).
123
STICS soil-crop model
This gradient is calculated from the effective root density profile (𝗅𝗋𝖺𝖼𝗓), assuming that
roots are vertical and equidistant and that mineral N concentration decreases linearly
from the middle of two adjacent roots up to root surface (mineral N concentration is
nil at the root surface). These assumptions lead to:
The effective diffusion coefficient is a function of soil water content and bulk density
(Cockborne et al., 1988). Only the moisture effect (which is the main effect) is consi
dered in STICS.
The uniform root distribution hypothesis maximises the diffusive flow. In fact,
roots are heterogeneously distributed so that the diffusive flow is lower. In order
to account for this effect, the diffusion coefficient at field capacity (𝐝𝐢𝐟𝐍G) used in
STICS (0.018 cm2 day–1) is lower than the measured values reported in the liter-
ature (0.10-0.25 cm2 day–1) (Barber and Silberbush, 2015; Cockborne et al., 1988;
Kersebaum and Richter, 1991).
Figure 6.3. N uptake capacity of each system (HATS and LATS) versus nitrate content in
soil (logarithmic scale). Parameter values: 𝐕𝐦𝐚𝐱1P = 0.0018; 𝐊𝐦𝐚𝐛𝐬1P = 50;𝐕𝐦𝐚𝐱2P = 0.050;
𝐊𝐦𝐚𝐛𝐬2P = 25000; 𝗅𝗋𝖺𝖼𝗓 = 0.20; zrac = 60; 𝗁𝗎𝗋 = 0.20, 𝐃𝐀𝐅S = 1.50.
124
Nitrogen acquisition by plants
In Eq. (6.26), 𝖼𝗈𝗇𝖼𝖭 is the molar concentration of mineral nitrogen (µmol l–1) and
the parameters 𝐕𝐦𝐚𝐱1P and 𝐕𝐦𝐚𝐱2P are expressed in µmol cm–1 h–1. Mineral N is
considered as a whole (NH4+NO3), so that any selectivity between ammonium and
nitrate absorption is not accounted for.
The potential uptake rate in each soil layer is 𝖿𝗅𝗎𝗑𝗋𝖺𝖼 (kg N ha–1 day–1). It is propor-
tional to the effective root density (Eq. (6.27)) which is limited by the threshold 𝐥𝐯𝐨𝐩𝐭G,
above which uptake is no longer limited by root density:
The coefficient 33.6 is the ratio of µmol cm–2 h–1 to kg ha–1 day–1. Figure 6.3 shows
the dynamics of 𝖿𝗅𝗎𝗑𝗋𝖺𝖼 versus the nitrate content in soil and the contribution of both
transport systems to the uptake capacity.
The integration of 𝗈𝖿𝖿𝗋𝖭 over the whole profile yields 𝖼𝗎𝗆𝗈𝖿𝖿𝗋𝖭. In each layer, the N
supply can be compared to the crop demand through the ratio demsup:
If demsup = 1, the soil N supply is the factor limiting N uptake. In this case the N
uptake in each layer is equal to the N supply 𝗈𝖿𝖿𝗋𝖭. Conversely, the demand is the
factor limiting N uptake if demsup < 1; in this case, the actual N uptake in each layer
is smaller than and proportional to the N supply.
In both cases, the actual N uptake in each soil layer (𝖺𝖻𝗌𝗓) and the total uptake over the
root profile (abso) can be written as a function of the demsup variable:
125
STICS soil-crop model
The 𝗉𝗋𝗈𝗉𝖿𝗂𝗑𝗉𝗈𝗍 coefficient varies according to growing degree-days (Eq. (6.33)) and is
calculated the same way as for root growth (see § 5.2.2). The fixation process begins at
the idno stage (defined by the thermal duration 𝐬𝐭𝐥𝐞𝐯𝐝𝐧𝐨P) and stops at the ifno stage
(defined by the thermal duration 𝐬𝐭𝐝𝐧𝐨𝐟𝐧𝐨P). The potential curve then decreases
until the death of nodules, corresponding to the ifvino stage, during the 𝐬𝐭𝐟𝐧𝐨𝐟𝐯𝐢𝐧𝐨P
thermal duration. The establishment rate of nodules between the idno and ifno stages
depends on the potential rate 𝐯𝐢𝐭𝐧𝐨P and on growing degree-days (Figure 6.4).
126
Nitrogen acquisition by plants
It may be inhibited by high mineral nitrogen levels, under the control of 𝗇𝗈𝖽𝗇 which is
nil when the soil mineral nitrate concentration exceeds the threshold 𝐜𝐨𝐧𝐜𝐍𝐧𝐨𝐝𝐬𝐞𝐮𝐢𝐥P
(in kg N ha–1 mm–1 water), and otherwise is equal to 1.
Figure 6.4. Evolution of Propfixpot versus thermal time, for a low level of soil nitrate content
(nodn = 1). Parameter values: = 500, = 1200, = 300, = 0.0025.
The maximum fixation capacity of the crop (fixmaxvar) can be either constant (equal
to 𝐟𝐢𝐱𝐦𝐚𝐱P) if 𝐜𝐨𝐝𝐞𝐟𝐢𝐱𝐩𝐨𝐭P is equal to 1, or variable depending on vegetative and
reproducive growth if 𝐜𝐨𝐝𝐞𝐟𝐢𝐱𝐩𝐨𝐭P is equal to 2. In the latter case, it is calculated as:
where dltams is the daily biomass increase and dltags is the daily increase of grain
biomass. The parameter 𝐟𝐢𝐱𝐦𝐚𝐱𝐯𝐞𝐠P is the maximum amount of N fixed per ton of
vegetative biomass and 𝐟𝐢𝐱𝐦𝐚𝐱𝐠𝐫P is the maximum amount of N fixed per ton of grain
biomass.
127
STICS soil-crop model
The water stress factor (fxw) is equal to the proportion of elementary soil layers in the
nodulation area which have a water content 𝗁𝗎𝗋 equal or greater than the permanent
wilting point:
Limitation by anoxia (fxa) is calculated according to the same principle (it is equal to
the proportion of elementary soil layers which are in anaerobic conditions) using the
anox variable (see § 5.3.3.3).
128
Nitrogen acquisition by plants
∑
Finally, fixation can be partially inhibited either by soil mineral nitrogen amount
(if 𝐜𝐨𝐝𝐞𝐟𝐱𝐧G = 2) or by mineral N concentration (if 𝐜𝐨𝐝𝐞𝐟𝐱𝐧G = 3). There is no inhi-
bition if 𝐜𝐨𝐝𝐞𝐟𝐱𝐧G = 1. In the case where 𝐜𝐨𝐝𝐞𝐟𝐱𝐧G = 2, N fixation is partly inhibited
(fxn < 1) when the mean amount of mineral nitrogen in the rooting zone (𝖺𝗓𝗈𝗋𝖺𝖼/zrac,
in kg N ha–1 cm–1 soil) exceeds the threshold 𝐜𝐨𝐧𝐜𝐍𝐫𝐚𝐜100P, and is fully inhibited
(fxn = 0) when the amount exceeds the threshold 𝐜𝐨𝐧𝐜𝐍𝐫𝐚𝐜0P (Figure 6.6).
Figure 6.6. Effect of soil mineral N content on fixation ( = 2). Parameter values are
= 0.80 and = 0.125.
129
Chapter 7
Biomass and nitrogen partitioning
Loïc Strullu, Jean-Louis Durand, Bruno Mary,
Nicolas Beaudoin and Nadine Brisson
7.1 Introduction
Allocation of assimilates is critical to the operation of some models (such as
SUCROS, described by Van Ittersum et al., 2003). The STICS module for allocation
of assimilates between organs was added at a late stage, mainly to help dimension
the N and C reserve pools. For annual plants with determinate growth, the parti-
tioning calculations allow users to dimension envelopes of harvested organs which
may play a trophic role and ensure information input for the senescence module.
For perennial plants or those with indeterminate growth, these calculations allow
users to dimension a reserve compartment. We will refer to this concept as “cumu-
lative partitioning” (𝐜𝐨𝐝𝐞_𝐚𝐜𝐭𝐢_𝐫𝐞𝐬𝐞𝐫𝐯𝐞P = 2). It is suitable for annual simulations
but not for long-term simulations of perennial plants. With the increasing interest
among researchers in studying perennial crops as a means for biofuels produc-
tion, C and N sequestration in soils or the effects of climate change on cropping
systems, cumulative partitioning has shown its limits. These limits are due to:
1. The lack of location of reserve compartments in the plant and the absence of paral-
lelism between biomass and nitrogen pools simulated by the model. This means the
model cannot accurately simulate biomass or nitrogen dynamics in perennial or
non-perennial organs.
2. The model’s incapacity to simulate biomass and nitrogen remobilisation from
non-perennial to perennial organs, and thus to simulate long-term perennial crops
functioning.
3. The absence of simulation of C and N fluxes from the plant to the soil due to peren-
nial organ or root death leading to bias in long-term simulations of cropping systems
with perennial crops.
We substantially modified the module for allocation of C and N assimilates in the plant
to be able to simulate cropping systems with perennial crops over the long term (Strullu
et al., 2020, 2014; Strullu et al., 2015). We will refer to this concept as “daily partitioning”
(𝐜𝐨𝐝𝐞_𝐚𝐜𝐭𝐢_𝐫𝐞𝐬𝐞𝐫𝐯𝐞P = 1) and will describe it in further detail in this chapter. The current
STICS model includes both the cumulative and daily partitioning concepts because all
crops have not yet been parameterised for the new daily partitioning concept.
131
STICS soil-crop model
Table 7.1. Various organs identified in STICS for cumulative biomass partitioning.
132
Figure 7.1a. Schematic representation of biomass partitioning. Black squares correspond to rate variables; black ovals correspond to input variables;
grey squares correspond to parameters; dashed black squares correspond to cumulative variables. Grey shaded areas represent new fluxes and pools
simulated by the model. Solid arrows indicate flows of energy or biomass and dashed arrows indicate relationship between components (adapted from
133
Biomass and nitrogen partitioning
Figure 7.1b. Schematic representation of nitrogen partitioning. Black squares correspond to rate variables; black ovals correspond to input variables;
grey squares correspond to parameters; dashed black squares correspond to cumulative variables. Grey shaded areas represent new fluxes and pools
simulated by the model. Solid arrows indicate flows of energy or biomass and dashed arrows indicate relationship between components (adapted from
Strullu et al., 2020).
Biomass and nitrogen partitioning
7.3.1.1 Biomass
Some of these yellow leaves may fall to the ground depending on the 𝐚𝐛𝐬𝐜𝐢𝐬𝐬𝐢𝐨𝐧P
parameter (between 0 and 1). The daily falling quantity dltamstombe is recycled in
the carbon and nitrogen balance; its cumulative value is mafeuiltombe (Eq. (7.3)).
7.3.1.1.2 Reserves
Reserves (restemp) are calculated as the difference between the total biomass and the
accumulated biomass of leaves, stems and harvested organs (Eq. (7.5)).
135
STICS soil-crop model
For perennial plants, at the beginning of the cropping season, the reserves (carbon)
can be initialised at a non-zero value (𝐫𝐞𝐬𝐭𝐞𝐦𝐩0I), to represent the role played by
reserves at growth resumption. The reserve pool is not located in the crop and the
variable restemp represents the non-structural biomass that can be mobilised. The use
of reserves applies to perennial or indeterminate plants. For determinate annuals, the
use of reserves for grain filling is not simulated as such, but taken into account overall
when calculating the ircarb variable (index of progressive harvest). The harvested
organs (grains, fruits or tuber) are the only ones characterised in terms of number and
biomass (see § 8).
Yet it is assumed that a limit exists regarding the reserve compartment size, which is
called 𝐫𝐞𝐬𝐩𝐥𝐦𝐚𝐱P. If this limit is reached, a “sink / source” effect is simulated (Eq. (7.6)).
The results of the above calculations are illustrated in the case of wheat and grapevine
in Figure (7.2):
Figure 7.2. Proportion of the shoot biomass allocated to leaves, fruits or grains and to the
virtual component of reserves for two different crops: a) wheat, b) grapevine.
7.3.1.2 Nitrogen
Nitrogen uptake and accumulation in the plant is a function of daily nitrogen uptake
and nitrogen fixation by the crop (see § 6). Nitrogen is not found in one specific l ocation
136
Biomass and nitrogen partitioning
in the plant and can be defined as reserves or structural nitrogen. The only organs for
which nitrogen partitioning is simulated are harvested organs (grains, fruits, tuber).
For perennial plants, at the beginning of the cropping season, the N reserves can be
initialised at a non-zero value (𝐐𝐍𝐩𝐥𝐚𝐧𝐭𝐞0I).
7.3.2.1.1 Biomass
When crop growth starts, the aerial biomass is initialised from an initial leaf area
index, 𝐢𝐧𝐢𝐥𝐚𝐢P:
The daily biomass available for plant growth (dltams and the remobilised biomass
dltaremobil+remobilj) is shared between the different organs following the principles
of i) priority of biomass allocation between the different organs, and ii) confrontation
between supply and demand. It is important to note here that potential leaf and stem
growth can be decrease due to a negative effect of the photoperiod on cell elongation
and division (Eq. (7.8)). For this example, we used the reduction factor photoperiod
index (rfpi) calculated for crop development (see § 3). The negative effect of the photo-
period on potential leaf area index growth occured when the photoperiod decreases.
If the photoperiod is lower than the base photoperiod of the crop (𝐩𝐡𝐨𝐛𝐚𝐬𝐞P), then
growth of stems and leaves is null (rfpi = 0).
We put forward the hypothesis that biomass is first used where it is produced and so
the priority for biomass allocation is given to the structural part of green leaves (dltafv).
If the daily biomass production (dltams) is insufficient to support the potential growth
of lai(t), then the latter is limited (Eq. (7.9)) to respect the maximum specific leaf area
(𝐬𝐥𝐚𝐦𝐚𝐱P) of the green leaves.
137
STICS soil-crop model
If there is still biomass available after leaf growth, then the growth of the structural
part of stem (dltat) is calculated (Eq. (7.10)).
Finally, after calculating the biomass actually allocated to the structural parts of the
green leaves (dltafv) and stems (dltat), the biomass surplus is allocated to temporary
reserves (dltares) (Eq. (7.11)).
During leaf senescence (Eq. (7.12)), some of the biomass can be remobilised
(dltaremobsen) to feed the plant’s temporary reserves pool (restemp).
Temporary reserves are limited and are accumulated during a growing cycle in living
non-perennial vegetative organs, although they are not precisely partitioned between
stems and leaves (Eq. (7.13)). The plant can remobilise these reserves (§ 7.4).
138
Biomass and nitrogen partitioning
7.3.2.1.2 Nitrogen
Nitrogen is allocated to the aerial non-perennial organs (absoaer) as a function
of their nitrogen demand (see § 6), once the nitrogen demand of perennial organs
(see § 7.3.2.2.2) and roots (see § 7.6.2) is met (Eq. (7.14)).
Prioritising nitrogen allocation between organs allows users to simulate the effect
of root and perennial organs growth on nitrogen availability for the growth of aerial
non-perennial organs. N is allocated to roots and non-perennial organs first, because
N demand of perennial organs is nil at the beginning of a growing cycle (Figure 7.4).
139
STICS soil-crop model
Figure 7.4. Simulation of nitrogen allocation in non-perennial organs, perennial organs and
roots of Medicago sativa L. with the STICS model v10.
The Figure 7.5 shows an example of N partitioning in non-perennial organs (stems and
leaves) for Medicago sativa L.
140
Biomass and nitrogen partitioning
7.3.2.2.1 Biomass
Biomass is allocated between reserves (resperenne) or structural parts of the peren-
nial organs using an allocation coefficient (alloresp) (Eq. (7.16)). The underlying
assumption is that the reserves are first stored in the pre-existing organs and then,
when the perennial organs reach their reserve storage limit, the structural part of the
perennial organs increases and receives new reserves. When developing the model,
we considered that the perennial organs do not operate as sink organs for the biomass
produced (unlike vegetative aerial organs or reproductive organs), but only as passive
receptacles for the biomass surplus produced by the plant.
141
STICS soil-crop model
7.3.2.2.2 Nitrogen
Nitrogen absorbed or fixed by the plant (absotot) is allocated first to the perennial
organs (absoper; Eq. (7.17)) as a function of their nitrogen demand (see § 6).
The nitrogen allocated to the perennial organs is shared between the structure and
reserves (dltaresperN) according to the carbohydrate allocation and the parameter
𝐩𝐫𝐨𝐩𝐫𝐞𝐬𝐩𝐧P (Eq. (7.18)).
The sink strength of vegetative organs fpv is a function of the daily real leaf area index
growth rate deltai and the potential growth rate of vegetative organs depending on
the minimal specific leaf area parameter (𝐬𝐥𝐚𝐦𝐢𝐧P) and the ratio between stems and
leaves (ratioTF) (Eq. (7.20)).
The sink strength of reproductive organs fpft is calculated only for indeterminate
crops and is described in § 8. It is important to note here that fpft and fpv are not
exactly the same: fpft refers to potential growth while fpv corresponds to real growth.
By design, this difference puts a priority on fruits and can generate a day-to-day insta-
bility of the variable sourcepuits by the feedback of sourcepuits on fpv via the stress
index splai (see § 4). A second source/sink ratio value (sourcepuits) is calculated to
account for possible reserve remobilisation (Eq. (7.21)). This variable drives trophic
stresses, useful for simulating indeterminate crop competition between vegetative and
reproductive sinks.
142
Biomass and nitrogen partitioning
7.4.2.1 Biomass
The daily flux of remobilised biomass dltaremobil is obtained by the remobilisation
of reserves in perennial plants. Each day the maximal proportion of the reserves
that can be remobilised is remob, until perennial stocks are exhausted (Eq. (7.22)).
𝐫𝐞𝐬𝐭𝐞𝐦𝐩0I only represents carbon reserves, and nitrogen reserves can only be
added through initiation of the 𝐐𝐍𝐩𝐥𝐚𝐧𝐭𝐞0I parameter. These remobilisations
contribute to increasing the source/sink ratio the following day because they are
counted in the variable dltams.
Reserves built up during the vegetative cycle (restemp) and reused later contribute to
the the source/sink ratio estimation for indeterminate crops. The maximum quantity
which can be remobilised per day (remobilj) is calculated similarly to dltaremobil
(Eq. (7.22)). If the plant is both perennial and indeterminate, the reserves originating
from the previous cycle are first used (dltaremobil), and when they are exhausted, the
current cycle’s reserves (remobilj) can be used.
7.4.2.2 Nitrogen
The remobilisation rate of initial nitrogen perennial reserves 𝐐𝐍𝐩𝐥𝐚𝐧𝐭𝐞0I is assumed
to be equal to the crop nitrogen demand, see § 6. The remobilisation stops when initial
reserves are exhausted.
7.4.3.1 Biomass
Each day, the flux of remobilised raw carbohydrates (dltaremobilbrut) is obtained by
remobilising perennial reserves (Eq. (7.23)), with temporary reserves given priority for
remobilisation.
143
STICS soil-crop model
Some of the carbohydrate reserves are burned through respiration of the perennial
organs during the remobilisation phase (dltaCO2resperenne) (Avice et al., 1996).
Although respiration is taken into account in the model for the production of newly
synthesised biomass via the radiation use efficiency (RUE), the respiration associated
with the remobilisation of C and N reserves stored in perennial organs is not. This
prevents an asymmetry when considering of respiration between the remobilisation
of carbohydrate reserves in spring (for which respiration needs to be modelled) and
carbohydrate storage (for which respiration is integrated into the RUE). The parameter
𝐞𝐟𝐫𝐞𝐦𝐨𝐛𝐢𝐥P corresponds to the remobilisation efficiency of carbohydrate reserves
and is used to calculate the net flow of remobilised carbohydrate reserves as well as
carbon losses in the form of CO2 (Eq. (7.24)).
If the remobilisation of the perennial reserves is not sufficient to meet the demand of
the growing vegetative or reproductive organs, the temporary reserves built up during
the vegetative cycle (restemp) can in turn be remobilised (remobilj) (Eq. (7.25)).
144
Biomass and nitrogen partitioning
7.4.3.2 Nitrogen
The amount of remobilised N (dltaremobilN) is a function of the amount of remo-
bilised carbohydrates (dltaremobilbrut) and the N concentration of source organs
(CNresperenne) (Eq. (7.27)). Nitrogen remobilisation cannot exceed the nitrogen
demand of non perennial organs (see § 6).
During senescence of vegetative organs and when the emission of new leaves is nil
(𝐢𝐥𝐚𝐱T stage reached), temporary N reserves in the non-perennial organs are remobi-
lised (dltarestempN) to the perennial organs. The 𝐢𝐥𝐚𝐱T stage can be reached for two
reasons:
– The degree days required to reach the stage have been accumulated;
– The photoperiod is shorter than the photoperiod triggering senescence and the
photoperiod is decreasing (when the option ‘effect of photoperiod on senescence’ is
activated).
The amount of remobilised N is a function of the amount of remobilised carbohydrates
(dltarestemp) and N concentration in temporary reserves (CNrestemp) (Eq. (7.28)).
The partition of the N allocated to the perennial organs between reserves and struc-
ture is a function of remobilised carbohydrate allocation (Eq. (7.29)).
145
STICS soil-crop model
If the 𝐬𝐞𝐚P parameter is not zero, then this biomass is transformed into an equivalent
leaf surface area, photosynthetically active from the idrp stage to the idebdes stage.
7.5.2 Nitrogen
The amount of nitrogen in harvested organs QNgrain, is an increasing proportion of
the amount of nitrogen in the biomass (see § 8).
7.6 Roots
7.6.1 Biomass
The model features two methods of root biomass calculation, depending on the chosen
simulation options. If the user choose the profile type option, the root biomass is calcu-
lated at harvest. It is a proportion (𝐩𝐫𝐨𝐩𝐫𝐚𝐜P) of the above ground biomass produced
by the crop (masecnp). The parameter 𝐲0𝐦𝐬𝐫𝐚𝐜P corresponds to the minimal amount
of root biomass which returns to the soil if above ground biomass is nil (Eq. (7.31)).
If the user chooses the true density option, the root biomass is calculated daily as a
function of total length of fine and coarse roots (rltotf and rltotg respectively) and
their specific root length (longsperacf and longsperacg respectively) (see § 5 for
more detail concerning root system simulation) (Eq. (7.32)).
146
Biomass and nitrogen partitioning
7.6.2 Nitrogen
Consequently, the model has two methods to calculate root nitrogen content,
depending on the chosen simulation options. If the user chooses the profile type or
true density option, the root nitrogen content is calculated at harvest in function of
root biomass and nitrogen concentration in the vegetative organs (Eq. (7.33)).
Daily nitrogen allocation to roots is calculated by the model if and only if the true
density and root deposition options are activated. In that case, if there is still nitrogen
available after nitrogen allocation to perennial organs, remaining nitrogen is allocated
to the roots (absorac) as a function of their nitrogen demand (see § 6) (Eq. (7.34)).
The residual leaf area index is calculated with the parameter 𝐤𝐡𝐚𝐮𝐭P to convert crop
height into leaf area, the previously defined parameter 𝐡𝐚𝐮𝐭𝐛𝐚𝐬𝐞P and the parameter
𝐡𝐚𝐮𝐭𝐦𝐚𝐱P, which corresponds to the maximum height of the crop (Eq. (7.36)).
The nitrogen content remaining in the residual biomass after harvest depends on the
residual biomass and parameters of the nitrogen dilution curve (Eq. (7.37)).
147
STICS soil-crop model
Finally, the residual lai(t) is calculated by multiplying the residual green leaf biomass
by the maximum specific leaf area of leaves (𝐬𝐥𝐚𝐦𝐚𝐱P) (Eq. (7.39)).
The amount of nitrogen in the aerial part (QNplantenp) is determined using the
amount of residual biomass after cutting (msres) and the N content of the plant prior
to cutting (CNplante) (Eq. (7.40)).
These C and N inputs are assumed to be homogeneously distributed within the biolo
gically active soil layer (𝐩𝐫𝐨𝐟𝐡𝐮𝐦P).
148
Chapter 8
Yield formation
Nadine Brisson, Iñaki García de Cortázar Atauri,
Françoise Ruget and Benjamin Dumont
Reviewed by: Nicolas Beaudoin
Yield prediction is a goal of most crop models. Let’s note that this chapter refers to the
‘biological yield’, as opposed to the ‘farmer’s yield’, the latter being generally affected
by losses of the combine harvester. By definition, yield is the weight and quality of
the harvested organs. These organs can be reproductive organs – either grains (dehy-
drated) or fruits (hydrated) – or vegetative storage organs – either stems (sugarcane)
or roots (tubers). Forage crops, where the total aboveground biomass (leaves, stems
and sometimes grains) is harvested in its entirety, is also considered in the following
sections. The determinate or indeterminate character (as defined within the STICS
model1) does not indicate the type of harvested organs. However, by convention, we
will call the harvested organs of determinate species ‘grains’ and the harvested organs
of indeterminate species ‘fruit’.
Warren-Wilson (1972) suggested that the plant should be considered as a set of
compartments playing the role of sources and/or sinks for assimilates. This concept
can be applied to carbon, water, nitrogen or any metabolite of interest. However,
hereafter we will use it only for carbon, though it is also thoroughly documented for
nitrogen (Barbottin et al., 2005; Jeuffroy et al., 2000; Sinclair and de Wit, 1976). The
source and sink compartments usually represent organs (e.g. roots, leaves, grains, etc.),
which can change their function during a cycle: ‘source then sink’ for roots and trunks
in perennial plants, or ‘sink then source’ for leaves. Application of this concept to crop
models generates self-regulation of the system between the growth of different types of
organs. It is particularly well-suited to crops with an indeterminate growth habit and
to perennial crops, in which trophic competition exists between growing and storage
organs (Jeuffroy and Warembourg, 1991; Munier-Jolain et al., 1998). Source capacity
includes both newly-formed assimilates and remobilised resources translocated from
vegetative organs. Carbon sink strength, i.e. potential growth rate, is usually repre-
sented by a continuous or discrete function of the physiological age of the organ. The
problems with this approach lie in determining the size of the source capacity and
1. Within STICS, ‘indeterminate’ denotes species for which there is significant trophic competition between
vegetative organs and harvested organs.
149
STICS soil-crop model
j
Σ j
The number of grains per m2 (𝗇𝖻𝗀𝗋𝖺𝗂𝗇𝗌) is defined at the 𝐢𝐝𝐫𝐩T stage (Eq. (8.2)). It
depends on i) the growth variable (vitmoy in ) that integrates the effect of the
prevailing stresses during the period preceding the 𝐢𝐝𝐫𝐩T stage, ii) three species-de-
pendent parameters 𝐜𝐠𝐫𝐚𝐢𝐧𝐯0P (unitless), 𝐜𝐠𝐫𝐚𝐢𝐧P (in ) and 𝐧𝐛𝐠𝐫𝐦𝐢𝐧P
(in ) and iii) a genetic-dependent parameter 𝐧𝐛𝐠𝐫𝐦𝐚𝐱P (in ). The
last two parameters define the limits of variation of 𝗇𝖻𝗀𝗋𝖺𝗂𝗇𝗌.
150
Yield formation
Figure 8.1. Proportion of grain number, for the maximum allowed by the variety ( ), as
a function of growth during the pre-grain filling period. Examples are provided for maize (solid
black line: =0.111, =0.050, =1500 , =4500), pea (dash-dotted
black line: =0.000, =0.031, =447 , =3500), rapeseed (solid
grey line: =0.000, =0.050, =50000, =850000) and wheat
(dash-dotted grey line: =0.000, =0.036, =6000, =30000).
After the 𝐢𝐝𝐫𝐩T stage, the grain number can be reduced in the event of frost (Eq. (8.3)
and § 4.4.4) and the daily proportion of grains affected is (1-fgelflo), whatever their
state of growth. The corresponding weight ( in ) is deducted from the
151
STICS soil-crop model
grain weight (Eq. (8.6)), using the elementary current grain weight (𝗉𝗀𝗋𝖺𝗂𝗇 in )
defined in Eq. (8.7).
Figure 8.2. Dynamics of the grain to shoot biomass ratio ( ), as a function of time since
the stage . Examples are provided for maize (solid black line: = 0.0103 and
= 0.53), pea (dash-dotted black line: = 0.022 and = 0.65), rapeseed
(solid grey line: = 0.008 and = 0.50) and wheat (dash-dotted grey line:
= 0.0107 and = 0.55).
152
Yield formation
t – idrpT + 1
A transposition of the formalism has also been developed to account for thermal time.
Under this formalisation, Eq. (8.4) is adapted: ircarb is computed by taking 𝐯𝐢𝐭𝐢𝐫𝐜𝐚𝐫𝐛𝐓P
as the input parameter, and the thermal time accumulation between 𝐢𝐝𝐫𝐩T and 𝐢𝐦𝐚𝐭T
as the dynamic variable of time.
Yet these dynamics may not be the actual grain filling dynamics since threshold
translocation temperatures defining the thermal stress 𝖿𝗍𝖾𝗆𝗉𝗋𝖾𝗆𝗉 (𝐭𝐦𝐢𝐧𝐫𝐞𝐦𝐩P and
𝐭𝐦𝐚𝐱𝐫𝐞𝐦𝐩P, see § 4.4.4) may stop the carbon filling of harvested organs. Conse-
quently, grain filling (dltags in ) is calculated at a daily time step, in order to
account for the effect of a potential thermal stress (as defined by Eq. (8.5)).
The daily grain filling is then accumulated within the mafruit (in ) variable, as
defined in Eq. (8.6).The mass of each grain is finally calculated as the ratio of the mass
to the number of grains, although this cannot exceed the genetic 𝐩𝐠𝐫𝐚𝐢𝐧𝐦𝐚𝐱𝐢P limit,
as defined in Eq. (8.7).
153
STICS soil-crop model
If the number of inflorescences is more than 1 (e.g. in the case of grapevine, inflores-
cences=bunches), it can either be prescribed (𝐧𝐛𝐢𝐧𝐟𝐥𝐨P), or calculated (nbinflo_recal)
as a function of the trophic status of the plant at an early stage (we have chosen iamfs).
In the latter case, nbinflo_recal is calculated using the 𝐩𝐞𝐧𝐭𝐢𝐧𝐟𝐥𝐨𝐫𝐞𝐬P and 𝐢𝐧𝐟𝐥𝐨𝐦𝐚𝐱P
parameters (see Eq. (8.9)).
where 𝗋𝖾𝗌𝗉𝖾𝗋𝖾𝗇𝗇𝖾 (§ 7.3.1.1.2) is the amount of carbon reserves for perennial species
coming from the previous cycle. Pruning is not accounted for in this calculation.
The fruit sink strength function is the derivative of the potential growth of a fruit
(𝗉𝗈𝗍𝖼𝗋𝗈𝗂𝖿𝗋𝗎𝗂𝗍) plotted against the fruit development stage (𝖽𝖿𝗋). There are two succes-
sive phases in fruit growth; the first corresponds to a cell division phase while the
154
Yield formation
Figure 8.3. Illustration of the dynamics of fruit cohorts using the boxcartrain simulation
technique. In this example = 130, = 5, = 300.
second is devoted to expansion of the cells already set. In order to account for this
double dynamic, the fruit potential cumulative growth is defined as the sum of two
functions, as specified in Eq. (8.11) and illustrated in Figure 8.4:
– an exponential-type function describing the cell division phase (using the parame-
ters 𝐜𝐟𝐩𝐟P and 𝐝𝐟𝐩𝐟P)
– a logistic-type function describing the cell elongation phase (using the parameters
𝐚𝐟𝐩𝐟P and 𝐛𝐟𝐩𝐟P)
In Eq. (8.11), and are calculated so that the following conditions are respected
(Eq. (8.12)):
where 𝐩𝐠𝐫𝐚𝐢𝐧𝐦𝐚𝐱𝐢P is the genetic-dependent maximal weight of the fruit and 𝖽𝖿𝗋
stands for the fruit development stage of each age class, varying between 0 and 1; this
value is calculated for each age class ( ) in a discrete way (Eq. (8.13)).
155
STICS soil-crop model
Figure 8.4. Normalised potential fruit growth ( ) versus fruit development status
( ) with its two components: the exponential dynamic representing cell division and the
logistic-type dynamic representing cell expansion. In this example, we used grapevine values:
= 0.55, = 18, = 15 and = 0.20.
If the potential fruit growth is represented by a simple logistic curve, Figure 8.5 shows
that when varying the parameters 𝐚𝐟𝐩𝐟P and 𝐛𝐟𝐩𝐟P, various dynamics, including linear
dynamic, can be represented.
The daily fruit sink strength function (fpft) is then calculated for each age class,
according to Eq. (8.14), accounting for the duration of fruit growth from setting to
maturity, expressed in developmental units (𝐝𝐮𝐫𝐞𝐞𝐟𝐫𝐮𝐢𝐭P).
The sensitivity of the model for subdividing fruit growth into discrete units (𝐧𝐛𝐨𝐢𝐭𝐞P
parameter) also depends on the 𝗉𝗈𝗍𝖼𝗋𝗈𝗂𝖿𝗋𝗎𝗂𝗍 dynamics, as shown in Figure 8.6. Conse-
quently, three elements must be taken into account to give a value to the parameter
𝐧𝐛𝐨𝐢𝐭𝐞P: the fruit setting duration, the fruit growth dynamics and the location of the
idebdess stage allowing the fruit water dynamics to be initiated.
156
Yield formation
Figure 8.5. Normalised potential logistic fruit growth ( ) versus fruit development
status ( ) with various parameterisations corresponding to and values.
157
STICS soil-crop model
Figure 8.6. Influence of the discretisation of fruit growth through the number of boxes
( ) in relation to the form of the dynamics: “S” shape in panel a): ( = 0.7 and = 9)
or nearly linear in panel b): ( = 0.5 and = 2).
158
Yield formation
Unlike annual crops, where harvest means the death of the plant, grasslands/pastures
are perennial crops. This means that after harvest of the exploitable yield, the plant will
grow back. Accordingly, there are some specificities regarding the crop (re-)growth
(§ 4) and yield elaboration (discussed in this section).
‘Exploited grasslands’ (agronomically speaking), also called mown grasslands, can be
defined as perennial crops partially harvested by ‘cutting’ operations. Cutting can be
achieved through machinery – here, we mean the actual operation of mowing using
human-made tools – or via animals – i.e. grazing.
It will therefore be necessary to determine what is exported during a cut (i.e. the
harvested part) and what remains in place, which is referred to as the remaining dry
matter or the residual dry matter (see § 7.7).
In its current version, the STICS model is distributed with species of the func-
tional type B, namely tall fescue (Lolium arundinaceum), orchard grass (Dactylis
glomerata) or permanent pasture. The functional type B is defined by Duru, Tallowin
and Cruz (2005) as a group of species with a capture strategy, rather than conserva-
tion, and with a slow organ recycling rate (Cruz et al., 2002). In a broader context,
according to their ‘habitat preference and use-value’, type B species are defined as
‘species of fertile environments, that are fairly large in size, have a medium-to-early
phenological development, are suitable to achieve a fairly early and good quality
forage’ (Cruz et al., 2010).
The results presented in the following section are based on the standard parameter
set distributed with the STICS soil-crop model and described in Ruget et al. (2006),
which is itself based on considerations proposed by Diaz and Cabido (1997) to account
for the functional type B species. Other functional types of grassland species could, of
course, be re-parameterised to run the model. Finally, the STICS model has recently
been successfully used to simulate C and N dynamics in alfalfa (Medicago sativa in
Strullu et al., 2020).
The following sections will describe how the decision to cut is taken (§ 8.1.3.2) and how
the biomass is pooled into the different organs to constitute the harvestable/harvested
organs and the residual dry matter (§ 8.1.3.3)
159
STICS soil-crop model
160
Yield formation
The variable masec represents the total aboveground biomass that exists on a given
day. This biomass is the result of different sources (Eq. (8.18)):
– the residual biomass from the previous cut (𝐦𝐬𝐫𝐞𝐬𝐢𝐝𝐮𝐞𝐥T) that is still photosyn-
thetically active;
– the newly produced biomass since the previous cut that is still photosynthetically
active (masecneo).
– a dead fallen pool, which refers to a fraction of the biomass that might experience
senescence and undergo abscission (mafeuiltombe, see § 7.3.1.1.1).
where is the number that corresponds to the last cut and masecneo is the sum of
the daily increase in biomass (dltams) since the day following the last cut ( ) till
day .
In the case of forage, the non-senescent potentially harvestable biomass (mafruit,
Eq. (8.19)) will be constituted by the non-senescent aboveground biomass which
offers a use-value, i.e. which is photosynthetically active. The senescent tissues are
considered as having no-value – they serve no purpose and/or will likely be lost during
harvest. They must be removed from the aboveground biomass. In line with the two
fractions comprising masec, the senescent tissues are pooled as follows:
– Senescent or dead pool coming from the residual biomass (msresjaune)
– The (non-fallen) senescent fraction of the newly formed biomass (msneojaune). It
should be noted that msneojaune is assimilated to mafeuiljaune (§ 7.3.1.1.1).
Finally, the harvested fraction msrec_fou (Eq. (8.20)) will be constituted by the
non-senescent potentially harvestable biomass (mafruit) from which the fraction that
will remain on the field/pasture after the next cut (𝐦𝐬𝐫𝐞𝐬𝐢𝐝𝐮𝐞𝐥T) must be removed.
An additional variable integrating the whole forage produced is also outputted from the
model (masectot). It accounts for all the biomass that has been produced (non-dead
or senescent) and harvested along the cropping season.
Figure 8.7 shows the evolution over a cropping season of the variables related to
forage production, including four cutting events, illustrating the differences between
total aboveground biomass, harvestable and harvested biomass – namely masec,
mafruit and msrec_fou – and showing the year-to-date production (masectot).
Figure 8.7 illustrates how the harvested dry matter (msrec_fou) increases every day,
and might, due to its mathematical construction, become negative when what is
harvestable (mafruit) is lower than what should remain on the ground at the next cut
(𝐦𝐬𝐫𝐞𝐬𝐢𝐝𝐮𝐞𝐥T). This usually occurs in periods of low productivity when senescence is
active, i.e. mainly during dry summer.
161
STICS soil-crop model
Figure 8.7. Evolution of biomass pools along a cropping season (Lusignan, France, 1991). In
this example, three cuts are performed.
162
Yield formation
that there is a ‘programmed’ time course in the water content of fruits, and so this
is expressed using the 𝐝𝐞𝐬𝐡𝐲𝐝𝐛𝐚𝐬𝐞P parameter (in ), which will modify
the fruit water content (𝗍𝖾𝖺𝗎𝗀𝗋𝖺𝗂𝗇) day after day from its initial value 𝐡2𝐨𝐟𝐫𝐯𝐞𝐫𝐭P.
For dehydration, 𝐝𝐞𝐬𝐡𝐲𝐝𝐛𝐚𝐬𝐞P is positive; if the programme evolution tends towards
hydration, 𝐝𝐞𝐬𝐡𝐲𝐝𝐛𝐚𝐬𝐞P is negative. Dehydration may be accelerated (or provoked)
by water stress, which is characterised by the difference between the crop and air
temperatures. The proportionality coefficient is called 𝐭𝐞𝐦𝐩𝐝𝐞𝐬𝐡𝐲𝐝P (in ).
In summary, the water content (𝗍𝖾𝖺𝗎𝗀𝗋𝖺𝗂𝗇) is the result of Eq. (8.21) where the index
(for the box number) is of no use for determinate plants. An example is provided for
grapevine in Figure 8.8.
Figure 8.8. Evolution of the thermal difference ( , black points) and the grape
water content (solid grey line) for the 2003 season in Montpellier, France, influenced by the
phenological course (the dynamic started on 26 July).
163
STICS soil-crop model
Figure 8.9. Evolution of sugar content in relation to fruit development for sugarcane
( = 0.00030) and sugar beet ( = 0.00035).
164
Yield formation
165
Chapter 9
Canopy microclimate
Nadine Brisson, Marie Launay and Gaetan Louarn
Reviewed by: Nicolas Beaudoin and Didier Combes
9.1 Introduction
9.1.1 Context and importance of the microclimate
Biological processes in the crop-soil system directly depend on several physical vari-
ables such as light, temperature and humidity. The physical properties of biological
entities within this system and the way they function also affect the environmental
variables at their immediate boundary. The climate to which the plants and soil are
exposed (i.e. the microclimate) can thus differ substantially from the standard plant
climate as defined for weather forecast (local climate). These changes result from an
exchange of force, momentum, energy, and mass within the boundary layer (Jones,
2013). These exchanges imply two important kinds of coupling: radiative coupling,
where energy is transferred through electromagnetic vibration, and diffusive coupling,
where heat, water vapor, are exchanged.
This chapter examines the modelling principle related to the plant microclimate
in STICS. The relevant variables (i.e. crop temperature, light interception and air
humidity), drive many processes that take place within the plant canopy such as
phasic development, photosynthesis and evapotranspiration. Moreover, they
provide the boundary conditions for the calculation of soil microclimate (see § 10)
and thus influence processes occurring within the soil, such as organic matter
mineralisation (see § 12), and plant germination (see § 3). Microclimate also alters
water balance (see § 11).
Most crop models do not go into detail about microclimate calculation and use the
standard measured weather variables as the driving variables for the model (Brisson
et al., 2006). The calculation of the temperature and air humidity within the canopy
from a daily energy balance is among the original components of STICS, and it allows
users to account for the combined effects of weather and water balance.
The energy balance calculations with a daily time step, although questionable in terms
of physics , have already been carried out as part of an operational estimation of crop
water requirements (Smith et al., 1998). The daily crop temperature is assumed to be the
167
STICS soil-crop model
arithmetic mean of the maximum and minimum crop temperatures. Two calculation
methods are proposed (depending on the availability of wind and air humidity input
data): one uses an empirical relationship from Seguin and Itier (1983) and the other
solves the energy balance. Both methods rely on the calculations of the daily sum of
evaporative fluxes and net radiation.
168
Canopy microclimate
169
STICS soil-crop model
STICS has two options with regard to radiation interception: a simple Beer’s law, recom-
mended for homogenous crops, and a more complex calculation for radiation transfers
within the canopy, recommended for crops in rows. If the leaf status variable is the ground
cover and not the leaf area index (LAI), then only the Beer’s law option is permitted. The
longwave radiation component of the net radiation may be calculated according to the
Brunt’s formula or the Brutsaert’s formula; the latter formula clearly illustrates the soil
and atmospheric components of the longwave radiation. Two calculation methods are
proposed for crop temperature, depending on weather data availability: users may choose
either an empirical approach or the energy balance. Finally, evaporative fluxes can be
assessed using the simple crop coefficient approach, the potential evapotranspiration as
the driving variable to calculate the plant water requirement, or the resistance approach
using empirical resistance parameters to estimate the energy budget of canopies.
170
Canopy microclimate
Figure 9.1. Proportion of global radiation (raint/trg) intercepted for maize, sunflower and
wheat (extinP values of 0.7, 0.9 and 0.5, respectively).
The height of homogeneous crops is deduced from the leaf area index or the ground
cover (Eq. (9.2)). It is used especially in the calculation module for water requirements
via the resistive option. The 𝐤𝐡𝐚𝐮𝐭G parameter is assumed to be plant independent (a
general value of 0.7 is proposed), while the potential height of foliage growth (hauteur)
is mostly plant dependent and defined by the two limits 𝐡𝐚𝐮𝐭𝐛𝐚𝐬𝐞P and 𝐡𝐚𝐮𝐭𝐦𝐚𝐱P.
171
STICS soil-crop model
Figure 9.2. Simplified representation of the plant canopy and the principles used to calculate
daily radiation received by the inter-row ( ): is the base height of the canopy,
E its thickness, largeur its width, is any point located in the inter-row and H1 and H2 are the
two sun height angles corresponding to the daily positions 1 and 2 of the sun between which
is directly illuminated (adapted from Brisson et al., 2004).
Each part of the radiation received at includes a direct component and a diffuse
component. Let us assume that, for the transmitted part, the same extinction coefficient
(𝐤𝐭𝐫𝐨𝐮P) applies to both fractions (which is generally accepted to be the case when the
172
Canopy microclimate
general Beer’s law is used with a daily time scale). In Eq. (9.3), the parameter 𝐤𝐭𝐫𝐨𝐮P
corresponds to a gap fraction (Baret et al., 1993).
In contrast, for 𝗋𝖽𝗋𝗈𝗂𝗍 (Eq. (9.4)), direct and diffuse components should be separated
because of the directional nature of the direct component, which requires the calcu-
lation of separate proportions of radiation reaching the soil (𝗄𝗀𝖽𝗂𝖿𝖿𝗎𝗌 and 𝗄𝗀𝖽𝗂𝗋𝖾𝖼𝗍 are
the proportions of diffuse radiation, 𝗋𝖽𝗂𝖿𝖿𝗎𝗌 , and direct radiation, 𝗋𝖽𝗂𝗋𝖾𝖼𝗍 , respectively,
that reach the soil):
In order to calculate the angles, we must solve the following set of equations
(Eq. (9.6)):
where is the height of the sun, its azimuth, 𝐥𝐚𝐭𝐢𝐭𝐮𝐝𝐞C is the latitude of the loca-
tion, the declination angle which depends on the day (Varlet-Grancher et al.,
1993), and 𝐨𝐫𝐢𝐞𝐧𝐭𝐫𝐚𝐧𝐠T is the azimuth angle of the rows. , the apparent tangent of
in Figure 9.2, depends on canopy geometry (largeur, and 𝐡𝐚𝐮𝐭𝐛𝐚𝐬𝐞P defined in
Figure 9.2) and the position of the given point within the inter-row ( ).
For example, assuming and the angle ,
The borderline between sun (surf(as)) and shade (surf(ao)) is arbitrarily considered as
largeur/2. The above set of equations cannot be solved using analytical methods, and
must therefore be solved numerically (loop over with a basic variation of 3 degrees
followed by linear interpolation).
– The case of diffuse components
We take 46 directions (azimuth, height) and their corresponding percentage of diffuse
radiation (SOC standard). For each direction, point is checked to see if it is directly
illuminated, depending on canopy geometry. The variable 𝗄𝗀𝖽𝗂𝖿𝖿𝗎𝗌 corresponds to the
cumulative proportion of radiation received at point for the 46 directions.
The diffuse to total radiation ratio (rdif ; Eq. (9.7)) is calculated according to Spitters
et al. (1986) on the basis of the total to extraterrestrial radiation ratio (𝖱𝗌𝖱𝗌𝗈), with
the extraterrestrial radiation being calculated from the classical astronomical formula
(Varlet-Grancher et al., 1993) represented in Figure 9.3.
173
STICS soil-crop model
Figure 9.3. Relationship between the diffuse to total radiation ratio ( ) and the total to
extraterrestrial radiation ratio ( ).
The above equations are applied to points spread equally along the inter-row, and
the transmitted radiation values are then averaged for the shaded fraction (rombre)
and the sunlit fraction (rsoleil). The complementary part to the global radiation
corresponds to the radiation intercepted by the crop (raint ; Eq. (9.8)).
174
Canopy microclimate
between two limits (𝐝𝐟𝐨𝐥𝐛𝐚𝐬P and 𝐝𝐟𝐨𝐥𝐡𝐚𝐮𝐭P). It depends on the foliage produced ,
which accounts for lai, eai , laisen and leaves suppressed by specific techniques such
as topping (𝗅𝖺𝗂𝗋𝗈𝗀𝗇𝖾𝖼𝗎𝗆) or leaf removal (𝐥𝐚𝐢𝐞𝐟𝐟𝐞𝐮𝐢𝐥T) (Eq. (9.9); see § 13.2.3) and the
slope 𝐚𝐝𝐟𝐨𝐥P. If we assume a constant foliage density, then 𝐝𝐟𝐨𝐥𝐛𝐚𝐬P = 𝐝𝐟𝐨𝐥𝐡𝐚𝐮𝐭P.
– This formalisation of leaf density makes it possible to represent both foliage getting
denser while growing (e.g. grapevine) or conversely becoming less dense while growing
(e.g. cereals, Figure 9.4).
Figure 9.4. Leaf density (dfol) evolution for grapevine and barley according to the cumulative
foliage produced (FP), as two opposite examples.
– The plant canopy width largeur is calculated according to its rectangular or trian-
gular shape (Eq. (9.10)).
175
STICS soil-crop model
– Two types of triangles can be chosen: ‘right way up’ or ‘upside down’. The more
appropriate shape for radiative transfer is the ‘right way-up’ triangle (Brisson et al.,
2004) because this shape suggests that the low leaf density (in the classical sense of
leaf area per m3) measured in the upper parts allows more radiation to be transmitted
than in the lower parts where the leaf density is higher. With our simple model based
on a constant leaf density within the shape, this can be accounted for only by a triangle.
Thus the shape required as a parameter in the model is more closely linked to the leaf
density profile than to the external shape of the plant foliage.
A maximum limit, 𝐡𝐚𝐮𝐭𝐦𝐚𝐱P , is imposed on the plant height value (𝐡𝐚𝐮𝐭𝐛𝐚𝐬𝐞P ).
Thereby, in the first stage, the shape of the plant evolves isotropically. Once the 𝐡𝐚𝐮𝐭𝐦𝐚𝐱P
value is reached, the only way in which the shape can evolve is in terms of width.
Height and width can also be limited by topping.
9.2.2 Intercrop
In intercropping systems (see 2.2.3 for more information on the conceptual frame-
work), the objective is to estimate, on a daily time step, the part of the radiation
intercepted by the dominant crop and the part transmitted to both components
of the understorey crop: the shaded (rombre) and the sunlit (rsoleil). To solve this
problem, STICS uses the most complex method for radiation transfers within the
canopy (see details in § 9.2.1.2). While the basic level of calculation for sole crops is
the soil, for intercropping, it is the top of the understorey canopy. On a daily time
step, the shaded part of the understorey canopy corresponds to the vertical projec-
tion of the dominant foliage at the soil surface. The elementary pixel for calculation
consists of the largeur part of the dominant crop (see § 9.2.1.2.2 and Figure 9.2),
the shaded surface of the understorey crop (surf(ao)) and the sunlit surface of the
understorey crop (surf(as)) (Figure 9.5).
176
Canopy microclimate
Figure 9.5. Simplified representation of an elementary pixel of the system. largeur/2 represents
the half-width of the dominant crop part, SURFAO represents the shaded surface of the
understorey crop part, and SURFAS represents the sunlit surface of the understorey crop part
of this elementary pixel.
The proportion of income radiation intercepted by the dominant (faparD) and the
understorey crop (faparU) can then be easily obtained from Eq. (9.11) coupled to
Eq. (9.1) (Beer’s law analogue applied to the understorey crop) (Eq. (9.12)).
The fapar of both crops greatly depends on their respective heights, which not only
depend on the plant characteristics but also on the growth conditions as demonstrated
in Figure 9.6.
177
STICS soil-crop model
9.3.1.1 Albedo
The surface albedo (albedolai; Eq.(9.14)) varies between the soil value (𝖺𝗅𝖻𝗌𝗈𝗅) and the
vegetation value (𝐚𝐥𝐛𝐯𝐞𝐠C) which is equal to 0.23 (Ritchie, 1985).
The soil albedo (𝖺𝗅𝖻𝗌𝗈𝗅) varies as a function of soil type (𝐚𝐥𝐛𝐞𝐝𝐨S of dry soil),
moisture in the surface layer (Figure 9.7), and the presence of any plastic or plant
cover (see § 11.4.7). It decreases linearly with the water content of the surface layer
(𝗁𝗎𝗋 (1)) according to a relationship established from experimental results obtained
for different types of soil (𝗁𝗎𝖼𝖼 (1) and 𝗁𝗎𝗆𝗂𝗇 (1) being the water content at field
capacity and wilting point, respectively). The albedo of the plastic or plant cover
(𝐚𝐥𝐛𝐞𝐝𝐨𝐦𝐮𝐥𝐜𝐡𝐩𝐥𝐚𝐬𝐭𝐢𝐪𝐮𝐞T and 𝐚𝐥𝐛𝐞𝐝𝐨𝐦𝐮𝐥𝐜𝐡𝐫𝐞𝐬𝐢𝐝𝐮𝐬G , respectively), is taken into
account, as well as the proportion of soil cover (parameter 𝐜𝐨𝐮𝐯𝐞𝐫𝐦𝐮𝐥𝐜𝐡𝐩𝐥𝐚𝐬𝐭𝐢𝐪𝐮𝐞T
for a plastic cover, and variable couvermulch(t) for a plant cover).
178
Canopy microclimate
Figure 9.7. Variation in a loam-sandy soil albedo based on its surface characteristics: water
content, colour (dry albedo of 0.18 for a dark soil or 0.28 for a light soil) and bulk density
(1.2 for a loose soil or 1.5 for a compacted soil).
179
STICS soil-crop model
The insolation fraction is estimated using Angström’s formula (Eq. (9.18)), the parame-
ters of which are 𝐚𝐚𝐧𝐠𝐬𝐭C and 𝐛𝐚𝐧𝐠𝐬𝐭C . Extraterrestrial radiation (𝖱𝖦𝖤𝖷)
is calculated using standard astronomic formulae (Grebet, 1993). If the vapour pressure
is not available, it is estimated as the saturated vapour pressure at the temperature
𝖳𝖣𝖤𝖶 𝗍𝗆𝗂𝗇 𝐜𝐨𝐫𝐞𝐜𝐓𝐫𝐨𝐬𝐞𝐞C.
Figure 9.8. Variation in the saturated vapour pressure as a function of temperature according
to Alt (1978) referred to by Guyot (1997). The water status in the air is vapour represented by
the point and the temperature corresponding to the same pressure on the curve is the dew
temperature (TDEW).
180
Canopy microclimate
Figure 9.9. Visual evaluation of the estimate of the actual vapour pressure using the hypothesis
TDEW=tmin in Avignon, France.
In both calculations, wich are compared in Figure 9.10, the crop temperature is
subjected to an iterative convergence procedure (explained in § 9.3.2), meaning that
these calculations need to be performed several times in succession.
Figure 9.10. Comparison of Brunt’s and Brutsaert’s formulae for the calculation of net
radiation in Avignon, France.
181
STICS soil-crop model
where rnet is the net daily radiation in MJ m–2, et is the daily evapotranspiration in
mm, and hauteur is the canopy height (see § 9.2.1.1). The tcultmax value cannot be
lower than tmax. In this approach, we assume that tcultmin = tmin.
Figure 9.11. Comparison between the empirical relationship and the energy balance for the
calculation of crop temperature for two different surfaces (bare soil and a sunflower crop) in
Avignon, France, using two bare soil roughness factors : a) 1 mm and b) 5 mm).
182
Canopy microclimate
Eq. (9.21) shows the minimum and maximum values of the various fluxes: net radia-
tion (𝗋𝗇𝖾𝗍𝗆𝗂𝗇 and 𝗋𝗇𝖾𝗍𝗆𝖺𝗑), soil heat (𝗀𝗆𝗂𝗇 and 𝗀𝗆𝖺𝗑) and evapotranspiration (𝖤𝖳𝖬𝖨𝖭
and 𝖾𝗍𝗆𝖺𝗑), as well as the minimum and maximum values of the aerodynamic resist-
ance (𝗋𝖺𝖺𝗆𝗂𝗇 and 𝗋𝖺𝖺𝗆𝖺𝗑).
To calculate long wave radiation, i) atmospheric radiation is assumed to remain constant
throughout the day and is estimated using Brutsaert’s formula (Eq. (9.17)), and ii) surface
radiation is calculated using tcultmax and tcultmin , requiring the iterative convergence
procedure. At the end of the night, 𝖤𝖳𝖬𝖨𝖭 and 𝗋𝗀𝗆𝗂𝗇 are zero, while 𝗋𝗀𝗆𝖺𝗑 and 𝖾𝗍𝗆𝖺𝗑
are estimated assuming sinusoidal changes during the day. The 𝗀𝗆𝗂𝗇 value is calculated
as an empirical function of the wind speed under the cover (Cellier et al., 1996). The
𝗀𝗆𝖺𝗑 value is assumed to be 25 % of the maximum net radiation below the cover. In
addition to the canopy height (hauteur) and the bare soil roughness (𝐳0𝐬𝐨𝐥𝐧𝐮S), the
calculation of 𝗋𝖺𝖺𝗆𝖺𝗑 and 𝗋𝖺𝖺𝗆𝗂𝗇 requires wind speed values (see § 11.3.2): the night-
time wind speed is assumed to be equal to 50 % of the daily mean wind speed, and the
daytime wind speed is assumed to be 150 % of the daily mean wind speed.
Figure 9.11 shows the impact of surface type and soil roughness on the temperature
calculation: the rougher the soil, the greater the soil evaporation. Meanwhile, Figure 9.12
shows that the energy balance method, for the minimum temperature, produces results
which are identical to the driving hypothesis of the empirical method (tcultmin tmin).
183
STICS soil-crop model
Figure 9.13. Diagrams representing the iteration loop of tcult calculations implementaing a test
on the value of difftcult for each option: a) reference evapotranspiration and b) Shuttleworth
and Wallace. difftcult is the variation of between two successive days, eop is the maximum
plant transpiration, EP the actual plant transpiration, SWFAC the EP/eop ratio from the
previous day, and esol the actual soil evaporation.
184
Canopy microclimate
Figure 9.14. Drawing of the resistance scheme applied to the soil crop system (adapted from
Brisson et al., 2004).
– Limited by 𝗆𝗈𝗎𝗂𝗅𝗅𝗆𝗎𝗅𝖼𝗁:
– Limited by 𝗆𝗈𝗎𝗂𝗅𝗅:
and:
185
STICS soil-crop model
The amount of energy required for the direct evaporation of water on leaves (𝖤𝗆𝖽)
is RNETP1 while this energy used for direct evaporation must be deducted to eval-
uate the resulting energy available for transpiration (RNETP2). Energy distribution
between bare soil and the soil cover (mulch) depends on couvermulch (Eq. (9.22)),
with 𝗆𝗈𝗎𝗂𝗅𝗅 being the water retained on the foliage and 𝗆𝗈𝗎𝗂𝗅𝗅𝗆𝗎𝗅𝖼𝗁 the water retained
by the vegetal mulch. The 𝖽𝖾𝗅𝗍𝖺𝗍 value is the gradient of the relationship between satu-
ration vapour pressure and temperature, rnetS is the net radiation at the soil surface,
and is the psychrometric constant (mbar °C–1). The 𝖽𝗈𝗌 variable is the saturation
deficit within the vegetation and links all the fluxes. It is calculated (Eq. (9.25) using
the method by Shuttleworth and Wallace (1985).
where 𝖽𝗌𝖺𝗍 is the air saturation deficit (mbar); rnet (MJ m–2) is the net daily radia-
tion; 𝖫 is the latent heat of vaporisation (MJ kg–1); 𝗋𝖺𝖺 is the aerodynamic resistance
between the canopy and the reference height of weather measurements (𝐳𝐫C gene
rally 2 m) calculated from the canopy height and wind speed (see § 11.3); EVAPO
is the total transpiration and evaporation; esol, emulch and 𝖾𝗆𝖽 are evaporation
(in mm) from soil, mulch, and free water on leaves, respectively; and eop is the daily
maximum plant transpiration flux (in mm).
Eq. (9.25) is very similar to Eq. (9.37) (see § 9.4.1), except that the evaporative term
EVAPO implies a potential transpiration flux: EVAPO is the accumulation of tran-
spiration fluxes eop and all the direct evaporation fluxes, i.e. esol, 𝖤𝗆𝖽 and emulch.
The value of this direct evaporation affects 𝖽𝗈𝗌 and can cause the evaporative demand
of the plant to fluctuate. The three components of the direct evaporation are calcu-
lated from an intermediate value of the saturation deficit 𝖽𝗈𝗌 based on the hypothesis
that, at complete saturation of the surfaces, the evaporation can be treated using a
Priestley-Taylor type formalisation (Brisson et al., 1998a).
To solve the above equations, several terms must be calculated: i) the distribution of
the energy sources between the soil and the plants (rnetS , 𝖱𝖭𝖤𝖳𝖯 and 𝖱𝖭𝖤𝖳𝖯), ii) the
water retention on the foliage and in the mulch (𝖤𝗆𝖽 and emulch), iii) the resistances
to diffusion (𝗋𝖺𝖺 and ras), and iv) the surface resistances (𝗋𝖼 and 𝗋𝖺𝖼) (see § 9.3.3.3), and
v) the soil evaporation (esol).
186
Canopy microclimate
The bare soil value of roughness can vary from 10–2 for very rough ploughing to 10–4 m
for a very flat soil surface. Figure 9.15 shows the effect of this parameter on soil evapo-
ration: the greater the roughness, the higher the soil evaporation in the given example,
the annual difference between extreme values is about 70 mm. The reference height
taken for meteorological data is 2 m. If the plant canopy height exceeds this threshold,
a wind speed value is recalculated at a reference height of over 2 m (parameter 𝐳𝐫C) by
187
STICS soil-crop model
applying a logarithmic profile. The other meteorological values are not recalculated.
The calculations of diffusive resistances are different for bare soil (Eq. (9.28)) and cover
crops (Eq. (9.30) for lai ), while for non-cover crops ( lai ) an LAI-dependent
linear combination of the two first values is used (Eq. (9.29)). They all depend on wind
speed (tvent).
The canopy resistance (𝗋𝖼 ; Eq. (9.32)) is the product of four factors:
where 𝐫𝐬𝐦𝐢𝐧P is the minimal stomatal resistance of leaves, 𝖽𝗌𝖺𝗍 (in mbars) is the
saturation deficit, trg (in MJ m–2 s–1) is the global radiation and fco2s is a CO2
dependent variable.
188
Canopy microclimate
Due to the daily time step, the parameter 𝐫𝐬𝐦𝐢𝐧P cannot be inferred from the instan-
taneous values of measurements but must be obtained using a top-down approach
(Brisson et al., 1998a; Tolk et al., 1996; Baldocchi et al., 1991). Values of 250, 215 and
220 sm–1 where found for soybean, maize and sorghum respectively (Brisson et al.,
1998a; Brisson et al., 2004). The sensitivity of ep to those variables are shown in
Figures 9.16 and 9.17
The ‘saturation deficit’ and ‘radiation’ components are taken from research by Stockle
and Kjelgaard (1996). With regard to the conditions for applying the proposed formulae,
the saturation deficit is calculated at the meteorological scale and the radiation is the
incident radiation above the crop.
Figure 9.16. Influence of the minimal stomatal resistance ( in m s–1) on the cumulative
transpiration of a rainfed vineyard in Avignon, France.
If the atmospheric CO2 is high, stomatal conductance falls. Idso (1991) demon-
strated the existence of proportionality between the CO2 effect on conversion
efficiency and the CO2 effect on stomatal conductance, at a ratio of 2.5 for the addi-
tion of 300 ppm in the nominal concentration. Furthermore, Stockle et al. (1992)
proposed a species-dependent formalisation (Figure 9.18). We propose combining
these two approaches to take into account the species and ensure a continual effect
of CO2 (Eq. (9.33)). The fco2 value is the species-dependent CO2 effect on conver-
sion efficiency (Eq. (4.20)), which affects the value of the species-dependent factor
on stomatal closure (fco2s).
189
STICS soil-crop model
Figure 9.18. Influence of the species on the stomatal resistance CO2 dependent effect.
190
Canopy microclimate
9.3.4 Intercrop
To simulate intercropping, the later energy budget, calculated using the ‘resistance
approach’ option is used to estimate the crop water requirement. This approach is
particularly relevant for intercrops because it allows for microclimatic effects on water
requirements: convection beneath the dominant canopy and a decrease in the vapour
pressure deficit due to transpiration from the understorey plants. Actual soil evapora-
tion and plant transpiration are then calculated independently by means of a soil water
balance (see § 11.2 and 11.5). These fluxes are then re-introduced into the energy budget
to calculate crop temperature, which is a driving variable for plant growth and develop-
ment (see § 3, 4, and 5). The required adaptations for intercrops concern the first stage.
191
STICS soil-crop model
But the energy actually available for crop transpiration must also account for possible
direct water evaporation from the leaves.
Figure 9.19. The two possible schemes of resistance networks used to estimate water
requirements for intercrops (right-hand side of the schemes) and the fluxes (left-hand side of
the schemes). a) The understorey crop is near the ground, b) the understorey crop is nearly as
high as the dominant crop (adapted from Brisson et al., 2004).
192
Canopy microclimate
For the high understorey crop (Eq. (9.36); see Figure 9.19b):
where hauteur D and hauteur U are the heights of the dominant and the understorey
crops respectively. The threshold height for the low understorey crop is arbitrarily
fixed at 0.2 m. The reference height taken from meteorological data is 2 m. If the plant
canopy height exceeds this threshold, a wind speed value is recalculated at a reference
height of over 2 m by applying a logarithmic profile. The other meteorological values
are not recalculated.
193
STICS soil-crop model
where 𝖽𝖾𝗅𝗍𝖺𝗍 is the gradient of the relationship between saturation vapour pressure and
temperature; 𝗍𝖺𝗂𝗋 (°C) is the average daily temperature; rnet (MJ m–2) is the net daily
radiation; 𝖫 is the latent heat of vaporisation (MJ kg–1); is the psychometric constant
(mbar °C–1) depending on atmospheric pressure 𝐩𝐚𝐭𝐦C; 𝖽𝗌𝖺𝗍 is the air saturation deficit
(mbar); 𝖳𝖵𝖠𝖱 is the saturated vapour pressure based on the temperature (mbar) (see
§ 9.3.1.2); 𝗋𝖺𝖺 is the aerodynamic resistance between the canopy and the reference
height of weather measurements (𝐳𝐫C generally 2 m) calculated from the canopy height
and wind speed (see § 11.3); esol , emulch and 𝖾𝗆𝖽 are evaporation (in mm) from soil,
mulch, and free water on leaves, respectively; and ep is plant transpiration (in mm).
The average daily moisture (humidite) is then calculated based on the crop tempera-
ture (Eq. (9.38)):
If the wind speed weather variable is not available, a default value of 𝗋𝖺𝖺 is used (𝐫𝐚C).
If air humidity is not available the same assumption is made as before, using the
parameter 𝐜𝐨𝐫𝐞𝐜𝐓𝐫𝐨𝐬𝐞𝐞C (see § 9.3.1.2). The moisture variable can thus be calculated
if actual weather data is not available.
194
Canopy microclimate
Figure 9.20. Diagrams representing the iteration loop of hourly humidity (Hum(h)) calculations
based on the convergence between the average hourly values and the daily value, the fitted
variable being the minimum dewpoint value.
For an unheated (‘cold’) shelter, water requirements are estimated using the reference
evapotranspiration approach. The potential evapotranspiration is easily estimated
(Eq. (9.40)) using a multiplicative coefficient of radiation, 𝐜𝐨𝐞𝐟𝐝𝐞𝐯𝐢𝐥C (De Villele, 1974).
Rainfall is assumed to be zero and thus the crop must be watered by irrigation.
195
STICS soil-crop model
Figure 9.21. Comparisons between average hourly values and daily values of a) crop tempera-
ture and b) canopy humidity.
Figure 9.22. Hourly dynamics of the microclimatic variables over eight days: b) crop tempera-
ture, dew point and a) humidity.
196
Canopy microclimate
Temperature variations under a cold shelter are estimated using an energy balance based
on the work by Boulard and Wang (2000). On a daily time step, the heat flux in the soil
is ignored, assuming that the losses and gains balance out. The difference in mean daily
temperature inside and outside (𝖽𝖾𝗅𝗍𝖺𝗍𝖾𝗆𝗉) is thus expressed in Eq. (9.41).
where 𝗄𝗁 is the coefficient of heat transfer (W m–2 K–1) of the shelter, 𝗄𝗌 is the coef-
ficient of energy losses between the outside and inside of the shelter (W m–2 K–1),
𝐜𝐨𝐞𝐟𝐫𝐧𝐞𝐭C is a synthetic coefficient which converts external global radiation into net
interior radiation (with a standard value of 0.59), 𝖫 is the latent heat of vaporisation and
𝖾𝗌𝗍𝗂𝗆𝖾𝗍 is the evapotranspiration estimated from the water balance for the previous
day (et) and the evaporative demand for the day (tetp , Eq. (9.42)).
where 𝗄𝗌 (Eq. (9.43)) increases with the external wind speed (tvent) using the param-
eters 𝐚𝐤𝐬C and 𝐛𝐤𝐬C, equal to 6.0 and 0.5 respectively. The 𝗄𝗁 variable (Eq. (9.43))
depends on the proportion of vents related to the total surface area of the greenhouse
(𝗌𝗎𝗋𝖿𝗈𝗎𝗏𝗋𝖾 T) and the wind speed. The values of the constants 𝐜𝐯𝐞𝐧𝐭C and 𝐩𝐡𝐢𝐯0C are
0.16 and 4.10–3, respectively. The 𝗌𝗎𝗋𝖿𝗈𝗎𝗏𝗋𝖾 input can take three values during the
growth cycle.
197
STICS soil-crop model
198
Canopy microclimate
199
Chapter 10
Transfers in soil:
water, nitrate and heat fluxes
Joël Léonard, Bruno Mary, Guillaume Jego,
Nicolas Beaudoin and Nadine Brisson
Reviewed by: Patrice Lecharpentier
201
STICS soil-crop model
(e.g. 30 cm) in order to compare with water or mineral N measurements that are
made at the same depths. Each layer is characterised according to five basic physical
properties:
– the gravimetric water content at field capacity (𝐡𝐜𝐜𝐟S , in ),
– the gravimetric water content at permanent wilting point (𝐡𝐦𝐢𝐧𝐟S , in ),
– the volumetric content of pebbles (𝐜𝐚𝐢𝐥𝐥𝐨𝐮𝐱S , in ),
– the bulk density of the fine earth fraction (𝐃𝐀𝐅S , in g cm–3),
– the dispersion thickness (𝐞𝐩𝐝S , in cm).
These properties must be defined in the soil input file. They are used to derive total
soil porosity assuming a constant solid density (2.66 g cm–3). For computational
purposes, the five layers are divided into “elementary layers” (1 cm thick) in order to
simulate water, nitrate and heat transfers in soil as well as root functioning. However,
model outputs are mostly available at the scale of each soil layer described in the soil
input file. The soil pore space can involve up to five compartments: microporosity,
macroposity, cracks, pebbles and artificial drains:
– Microporosity is the main compartment and the only one which is mandatory
because this is where the non-preferential flow of water and nitrate occurs. At the
daily scale, water content in microporosity cannot exceed field capacity. It can exceed
field capacity temporarily at an hourly scale when the hourly water filled pore space
(WFPS) option is active (§ 12.3.5).
– Macroporosity allows water in excess over field capacity to accumulate at the
bottom of any soil layer when its infiltration rate is limiting, i.e. lower than the down-
wards water flow. The volume occupied by the macroporosity, which is assumed to be
nil by default, is calculated if 𝐜𝐨𝐝𝐞𝐦𝐚𝐜𝐫𝐨𝐩𝐨𝐫S = 1, as follows:
– Cracks are intended to simulate water flow in swelling clay soils and bypass flow in
soils having a variable porosity. Cracks, which are not activated by default, are active if
𝐜𝐨𝐝𝐞𝐟𝐞𝐧𝐭𝐞S = 1 and if the macroporosity option is also active.
– pebbles can be explicitly taken into account when they represent a large percentage
of the soil volume because their properties strongly differ from those of the soil matrix.
Pebbles, which are not activated by default, are active if 𝐜𝐨𝐝𝐞𝐜𝐚𝐢𝐥𝐥𝐨𝐮𝐱S = 1.
– artificial drains are sometimes installed on poorly draining soils. Their effect is
not simulated by default but is activated if 𝐜𝐨𝐝𝐫𝐚𝐢𝐧𝐚𝐠𝐞S = 1 and if the macroporosity
option is also activated.
Among these five pore space compartments, which are summarised in Figure 10.1,
only the first is compulsory. The four other optional compartments should be used
in specific situations that cannot be handled by the standard soil description and
associated transfer model, which is the only one that has been extensively tested.
The optional compartments and their associated formalisms require additional
parameters which are not easy to provide. It should be noted that the nitrification,
denitrification and N2O emissions routines incorporate a hourly WFPS option
which allows users to take into account occurrence of soil water content values
between field capacity and saturation near the soil surface on rainy days (§ 12.3.5).
However, these water content values do not appear in daily outputs: it is assumed
that they occur only within a 24 hour period and do not influence the return to field
capacity the next day.
202
Transfers in soil: water, nitrate and heat fluxes
203
STICS soil-crop model
1999; Van Der Ploeg et al., 1995). In the first STICS versions, the thickness of mixing
cells was equal to that of elementary layers (1 cm) which often resulted in an under-
estimation of nitrate dispersion in the soil profile. This is why the thickness of mixing
cells is now considered independently from the thickness of elementary layers and
provided in the soil input file as a specific dispersion parameter of each layer (𝐞𝐩𝐝S).
This parameter varies between 1 cm and 30 cm, depending on the type of soil porosity.
The relationship between dispersivity and soil textural class, as shown for example in
Perfect et al. (2002) can be used to estimate the mixing cell thickness. STICS does not
simulate the upward nitrate movement.
When water or nitrate reaches a layer deeper than maximum root depth, it becomes
unavailable for plants. It may be retrievable by a subsequent crop if it remains above
the base of the soil profile. If not, it is definitively lost. The model allows users to define
a minimum nitrate concentration level which cannot be leached or taken up by the
crop. This threshold (𝐜𝐨𝐧𝐜𝐬𝐞𝐮𝐢𝐥S , in ) allows users to simulate soils having
an anionic exchange capacity. It is provided in the soil input file.
When they represent a significant percentage of the soil volume, pebbles can affect
the properties of the soil microporosity and then water and nitrate flows. In STICS,
the volumetric percentage of pebbles in each soil layer 𝐜𝐚𝐢𝐥𝐥𝐨𝐮𝐱S ( ) can be provided,
together with the type of pebble. Each type of pebble has a specific bulk density
𝐦𝐚𝐬𝐯𝐨𝐥𝐜𝐱G ( ) and moisture at field capacity 𝐡𝐜𝐜𝐜𝐱G ( ), while the minimal
moisture content of pebbles is assumed to be equal to . Using this infor-
mation, equivalent properties of the whole soil are calculated as the volume-weighted
average of the fine earth fraction and pebble properties. When the percentage of
pebbles is high, fluxes such as evaporation can be strongly affected (Figure 10.2).
204
Transfers in soil: water, nitrate and heat fluxes
Figure 10.2. Effect of pebbles (70 % in volume) on soil evaporation and the consequences on
mean spring crop temperature in a vineyard in south-eastern France.
than the water content threshold 𝐡𝐮𝐦𝐜𝐚𝐩𝐢𝐥S ( ). Capillary rise occurs at a constant
rate 𝐜𝐚𝐩𝐢𝐥𝐣𝐨𝐮𝐫S (mm day–1) until the water content of the deepest layer reaches the
𝐡𝐮𝐦𝐜𝐚𝐩𝐢𝐥S threshold (in ).
Bypass flow can be considered in the case of swelling clay soils, where water flow can
enter the soil through shrinkage cracks. In the model, cracks are represented in the
form of a unique compartment, extending from the soil surface to the maximum soil
depth, which replaces the macroporosity described above. The macroporosity volume
(in mm cm–1 soil) is then redefined as a proportion (50%) of the available water:
Cracks can only be filled from the soil surface, due to overflow from the surface layer
or runoff when present. Direct feeding by rainfall interception is not considered.
The impact of an artificial drainage system (e.g. mole drains) installed on poorly
draining soils can be taken into account in the model. Although artificial drainage
is better described in three dimensions rather than one and with shorter time steps
than the day, it can be approximated by steady state drainage equations such as the
205
STICS soil-crop model
Figure 10.3. Effect of a strong decrease in infiltrability ( ) at the base of the second layer
located at 30 cm on the soil water content over 0-30 cm.
206
Transfers in soil: water, nitrate and heat fluxes
Figure 10.4. Water and nitrate flows in the soil: they are mainly downward but can be upward
if cracks are present or if capillary rise occurs.
207
STICS soil-crop model
with 𝗐𝗌𝖺𝗍 depending on whether the 𝐜𝐨𝐝𝐞𝐟𝐞𝐧𝐭𝐞S option is active or not (§ 10.1.1 and
10.1.3).
The parameters 𝖻𝗉𝗌𝗂𝗌𝗈𝗅 (unitless) and 𝗉𝗌𝗂𝗌𝗈𝗅𝗌 ( ) are calculated using the two points:
( , –0.03 MPa) and ( , –1.5MPa).
The predawn leaf water potential (psibase) is calculated as:
since the roots contributing to predawn water potential are only those located in moist
layers, in which 𝗉𝗌𝗂𝗌𝗈𝗅 is above –1.5 MPa.
The predawn leaf water potential depends on the rooting system and rainfall events as
illustrated in Figure 10.5.
208
Transfers in soil: water, nitrate and heat fluxes
Figure 10.5. Influence of rooting depth and rainfall events on psibase. Example of a vineyard
in the Rhône valley: the simulated production of the shallow rooted vineyard is half of that of
the deeply rooted vineyard.
209
STICS soil-crop model
Figure 10.6. Simulated evolution of the soil temperature in two soil layers for two values of
albedo at soil surface, in spring and northern France.
210
Transfers in soil: water, nitrate and heat fluxes
depth calculations, e quations derived from Thorsen et al. (2010) were used. Model
parameterisation and validation under Canadian climatic conditions are presented
in Jégo et al. (2014).
Regarding runoff, STICS does not include a component dedicated to its prediction, such
as depending on soil surface state or rainfall properties. However, STICS does integrate
a linear relationship between runoff and daily rainfall that can be used to reproduce the
water loss from runoff by providing adequate parameters (see § 11.4.3). The presence of
vegetation cover or mulch modifies runoff behaviour by intercepting some of the rainfall.
The snow cover is either affected by the process of refreezing or the process of melting.
The amount of snow cover in the refreezing process (𝖬𝗋𝖿, in mm d–1) depends on the
average daily air temperature (tmoy) and the threshold temperature 𝐓𝐦𝐟C (in °C):
The refreezing process is controlled by the parameter 𝐒𝐖𝐫𝐟C (in mm °C–1 d–1). 𝖬𝗋𝖿 is
nil if tmoy is higher than 𝐓𝐦𝐟C.
The amount of snow cover in the process of melting (𝖬, in mm d–1) also depends on
tmoy and 𝐓𝐦𝐟C:
where 𝖪 (in mm °C–1 d–1) is the melting rate. Due to the seasonal variation in incoming
solar radiation and albedo, 𝖪 follows a sinusoidal function over a year period, with
a minimum melting rate (𝐊𝐦𝐢𝐧C, in mm °C–1 d–1) reached on December 21, and a
maximum amplitude (𝐃𝐊𝐦𝐚𝐱C, in mm °C–1 d–1) reached on June 21:
The snow cover which consists in a mixture of water in a solid state and liquid state is
estimated according to the model proposed by Thorsen et al. (2010), using the daily
precipitation. The amount of snow cover in a solid state (𝖲𝖽𝗋𝗒, in mm) is calculated as
follows:
211
STICS soil-crop model
The amount of snow cover in a liquid state (𝖲𝗐𝖾𝗍, in mm) is calculated similarly, but
cannot exceed 10% of the amount of solid snow:
The first term represents the remaining snow after sublimation and compaction of the
accumulated snow. The empirical parameter 𝐄C (in ) is assumed to capture
the combined effect of the metamorphosis of snow crystals and the densification of
the lower snow layers. The second term corresponds to the addition of new snow, the
density of which is assumed to be 100 kg m–3, i.e. 10 times lower than water density.
The third term stands for the snow melting on the current day. It is determined using
the density of the snow cover (𝗉𝗌 , in kg m–3), itself calculated as follows:
The snow cover behaves as a thermal insulator: it prevents the soil temperature to drop
as low as the minimum air temperature (tmin) and to get as high as the maximum air
temperature (tmax). In order to simulate this thermal effect, the climate input data
are modified on days with a snow cover: the minimum and maximum air temperature
data are replaced by re-calculated values (called tminrec and tmaxrec) as indicated
below, and the model run again with these new climatic data.
When the snow depth Sdepth ( ) reaches or exceeds a user-defined threshold
, the re-calculated minimum air temperature tminrec reaches a threshold called
𝐭𝐦𝐢𝐧𝐬𝐞𝐮𝐢𝐥C (in °C), and the re-calculated maximum air temperature tmaxrec reaches
𝐭𝐦𝐚𝐱𝐬𝐞𝐮𝐢𝐥C (in °C).
When the snow depth lies between 0 and , a linear function is used to calculate
the new temperatures: tminrec varies linearly between tmin and 𝐭𝐦𝐢𝐧𝐬𝐞𝐮𝐢𝐥C; tmaxrec
varies linearly between tmax and 𝐭𝐦𝐚𝐱𝐬𝐞𝐮𝐢𝐥C. If there is no snow cover, tminrec and
tminrec are equal to tmin and tmax:
212
Transfers in soil: water, nitrate and heat fluxes
The snow cover also influences the soil water budget. If both rainfall and snowfall
occur, only the liquid water fraction enters the soil and moves down in soil. The water
equivalent of the snowfall is assumed to enter the soil profile only after melting.
The snow water equivalent (𝖲_𝖶𝖤, in mm) is the sum of 𝖲𝖽𝗋𝗒 and 𝖲𝗐𝖾𝗍:
The daily rainfall preciprec (in ) is re-calculated based on the total daily preci
pitation and the snow cover depth variation, as follows:
In order to simulate these effects on soil water, the climate input data are modified
on days with a snow cover: the daily precipitation data are replaced by re-calculated
values (preciprec) as indicated above for temperature data, and the model run again
with these new climatic data.
213
Chapter 11
Water balance
Nadine Brisson, Remi Vezy, Dominique Ripoche-Wachter
and Patrick Bertuzzi
Reviewed by: Nicolas Beaudoin and Jean-Louis Durand
215
STICS soil-crop model
capacity of the plant to extract this water due to its root characteristics. This is the frac-
tion of transpirable soil water (Lacape et al., 1998; Pellegrino et al., 2002; Sinclair, 1986),
which also corresponds to the concept of maximum available water content (amount of
water between field capacity and wilting point). This approach does not allow a precise
estimation of the location of root absorption in the soil layer (on a daily time step, all
models assume that transpiration equals absorption), but it does have the advantage of
implicitly taking into account capillary rise within the root zone. However, the threshold
of sensitivity may vary over time. This global estimate of transpiration is used in STICS,
while in other models (e.g. CERES) the calculation of uptake is differentiated in terms of
the soil layer, because of the need to simulate capillary rise. In that alternative approach,
water uptake per unit root length is based on the equation for radial flow to roots.
or, if the cover rate is used instead of the lai (§ 4.1.4; Eq. (11.2)):
When using the radiation transfer option, the values of 𝐞𝐱𝐭𝐢𝐧P and 𝐝𝐞𝐥𝐭𝐚P are dynam-
ically recalculated as a function of the canopy geometry and the quality of radiation
(direct/diffusive radiation). However, for row crops, which justify the use of the radia-
tion transfer calculations, it is highly recommended to use the second method, known
as the energy balance approach, which is explained as follows.
The energy balance method (Eq. (11.3)) is available only if the lai is explicitly calculated:
216
Water balance
Figure 11.1. Relative potential evaporation as a function of LAI for three different crops.
where rnetS is the net radiation at soil level (Eq. (9.26)), 𝖽𝖾𝗅𝗍𝖺𝗍 is the slope of the rela-
tionship between saturation vapour pressure and temperature (Eq. (11.4)) using the
function 𝖳𝖵𝖠𝖱 explained in chapter 9 (Eq. (9.19)), 𝖫 is the latent heat of vaporisation,
𝗀𝖺𝗆𝗆𝖺 is the psychrometric constant (Eq. (9.37)), ras is the aerodynamic resistance
between the soil and the vegetation (Eq. (9.28), (9.29), (9.30)) and 𝖽𝗈𝗌 is the saturation
deficit in the vegetation (Eq. (11.5)).
The 𝖽𝗈𝗌 variable is calculated assuming that, under soil conditions which are kept
moist, total evapotranspiration (𝖾𝖯𝖳: soil+canopy) can be written in the form of
evaporation according to Priestley and Taylor (Brisson et al., 1998a) as:
where 𝗋𝖺𝖺 is the aerodynamic resistance between the vegetation and the reference
level (Eq. (9.28), (9.29), (9.30)), and 𝖽𝗌𝖺𝗍 is the air saturation deficit at the same level.
217
STICS soil-crop model
During the first phase, evaporation is potential until the daily evaporation accumu-
lation reaches the 𝐪0S threshold. During the second phase, evaporation decreases;
this decrease depends on the weather and soil type through the parameter 𝖺𝖾𝗏𝖺𝗉
(Eq. (11.6)):
where 𝐚𝐜𝐥𝐢𝐦C is a weather parameter which depends mainly on the average wind
speed, 𝗁𝗎𝖼𝖼 is the water content at field capacity of the surface layer, 𝐃𝐀𝐅S is the
bulk density of the surface layer and 𝐚𝐫𝐠𝐢S is the clay content used here to estimate
the residual moisture, 𝗁𝖺. Nevertheless, the sensitivity of soil evaporation to these
parameters (Figures 11.2 and 11.3) is rather low compared to the sensitivity to 𝐪0S
(Figure 11.4). Although 𝐪0S depends on the soil texture and structure, it is diffi-
cult to infer it from soil particle size distribution or bulk density. It generally varies
between 0 to 30 mm.
Figure 11.2. Sensitivity of cumulative soil evaporation (ces) as function of the cumulative
evaporative demand (𝗌𝗎𝗆_𝖾𝗈𝗌) for two soil types (Clay soil: 𝐚𝐫𝐠𝐢S = 60 and 𝗁𝗎𝖼𝖼 = 0.4, Sandy
soil: 𝐚𝐫𝐠𝐢S = 10 and 𝗁𝗎𝖼𝖼 = 0.2).
The formalisation (Eq. (11.6) and Eq. (11.7)) also provides an estimate of the thick-
ness of the dry layer in the surface (or natural mulch: 𝖷𝖬𝖴𝖫𝖢𝖧), which is accounted
for in the water profile in the soil, since this layer is assumed to not participate in
evaporation.
218
Water balance
Figure 11.3. Sensitivity of cumulative soil evaporation (ces) as function of the cumulative
evaporative demand (𝗌𝗎𝗆_𝖾𝗈𝗌) for two weather conditions (No windy location, for wind speed
of 1 m s–1, 𝐚𝐜𝐥𝐢𝐦C = 20, Windy location, for wind speed of 2 m s–1, 𝐚𝐜𝐥𝐢𝐦C = 14).
Figure 11.4. Cumulative soil evaporation from 1 August until 31 December in northern France
for three values of 𝐪0S.
219
STICS soil-crop model
Figure 11.5. Partitioning of soil evaporation with depth as a function of the parameter 𝐳𝐞𝐬𝐱S
and 𝐜𝐟𝐞𝐬S = 5, assuming 𝖶𝗂 = 1 and 𝖷𝖬𝖴𝖫𝖢𝖧 = 0.
220
Water balance
Figure 11.6. Partitioning of soil evaporation with depth as a function of the parameter 𝐳𝐞𝐬𝐱S
and 𝐜𝐟𝐞𝐬S = 1, assuming 𝖶𝗂 = 1 and 𝖷𝖬𝖴𝖫𝖢𝖧 = 0.
221
STICS soil-crop model
Figure 11.7. Relative evaporative demand applied to soil (eos/tetp) and crop (eop/tetp)
accounting for the actual soil surface water state (esol = eos or esol = 0) for canopy qualified
in lai.
Figure 11.8. Relative evaporative demand applied to soil (eos/tetp) and crop eop/tetp)
accounting for the actual soil surface water state (esol = eos or esol = 0) for canopy qualified in
ground cover (tauxcouv).
222
Water balance
If the leaves (𝗆𝗈𝗎𝗂𝗅𝗅 ≠ 0) or the plant mulch laid on the soil surface (𝗆𝗈𝗎𝗂𝗅𝗅𝗆𝗎𝗅𝖼𝗁 ≠ 0)
have intercepted water (§ 9), this water will evaporate depending on the reference
evaporative demand (tetp): Emd for leaves (Eq. (11.10) and (11.11)) and emulch for
mulch (Eq. (9.22)). Naturally, the Emd threshold is set by the amount of water retained
on the foliage (𝗆𝗈𝗎𝗂𝗅𝗅), while the emulch threshold is set by the amount of water
retained in the mulch (𝗆𝗈𝗎𝗂𝗅𝗅𝗆𝗎𝗅𝖼𝗁). The evaporated water contributes to reducing
evaporative demand at the plant level.
223
STICS soil-crop model
224
Water balance
To account for this biochemical role, it is essential to also consider the mulch as a crop
residue left on the soil surface, and to determine its chemical composition with appro-
priate parameters (§ 13.5.3).
Figure 11.9. Simulated effects of a mulch on the various processes involved in the water and
nitrogen balances, and their consequences on the water fluxes.
225
STICS soil-crop model
Figure 11.10. Variation in the proportion of the soil covered by a plant mulch, whose quantity
(qmulch in t DM ha–1, initial mulch amount = 6 t DM ha–1) decreases as a function of the type
of the crop residue given by the parameter 𝐤𝐜𝐨𝐮𝐯𝐦𝐥𝐜𝐡G (a high value for entire fresh plants
and a low value for cut stalks).
where 𝐪𝐦𝐮𝐥𝐜𝐡𝐫𝐮𝐢𝐬0G is the mass of mulch above which runoff starts. For values
between qmulch = 0.1 and qmulch = 𝐪𝐦𝐮𝐥𝐜𝐡𝐫𝐮𝐢𝐬0G, we use the relationship estab-
lished by Scopel et al. (1998) to calculate FRUIS: above 𝐪𝐦𝐮𝐥𝐜𝐡𝐫𝐮𝐢𝐬0G, FRUIS is zero,
and below 0.1 we take the value of 𝐫𝐮𝐢𝐬𝐨𝐥𝐧𝐮S (Figure 11.11).
226
Water balance
Figure 11.11. Proportion of runoff water as a function of the soil surface (𝐫𝐮𝐢𝐬𝐨𝐥𝐧𝐮S) and the
plant mulch (𝐪𝐦𝐮𝐥𝐜𝐡𝐫𝐮𝐢𝐬0G and qmulch).
The effect of vegetation being present above the soil (lai) is accounted for via mecha-
nisms for the flow of water along stems (stemflow) as the 𝖿𝗋𝗎𝗂𝗌 proportion only applies
to the amount of water not involved in stemflow (Eq. (11.17)).
The 𝐬𝐭𝐞𝐦𝐟𝐥𝐨𝐰𝐦𝐚𝐱P parameter may vary from 0.2 to 0.5, depending on the species.
The 𝐤𝐬𝐭𝐞𝐦𝐟𝐥𝐨𝐰P parameter is less well known: it can initially be equal to the solar
radiation extinction coefficient (𝐞𝐱𝐭𝐢𝐧P).
The amount of water directly evaporated from the mulch (emulch) can be calculated
in two ways, using either the reference evapotranspiration (Eq. (11.19)) given in the
weather input file (tetp intended to be the Penman value) or the raw weather vari-
ables, which include wind speed and air humidity. The emulch value is limited by
𝗆𝗈𝗎𝗂𝗅𝗅𝗆𝗎𝗅𝖼𝗁:
227
STICS soil-crop model
With the first method, it is assumed that the water contained in mulch evaporates in
the same way as from a grass canopy, according to a resistance/height compensation
phenomenon. This phenomenon corresponds to the ‘extinction of energy at the soil
level’ by the vegetation (as is the case for soil). If the 𝐞𝐱𝐭𝐢𝐧P parameter is not active
(because the radiation intercepted by the canopy is calculated with the radiation
transfer model and not using the Beer’s law approach as explained in § 9.2.1.1), the
value is recalculated and varies depending on the crop geometry and the quality
of radiation.
With the second method, the Shuttleworth-Wallace formalisation is applied as
explained in § 11.3.2, and emulch evaporates in the same way as free water located at
the soil level receiving energy. It takes into account the proportion of soil covered by
the mulch (couvermulch).
In both cases emulch is limited by the amount of water intercepted by the plant mulch,
𝗆𝗈𝗎𝗂𝗅𝗅𝗆𝗎𝗅𝖼𝗁.
228
Water balance
The water content threshold separating the maximum transpiration stage and the
reduced transpiration stage (tetstomate) depends on root density, the stomatal func-
tioning of the plant, and the evaporative demand as expressed in Eq. (11.22), according
to Brisson et al. (1998a), who showed that that this threshold does not depend on the
soil type, for example via the maximal available soil water content.
where cumlracz is the sum over the whole rooting depth (zrac), of effective root length
density 𝗅𝗋𝖺𝖼𝗓 (in cm cm–3), 𝐩𝐬𝐢𝐬𝐭𝐨P (positive in bars) is the critical leaf water potential
229
STICS soil-crop model
of stomatal closure, and 𝐫𝐚𝐲𝐨𝐧P is the mean root radius, which is assumed to be equal
to 0.02 cm. When using this formula, we find that the tetstomate threshold tends to
be stable beyond a certain root depth (Figure 11.12). Table 11.1 summarises the role
played by the various factors influencing this value. It highlights the dominant effect of
evaporative demand (EOP), and the parameter of stomatal closure, 𝐩𝐬𝐢𝐬𝐭𝐨P.
Figure 11.12. Influence of the rooting depth (zrac) and the maximum daily transpiration
(eop) on the threshold of soil water content above wilting point (tetstomate) below which the
transpiration is reduced.
Table 11.1. Sensitivity analysis of the threshold tetstomate (in cm3 water cm–3 soil above
the wilting point).
230
Water balance
An iterative calculation distributes the transpiration within the soil profile, ensuring
that the whole of ep is shared out according to both root and water availability.
231
Chapter 12
Carbon and nitrogen transformations
in soil and balances
Bruno Mary, Fabien Ferchaud, Hugues Clivot
and Joël Léonard
Reviewed by: Nicolas Beaudoin
233
234
STICS soil-crop model
Figure 12.1. Soil C and N compartments and incoming and outgoing C and N fluxes in the STICS model (adapted from Autret et al., 2020). STICS
variable names are in italics.
Carbon and nitrogen transformations in soil and balances
which is highly variable over time. During a first phase after the addition of resi-
dues, the mineralisation can be positive or negative (immobilisation of soil mineral
N). During the second phase, it is positive through the ‘re-mineralisation’ process
which releases N coming from either the residue or the microbial biomass which
has broken it down.
The humified organic N in soil (Nhumt, in kg ha–1) is composed of two pools: an active
pool (Nhuma) and an inert pool (Nhumi). The first pool is fed by the humification of
fresh organic residues and contributes to mineralisation, whereas the second is assumed
to be inert on the time scale of a century. Such an inert pool is included in most models
simulating the long-term evolution of soil organic matter. However the models differ
greatly in the size of this pool, which may vary from less than 10% in ROTHC (Coleman
and Jenkinson, 1996) to 50% in CENTURY (Parton et al., 1987), 40% to 80% in AMG
(Clivot et al., 2019) and even 60% to 80% (Ludwig et al., 2007, 2003).
In STICS, the initial proportion of the inert pool (𝐟𝐢𝐧𝐞𝐫𝐭S, ratio of Chumi to Chumt) is
set to a default value of 0.65. This value appeared to be adequate to simulate the evolu-
tion of soil organic carbon in long-term arable experiments with the AMG model
(Saffih-Hdadi and Mary, 2008), which includes a mineralisation function similar to
STICS (Clivot et al., 2019). However this value depends on the cropping history of the
field. It should be smaller (down to 0.40) if the soil had a recent history of grassland or
forest land use or if it received large amounts of organic amendments.
The decomposition and mineralisation processes are dependent on soil and envi-
ronmental conditions, and may be affected differently by them, particularly soil
temperature and water content. The effect of temperature on C or N mineralisation
in soil is still a matter of controversy, as pointed out by Kirschbaum (2006). We
attribute some of the disagreement between authors to the fact that the tempera-
ture response differs according to the type of organic matter decomposed. In STICS
we use a different function for decomposition of humified soil organic matter
(SOM) and fresh organic residues. The effect of soil moisture might also be different
for the two processes, but little is known about this aspect. Therefore we use a
single function to describe the effect of water content on decomposition and C and
N mineralisation.
235
STICS soil-crop model
Σ
where 𝐩𝐫𝐨𝐟𝐡𝐮𝐦S is the depth of the biologically active layer (in cm), 𝐍𝐨𝐫𝐠S is its
organic N content (in %), 𝐟𝐢𝐧𝐞𝐫𝐭S is the fraction of inert carbon in SOM, 𝐃𝐀𝐅S
(in g cm–3) is the bulk density of the fine earth and 𝐜𝐚𝐢𝐥𝐥𝐨𝐮𝐱S is the volumic percentage
of pebbles in soil (proportion of material > 2 mm, in %).
The actual mineralisation rate constant K2 is calculated as:
with
𝖪𝗁𝗎𝗆 is the potential mineralisation rate constant (day–1), i.e. the mineralisation rate
constant of a soil maintained at the reference temperature value (𝐭𝐫𝐞𝐟𝐡G = 15 °C) and
reference moisture content (field capacity). The 𝗍𝗇𝗁 factor (unitless) is the norma
lised time for humus mineralisation, accounting for the effect of soil moisture and
temperature compared to reference conditions.
The soil water content (𝗁𝗎𝗋) modifies the mineralisation rate according to a multi-
linear function (Eq. (12.5)). Mineralisation is nil when the soil water content is below
𝐡𝐦𝐢𝐧𝐦G (Figure 12.2). It increases when the water content increases and then
decreases. The maximum rate is reached when the soil water content (expressed as a
proportion of field capacity) is between 𝐡𝐨𝐩𝐭𝐦G and 1. It decreases and reaches the
rate 𝐟𝐡𝐦𝐢𝐧𝐬𝐚𝐭G when the soil is saturated. Values of these parameters can be different
for temperate and tropical soils.
where 𝗐𝖿𝗉𝗌 and 𝗐𝖿𝗉𝗌𝖼𝖼 are the values of the water-filled pore space at actual water
content and at field capacity, respectively.
The effect of soil temperature on basal mineralisation (𝖿𝗍𝗁) can be described either by an
Arrhenius or a logistic function (Valé, 2006). We have chosen a logistic function which
enables to simulate the slower increase in mineralisation rate at high temperatures
when microbial activity slows down (Figure 12.3).
The proposed function is roughly exponential from 0°C to 25°C and increases more
slowly above this temperature (Eq. (12.6)). It relies on three parameters including the
236
Carbon and nitrogen transformations in soil and balances
Figure 12.2. Influence of relative soil water content (hur/hucc) on decomposition rate of soil
organic matter in two soils. Values of and = are 0.90 and 0.25 for the temperate
soil (standard parameters, adapted from Rodrigo et al. (1997)) and 0.67 and 0.22 for the tropical
soil (Sierra et al., 2003). = 0.50 and = 1.50 in both soils.
Figure 12.3. Influence of temperature on decomposition rates of organic matter: factor fth for
humus and ftr for organic residues. Parameter: 𝐭𝐦𝐢𝐧_𝐦𝐢𝐧𝐞𝐫𝐚𝐥𝐢𝐬𝐚𝐭𝐢𝐨𝐧G = 0.
237
STICS soil-crop model
tmin_mineralisationG
The effects of clay content, CaCO3 content, pH and C/N ratio on 𝖪2𝗁𝗎𝗆 are illustrated
in Figure 12.4.
238
Carbon and nitrogen transformations in soil and balances
Figure 12.4. Influence of clay content ( in %), CaCO3 content ( in %), pH ( ) and
C/N ratio ( ) on the potential decomposition rate of soil organic matter.
239
STICS soil-crop model
(𝐲𝐫𝐞𝐬G and hres) and two C/N ratios (CNbio and CNhum). For a given category, the
parameters are either constant (𝐤𝐛𝐢𝐨G, 𝐲𝐫𝐞𝐬G, and CNhum=1/𝐖𝐡G) or dependent
upon the C/N ratio of the organic residue (𝐂𝐬𝐮𝐫𝐍𝐫𝐞𝐬T), according to the following
relationships:
Figure 12.5. Flow diagram of the decomposition of organic residues in soil (Nicolardot et al.,
2001). The continuous lines indicate carbon flows and the dashed lines nitrogen flows.
where 𝐟𝐫𝐞𝐝𝐤𝐍G is a reduction factor of the decomposition rate of residues due to lack
of nitrogen.
The change in the associated microbial biomass (Cbio) and the rate of humus forma-
tion ( ), both in kg C ha–1 day–1, are calculated as follows:
240
Carbon and nitrogen transformations in soil and balances
peration is assumed to mix the residues uniformly with the soil over the depth
o
defined by a minimal value (𝐩𝐫𝐨𝐟𝐫𝐞𝐬T) and a maximal value (𝐩𝐫𝐨𝐟𝐭𝐫𝐚𝐯T). The
lower depth is a rather straightforward parameter to define. The upper depth has
no true physical meaning (it should be 0) but is an equivalent depth defining the
“residuesphere”, i.e. the volume of soil in close contact with the residues. It has been
shown that decomposition and N immobilisation are markedly reduced when resi-
dues are stratified (i.e. located in a fraction of soil volume) compared to a situation
where residues are homogeneously distributed throughout the soil (Magid et al.,
2006). Oorts et al. (2007) obtained a good simulation of C and N mineralisation
under field conditions after addition of straw by considering that the residuesphere
represented 10% of the total soil volume. Since STICS is a 1D model, it is not possible
to simulate this effect explicitly: it must be done indirectly by considering the ratio
𝐩𝐫𝐨𝐟𝐫𝐞𝐬T / 𝐩𝐫𝐨𝐟𝐭𝐫𝐚𝐯T as the fraction of soil volume occupied by the residuesphere
(see § 13.3.2).
The net N mineralisation rate ( , in kg N ha–1 day–1, positive or negative) resulting
from residue decomposition is calculated as the complement of the variation in the
three organic pools:
Examples of C and N dynamics predicted by the model are given in Figure 12.6. The
variation in organic pools may exceed the amount of N added by the residue when
mineral N is immobilised.
The C and N mineralisation kinetics differ according to the type of organic residue
(Figure 12.7). Decomposition results in net release of N for residues with a low C/N
ratio (vinasse, C/N = 8; catch crop C/N = 15) and net immobilization with residues
poor in N (rapeseed straw, C/N = 45).
241
STICS soil-crop model
Figure 12.6. Evolution of C and N pools simulated during the decomposition of a crop residue
(rapeseed straw) finely mixed in the soil, assuming no limitation by mineral N. The abscissa
represents the normalised time (constant temperature and moisture). The ordinate is expressed
in % of C or N added by the residue.
Figure 12.7. Evolution of C and N mineralised due to the decomposition of three types of
organic residues (rapeseed straw, catch crop shoots and vinasse). The abscissa represents
the normalised time (constant temperature and moisture, no limitation by mineral N). The
ordinate is expressed in % of C added or kg N ha–1.
242
Carbon and nitrogen transformations in soil and balances
12.3 Nitrification
Nitrate production in soil results from two successive processes: mineralisation (or
ammonification) and nitrification. Nitrification is often a rapid process in cultivated
soils under temperate climates, which may justify avoiding describing nitrification and
assimilating mineral N to nitrate-N. However in some soil and climatic conditions
(e.g. acidic, hydromorphic or tropical soils), the nitrification process may be much
slower and ammonium ions may persist in soil. Furthermore, the simulation of N2O
emissions due to nitrification requires a description of nitrification, as well as the
simulation of ammonia volatilisation which is highly dependent on NH4+ concentra-
tion. Accordingly, the present version of STICS can simulate nitrification and the two
forms of mineral N separately.
Nitrification is calculated in the biologically active layer (𝐩𝐫𝐨𝐟𝐡𝐮𝐦S). If the nitrifica-
tion option is not activated, the ammonium produced by the soil or added as fertilizer is
instantly converted into nitrate. If the nitrification option is activated, the nitrification
rate (𝗏𝗇𝗂𝗍, in mg N kg–1 day–1) is calculated as the product of the potential nitrifica-
tion rate (𝗏𝗇𝗂𝗍𝗉𝗈𝗍 , in mg N kg–1 day–1) and three factors depending on environmental
conditions: soil pH (fpHn), temperature (fTn) and water content (fhn).
243
STICS soil-crop model
where 𝐟𝐧𝐱G is the maximum fraction of NH4 which can be nitrified daily (day–1) and
NH4min (mg kg–1) is the minimal concentration of exchangeable NH4 in soil.
With the second option (𝐜𝐨𝐝𝐞_𝐯𝐧𝐢𝐭G = 2), vnitmax is an input value (𝐯𝐧𝐢𝐭𝐦𝐚𝐱G). The
effect of ammonium concentration is described by a Michaelis-Menten function:
where w is the gravimetric soil water content (L kg–1) and 𝐊𝐚𝐦𝐦G (mg N L–1) is the
affinity constant of nitrifiers for NH4.
Figure 12.8 shows 𝗏𝗇𝗂𝗍𝗉𝗈𝗍 when 𝐜𝐨𝐝𝐞_𝐯𝐧𝐢𝐭G = 1 and when 𝐜𝐨𝐝𝐞_𝐯𝐧𝐢𝐭G = 2 for two
values of 𝐊𝐚𝐦𝐦G (24 and 202 mg N L–1) and two values of w (0.15 and 0.24 g g–1).
Figure 12.8. Influence of the soil NH4 content on the potential nitrification rate. Parameters:
= 0.8, = 27.3 and NH4min = 1.
244
Carbon and nitrogen transformations in soil and balances
However, the second option has its drawbacks. Indeed, after a classical fertilisation at
soil surface, the nitrogen supplied is located in the first elementary soil layer. Since the
thickness of this layer is only 1 cm, ammonium concentration becomes very high and
fNH4 reaches 1. The nitrification rate is maximum in the first centimetre of soil and can
be very low in the soil below 1 cm. Total nitrification may be greatly under-estimated.
This effect can be corrected by placing the fertiliser over a greater depth, but this
requires one more parameter to estimate. The linear approach avoids this difficulty
thanks to the absence of a saturation of the nitrification rate that compensates for the
small thickness of the soil layer concerned.
12.3.2 pH effect
The effect of pH on nitrification is linear between 𝐩𝐇𝐦𝐢𝐧𝐧𝐢𝐭G where nitrification
is zero and 𝐩𝐇𝐦𝐚𝐱𝐧𝐢𝐭G where nitrification is maximum. Figure 12.9 illustrates this
relationship with the default values in STICS.
Figure 12.9. Influence of soil pH on the relative nitrification rate. Parameters: = 4.0,
= 7.2.
245
STICS soil-crop model
It is easier to use the Gaussian function to manipulate the optimum temperature, such
as when shifting it according to the average temperature. The two functions are illus-
trated in Figure 12.10.
Figure 12.10. Influence of soil temperature on the relative nitrification rate. Parameters:
tnitopt_gaussG = 32.5, = 16.0.
It must be noticed that the optimal water contents for nitrification (𝐡𝐨𝐩𝐭𝐧G) and miner-
alisation (𝐡𝐨𝐩𝐭𝐦G) can be different, which can lead to NH4+ accumulation in soil.
246
Carbon and nitrogen transformations in soil and balances
Figure 12.11. Influence of the relative water content (hur/hucc) on the relative nitrification
rate. Parameters: = 0.30, = 0.90. Bulk density = 1.30 g cm–3, water content at field
–1
capacity = 0.24 g g .
247
STICS soil-crop model
assumed to be driven by drainage, the rate of which is proportional to the water excess,
𝗍𝗁𝖾𝗍𝖺_𝗌𝖺𝗍 - 𝗁𝗎𝖼𝖼. This results in an exponential evolution of soil water content over time
(every hour), in which 𝐤𝐝𝐞𝐬𝐚𝐭G is the desaturation rate constant (day–1):
Figure 12.12 illustrates the hourly evolution of the volumetric water content of a top
soil layer after initial saturation by rainfall, using the default 𝐤𝐝𝐞𝐬𝐚𝐭G value = 3 day–1.
Figure 12.12. Evolution of the volumetric water content of a top soil layer during a day after
initial saturation by rainfall ( = 1.50 g cm–3, = 0.36 cm3 cm–3).
The default value of 𝐤𝐝𝐞𝐬𝐚𝐭G enables 95% of the water to be drained during one day
and thus to reach at the end of the day a value which is very close to field capacity, in
line with the daily soil water content calculated by the STICS water transfer submodel.
Each function of soil moisture or 𝗐𝖿𝗉𝗌 is then integrated from t = 0 to t = 1 instead of
being applied to the daily soil moisture content, thus accounting soil water contents
exceeding field capacity and simulating active denitrification.
12.4 Denitrification
Denitrification is assumed to occur only in the top soil, over the depth defined by the
parameter 𝐩𝐫𝐨𝐟𝐝𝐞𝐧𝐢𝐭S. The denitrification rate (vdenit, in kg N ha–1 day–1) is calcu-
lated as the product of a potential rate (vpotdenit, in kg N ha–1 day–1) and three control
factors: the substrate availability (NO3), the soil temperature (fTd) and the water-filled
pore space ( ):
248
Carbon and nitrogen transformations in soil and balances
There is no pH effect taken into account on the denitrification rate itself, although
some studies suggest that the optimum pH may be soil dependent.
Figure 12.13. Influence of the soil organic carbon content (%) on the potential denitrification
rate. Parameters: = 1, = 6, = 1, = 20.
249
STICS soil-crop model
where w is the gravimetric soil water content (L kg–1), NO3 is the nitrate content
(mg N kg–1 soil) and 𝐊𝐝G (mg N L–1) is the affinity constant of denitrifiers for nitrate
(Figure 12.14).
Figure 12.14. Influence of the soil nitrate concentration on the relative denitrification rate for
two values of (0.15 and 0.24 g g–1). Parameter: = 148.
250
Carbon and nitrogen transformations in soil and balances
Figure 12.15. Influence of the soil temperature on the relative denitrification rate. Parameters:
= 47 and = 25.
The 𝐰𝐟𝐩𝐬𝐜G threshold depends on soil texture. Some work suggests that it varies from
about 0.50 for a coarse texture to 0.70 for a fine texture. However, there is currently
no function linking this threshold to soil texture in STICS and the default value is
𝐰𝐟𝐩𝐬𝐜G = 0.62.
An option is available to simulate the soil water content and the 𝗐𝖿𝗉𝗌 at an hourly
time step. If it is activated (𝐜𝐨𝐝𝐞_𝐡𝐨𝐮𝐫𝐥𝐲_𝐰𝐟𝐩𝐬_𝐝𝐞𝐧𝐢𝐭G = 1), water content and 𝗐𝖿𝗉𝗌
are calculated every hour, assuming that each layer being saturated after a rainfall
event then desaturates at a exponential rate (𝐤𝐝𝐞𝐬𝐚𝐭G = 3 day–1). The values of the
hourly water content and water-filled pore space are calculated (see Eq. (12.30)) and
integrated at the daily scale.
251
STICS soil-crop model
Figure 12.16. Influence of the water-filled pore space on the relative denitrification rate, with
the default value of = 0.62 and two alternative values (0.50 and 0.70).
As 𝗐𝖿𝗉𝗌 approaches 1, anoxia becomes dominant in soil and favours losses in the
N2O form (Figure 12.17). For low 𝗐𝖿𝗉𝗌 values, the value of z is 0.16% (like the default
constant value).
252
Transfers in soil: water, nitrate and heat fluxes
Figure 12.17. Influence of the water-filled pore space on the fraction of nitrified nitrogen
emitted as . Parameter: = 0.0016.
where the factor r0 depends on the value of the 𝗐𝖿𝗉𝗌 threshold for denitrification
(𝐰𝐟𝐩𝐬𝐜G). With the standard value 𝐰𝐟𝐩𝐬𝐜G = 0.62, r0 is equal to 2.05.
253
STICS soil-crop model
Figure 12.18. Effect of soil pH on the proportion of N2O emitted by denitrification. Parameters:
= 5.6, = 9.2.
254
Transfers in soil: water, nitrate and heat fluxes
Figure 12.19. Effect of water-filled pore space on the proportion of N2O emitted by
denitrification. Parameter: = 0.62.
Figure 12.20. Effect of soil nitrate concentration on the proportion of N2O emitted by
denitrification.
255
STICS soil-crop model
The combined effects of pH, and nitrate on the molar fraction r are illustrated in
Figure 12.21.
Figure 12.21. Influence of combined factors on the fraction of denitrified nitrogen emitted
as N2O.
256
Transfers in soil: water, nitrate and heat fluxes
NH4l and NH3l are linked through a chemical equilibrium which is pH and temper-
ature dependent. The solubility equilibrium between NH3l and NH3g forms mainly
depends on temperature.
The first step consists in defining the volatilisable NH4+ immediately after the applica-
tion. The exchangeable NH4+ (𝐍𝐦𝐢𝐧𝐫𝐞𝐬T, in kg N ha–1) is split into two pools: a pool
which remains at the soil surface and which can be volatilised (𝖭𝗏𝗈𝗅𝖺𝗍𝗈𝗋𝗀, in kg N ha–1)
and a pool which infiltrates and is not volatilisable. The volatilisable NH4+ at the time
of application (t0) is:
The proportion of the volatilisable fraction (𝗉𝗋𝗈𝗉𝗏𝗈𝗅𝖺𝗍) decreases with the water
content of the manure (𝐞𝐚𝐮𝐫𝐞𝐬T), as follows in Figure 12.22.
Figure 12.22. Effect of soil tillage and water content of slurry on its volatilisable fraction.
The proportion of the volatilisable fraction is also affected by soil tillage: it decreases if
the soil has been tilled during the last 7 days before manure spreading (trsolvolat = –1)
and increases otherwise (trsolvolat = +1) (Morvan, 1999). It is calculated as follows:
257
STICS soil-crop model
Using the data given by Morvan and Leterme (2001) and Chantigny et al. (2004), we
can propose the following parameter values: 𝐚𝐥𝐩𝐡𝐚𝐩𝐇G = 0.005, 𝐝𝐩𝐇𝐯𝐨𝐥𝐦𝐚𝐱G = 1.0
and 𝐩𝐇𝐯𝐨𝐥𝐬G = 8.6. Based on these values, the soil pH variation immediately after
manure application is shown below:
Figure 12.23. Effect of mineral N content of slurry and soil pH on the initial pH increase
following slurry application.
During the following days, the pH at the soil surface (𝗉𝖧𝗏𝗈𝗅) returns to the soil pH
value (𝐩𝐇S), at a rate proportional to the decrease in the volatilisable pool:
The model then calculates the amounts of the four forms: NH4s, NH4l, NH3l and
NH3g (in mol m–2), using the acido-basic equilibria equations, the Henry solubility
equation and the transfer equations of Beutier and Renon (1978). These amounts
depend on soil temperature, water content, soil porosity, pH at soil surface and wind
speed. The ammonia concentration at the soil surface (𝖭𝖧𝗌𝗎𝗋𝖿, in µg N m–3) is:
258
Transfers in soil: water, nitrate and heat fluxes
259
STICS soil-crop model
Table 12.1. Components of the soil mineral nitrogen balance (kg N ha–1).
Initial pools Final pools
NH4-N 3 4
NO3-N 30 25
Total mineral N 33 29
Input fluxes Output fluxes
Plant N uptake 273
Precipitation 17 Nitrate leaching 8
Irrigation 10 Leaching in mole drains 0
Fertiliser 110 Fertiliser immobilisation 32
Fertiliser volatilisation 12
Humus mineralisation 187 Manure volatilisation 0
Residue mineralisation 10 (N2+N2O)-N losses 13
TOTAL INITIAL 367 TOTAL FINAL 367
Table 12.2. Components of the soil organic nitrogen balance (kg N ha–1).
260
Transfers in soil: water, nitrate and heat fluxes
Table 12.3. Components of the soil organic carbon balance (kg C ha–1).
The total N surplus (Nsurplus) is similar, but also includes the supply of organic fertil-
iser (QNresorg):
The gaseous N losses (QNgaz) are the sum of N denitrified (QNdenit), N volatilised
by mineral fertilisers (QNvoleng), N volatilised by organic fertilisers (QNvolorg) and
N emitted as N2O by nitrification (Qem_N2Onit):
where DSON is the variation in soil organic N, DSMN is the variation in soil mineral N,
DQNtot2 is the variation in crop N content (one or two crops in the case of intercrop-
ping) and Qles is the N leaching loss during the whole simulation period.
261
STICS soil-crop model
If we consider annual crops grown for several years, the terms DSMN and DQNtot2
are negligible. In that case, Eq. (12.54) can be simplified as:
This equation shows that: i) a negative N surplus implies that soil organic N is neces-
sarily decreasing over time and ii) a positive N surplus indicates that N losses (NO3,
NH3, N2, N2O) occur and/or that N has been stored in soil organic matter.
where DSOC is the variation in soil organic carbon and Qem_N2O is the amount
of N emitted as N2O during the simulation period. Other emissions due to ferti
liser production and fuel consumption must be added to this value in order to have a
complete GHG assessment (e.g. Autret et al., 2020).
262
Chapter 13
Soil-crop management effects
Nadine Brisson, Bruno Mary
and Dominique Ripoche-Wachter
Reviewed by: Nicolas Beaudoin
13.1 Introduction
The very first crop models (de Wit, 1978; Weir et al., 1984) sought to describe the
ecophysiology of crops in detail, often for didactic purposes; however, they paid little
attention to agronomic objectives. The next generation of models started to include
farming practices in the inputs (Ritchie and Otter, 1985; Williams et al., 1984), particu-
larly those related to irrigation and fertilisation. Accounting for farming practices in
the models requires an explicit and adequate simulation of the main state variable(s)
that the practice is assumed to modify, such as soil water content for irrigation, organic
residue dynamics for manure application or annual wood production for pruning. In
STICS, emphasis has been placed on crop and soil management which is important
for simulating industrial crops and essential for high value-added crops. However,
some techniques are not yet accounted for in the model when the corresponding state
variables are poorly described (e.g. those relating to soil structure) or not calculated at
all (e.g. those relating to crop health status).
263
STICS soil-crop model
soil (germination and underground shoot growth) are simulated (§ 3.4.1), depending
on sowing depth. With the planting option, the model simply considers a lag time from
transplanting to the start of actual plantlet growth (§ 3.4.2); this option requires the
user to initialise the plant state in terms of leaf area index, biomass, nitrogen status,
rooting depth and root density profile.
For STICS the uncertainty on sowing depth is assumed to be 2 cm, so that the
prevailing soil conditions for emergence (humidity in the seed bed) are those found in
the layer [x–2, x+2], where x is the prescribed sowing depth. However, this variation
in depth does not lead to variability in the emergence rate: all plants that emerge do
so at the same time.
The model also requires information about the crop’s geometrical pattern, which is
very important for radiation interception. The pattern can either be homogeneous
or in rows. In the latter case, the model requires geometrical parameters, such as the
inter-row distance and row orientation (§ 9.2.1.2).
STICS can also simulate a companion crop such as grass in vineyards. In this case, the
system is simulated as an intercropping system (§ 9.3.4).
13.2.2 Sowing
Users have the option to prescribe the sowing date (𝐢𝐩𝐥𝐭0T) or calculate it (iplts)
using rules relative to the weather and soil water status. To calculate the sowing
date, a period when sowing is allowed and defined as the interval [𝐢𝐩𝐥𝐭j, 𝐢𝐩𝐥𝐭j +
𝐧𝐛𝐣𝐦𝐚𝐱𝐚𝐩𝐫𝐞𝐬𝐬𝐞𝐦𝐢𝐬T]. Four criteria are taken into account to postpone sowing
within this sowing window:
– the soil must be wet enough: the water content in the seedbed layer must be
greater than the water content at wilting point. The depth of the seedbed is defined as
𝐩𝐫𝐨𝐟𝐬𝐞𝐦T ± 2 cm;
– the soil must be warm enough: the air temperature (𝗍𝖺𝗂𝗋) must exceed 𝐭𝐝𝐦𝐢𝐧P for
several days (𝐧𝐛𝐣𝐬𝐞𝐮𝐢𝐥𝐭𝐞𝐦𝐩𝐫𝐞𝐟T) to avoid germination delay or the stopping of plant
emergence (§ 3.4.1);
– the risk of freezing must be low: the minimum air temperature (𝗍𝗆𝗂𝗇) must be
greater than 𝐭𝐝𝐞𝐛𝐠𝐞𝐥P for at least 𝐧𝐛𝐣𝐬𝐞𝐮𝐢𝐥𝐭𝐞𝐦𝐩𝐫𝐞𝐟T days (§ 3.4.1.3);
– the soil must be not too wet so as to avoid compaction risks: the soil water content is
considered as damaging if it exceeds 𝐩𝐫𝐨𝐩𝐡𝐮𝐦𝐭𝐚𝐬𝐬𝐬𝐞𝐦G x 𝐡𝐜𝐜𝐟S in the zone between
the soil surface and the depth 𝐩𝐫𝐨𝐟𝐡𝐮𝐦𝐬𝐞𝐦𝐨𝐢𝐫T (§ 13.3.1.1).
264
Soil-crop management effects
Topping only concerns crops having a row structure and consists in restricting plant
structure in terms of height (𝐡𝐚𝐮𝐭𝐫𝐨𝐠𝐧𝐞T) and/or width (𝐥𝐚𝐫𝐠𝐫𝐨𝐠𝐧𝐞T). Two topping
options are available. With automatic calculation, topping occurs as soon as the plant
height exceeds 𝐡𝐚𝐮𝐭𝐫𝐨𝐠𝐧𝐞T+𝐦𝐚𝐫𝐠𝐞𝐫𝐨𝐠𝐧𝐞T. With the second option, topping is
performed at an imposed date (𝐣𝐮𝐥𝐫𝐨𝐠𝐧𝐞T). In this case, topping is carried out at the
specified height (𝐡𝐚𝐮𝐭𝐫𝐨𝐠𝐧𝐞T) or width (𝐥𝐚𝐫𝐠𝐫𝐨𝐠𝐧𝐞T). The topped LAI (𝗅𝖺𝗂𝗋𝗈𝗀𝗇𝖾)
depends on these parameters and the foliage density.
The topped biomass (𝖻𝗂𝗈𝗋𝗈𝗀𝗇𝖾) is calculated as:
where 𝗌𝗅𝖺 is the specific leaf area and 𝐭𝐢𝐠𝐞𝐟𝐞𝐮𝐢𝐥P is the ratio of stem to leaf area.
Topping is performed only if the topped biomass exceeds a specified threshold
(𝐛𝐢𝐨𝐫𝐨𝐠𝐧𝐞𝐦T). The topped biomass and the corresponding LAI are subtracted
from the biomass and LAI of the plant. The topped biomass returns to the soil and is
recycled in C and N soil pools.
Leaf removal is another technique which directly reduces the leaf area and does not
depend on the canopy geometry like topping does. Two options are possible for leaf
removal. With automatic calculation, a constant proportion (𝐞𝐟𝐟𝐞𝐮𝐢𝐥T) of the new
foliage generated each day (deltai) is removed as soon as the LAI reaches a threshold
value (𝐥𝐚𝐢𝐝𝐞𝐛𝐞𝐟𝐟T). With the second option, leaf removal is performed only once on
day 𝐣𝐮𝐥𝐞𝐟𝐟𝐞𝐮𝐢𝐥T and the specified leaf area 𝐥𝐚𝐢𝐞𝐟𝐟𝐞𝐮𝐢𝐥T is removed. The corresponding
biomass is calculated from the specific leaf area (𝗌𝗅𝖺) and substracted from the living
plant biomass. The location of leaf removal may be either the top or the bottom of the
canopy, which affects the radiation interception and water balance of crops in rows.
265
STICS soil-crop model
Forage crops can be cut according to one of the following three methods:
– With automatic calculation, as soon as the crop reaches the stage defined by the
parameter 𝐬𝐭𝐚𝐝𝐞𝐜𝐨𝐮𝐩𝐞𝐝𝐟T, the forage crop is cut at the height corresponding to
𝐡𝐚𝐮𝐭𝐜𝐨𝐮𝐩𝐞𝐝𝐞𝐟𝐚𝐮𝐭T. The biomass of the cutting is calculated using the crop-specific
conversion parameter 𝐜𝐨𝐞𝐟𝐦𝐬𝐡𝐚𝐮𝐭P (in t ha–1 m–1).
– With imposed dates, a table providing all cutting dates is entered, linked to
either a single parameter (the cutting height 𝐡𝐚𝐮𝐭𝐜𝐨𝐮𝐩𝐞T) or five other parame-
ters: the residual LAI after cutting (𝐥𝐚𝐢𝐫𝐞𝐬𝐢𝐝𝐮𝐞𝐥T), the residual biomass after cutting
(𝐦𝐬𝐫𝐞𝐬𝐢𝐝𝐮𝐞𝐥T), the N fertilisation rate after cutting (𝐚𝐧𝐢𝐭𝐜𝐨𝐮𝐩𝐞T), the minimum
aerial biomass required for cutting (𝐦𝐬𝐜𝐨𝐮𝐩𝐞𝐦𝐢𝐧𝐢T) and the fraction (0-1) of the
cut which is exported (𝐭𝐚𝐮𝐱𝐞𝐱𝐩𝐨𝐫𝐭𝐟𝐚𝐮𝐜𝐡𝐞T). This last parameter allows users to
simulate leys with a partial or total return of the grass biomass. The default value
for this p
arameter is 1 (the whole cut is exported). If the user does not supply these
five parameters, they are calculated based on the cutting height using the parameter
𝐜𝐨𝐞𝐟𝐦𝐬𝐡𝐚𝐮𝐭P and the height/LAI ratio.
– With imposed crop stages, the procedure is similar but uses physiological dates
instead of Julian days, with the cutting dates defined by cumulative development units.
Whatever the method chosen, the amount of biomass harvested equals the difference
between the non-senescent living biomass and the biomass left on the ground. It must
be greater than a minimum fixed by the user. Once the cutting date or stage has been
reached, these conditions define the remaining and harvestable parts. Cutting occurs
if three conditions are fulfilled:
– lai is greater than the minimum harvestable LAI (𝐥𝐚𝐢𝐫𝐞𝐬𝐢𝐝𝐮𝐞𝐥T);
– the aerial biomass (masec) is greater than the minimum harvestable biomass
(𝐦𝐬𝐫𝐞𝐬𝐢𝐝𝐮𝐞𝐥T);
– the harvestable biomass (msrec_fou) is greater than the minimum amount that we
accept to harvest (𝐦𝐬𝐜𝐨𝐮𝐩𝐞𝐦𝐢𝐧𝐢T).
These three criteria (residual LAI, residual biomass and harvestable biomass) are
defined for each cut and allow cuts to be postponed. The third criterion makes it
possible to change the threshold according to the season or the farmer’s decisions,
according to the harvest needs (e.g. small quantities if there is stress on herd feeding).
266
Soil-crop management effects
the decision to harvest is taken based on crop maturity status or depending on other
criteria such as soil water status, crop health or even economic criteria. The harvest
date is calculated according to one of the six following criteria:
– the physiological maturity (end of growth-development period) is reached,
– the maximum fruit water content is reached, as shown by the dehydration dynamics
(𝐡2𝐨𝐠𝐫𝐚𝐢𝐧𝐦𝐚𝐱T) after the stage idebdess,
– the minimum fruit water content is reached, as shown by the hydration dynamics
(𝐡2𝐨𝐠𝐫𝐚𝐢𝐧𝐦𝐢𝐧T) after the stage idebdess,
– the minimum fruit sugar content (𝐬𝐮𝐜𝐫𝐞𝐫𝐞𝐜T) is reached,
– the minimum fruit nitrogen content (𝐂𝐍𝐠𝐫𝐚𝐢𝐧𝐫𝐞𝐜T) is reached,
– the minimum fruit oil content (𝐡𝐮𝐢𝐥𝐞𝐫𝐞𝐜T) is reached.
If the soil is too wet at this date, it is possible to postpone harvest to avoid soil compac-
tion. In this case, a period (𝐧𝐛𝐣𝐦𝐚𝐱𝐚𝐩𝐫𝐞𝐬𝐫𝐞𝐜𝐨𝐥𝐭𝐞T) following the calculated harvest
date is defined during which the average soil water content is tested. The average soil
water content in the zone between soil surface and the maximum depth affected by the
harvesting equipment (𝐩𝐫𝐨𝐟𝐡𝐮𝐦𝐫𝐞𝐜𝐨𝐥𝐭𝐞𝐮𝐬𝐞T) is considered damaging if it is greater
than 𝐩𝐫𝐨𝐩𝐡𝐮𝐦𝐭𝐚𝐬𝐬𝐫𝐞𝐜G x 𝐡𝐜𝐜𝐟S. However, harvest cannot occur after 𝐢𝐫𝐞𝐜𝐛𝐮𝐭𝐨𝐢𝐫T
which is the latest date for harvesting. Several reasons lead to fix this deadline: the risk
of crop health problems or yield losses or the need to free up the field for the following
crop or due to economical constraints.
13.2.5 Pruning
This option is activated using the code 𝐜𝐨𝐝𝐞𝐭𝐚𝐢𝐥𝐥𝐞T in the .tec file.
Winter pruning is used for perennial woody crops like grapevine. On the prescribed day
of winter pruning (𝐣𝐮𝐥𝐭𝐚𝐢𝐥𝐥𝐞T), the biomass of stems and leaves remaining on the plant
(mabois) is removed from the living plant and returned to soil so that the following
cycle starts with the reserves only. The C and N concentrations in these organs are
assumed to be 420 g C kg–1 and 5 g N kg–1 respectively, yielding a C/N ratio of 84.
In reality, pruning is also a technique to regulate yield through the number of remaining
buds. However the model does not establish a relationship between pruning and the
number of inflorescences (𝐧𝐛𝐢𝐧𝐟𝐥𝐨) or the number of fruits removed (𝐧𝐛𝐢𝐧𝐟𝐥𝐨𝐞𝐜𝐥T),
which are predicted independently.
267
STICS soil-crop model
𝐜𝐨𝐝𝐞𝐩𝐚𝐢𝐥𝐥𝐚𝐠𝐞T in the tec file. The user must specify two parameters: the albedo of the
plastic cover, which is related to its colour (𝐚𝐥𝐛𝐞𝐝𝐨𝐦𝐮𝐥𝐜𝐡𝐩𝐥𝐚𝐬𝐭𝐢𝐪𝐮𝐞T) and the propor-
tion of soil cover (𝐜𝐨𝐮𝐯𝐞𝐫𝐦𝐮𝐥𝐜𝐡𝐩𝐥𝐚𝐬𝐭𝐢𝐪𝐮𝐞T). Another option (𝐜𝐨𝐝𝐞𝐚𝐜𝐭𝐢𝐦𝐮𝐥𝐜𝐡G) can
also be activated. It re-inforces the simulation of the “natural” mulch effect, i.e. drying
out at the soil surface.
Figure 13.1 simulates a winter wheat crop growing in the north of France, with a cover
crop used as mulch residue to provide an example showing an order of magnitude
related to the above mentioned mulch effects, which are described in further detail
later (§ 11.4.1 and § 12.2.3). The parameter 𝐚𝐥𝐛𝐞𝐝𝐨𝐦𝐮𝐥𝐜𝐡𝐩𝐥𝐚𝐬𝐭𝐢𝐪𝐮𝐞T is 0.49 (Hidayat
et al., 2019) and the fraction of soil covered (𝐜𝐨𝐮𝐯𝐞𝐫𝐦𝐮𝐥𝐜𝐡𝐩𝐥𝐚𝐬𝐭𝐢𝐪𝐮𝐞T) is 0.50. The
increase in N mineralisation is simply due to the increase in soil temperature.
Figure 13.1. Simulated effects of various mulches on the crop yield, crop biomass and net N
mineralisation
The simulated effects of the same mulches on the various processes involved in the
water and their consequences on the water fluxes are presented in Figure 11.9.
268
Soil-crop management effects
269
STICS soil-crop model
Table 13.1. Values of parameters linked to the bulk density of the ploughed soil layer in the
Paris Basin, established from the data taken from (Mumen, 2006).
270
Soil-crop management effects
The modification in bulk density affects infiltrability, water and nitrogen profiles,
following the rules previously defined. So ploughing tends to increase soil evaporation
by increasing the roughness of soil surface, but the water balance generally remains
positive due to the increase in water storage as a consequence of greater infiltrability.
13.4.1 Irrigation
Depending on the irrigation system used, water may be applied above the foliage
(𝐜𝐨𝐝𝐥𝐨𝐜𝐢𝐫𝐫𝐢𝐠T = 1), below the foliage (𝐜𝐨𝐝𝐥𝐨𝐜𝐢𝐫𝐫𝐢𝐠T = 2) or in the soil (𝐜𝐨𝐝𝐥𝐨𝐜𝐢𝐫𝐫𝐢𝐠T = 3)
271
STICS soil-crop model
at a given depth (𝐥𝐨𝐜𝐢𝐫𝐫𝐢𝐠T) intended to mimic drip irrigation. For irrigation below
the foliage, water supply is not subject to the mechanism of rainfall interception by
the foliage. For underground irrigation, water supply is also withdrawn from the soil
evaporation calculation. The parameter 𝐞𝐟𝐟𝐢𝐫𝐫T stands for irrigation efficiency (0-1),
which accounts for water losses during irrigation. It is applied as a multiplier to each
irrigation amount.
The amounts of water applied can be entered from an irrigation calendar or calculated
automatically by the model if the code 𝐜𝐨𝐝𝐞𝐜𝐚𝐥𝐢𝐫𝐫𝐢𝐠T is activated.
In automatic mode, a specified irrigation rate (𝐢𝐫𝐫𝐥𝐞𝐯G) of about 20 mm is applied at
the sowing date if rainfall on that day is smaller than this value, in order to ensure good
germination. The model then calculates the water inputs needed over time to satisfy a
fraction (𝐫𝐚𝐭𝐢𝐨𝐥T) of crop water requirements: the model triggers irrigation when the
stomatal stress index (swfac) becomes smaller than 𝐫𝐚𝐭𝐢𝐨𝐥T. The amount of irrigation
applied (airg) allows to raise the soil water content between soil surface and rooting
front up to field capacity. It cannot exceed a maximum daily irrigation rate dose autho
rised by the irrigation system (𝐝𝐨𝐬𝐢𝐦𝐱T) and is nil if it is smaller than 𝐝𝐨𝐬𝐞𝐢𝐫𝐫𝐢𝐠𝐦𝐢𝐧T:
The irrigation schedule calculated by the model can be restricted to a specific time
period. If the code 𝐜𝐨𝐝𝐞𝐝𝐚𝐭𝐞_𝐢𝐫𝐫𝐢𝐠𝐚𝐮𝐭𝐨T is activated, irrigation is initiated on day
𝐝𝐚𝐭𝐞𝐝𝐞𝐛_𝐢𝐫𝐫𝐢𝐠𝐚𝐮𝐭𝐨T and ends on day 𝐝𝐚𝐭𝐞𝐟𝐢𝐧_𝐢𝐫𝐫𝐢𝐠𝐚𝐮𝐭𝐨T.
272
Soil-crop management effects
273
STICS soil-crop model
Ammonium
Ammonium
Ammonium
Anhydrous
phosphate
efficiency1
ammonia
Calcium
solution
nitrate
nitrate
sulfate
Fixed
UAN
Type
Urea
engammt 0.50 0.75 1.00 1.00 1.00 1.00 0.0 0.50
denengt 0.11 0.13 0.10 0.10 0.10 0.10 0.2 0.05
volengt 0.12 0.30 0.35 0.35 0.25 0.25 0.0 0.05
orgengt 30.00 33.80 37.70 37.70 37.70 37.70 25.0 0.20
1 With this option the denengt volengt and orgengtvalues represent the proportion of fertiliser which is
denitrified, volatilised and immobilised in soil, respectively.
274
Soil-crop management effects
where the parameter 𝐕𝐚𝐛𝐬2G corresponds to the crop uptake rate (kg N ha–1 day–1) at
which losses represent 50% of their maximum.
The daily N loss through NH3 volatilization (Nvoleng) is calculated in a similar way,
but also depends on soil pH: it increases linearly when the soil pH increases from
𝐩𝐇𝐦𝐢𝐧𝐯𝐨𝐥G to 𝐩𝐇𝐦𝐚𝐱𝐯𝐨𝐥G:
The Figure 13.2 is an example of N losses predicted by the model versus N uptake rate:
Figure 13.2. Predicted fertiliser-N losses versus crop N uptake rate at the time of fertiliser
application. The example applies to UAN (urea and ammonium nitrate) fertiliser added at the
rate of 100 kg N ha–1 in a soil with a pH of 7.5.
The N fertiliser losses through immobilisation and volatilisation are always calcu-
lated as indicated above. However the N losses through denitrification (coming from
soil and fertiliser) can be calculated more mechanistically by activating the option
𝐜𝐨𝐝𝐞𝐝𝐞𝐧𝐢𝐭S (§ 12.4).
275
STICS soil-crop model
Figure 13.3. Relationship between fertiliser-N immobilised in soil and the amount of fertiliser,
for three types of 15N-labelled mineral fertilisers. (Data from: [1] Recous and Machet, 1999; [2]
Limaux et al., 1999; [3] Bronson et al., 1991; [4] Powlson et al., 1992; [5] Recous et al., 1992).
276
Soil-crop management effects
where QNvoleng, QNdenit and QNorgeng represent the cumulative losses by deni-
trification, volatilisation and immobilisation.
277
STICS soil-crop model
278
Soil-crop management effects
279
STICS soil-crop model
Plant residues are assumed to remain at the soil surface until being buried by the next
soil tillage, except for dead roots which remain at depth in the layer where they were
produced.
For chained simulations, the model simulates the characteristics of the crop residues
being returned to the soil and takes them into account automatically in the following
simulation (§ 12.1).
280
Chapter 14
Ways of STICS use
Nicolas Beaudoin, Julie Constantin, Anne-Isabelle Graux,
Gatien Falconnier, François Affholder,
Françoise Ruget and Laurent Ruiz
This chapter will discuss the purposes of STICS uses along with scale variability and
coupling possibilities.
Some 328 peer-reviewed papers involving the use of STICS were published from 1998
to 2020 (not counting numerous operational papers and reports) that showed that
STICS specifications allowed for a large range of uses. The topics of these papers were
analysed according to their abstracts, with a focus on the last decade (Figure 14.1). In
58% of them, STICS use dealt with agronomy and crop and environmental sciences,
while statistics and modelling sciences, bioclimatology and soil science also repre-
sented a substantial share (32%). STICS applications in microbiology, ecology, animal
science and rural economy were less frequent.
Since it was first developed in 1996, STICS has evolved to deal with an larger range of
issues requiring new skills (Beaudoin et al., 2021). STICS has been used in various deci-
sion-support systems and participatory approaches in several countries (e.g. France,
Canada, Brazil, India) to: assess regional production potential, quantify crop water
and nitrogen needs, identify opportunities to reduce nitrate leaching through inno-
vative agricultural practices, support precision agriculture (including by assimilation
of remote sensing data), forecast forage production, and adapt to climate change. In
recent years, the focus has shifted to evaluating agronomic and environmental perfor-
mances of agroecological cropping systems, intercrops, perennial crops for bioenergy.
It has also focused more on greenhouse gas (GHG) emissions and long-term soil
carbon sequestration. STICS is now one of the five most cited soil-crop models in
academic literature (Keating and Thorburn, 2018). A majority of publications (58%)
over the last decade deal with case studies outside France.
This chapter explains the many possible uses of STICS, while the chapters on ‘tools’
and ‘model extension’ provide more practical details. This chapter starts by defining
the ‘unit of simulation (USM)’, a core concept in STICS, and then shows possible
STICS applications for simulations on large temporal and spatial scales and on objects
beyond soil-crop systems through model coupling. A typology of STICS uses for
diagnosis, scenario exploration, support for decision-making and testing of scientific
281
STICS soil-crop model
hypotheses is then described, with examples from case studies. The chapter concludes
with options for model improvements, including concepts and methods to integrate
issues related to agroecology. Some paragraphs of this chapter come from the trans-
lation of the book chapter “Modélisation du fonctionnement des agro-écosystèmes:
l’épopée STICS” (Beaudoin et al., 2019), with the permission of QUAE Editions.
Figure 14.1. Discipline classification of papers involving STICS from 2010 to 2020 (n=231).
282
Ways of STICS use
Considering spatial scale, STICS is usually used to simulate plots. However, it can be
used to simulate a few square meters and up to several square kilometers, depending
on the realism of the assumption relative to the invariance of the parameters over the
area considered.
Figure 14.2. Different types of unit of simulation (USM) in STICS. Bare soil periods are in
brown and crop growing periods in green. Hatches mean that simulating a bare soil period is
optional, depending on the objectives of the simulation. DOY: day of the year.
14.1.3 Intercrops
Intercropping consists of growing several crops (annual or perennial) simultaneously,
each crop developing and growing at its own rate because of resource partitioning
(§ 2.2.3). USMs are created to simulate intercropping using one unique soil type and
283
STICS soil-crop model
weather data, and agro-physiological and management properties specific to the two
intercropped species. Although various spatial arrangements exist (strip intercrops,
alley crops, mixed intercrops or even windbreaks), STICS is only able to simulate row
intercropping.
The two management files of the USM and separated initialisation of the two crops
allow users to simulate i) different crop statuses at the start of the simulation and
ii) different sowing and harvest dates (Figure 14.2). N fertilisation, irrigation and soil
tillage apply to both crops.
Figure 14.3. Temporal scale investigated in papers involving STICS from 2010 to 2020.
284
Ways of STICS use
STICS can be used over the long term either i) with regular re-initialisation (reset) using
in situ observations or ii) with continuous simulations, where the initial water and N
content for only the first USM are provided. In the first case, reset runs allow a close fit of
the model to the observations. This close fit allows the model to simulate accurately vari-
ables that are difficult to measure, such as N and C mineralisation, nitrate leaching, water
drainage and N2O emissions. In the second case, the model can simulate the impacts
of alternative crop management, cropping systems and future plausible climates. The
complementarity of these two possible uses has been well illustrated (Beaudoin et al.,
2008; Constantin et al., 2012). A lack of observations on initial soil status for each USM
shows the advantages of the continuous option over the reset option in long-term
simulations (Kollas et al., 2015); their respective interests are detailed in 14.2.2.
The comparison of soil carbon and nitrogen turnover simulations with a range of
soil-crop models (CERES, NCSOIL, SUNDIAL and STICS) highlighted a trade-off
between short-term (day to year) prediction of N mineralisation and long-term (year
to decade) SOM dynamics in arable cropping systems (Gabrielle et al., 2002). STICS
correctly simulated the SOM mineralisation rate when the amount of incorporated
residues was known. Similarly, long-term predictions were sensitive to root biomass
parameters (Yin et al., 2020). This topic will be further discussed in § 14.2.4.
Figure 14.4. Example of three rotations defined to simulate the ORE-ACBB experiment (long
term experimental observatory system for environmental research of biologeochimical cycle
and biodiversity in agrosystems) over the period 2005-2017 (https://www.soere-acbb.com).
The user needs to define the schedule of each of the USMs involved in the rotation
(Figure 14.4). The simulated period can be divided into USMs in several ways; the
same rotation and its repetition over time can be defined differently depending on
285
STICS soil-crop model
the user. However, each USM must start the day after the previous USM ended. In the
case of a sown annual crop, the USM can start a few days before sowing and end a few
days after harvest. The user can also choose to define independent USMs for bare soil
periods only, as discussed previously.
When chaining USMs, STICS automatically considers the final simulated values of SOM
and recycled crop residues from the previous USM as initial conditions for the following
USM (§ 2.2.2). A USM cannot exceed two years. Thus, for grassland, for example, the
user must create several USMs. The first USM could start a few days before sowing of
the grassland or January 1st in the case of established grasslands. The USM can end a
few days after the final autumn or winter herbage removal of the first year. The following
USMs may start the following day and end a few days after the final autumn or winter
herbage removal of the second year, and so on, for the rest of the simulation period.
286
Ways of STICS use
USM are missing, the model can be run for some years preceding the first year of
interest to calculate reliable simulated initial values, such as initialization of multi-year
STICS prediction in organic alfalfa systems (Strullu et al., 2020).
Table 14.1. Summary of key differences between reset and continuous options in
simulations.
Option for long-term simulations
("climatic series" in the param_gen files
or rotation with different USMs)
Reset Continuous
Link between USMs Independent, no link Dependent on the previous one
Values from Last values simulated
Crop status
the "*_ini.xml" file in the previous USM
Soil water Values from Last values simulated
and N mineral content the "*_ini.xml" file in the previous USM
Soil Organic N Value from 'norg' Last 'Nhumt' simulated
initialization in the soil file in the previous USM
Values from the soil file except
Other soil parameters Values from the soil file
organic N (see above)
Last values simulated in the previous USM,
Crop residues Values provided
according to ressuite parameter
at the start of the USM in the "*_tec.xml" file
in the "*_tec.xml" file
In the following sections, we will illustrate how these two options can influence model
outputs.
287
STICS soil-crop model
Figure 14.5. Carbon reserve (resperenne) dynamics for a temperate grass (Festuca arundinacea),
with reset and continuous simulations from 1994 to 2003 (Ruget and Brisson, 2007).
Table 14.2. Comparison of STICS predictions of annual residue biomass, N content and
mineralisation for reset and continuous simulations in the Bruyères catchment from
1991 to 1999. Measurements were taken from 36 sampling sites with various soil types
(Beaudoin et al., 2008).
Mean prediction soil type
deep loam shallow sandy shallow loamy shallow loamy
loam on clay on marl sand on sand.
limestone and rock
reset simulations
Residue biomass mg ha–1 y–1 8 7.3 5.6 5.7
Residue N content % 0.8 0.9 0.8 1.3
Residue mineralization kg ha–1 y–1 –5 –9 –5 –8
continuous simulations
Residue biomass mg ha–1 y–1 7.7 6.6 5.2 5.6
Residue N content % 0.8 0.8 0.8 1.3
Residue mineralization kg ha–1 y–1 –21 –22 –17 –6
288
Ways of STICS use
Simulated and measured soil nitrate contents for continuous and reset simulations were
compared for two contrasting soils within the same field (Figure 14.6): a) shallow sandy
stony loam overlying limestone and b) deep loamy soil. In the reset simulations, soil
nitrate values were reset twice a year against data and the model correctly mimicked the
remaining part of observed data. Continuous simulations were less in agreement with
measurements, especially in the case of the deep loamy soil. The largest discrepancies
occurred for long fallow periods during dry winter conditions. Hence, using measure-
ments to force the initial soil nitrate USM provides better results at the year scale. This
study also showed that model sensitivity to soil parameters such as 𝐩𝐫𝐨𝐟𝐡𝐮𝐦S (thickness
of mineralisation layer) or 𝐨𝐛𝐬𝐭𝐚𝐫𝐚𝐜S (depth of obstacle to root system) greatly depends
on both the simulation duration and the nature of the variable of interest.
Figure 14.6. Observed and simulated soil nitrate contents during the 1991–1999 period for two
use options (reset vs continuous) and two contrasting soils: a) shallow sandy loam overlying
limestone and b) a deep loamy soil from the same field. The successive crops include sugarbeet,
winter wheat, winter barley, catch crops, spring peas (Beaudoin et al., 2008).
289
STICS soil-crop model
290
Ways of STICS use
with climate change and STICS were scarce at first (Ducharne et al., 2007; Perarnaud
et al., 2005) but that quickly changed: climate change became a major topic and was
covered in 21% of the papers involving STICS published from 2010 to 2020. Crop
models can be used to study the impact of climate change on crops through changes in
temperature, radiation, CO2 concentration and water availability, as well as alternative
and more adapted cropping systems.
To estimate the possible contribution of agricultural systems to climate change miti-
gation, accurate simulations of greenhouse gas emissions (GHG), such as N2O, is
essential (Bessou et al., 2010; Brilli et al., 2017; Ehrhardt et al., 2018; Peyrard et al.,
2017). Carbon storage simulation is also crucial to assess cropping system GHG
balance (Pellerin et al., 2020; Tribouillois et al., 2018b). Accurate and meaningful
simulations imply that the effect of climate change on C inputs and soil C mineralisa-
tion are accurately predicted.
STICS can take into account climate change-related issues in three ways by i) esti-
mating the effects of climate change and increased atmospheric CO2 concentrations
on changes in agricultural productivity and environmental impacts; ii) designing and
evaluating strategies to adapt management practices and cropping systems; and iii)
proposing ways to mitigate GHG emissions from agriculture.
Using STICS in this changing climatic context has meant revisiting some of the
model’s formalisms to address extreme temperature phenomena, prolonged water
stress, increased atmospheric CO2 concentrations, and the way these phenomena
interact. It has also led to coupling STICS with other models, especially when it comes
to considering biotic factors. This section provides some examples.
291
STICS soil-crop model
Figure 14.7. Approach of the ANR CLIMATOR project: multi-site implementation of a set
of models and multi-criteria analysis of climate change impacts (Brisson and Levrault, 2010).
In the ANR CLIMATOR project, STICS highlighted the interaction between crop
phenology shifts due to global warming, and the occurrence of increasing abiotic
factors such as heat and drought. The model behaved differently between spring and
winter crops in terms of their development and the consequences on flowering and
maturity dates (Figure 14.8). The model also showed a shift and increase in total water
requirements, which increased the need for irrigation at a time of the year when aqui-
fers are vulnerable. This in silico experiment showed that sorghum was a relevant
alternative to maize in southern cropping systems facing increasing droughts caused
by climate change.
It also is crucial to properly anticipate the direct and indirect effects of climate
change (via a host) on crop diseases. The impact of climate change on biotic factors
(pests and diseases) often remains poorly understood, particularly in the case of
fungal diseases, which are nowadays responsible for 16% of harvest losses and thus
already threatening the food security (Flood, 2010; Savary et al., 2012). Coupling
STICS to integrate pest and disease pressure is therefore challenging when it comes
to dealing with climate deregulation (see § 14.4.2 for the method). For instance, in
the ANR CLIMATOR project, a dedicated disease module (MILA) was developed
and coupled with STICS to reproduce fungal disease development and damage on
crop development (Caubel et al., 2014, 2012). In a recent study on the evolution of
leaf rust of wheat in France, STICS-MILA showed a much earlier disease onset and
an increase in final disease severity for the 2070–2100 period (Caubel et al., 2017;
Caubel et al., 2012).
292
Ways of STICS use
Figure 14.8. STICS simulations for the CLIMATOR ANR project: global warming impacts
differently winter and spring crops, at a fixed sowing date (Brisson and Levrault, 2010).
293
STICS soil-crop model
temperature and/or water stress and maize (Durand et al., 2018), eCO2 combined with
water stress. These studies showed that models generally captured the impact of eCO2
and its interaction with heat and drought stress quite well, despite considerable varia
bility across models. STICS has participated in these studies and benefited from the
lessons drawn from these activities so users can get information about STICS predic-
tion reliability of the impact of eCO2, drought and heat on wheat and maize. To date,
no systematic appraisal of model accuracy exists when dealing with eCO2, tempera-
ture or water stress for crops other than wheat and maize. Initiatives are ongoing for
crops including rice, soybean, potato and canola, and STICS is participating in some of
these initiatives, but great caution is required when analysing the outcomes of a climate
change impact study using STICS for crops other than wheat or maize.
Existing ensemble studies have mainly focused on temperate, high-input environ-
ments. Recently, the AgMIP ‘low-input smallholder systems’ study has given users the
ability to assess the ability of a crop model to accurately predict maize yield (and other
intermediary variables) in low-input environments in the tropics. STICS was ranked
among the more accurate and consistent models (Figure 14.9). The study in question
explored the potential impact of climate change (i.e. changes in CO2, temperature and
rainfall), but the simulated impacts of eCO2, heat stress and water stress were not
compared against experimental data.
Figure 14.9. Sum of ranks (based on relative root mean squared error for maize grain yield,
total aboveground plant biomass at maturity, maximum leaf area index, aboveground plant
nitrogen at maturity, harvest index and in-season soil water content) for 20 fully calibrated
models on ten experimental plots across sub-Saharan Africa (Falconnier et al., 2020). The
vertical dotted line indicates the median sum of ranks. ST=STICS model. The most consistent
models have a low sum of rank (i.e. they are among the best models, for all simulated variables).
294
Ways of STICS use
Climate change can also impact soil processes (e.g. the dynamics of soil organic matter
mineralisation). Despite a review of the strengths and weaknesses of different crop
models to simulate C and N fluxes (Brilli et al., 2017), we are unaware of any study
to date that has tested the accuracy of a simulated impact of eCO2, heat stress and
drought stress on soil processes against experimental data. However, the hypothesis of
the effect of a probable microbial biomass adaptation to climate change was tested in
tropical conditions (Sierra et al., 2010; § 14.5.4).
295
STICS soil-crop model
The use of STICS at scales larger than a homogenous field requires expertise and infor-
mation about soil and agricultural practices. A useful example is the case study of the
Seine River basin (PIREN-Seine) (§ 14.3.2). In this study, pedotransfer functions were
developed to infer STICS inputs (soil parameters such as field capacity and wilting
point) from a soil map that included qualitative soil description (Le Bas, 2016; Lefe-
bvre, 2010). Using data from remote sensing or digital soil mapping can also be useful
to access, through model inversion, to: i) soil properties, since the climatic condi-
tions allowed to exhibit these properties, e.g. water storage versus drought (Varella
et al., 2010a); ii) technical parameters that are not accurately available everywhere
(Courault et al., 2010).
Finally, STICS simulation accuracy is mainly evaluated at plot scale and there are
limited possibilities to evaluate this accuracy at larger scales such as the region,
watershed or country (Beaudoin et al., 2018; Beblik and Kersebaum, 2001; Loague
and Corwin, 1996). In this case, analysis of model output sensitivity to varying spatial
resolution and accuracy of inputs can guide the parameterisation work and indicate
the required improvement in input accuracy depending on the focus of the study.
296
Ways of STICS use
We believe that these outcomes can help users make relevant choice to represent
climate, soil and cropping systems in accordance with the objectives of the large-scale
assessment.
297
298
STICS soil-crop model
Figure 14.10. The workflow developed to represent current cropping systems in France. PCU: pedoclimatic units, LPIS: Land Parcel Identification
System, TG: temporary grasslands. Only the three main irrigated crops in France (maize, wheat and sunflower) were considered to be irrigated in the
simulations (from Launay et al., 2021).
Ways of STICS use
most common crop rotations were chosen, including only crops that were calibrated
in STICS. For maize, we adapted the crop cycle duration of cultivars to each PCU
growing-degree days available between sowing and harvest. For the other crops, we
chose the most common cultivar in France available in STICS.
We used the Agricultural Practices Survey conducted in 2006 and 2011 by the French
Ministry of Agriculture, Agrifood and Forestry for around 14,000 fields throughout
France to determine crop management. Since the data was available and considered
representative at the administrative region level (NUTS II; Eurostat, 2018), we used
the median of observations per arable crop and administrative region for sowing
and harvest dates, mineral or organic fertiliser dose, dates of mineral N fertilisation,
percentage of the crop with organic N fertilisation, type and frequency of organic N
fertilisation, and frequency of tillage.
In accordance with the French application of the European Union Nitrates Direc-
tive (Council Directive 91/676/EEC), cover crops were included in rotations in PCUs
located in Nitrate Vulnerable Zones in 2012. Nitrate Vulnerable Zones are “areas
of land which drain into polluted waters or waters at risk of pollution and which
contribute to nitrate pollution”. Cover crops were sown before spring crops, except
when the previous crop was grain maize or sugar beet, after which the soil was left
bare. Cover crop species (white mustard or Italian ryegrass) and dates of sowing and
destruction by soil tillage were defined per region based on the Agricultural Prac-
tices Survey. Results were expressed at a spatial scale larger than the field by averaging
results, weighted by their areas. Figure 14.11 shows one example of the results obtained
from the 4per1000 initiative.
The first two years were used only to initialise soil mineral N and water contents
according to agro-pedoclimatic conditions. Yearly outputs were then analysed for
1984–2013 and averaged over the 30 years at the PCU level.
– Case study of type C:
The Interdisciplinary Research Programme on water and the environment in the Seine
basin (PIREN-Seine) started in 1989 to protect and manage the water resources of the
Seine river and Normandy basins. Multiscale evaluations of the impact of either actual
or improved cropping systems on yield and water resource quality aimed to prevent
diffuse pollution from nitrate and pesticides affecting drinking water, which mainly
comes from deep aquifers. A modelling chain was created to quantify the impacts
of scenarios of farming practices (Ledoux et al., 2007; Tavakoly et al., 2019) on crop
yield and water resource quality. STICS was calibrated and tested independently
before being integrated into the modelling chain. The STICS inputs were provided by
cross-referencing several datasets that integrated all the spatiotemporal variability at
a given resolution (Figure 14.12).
The Figure 14.12 indicates the following:
– Upper line, information layers (from left to right):
– SAFRAN grid providing local meteorological data (Quintana Segui et al., 2009);
– French soil geographical database at 10–6 scale (INRA, 1998) containing the soil
mapping units (SMUs); each SMU can contain several soil type units (STUs);
– Agricultural database (Mignolet et al., 2007; Mignolet et al., 2004; Puech et al.,
2015; Schott et al., 2010) containing both the 95 agricultural modelling units (AMUs)
299
STICS soil-crop model
Figure 14.11. Maps of a) annual soil organic carbon (SOC) storage in the baseline scenario and
b) to f ) additional SOC storage in mitigation scenarios (kg C ha–1 yr–1) in the topsoil (0–0.3 m
deep) in cropping systems simulated over 30 years with STICS. Additional SOC storage
potential relative to the baseline scenario when implementing mitigation scenarios: b) cover
crop insertion, c) insertion of temporary grasslands instead of silage maize, d) extension of
temporary grassland duration, e) improved recycling of organic resources and f ) consistent
combination of these mitigation scenarios when possible at the cropping system scale (from
Launay et al., 2021).
300
Figure 14.12. Method of distribution of the STICS agronomic model over the Seine-Normandie basin (from Beaudoin et al., 2018).
301
Ways of STICS use
STICS soil-crop model
and the seven agricultural districts, which originated from expert enquiries and crop
succession modelling using the Markow chain.
– Lower line, distribution process of the model run in time and space.
Finally, a quality assurance protocol (Refsgaard et al., 2005) was used to qualify the
reliability of STICS outputs over the Seine-Normandie basin (Beaudoin et al., 2018).
The modelling chain predictions were tested and submitted for expert appraisal before
being discussed with stakeholders (Figure 14.13). In this project, stakeholders needed
to explore the consequences of innovative scenarios that would ensure food security
while reducing the transfer of nitrates and pesticides to surface and deep water, and
of gaseous pollutants to the atmosphere. These scenarios were simulated at the scale
of local catchments (Seine tributary basin and Seine River Basin). They showed, for
example: no change in fertilisation, a 20% reduction in N fertilisation, adoption of
best management practices, and the generalisation of organic farming. Scenarios were
developed with involvement from farmers’ groups, agricultural extension services
and water resource stakeholders. Using STICS required a rigorous calibration and
evaluation of the model ability to reproduce the impact of various N management
practices on N balance. Moreover, the ability of STICS to simulate GHG emissions and
N balance of conventional or agroecosystems was improved both for experimental
conditions (Autret et al., 2020; Peyrard et al., 2017) and “on farm” research (Benoit
et al., 2016; Rakotovololona, 2018). Scenarios highlighted strong differences in terms
of N balance and N gaseous losses between systems.
302
Ways of STICS use
Figure 14.13. Maps of spatially mean outputs of the chain modelling (Beaudoin et al., 2018): a)
weighted mean nitrate concentration in drained water simulated in agricultural areas; b) mean
difference (simulated value – observed value) in nitrate concentration at monitored outlet of
the aquifers represented by different colors.
303
STICS soil-crop model
in the following paragraph. The use of the lone GIS seems limited to biological or
biogeochemical studies. The use of the statistical response function appeared most
frequently in socio-economic studies. However, these studies can also mobilise internal
links. This short overview indicates that the coupling specifications must be driven by
considerations of scale, operational skills and the expectations of the project partner
rather than by standard rules. However, the scientific basis for linking models across
disciplines and scales is still weak and requires specific attention in future research
(Ewert et al., 2009).
Table 14.4. Typology of the STICS coupling according to the publications involving STICS
for the 2016-2020 period.
Application domain
Geographical
Agronomical
biochemical
economical
economical
and socio-
Biological
Nature of the link Code link Total
and/or
Socio-
(GIS)
304
Ways of STICS use
Figure 14.14. Diagram of the coupling of STICS and hydrologic (MODCOU) and economic
(AROPAj) models in the PIREN-Seine modelling platform (Beaudoin et al., 2019).
305
STICS soil-crop model
to calculate vegetation indices and ultimately LAI using statistical relationships. The
computed LAI can be forced into the model, or used to re-optimise unknown model
inputs (e.g. field capacity). Such an approach has been implemented to estimate forage
and annual crop production (Courault et al., 2010; Jégo et al., 2012a) and optimise soil
parameters (Ferrant et al., 2014; Varella et al., 2010a).
– Dynamical downstream external link
This type of link was more frequent in published studies involving STICS. The
exchange between models concerns their respective outputs and retroaction cannot
be implemented. A good example is the STICS-MACRO chain, which aims to simu-
late pesticide transfer within the soil profile (Figure 14.15). It combines the strengths
of each model: STICS can account for the effects of varying agricultural practices
(fertilisation, waste, mulch or crop residue management) on soil organic C/N changes;
MACRO integrates the roles of microporosity and macroporosity on the retention and
transfer of pesticide molecules. The coupling implements a sequential transfer of data
on the soil parameters of both models. STICS outputs (potential evapotranspiration,
LAI, height, root depth) are then transferred to MACRO; some MACRO parameters
are re-estimated. A MACRO run will eventually predict pesticide behaviour.
Another example of the downstream external link is the chaining of STICS with ecolo
gical and hydrological models, within a platform including a GIS. Here, the pixels
cannot spatially interact and the outputs of the agronomical model become inputs for
the hydrological one. For example, in the PIREN-Seine project, the platform simulates
crop production and soil C and N balances in cropping systems as well as their long-
term impacts on water resources (§ 14.3.2). Another example is the ORCHIDEE land
surface model, which predicts water and C flows at the European scale (Wu et al., 2016).
Figure 14.15. Sequential use of the STICS crop model and the MACRO pesticide fate model to
simulate pesticide leaching in cropping systems (from Lammoglia et al., 2017).
– Internal link
This type of link allows the retroactions between models by coupling the daily loops
of the models. Examples of stand-alone STICS coupling cover a large range of topics:
– MILA-STICS simulates in 1D, the development and impact of pathogenic fungi on
LAI and crop growth. The crop-disease model MILA-STICS was able to simulate a
range of airborne fungal diseases (Figure 14.16).
The climate change is an emblematic stake for this couple (§ 14.2.5.3).
306
Ways of STICS use
Figure 14.16. Calculation of the susceptible, exposed, infectious and removed surface areas in
the conceptual design of and feedback to a process-based crop model (MILA-STICS coupling).
The available surface is the green leaf surface minus the exposed surface. The infectious surface
is deducted from the green surface (from Caubel et al., 2012).
– PE-STICS simulates, in 1D, the retention and transfer of pesticides and nitrate
within the daily loop (Queyrel et al., 2016).
– FLORSYS predicts crop - weed interactions in 3D; seed stock, seed germination and
seed stock growth are simulated at plant level, while STICS predicts soil temperature,
moisture, nitrate availability and organic matter dynamic in 1D (Moreau et al., 2021).
– Hi-sAFe simulates agroforestry system behaviour in 3D by coupling STICS with a
tree model (sAFe-Tree) on a daily time step (Dupraz et al., 2019).
Encapsulating STICS in the RECORD platform results in a turnkey simulator for crop-
ping system functioning and management at varying spatiotemporal scales (Bergez
et al., 2013). Encapsulation relies on the modularity of the STICS code within its
daily loop (§ 16). It addresses the following issues: (1) water resource management at
farm and regional scales with MouSTICS (Bergez et al., 2012); (2) ecosystem service
assessment with the EFESE project (Therond et al., 2017); (3) livestock farm studies
with MELODIE (Moreau et al., 2013); and (4) water and N cycles within the agro-
hydrosphere at landscape scale, with the TNT model, which allows users to deal with
interactions between pixels (Beaujouan et al., 2002; Ferrant et al., 2014).
– Combining multiple links
The case study of the Adaptation of Irrigated Agriculture to Climate Change (AICHA)
and Accompanying The adaption of irrigated agriculture to climate CHAnge (ATCHA)
projects, in India, is an emblematic example where multiple links were combined. The
‘groundwater revolution’ started three decades ago and allowed millions of small-
holder farmers to access irrigation with individually owned bore wells. It increased
crop production but resulted in a well-identified ‘groundwater crisis’ with tremen-
dous impact on water resources and ecosystems. In such a highly dynamic system, a
multi-link approach was necessary to account for the critical interactions and feed-
back between the variations of groundwater availability and crop prices, and farmer
decisions for crop choice and irrigation scheduling. The STICS model is a component
307
STICS soil-crop model
of the integrated biophysical and economic model NAMASTE (Robert et al., 2018)
created to design and assess scenarios of farming system adaptation, with a partici-
patory approach involving farmers and policymakers, in a network of experimental
watersheds in southern India (Figure 14.17).
Using this soil-crop model to simulate the impact of management scenarios in the
Indian context is a scientific and technical challenge due to the high diversity of crop
species found in Indian farming systems. In addition to several intensively farmed
cash crops (e.g. maize, sunflower) for which default calibration is available in STICS,
there are many other crops (e.g. millet, pulses, spices such as turmeric) that need to
be calibrated by drawing on a reliable database (see the chapter on ‘model extension’).
Another challenge consists in checking the reliability of the water-C-N balance predic-
tions in the tropical context of South India, characterised by the monsoon regime, and
the impact of soil salinisation, which occurs in some places due to potash fertilisa-
tion (Buvaneshwari et al., 2020). Accordingly, the salinity impact on crop growth was
quantified (§ 14.5.1). STICS coupling with a groundwater model showed that ground-
water irrigation plays an important role in the accumulation of saline solute in soil
and groundwater, and threatens the sustainability of farming systems in the long term
(Buvaneshwari et al., 2018).
STICS was also used to estimate soil properties by model inversion using surface soil
moisture and LAI data assimilation (Sreelash et al., 2017, 2012). Finally, STICS appli-
cation at field scale relies on the classic coupling with a GIS and the data assimilation
from remote sensing to secure its predictions.
14.5.1 Diagnosis
Agronomic diagnosis mobilizes scientific knowledge to explain observations. For yield
gap analysis, simulated potential yield (P) or water limited or only N limited yield, are
compared to measured actual yield, namely farmer yield (F). STICS was used for yield
gap analysis in tropical regions, at the scale of France (Brisson et al., 2010) or at global
scale (Kvaki’c et al., 2018). STICS was also used to estimate yield losses due to water and
N limitations in complex contexts, as following. In tropical smallholder farms, STICS
was used to quantitatively determine the respective share of suboptimal stand density,
water stress and nitrogen stress in deviations of actual yields to potential yield across a
region, and helped determine the role of other biotic and abiotic stresses (Affholder et al.,
2003 and Figure 14.18). In temperate organic systems, STICS was used to predict the
potential yield and the water and N stresses in the actual situation; so it was assumed the
308
Figure 14.17. Diagram of the multidisciplinary approach in the ATCHA project (2016).
309
Ways of STICS use
STICS soil-crop model
N availability to the crop was well predicted when the actual crop yield was affected by
both abiotic and biotic stresses (Rakotovololona, 2018). In the aforementioned ATCHA
project, the impact of groundwater salinisation on turmeric yield was q uantified in a
watershed in Karnataka using yield gap analysis (Figure 14.19).
Figure 14.18. Boxplots of potential (P), water-limited (WL) and farmers’ (F) grain yield. “Up.
Rice”: upland rice. Predictions were made using an ad-hoc model for potential yield estimation
derived from STICS and SARRA models (Affholder et al., 2013).
Figure 14.19. Effect of groundwater salinity on turmeric production. X axis = electrical conduc-
tivity (Ece, deciSiemens per metre) of irrigated water; y axis = observed yield to potential yield
simulated by STICS ratio (Kizza et al., 2016).
310
Ways of STICS use
14.5.2 Prognosis
In prognosis, the model predicts the behaviour of a given system. Context and varia-
bles of interest can differ and include: i) crop yield response to sowing date, e.g. durum
wheat, in water scarcity contexts(Debaeke and Aboudrare, 2004), ii) nitrogen leaching
with current practices (Schnebelen et al., 2004), iii) nitrate in drained water according
to varying catch crops. This third situation provides the following emblematic example
of STICS relevance for prognosis.
Figure 14.20. Simulated response of nitrate concentration in drained water to catch crop
management versus emergence and termination dates and species: a) mustard, b) ryegrass,
c) vetch. The line is the EU standard of water drinkability (Justes et al., 2012).
311
STICS soil-crop model
Figure 14.21. Permanent grassland yield indicator, by forage region, as of April 20, 2021.
(Agreste Infos rapides, 2021)
Several other projects have explored the impacts of climate change on grassland
production. Knowing the response of forage production to pedoclimatic variability is
crucial for the feed security of livestock farms and for the dairy and meat industries.
Studying the effect of global climate change relied on an accurate representation of the
changed environmental conditions with regard to primary production and soil water
consumption (Gonzalez-Camacho et al., 2008; Ruget et al., 2012). The model thus
helped farmers adapt their livestock feeding strategies (Moreau et al., 2012), which
illustrates the following section: the model can be used to support decision making.
312
Ways of STICS use
313
STICS soil-crop model
Figure 14.22. Difference in simulated minus observed days of year of beginning of irrigation in
2002 and 2003 (Nesme et al., 2006).
Figure 14.23. Optimal sowing windows calculated to comply with criteria on yield (Yield),
yield variability (Yield var.), drainage (Drain.) and runoff (Run.) for the conventional tillage
(CT), no-tillage (NT), no-tillage with millet residue mulch (NTM) cropping systems, for two
maximum rooting depths (90 and 180 cm) at the Goiânia experimental site. Grey bars indicate
that no sowing dates matched the criteria (from Silva et al., 2019).
314
Ways of STICS use
Maximising yield while minimising yield variability, water run-off and drainage could
only be achieved for a narrow optimal sowing window. The approach developed in this
study can form the basis of a comprehensive decision support system that accounts
for soil characteristics, mulch effects, and the occurrence of dry spells during the rainy
season.
Figure 14.24. Comparison of simulated and observed total and partial land equivalent ratio
(LER) values for dry matter and N accumulation, in pea-barley intercropping. Average values
obtained in intercrops grown in 2002 and 2003 with N or without N fertilization. Error bars
represent the standard error of observed values (from Corre-Hellou et al., 2009).
A second example deals with the controversial effects of climate deregulation on SOM
behaviour in the tropics (Sierra et al., 2010). The impact of climate change on the
315
STICS soil-crop model
Figure 14.25. Impact of thermal adaptation of soil microorganisms and crop system on the
dynamics of organic matter in a tropical soil under a climate change scenario (Sierra et al., 2010).
For perennial crop behaviour, the hypothesis was double: i) the remobilisation of the
perennial pool of N reserves represents the main N supply of the growing canopy,
when senescence of the canopy permitted this pool to refill, and ii) the crop N status
316
Ways of STICS use
is the main driver of canopy growth as opposed to crop ontology; thus, the nitrogen
nutrition index (NNI) allows to mimic its behaviour, with a unique corpus of forma
lisms and parameters from sowing to destruction. This hypothesis was positively tested
with the bioenergy crop Miscanthus G. and then for the perennial leguminous alfalfa
(Strullu et al., 2020, 2014). In unfertilised treatment of Miscanthus G., the perennial
reserve can reach 200 kg N ha–1. Periodic depletion affects the perennial biomass pool,
when aerial biomass is accumulating. The model was calibrated against treatment with
late harvest and tested against the ones with early harvest. The periodic pattern was
satisfactorily simulated by the model during four growing seasons (Figure 14.26).
317
STICS soil-crop model
model specifications are encouraged. The STICS project team has been involved in
training sessions on modelling and the use of STICS, in order to promote the ‘suitable
and reasonable’ use of the model (§ 1 and § 15). Continued discussions about STICS
conceptual bases are vital in order to renew the model.
Evaluating an extension of the scope of STICS requires specific efforts with a rigorous
collection of input data, observations and metadata. Such efforts are made through a
scientific project if the project relates to processes, or as an engineering project (§ 16). A
comprehensive generic test was completed based on a large dataset covering 15 single
crops and a wide range of agro-pedoclimatic conditions in France (Figure 14.27).
STICS was also tested at the crop rotation scale (Beaudoin et al., 2008; Constantin
et al., 2012; Jégo et al., 2012b; Plaza-Bonilla et al., 2015), including a comparison with
other models (Yin et al., 2020).
318
Ways of STICS use
319
STICS soil-crop model
STICS can help to quantify the ecosystem services linked to the water, C and N cycles
in the soil-crop system. It allows users to test the agronomical and environmental
interests of cover crops (Justes et al., 2017; Plaza-Bonilla et al., 2015; Tribouillois
et al., 2015) and to lesser extent, intercrops (Bedoussac et al., 2017; Launay et al.,
2009; Shili-Touzi et al., 2010) and grasslands (Graux et al., 2020), which are pillars
of agro-ecology. STICS has been mostly applied to conventional farming systems but
is also now used to simulate low-input and organic farming systems (Figure 14.28).
This application makes explicit assumptions about the role of biotic factors, and more
specifically, that they are sufficiently controlled to allow STICS to accurately predict
crop biomass and N balance. The first studies where STICS was adapted to arboricul-
ture (Demestihas et al., 2018), agroforestry (Dufour et al., 2013; Dupraz et al., 2019)
and agrivoltaic systems (Dinesh and Pearce, 2016) were recently conducted, namely by
coupling STICS to other models.
Modelling the phosphorus cycle is a new priority for the STICS project team in order
to go beyond the current field of application and address questions linked to, for
instance, the eutrophication of environments. Other investigations are in progress
to i) improve the formalism of crop height kinetics and their consequences on inter-
crop behaviour; ii) account for the roles of organic amendments on soil C-N storage
(Levavasseur et al., 2021) and iii) better represent the functioning perennial crops
(Strullu et al., 2020; Strullu et al., 2015), especially the diversity of grassland types and
management strategies (including animal returns at grazing) and grasslands’ ability to
store C in their soils (Graux et al., 2020).
Figure 14.28. Simulation with the research version of STICS (tagged v10) of the evolution of
organic N stocks (mg N ha–1) in a) the long-term DOK (Thervil, Switzerland) and b) Organic
(Foulum, Denmark) trials, comparing conventional (CONMIN, CONFYM, C4) and agri-biological
systems (NOFERT, BIOORG, O2, O4); from Beaudoin et al. (2019) and Autret et al. (2020).
320
Ways of STICS use
The desired agroecological transition has called into question the ability of research
to build and implement solutions to tackle future climate variations and their abiotic
and biotic consequences. Combined practices limiting both input use and the risks
of additional stresses on crops have emerged as ways to address today’s food security
challenges.
321
Chapter 15
Tools for smart use
of the standard STICS model version
Dominique Ripoche-Wachter, Christine Le Bas
and Nadine Brisson
Reviewed by: Nicolas Beaudoin and Florent Levavasseur
This chapter aims to guide users in their use of the model. Building the STICS model
has always been a twofold scientific and technical challenge, as explained in Beaudoin
et al. (2019) (§ 2.2.1).
The STICS conceptual scheme is built like a set of modular bricks, which means that
many processes are available in the standard version and new ones can be easily added
in the research branches (§ 1.5.2). These processes can be activated or deactivated as
options to reproduce actual farming practices or to describe soil physics, crop devel-
opment and growth processes. The first three sections deal with these options: § 15.1
covers the driving options, § 15.2 the strategy options and § 15.3 the formalisms.
As we mentioned above, users must determine the relevant local parameters, but we
do provide some tips in § 15.4 to help users fill in these values. In the last part of
the chapter, we present a brief overview of the STICS software tools (the JavaStics
interface and Stics R packages) and the STICS forum where users can get additional
support and information.
323
STICS soil-crop model
(°C), global radiation (MJ m–2 d–1) and rainfall (mm d–1). There are four possible ways
of estimating evapotranspiration (tetp(t)), for which additional variables are required.
The least demanding option is to use the calculation developed by Priestley and Taylor
(1972) (§ 15.1.1.2), followed by the calculation that imposes a pre-calculated value
(§ 15.1.1.1). The other two options require two additional primary weather vari-
ables: wind speed (m s–1) and vapour pressure (mbars). The first one is to use the
Penman formula (§ 15.1.1.3) and the second does not rely on evapotranspiration but
directly calculates water requirements at the plant level through a resistive approach
(§ 15.1.1.4). For the three first options (Priestley-Taylor, pre-calculated value and
Penman formula), users should note the close dependence between the reference
evapotranspiration value and the 𝐤𝐦𝐚𝐱P value (because 𝐤𝐦𝐚𝐱P is experimentally
calculated with a given reference evapotranspiration). A change of option should
theoretically lead to a change in 𝐤𝐦𝐚𝐱P. In a comparative work cited by Smith et al.
(1998), many reference evapotranspiration calculations were compared to lysimeter
measurements from 11 locations. The Penman formula exhibits a 0.60-0.70 mm error
regardless of the environment, while the Priestley-Taylor formula appears to perform
far better in humid environments (0.68 mm) than in arid ones (1.89 mm).
324
Tools for smart use of the standard STICS model version
where 𝖲𝖳_𝖫𝖠𝖨 is the growing degree-days since emergence, 𝖪_𝖫𝖠𝖨 is the maximal LAI
produced (which is higher than the maximal measured LAI because of the effect of
senescence), 𝖳𝖨_𝖫𝖠𝖨 and 𝖳𝖥_𝖫𝖠𝖨 are the growing degree-days for the point of inflexion
of the growth curve and complete senescence, respectively, and 𝖠_𝖫𝖠𝖨 and 𝖡_𝖫𝖠𝖨
describe the shapes of the growth and senescence curves (Figure 15.1).
The first term of Eq. (15.3) stands for the green leaf appearance and the second for leaf
senescence.
The example in Figure 15.2 shows the three LAI curves for a wheat crop. The daily
prescribed LAI data were calculated by fitting the five parameters of Eq. (15.3) to the
325
STICS soil-crop model
LAI measurements. Next, the daily prescribed LAIs were used to drive the model,
improving the simulation of above-ground biomass (masec) (Figure 15.3).
Figure 15.1. Empirical relationship representing the time course of LAI, and its two
components, green leaf appearance and leaf senescence.
Figure 15.2. Example of simulated LAI versus fitted LAI using this method.
326
Tools for smart use of the standard STICS model version
Figure 15.3. Improvement of the above ground biomass values (masec) when the model is run
by prescribed LAI versus the free simulation of LAI.
327
STICS soil-crop model
Figure 15.4. Comparison of stressed and unstressed LAI dynamics and the evolution of water
and nitrogen stress indices over the cropping season (calculated when influencing crop growth).
Figure 15.5. Nitrogen content partitioning with soil depth, with or without the smoothing
option activated.
328
Tools for smart use of the standard STICS model version
1. https://agmip.org/
329
STICS soil-crop model
Figure 15.6. Time scale for a single USM (over one or two calendar years).
The beginning time scale could start or end outside the crop period (Figure 15.7).
Figure 15.7. Three examples of time scale for a single USM (with or without a bare soil period).
To run over two years (two successive crops), a second USM must be built, as shown
in Figure 15.8.
330
Tools for smart use of the standard STICS model version
that the model handles the inversion of dominance if the height of the associated crop
exceeds that of the main crop. The harvest can take place on the same date (maturity
of the earliest) or at two different dates (maturity of both) (Table 15.9). The system is
described briefly in the § 14.1.3 and a use case is given in § 14.5.4.
For this kind of simulated system, the Shuttleworth-Wallace PET option based on the
energy balance approach is mandatory (§ 9.3.3).
331
STICS soil-crop model
332
Tools for smart use of the standard STICS model version
average
codhnappeG calculation of groundwater if drainage
interdrain localisation
333
STICS soil-crop model
Transfers in nitrification
Nitrification codenitrifS See chap. 12.3
soil: water, (yes/no)
nitrate and
heat fluxes
denitrification
Denitrification codedenitS See chap. 12.4
(yes/no)
Table 15.3. Formalisation options that address plant and variety ecophysiology for
development.
Code name:
Chapter Formalisation Option choice Comments
option name
Not really an option
but a backdoor option
codeplante
which imposes call of
functions in the model
air temperature
Development codetemp: driving
driving temperature
temperature
within canopy
codeperenne: annual
Emergence
Development annual or
and starting
perennial
perennial
sensiphot, phobase
codephot: yes and phosat in genotype
photoperiodic parameter
plant
Aboveground no
biomass
yes
development
coderetflo: delay
no
effect of stress
(yes/no) vernalisation jvc in genotype
(herbaceous) parameter
334
Tools for smart use of the standard STICS model version
ments. There are other formalisation options for each module so, the model can take
into account the specific ecophysiology of different crops. These various options are
listed in the following tables and grouped by formalisations.
All parameters in these tables are in the ‘crop name’_plt.xml file.
daily scale
codegdh: time scale
hourly scale
335
STICS soil-crop model
Code name:
Chapter Formalisation Option choice Comments
option name
Aboveground stdordebour in
Development codebfroid: cold dormancy (woody)
biomass genotype parameter
requirements
development
no
Table 15.4. Formalisation options that address plant and variety ecophysiology for shoot growth.
Chapter Formalisation Code name: option name Option choice
ground cover
LAI
Shoot growth
336
Tools for smart use of the standard STICS model version
337
STICS soil-crop model
Table 15.5. Formalisation options that address plant and variety ecophysiology for yield
determinate
growing plant
codeindetermin:
growing dynamics
indeterminate
growing plant
Yield formation
Yield codetremp: thermal stress
on filling (yes/no)
codgelflo (yes/no)
Table 15.6. Formalisation options that address plant and variety ecophysiology
for root growth.
Root growth
Growth in root
coderacine: root density
density true density
338
Tools for smart use of the standard STICS model version
339
STICS soil-crop model
Table 15.7. Formalisation options that address plant and variety ecophysiology
for water balance.
Code name:
Chapter Formalisation Option choice
option name
crop coefficient
Table 15.8. Formalisation options that address plant and variety ecophysiology for nitrogen transform
Code name:
Chapter Formalisation Option choice
option name
dense canopies
Nitrogen uptake codeplisoleN: calculation (initial)
by plant nitrogen requirements isolated plants
Nitrogen
acquisition (new calculation)
by plants
yes
Nitrogen fixation codelegume: leguminous
no
Table 15.9. Formalisation options for crop management and crop environment through
the effect on plants (crop design and cover crop regulation).
Code name:
Chapter Sub chapter Formalisation
option name
340
Tools for smart use of the standard STICS model version
Sub-option Sub-option
Comments code name: Option choice code name:
sub-option name sub option name
mations.
Sub-option code name: Sub-option code name:
Comments Option choice
sub-option name sub option name
Classical dilution
curve
341
STICS soil-crop model
Code name:
Chapter Sub chapter Formalisation
option name
no establishment
(perennial plants)
codrognage: topping
Soil-crop
management Crop management
effects
Yield regulation
codeffeuil: leaf removal
codeclaircie: thinning
Table 15.10. Formalisation options for crop management, in particular for the harvest.
Code name:
Chapter Sub.Chapter Formalisation
option name
codcueille:
method of harvest
Soil-crop
management Crop management Harvest policy
effects codefauche:
cut crop
(particular case
of forage crop)
342
Tools for smart use of the standard STICS model version
codhauteff: bottom
location of leaf of the canopy
removal top of the canopy
yes
automatic
codcaleffeuil: calculation
leaf removal calculation
fixed date
no
yes
no
yes
no
343
STICS soil-crop model
Code name:
Chapter Sub.Chapter Formalisation
option name
codrecolte:
harvest decision
Soil-crop
management Crop management Decision of harvest
effects codedecirecolte: rules of
harvest/moisture status
of the soil harvest delay
(depending on the soil
physical conditions)
coderecolteassoc: decision of
harvest for associated crops
Table 15.11. Formalisation options for crop management and crop environment through soil water s
Code name:
Chapter Sub.Chapter Formalisation Option choice
option name
yes
codecalirrig:
Irrigation amount
automatic calculation
and date
Soil-crop of irrigations
management Irrigation
effects
no
344
Tools for smart use of the standard STICS model version
yes/no
maturity of the
earliest crop Has to be chosen
for the Intercrop
maturity of both simulation
(2 dates)
Only for perennial
crops
supply.
Sub-option code name:
Comments Option choice Comments
sub-option name
codedate_irrigauto: dates
dates to drive automatic
irrigations (start and end) stages
Depending on the water
stress, recharges the soil
amounts calculated
water content to field
capacity
in sum of upvt
codedateappH2O: date of (physiological dates)
irrigation
in Julian days
amounts fixed
Available
in the plant file
345
STICS soil-crop model
Code name:
Chapter Sub.Chapter Formalisation Option choice
option name
Transfers in
Management
soil: water, Modification of codesnow:
of snow and its yes/no
nitrate and rainfall by snow Snow activation
consequences
heat fluxes
Table 15.12. Formalisation options for crop management and crop environment through net nitroge
Code name: option
Chapter Sub.Chapter Formalisation Option choice
name
N inputs
from rain
and irrigations
type of mineral
Nitrogen use fertiliser
efficiency (7 different given types
+ 1 "blank" type
yes
codecalfertiG:
Soil-crop N inputs
Fertilisation automatic
management from mineral
calendar calculation of
effects fertiliser
fertilisation
no
codlocferti: location of
at soil surface
mineral nitrogen inputs
Localisation
below soil
surface
N and C inputs type of organic residues
from organic (9 different given types
residues + rhizomes + roots)
N and C
inputs from types of crop residues
aboveground (whole_crop / straw
and / stubble /prunings /
belowground stubbleveg/ roots)
residues
346
Tools for smart use of the standard STICS model version
Available
in general
parameter file
en supply.
Sub-option code name:
Comments Option choice Comments
sub-option name
Mean
concentration
to define in crop
management file
for irrigations
in general
parameter file
for rainfall
Only available
impose fertiliser efficiency
with the type number 8
calculated fertiliser Only available
efficiency with the 7 first types
automatic N fertilisation
(1 = based on rainfall, minimum rainfall
Use with care, 2 = based on soil water threshold
because very content)
dependent on automatic N fertilisation
the water supply (1 = based on rainfall,
threshold
2 = based on soil water
content)
codedateappN: sum of upvt
date of fertilisation (physiological dates)
codedateappN: Julian days (calendar
date of fertilisation dates)
daily amounts fixed
347
STICS soil-crop model
Table 15.13. Formalisation options for crop management and crop environment through physical so
Code name:
Chapter Sub.Chapter Formalisation Option choice
option name
no plastic mulch
Presence
Plastic mulch codepaillage: mulch
of plastic cover
plastic mulch
Fragmentation
Soil-crop induced codeDST: activation
yes/no
management by tillage of fragmentation
effects operations
Soil structure
Compaction modification
induced by codeDSTtass: activation
sowing and of compaction at yes/no
harvesting sowing/harvest
operations
Calculation of codernetC: calculation of Brunt
net radiation net radiation Brutsaert
15.4 Parameterisation
This chapter presents some examples of plant, soil and crop management parameters.
The methodology for plant parameterisation (formalisation choice and parameter
values) is discussed in detail in the § 16.2.
348
Tools for smart use of the standard STICS model version
Mandatory when
water requirements
= crop coefficient
349
STICS soil-crop model
350
Tools for smart use of the standard STICS model version
Grapevine Grapevine
Pea Miscanthus
(Chardonnay) (Cabernet)
0.0000 6.0 10.00
25.0000 35.0 37.00
2.0000 1.0 1.00
1.0000 1.0 1.00
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
2.0000 2.0 2.00
2.0000 2.0 2.00
2.0000 1.0 2.00
0.1
1.0000 3.0 3.00
0.0 5.00
50.0 25.00
3.0000 1.0 3.00
15.0
2.17
213.00
2.0000 1.0 2.00
2.0000 1.0 2.00
11.7
230.0000 150.0 25.00
418.0000 2,400.0 1,123.00 926.60
900.0000 2,550.0 363.30 304.90
216.0000 0.0 50.00 51.00
700.0000 50.0 96.00 95.00
351
STICS soil-crop model
stdordebour
Development/ sensiphot 0.000
Phasic development* phobase 6.300
phosat 20.000
tgmin 8.000 0.000
codeperenne 1.000 1.000
codegermin 1.000 1.000
stpltger 35.000 50.000
potgermi –1.600 –1.600
nbjgerlim 50.000 50.000
propjgermin 1.000 1.000
codehypo 1.000 1.000
Development/ belong 0.022 0.012
Emergence and starting celong 2.040 3.200
elmax 8.000 8.000
nlevlim1 50.000 10.000
nlevlim2 50.000 50.000
vigueurbat 1.000 1.000
laiplantule
nbfeuilplant
masecplantule
zracplantule
The * in this table indicates genotype parameters.
352
Tools for smart use of the standard STICS model version
Grapevine Grapevine
Pea Miscanthus
(Chardonnay) (Cabernet)
70.0 9,174.30 6,576.70
Grapevine Grapevine
Pea Miscanthus
(Chardonnay) (Cabernet)
120.00000 70.0000 25.000
0.00000 3.0000 0.000
0.00000 6.0000 10.000
30.00000 35.0000 37.000
100.00000 40.0000 100.000
1.00000 1.0000 1.000
2.20000 1.5000 2.200
5.50000 10.0000 5.000
3.00000 2.0000 3.000
0.80000 0.5000 1.000
0.25000 0.3000 0.800
353
STICS soil-crop model
354
Tools for smart use of the standard STICS model version
Grapevine Grapevine
Pea Miscanthus
(Chardonnay) (Cabernet)
0.00000 0.8500 1.000
13.00000 120.0000 13.000
0.30000 –2.0000 0.300
0.00000 0.0100 0.000
1.00000 2.0000 2.000
0.70000
0.0000 0.100
2.00000 1.0000 1.000
12.0000 12.000
0.1000 0.010
0.0100 0.150
0.2500 200.000
355
STICS soil-crop model
356
Tools for smart use of the standard STICS model version
Grapevine Grapevine
Pea Miscanthus
(Chardonnay) (Cabernet)
1.00 1.0 2.00
2.00
1.50
3.16
1.50
11.50
0.85 0.7
1.70
Grapevine Grapevine
Pea Miscanthus
(Chardonnay) (Cabernet)
1.000 1 2.00000
0.031 0
0.000 0
0.000 0
1.000 2 1.00000
0.650
10.00000
1.00000
0.55000
18.00000
15.00000
0.20000
0.75000
1.00000
0.63000
1.00000
1.000 2 1.00000
1.000 2 1.00000
8.000 0.00000
40.000 37.00000
0.350 0 1.96000 2.00000
0.000 0 0.00290
0.000 0 0.00000
357
STICS soil-crop model
sensanox 0 1
stoprac lax sen
sensrsec 0 0.5
contrdamax 0 0.3
rayon 0.02 0.02
codetemprac 2 1
codemortalracine 2 2
coefracoupe
coderacine 2 2
Roots zlabour
zpente
zprlim
draclong 5000 80
debsenrac 1500 1000
lvfront 0.05 0.05
longsperac 11000 18182
codedisrac 2 2
kdisrac
alloperirac
358
Tools for smart use of the standard STICS model version
Grapevine Grapevine
Pea Miscanthus
(Chardonnay) (Cabernet)
0.022 0 0.00400
0.008 0 0.00159 0.00132
15.000 0
447.000 0
3,500.000 1,000
490.000 1,000
0.022 0.00000
0
1.15000 2.12000
1,472.00000 1,280.00000
91.00000 90.00000
15.00000 22.00000
Grapevine Grapevine
Pea Miscanthus
(Chardonnay) (Cabernet)
0 0 0
lax lax lax
0.4 1 0
0.34 0 0.3
0.02 0.025 0.02
2 2 1
2 2 2
2 2 2
30 400 40
1000 2250 2000
0.05 0.075 0.005
3300 1553 1021
2 1 2
0.002
0
359
STICS soil-crop model
codazorac 2 2
minefnra
minazorac
maxazorac
codtrophrac 3 3
repracpermax
repracpermin
krepracperm
repracseumax
Roots repracseumin
krepracseu
code_INN_root 2 2
code_stress_root 2 2
parazorac
code_diff_root 2 2
lvmax
rapdia
RTD
propracfmax
Roots* croirac 0.15 0.12
The * in this table indicates genotype parameters.
360
Tools for smart use of the standard STICS model version
Grapevine Grapevine
Pea Miscanthus
(Chardonnay) (Cabernet)
2 2 2
3 1 3
0.1
0.15
2
2 1 2
2 1 2
30
2 1 2
2.6
2.6
0.3
0.8
0.12 0.06 0
Grapevine Grapevine
Pea Miscanthus
(Chardonnay) (Cabernet)
0.012 0.0025 0.0018
50.000 20.0000 50.0000
0.120 0.0500 0.0058
20,000.000 4,000.0000 25,000.0000
5.080 2.7000 3.3000
0.320 0.4800 0.4400
1.000 0.1000 1.6000
0.300 0.3000 0.3000
–0.500 0.0000 –0.4900
0.300 2.0000 1.0000
0.900 2.0000 1.0000
361
STICS soil-crop model
362
Tools for smart use of the standard STICS model version
Grapevine Grapevine
Pea Miscanthus
(Chardonnay) (Cabernet)
0.320 0.4800 0.4400
2.000 1.0000 2.0000
4.5000
6.470 6.0000
0.040 0.0400
0.500 1.6000
1.000 1.0000 2.0000
2.000 1.0000 1.0000
0.000
2,000.000
0.000
1.000
30.000
6.000
2.700
0.040
0.000
15.000
25.000
35.000
2.000 2.0000 1.0000
30.000
9.500
1.000 1.0000 1.0000
Grapevine Grapevine
Pea Miscanthus
(Chardonnay) (Cabernet)
0.900 0.630 0.75000
0.150 0.100 0.50000
0.600 0.590 0.70000
0.700 0.620 0.70000
0.550 0.400 0.92500
0.005 0.005 0.00016
1.000 1.000 2.00000
1.300 1.000
363
STICS soil-crop model
rsmin
codeintercept 2.000 2.000
Water mouillabil
stemflowmax
kstemflow
psisto 12.000 15.000
Water* psiturg 5.000 4.000
swfacmin 0.100 0.100
tletale –5.000 –25.000
tdebgel 0.000 –4.000
codgellev 1.000 2.000
nbfgellev 2.000
tgellev90 –20.000
codgeljuv 1.000 2.000
Frost
tgeljuv90 –20.000
codgelveg 1.000 2.000
tgelveg90 –10.000
codgelflo 1.000 2.000
tgelflo10 –4.500
tgelflo90 –6.500
tgellev10 –1.000 –4.000
Frost* tgeljuv10 –1.000 –10.000
tgelveg10 –1.000 –4.500
The * in this table indicates genotype parameters.
The beginning of growth for forage crops and grapevine is usually simulated after the
winter rest period (dormancy and budding having being parameterised for grapevine,
see García de Cortázar Atauri (2006)), when perennial reserves are remobilised. The
root system for grapevine is considered to be already completely established (García
de Cortázar Atauri et al., 2009; García de Cortázar Atauri, 2006), whereas the root
system for forage crops is partially established and will continue to grow during the
cropping period. Forage crop parameterisation was done for a grass mixture with
an ecophysiology similar to tall fescue (Festuca arundinacea Schreb.) and cocksfoot
(Dactylis glomerata L.) (Ruget et al., 2006). Sugar beet is considered an annual crop by
the model because of how it is grown and despite the fact that it completes its vegeta-
tive cycle in two years (Launay and Brisson, 2004). For winter wheat only parameters
controlling the photoperiod slowing effect (𝐩𝐡𝐨𝐛𝐚𝐬𝐞P, 𝐩𝐡𝐨𝐬𝐚𝐭P and 𝐬𝐞𝐧𝐬𝐢𝐩𝐡𝐨𝐭P) and
the vernalisation requirement (𝐣𝐯𝐜𝐦𝐢𝐧𝐢P, 𝐣𝐮𝐥𝐯𝐞𝐫𝐧𝐚𝐥P, 𝐭𝐟𝐫𝐨𝐢𝐝P and 𝐚𝐦𝐩𝐟𝐫𝐨𝐢𝐝P) are
activated (Brisson et al., 2002).
364
Tools for smart use of the standard STICS model version
Grapevine Grapevine
Pea Miscanthus
(Chardonnay) (Cabernet)
250.00000
2.000 2.000 2.00000
Shoot growth, and especially leaf production, are unrestricted throughout the
cropping period for forage crops and sugar beet, which are simulated using a
high 𝐬𝐭𝐚𝐦𝐟𝐥𝐚𝐱P parameter value (Launay and Brisson, 2004). Considering the
row-planting arrangement of grapevine and the need to simulate intercropping
with peas, these two crops were parameterised in order to use the radiation transfer
formalisation (§ 9.2.1.2.1) and the associated resistive approach, involving the esti-
mation of 𝐤𝐭𝐫𝐨𝐮P, 𝐟𝐨𝐫𝐦𝐞P, 𝐫𝐚𝐩𝐟𝐨𝐫𝐦𝐞P, 𝐚𝐝𝐟𝐨𝐥P, 𝐝𝐟𝐨𝐥𝐛𝐚𝐬P, 𝐝𝐟𝐨𝐥𝐡𝐚𝐮𝐭P and 𝐫𝐬𝐦𝐢𝐧P
parameters (Table 15.16).
With regard to yield formation, forage crops, spring pea and winter wheat are simulated
as determinate crops, whereas sugar beet and grapevine are simulated as indetermi-
nate (§ 8). The parameterisation for forage crops was not targeted to grain production
but rather to the aboveground biomass prediction, since this is the harvested part of
the crop (Ruget et al., 2006). For sugar beet, we assumed that only one tuber (storage
and harvested root) was set by each plant (𝐧𝐛𝐢𝐧𝐟𝐥𝐨P=1), and the trophic stress effects
365
STICS soil-crop model
on tuber setting were cancelled by means of very low 𝐬𝐩𝐟𝐫𝐦𝐢𝐧P and 𝐬𝐩𝐟𝐫𝐦𝐚𝐱P
parameter values; storage root growth was assumed to be linear over the growth cycle
(𝐛𝐟𝐩𝐟P=1) (Launay and Brisson, 2004).
Finally, root length growth was simulated as trophic linked for spring pea, as demon-
strated in trials comparing sole and intercropped pea (Corre-Hellou et al., 2007).
Symbiotic N uptake formalisation was also parameterised for this legume (Corre-
Hellou et al., 2009). The instantaneous nitrogen nutrition index (INNI) based on the
daily accumulation of N (Eq. (4.24); § 4.4.2) was chosen to avoid the notable inertia of
the NNI dynamics in grapevine and winter wheat (Mary and Guérif, 2005).
Table 15.21. List of soil parameters with recommendations to assign them. PTF stands for
pedotransfer functions or rules. The sensitivity levels are only suggestions and depend on
the purpose of the simulation.
Recommended Links between Sensitivity
Parameter Default value
assigning method parameters level
Mandatory parameters
Reflectance
albedoS Table 15.24 *
measurements
Soil analysis (with
argiS **
decarbonatation)
Non calcareous=1
calcS Soil analysis Limestone=10 **
Chalk=60
Estimation by fitting
to water content profiles
cfesS 5 *
during evaporation
periods
0.01 in temperate
Estimation by fitting soil
concseuilS to the mineral nitrogen epdS(z) *
content profile 0.20 in tropical
soil
0 (if 0 this value
Initial C to N ratio is recalculated
CsurNsolS
of soil humus by the model
= 1./Wh)
366
Tools for smart use of the standard STICS model version
367
STICS soil-crop model
368
Tools for smart use of the standard STICS model version
principles for choosing a PTF. The first is the effort required to measure the predictors: an
efficient PTF is one where the predictors are more easily available (i.e. less expensive, less
time consuming) than the property to predict. The second principle is to choose a PTF
where uncertainty is evaluated (using first-order Taylor analysis or bootstrap method)
and with minimum variance. McBratney et al. (2002) also suggest avoiding extrapolation
and propose a method for determining if data are inside or outside a PTF’s training set
(the published PTF must include statistics on the training data set).
In France, the choice of which PTF to use could be linked to the availability of input
data among the most recent PTFs such as Al Majou et al. (2008), Bruand et al. (2004)
or Román Dobarco et al. (2019). Pedotransfer functions have also been established at
the European level (Tóth et al., 2015).
To enable STICS users to parameterise their soil (at least roughly for test runs), we
have created pedotransfer tables based on well-known literature. The tables mostly
use textural information so parameter values are likely to change with soil structure
and organic matter content (Figure 15.10 and Table 15.22).
Figure 15.10. Soil transfer functions to assess the Q0 parameter as a function of clay and
sand content.
In Table 15.22, the permeability classes proposed by Ritchie (1985) are arbitrarily asso-
ciated with textural classes and correspond to a percentage of the amount of water
stored in the macroporosity that infiltrates from one day to the next. The calculations
show that the effect of layer thickness on 𝐢𝐧𝐟𝐢𝐥S disappears as permeability decreases.
369
STICS soil-crop model
We can also refer users to the work of Lefebvre (2010), which proposes a method to
estimate various soil parameters using the Soil Geographical Data Base for France at
1:1,000,000 (INRA, 2018).
The values of 𝐫𝐮𝐢𝐬𝐨𝐥𝐧𝐮S (Table 15.23), derived from the USDA Runoff Curve
Number method, are rather low because they represent only Hortonian (surface)
runoff, which only depends on obstacles created by plants and on the water velocity
on a slope field. The STICS model takes into account other component of runoff,
i.e. resistance to i nfiltration, as well as the presence of a plant mulch (§ 11.4).
The Table 15.23 provides the values of the parameter 𝐫𝐮𝐢𝐬𝐨𝐥𝐧𝐮S as the proportion of
Hortonian runoff to incoming precipitation minus the 𝐩𝐦𝐢𝐧𝐫𝐮𝐢𝐬G threshold, based
on the USDA Runoff Curve Number approach described in Chapman and Lake (2003).
The 𝐚𝐥𝐛𝐞𝐝𝐨S parameter applies to dry soil, with the effect of water content being simu-
lated (Eq. (9.15)). There are two criteria to assign this parameter, texture or colour; the
latter is read from a Munsell chart (Table 15.24). The relationship between soil colour
and albedo can also be found in Post et al. (2000).
370
Tools for smart use of the standard STICS model version
20 cm 25 cm 30 cm 40 cm 50 cm 60 cm 80 cm 100 cm 120 cm
Table 15.24. Values of the dry soil albedo using either textural classes or colours, based on
Richard and Cellier (1998), Jacquemoud et al. (1992).
Soil type albedo
TEXTURE
Limestone 0.31
Loamy sand 0.25
Clayey loam 0.18 - 0.22
Loam 0.22 - 0.23
Crusted loam 0.28
Clay 0.28
COLOUR
Brown soil 0.27
Red soil 0.29
Black soil 0.17
Grey soil 0.29
Yellow soil 0.35
Pebbles are characterised according to their water retention ability. Tetegan et al.
(2011) assess hydraulic properties for sedimentary pebbles frequently found in French
agricultural fields (Table 15.25). For non sedimentary pebbles, we refer to Gras (1994).
371
STICS soil-crop model
Table 15.25. Average values of volumetric mass and gravimetric water content at field
capacity for various sedimentary pebbles in France from Tetegan et al. (2011) and Gras
(1994).
Pebbles type Volumetric mass Field capacity
in cm g–3 in g/100 g
Gaize 1.44 31
Chalk 1.76 21
Chert 2.07 13
Limestone 2.18 9
Flint 2.22 6
Sandstone or unaltered granite 2.65 0
Altered granite 2.3 10
Default types of pebbles of STICS
Beauce limestone 1 2,2 7
Beauce limestone 2 1,8 16
Lutetian limestone 2,1 11
Lutetian Brackish marl and limestone 2,3 5
Morainic gravels 2,5 3
Unweathered flint, sandstone or granite 2,65 1
Weathered granite 2,3 5
Jurassic limestone 2,2 5
Pebbles from Magneraud 1,5 26
Table 15.26. List of techniques included in the STICS model and the corresponding parameters.
Compulsory parameters
Technique
if the technique is applied
If fragmentation dependent:
(codeDST activated)
jultrav, If compaction dependent
Soil tillage (codeDSTtass activated)
profres,
and residue
proftrav, If sowing date calculated
incorporation1
coderes and compaction dependent
If codedecirecolte activated and compaction
dependent
densite If annual:
If row crop:
Sowing1
variete
If sowing date is calculated:
372
Tools for smart use of the standard STICS model version
prophumtassrec
iplt, profsem
interrang, orientrang
nbjmaxapressemis, nbj_pr_apres_semis,
variete All varietal parameters
humirac_decisemis,eau_mini_decisemis,
nbjseuiltempref
373
STICS soil-crop model
Compulsory parameters
Technique
if the technique is applied
If prescribed by dates:
If prescribed by phasic stages:
Irrigation1, 2 effirr
If calculated:
If in the soil:
If fert-irrigation:
dosen If calendar application:
Fertilisation1
engrais If phasic application:
If in the soil:
If non physiological:
If water content dependent:
If sugar dependent:
Harvesting1 ressuite If nitrogen dependent:
If oil dependent:
If several pickings:
If codecirecolte activacted:
If prescribed cuts:
Forage cutting1
If calendar prescription:
If phasic prescription:
If calculated:
If plastic:
Mulching
If plant (link to the residue properties):
hautmaxtec
Trellising crops
largtec
largrogne If prescribed:
Tactical shape
hautrogne If calculated:
control1
biorognem
If prescribed:
Leaf removal1 codhauteff
If calculated:
juleclair,
Fruit removal
nbfrote
374
Tools for smart use of the standard STICS model version
qmulchruis0, mouillabilmulch,
codetypres
kcouvmlch, albedomulchresidus,
(see Table 15.27)
Qmulchdec
julrogne
margerogne
juleffeuil, laieffeuil
laidebeff, effeuil
375
STICS soil-crop model
Compulsory parameters
Technique
if the technique is applied
Pruning jultaille
Crop destruction juldes
If shelter opened occasionally (3 times
Shelter transplastic
maximum):
1 Several cultivation operations can be planned to bury or mix in residues (of different types) in the soil.
These operations include adding new residues or simply modifying the structural and moisture conditions
of previously added residues.
2 Technique that can either be prescribed or partly calculated using decision rules (Table 15.28).
In this table, all the parameters are available in the crop management file excepted for the typology
parameters, which are available in the general parameters file.
Many of the above mentioned techniques can be calculated using simple decision
rules (Table 15.28).
376
Tools for smart use of the standard STICS model version
surfouvre, julouvre
With regard to the crop, if bare soil (i.e. before sowing), is selected as the initial situ-
ation, the initial values of the crop can be ignored. This may be desirable if there is
uncertainty about the initial crop status, see § 14.2.2
Regarding the water content value, if the status at sowing is unknown, users can start
the simulation at a date where it is known (i.e. at the end of the winter when the soil
moisture is at field capacity).
More generally, initialisation values of soil water content and mineral N content are
those without the pebble compartment (i.e. fine earth only, because the model will
integrate the pebble role) whereas the observed values must integrate the pebble
compartment.
377
STICS soil-crop model
JavaStics interface * * * *
JavaStics in command line *
Record platform *
Batch users *
SticsRfiles * **
SticsOnR *
CroptimizR
CropPlotR **
* available, ** preferable
378
Tools for smart use of the standard STICS model version
– In the plant file (the relevant variety file), choose the forcing option in the phasic
development/cold requirements/dormancy/dormancy calculation
– Enter a new value for the parameter 𝐬𝐭𝐝𝐨𝐫𝐝𝐞𝐛𝐨𝐮𝐫P (the value currently available
was calculated using the BRIN model (García de Cortázar Atauri et al., 2005))
– Open the USM to use with all the information:
• For the Begin date, introduce the date 1
• For the End date, introduce the date 365 (pruning date)
• In the Climate file (first year), select the first year (i.e. 2001)
• In the Climate file (second year), select the last year (i.e. 2015)
• Untick ‘2-year crop’ box.
If everything has been entered correctly, a single balance file will appear with all the
information from the first year to the last.
Please note: The same configuration is used for all the years; It cannot be changed
(adaptation). Option 2 is only possible if users do not take into account dormancy. In
this case, users must verify if the phenology is correctly simulated. This method is not
suited to climate change studies.
Starting
Optimise Multi-simulation Calibration Pre-requisites
with the model
* ** * * No
* ** No
** ** Record or web record
No
Rstudio
** Rstudio
** ** Rstudio
Rstudio
379
STICS soil-crop model
Users can use JavaSticsCmd.exe to run the model from a terminal or command
prompt. The command options can be used only when executing JavaSticsCmd.exe
(users should replace the text below in italics with the name of their own workspace,
USM or file path):
– "–run": Use this command to run STICS. The command should be launched in a
JavaStics workspace. For example: –run workspace USM or run workspace
– "–run-successive": Use this command to run successive USMs. The command
should be launched in a JavaStics workspace. For example: –run-successive workspace
USM1 USM2 etc.
2. https://www6.paca.inrae.fr/stics_eng/Download
380
Tools for smart use of the standard STICS model version
– "–generate-txt": Use this command to generate Fortran binary input files from Java-
Stics input files (all the files necessary for the simulation). For example: –generate-txt
workspace USM
Please note: for use on Linux, replace JavaSticsCmd.exe with java -jar JavaSticsCmd.
exe in the command line. Java version 11 is required for using JavaStics 1.5.0 command
line interface as for the graphical interface.
Sample batch files are available in the JavaStics root directory (for Windows: example_
batch.bat, or for Linux: example_batch.sh).
3. https://github.com/SticsRPacks
4. https://w3.avignon.inra.fr/forge/projects/stics_main_projecu/boards
381
382
STICS soil-crop model
16.1 Introduction
A model’s domain of validity is the range of conditions for which the model is intended
to adequately represent the system under study (see § 2.2).
If the model is to be applied to conditions outside this domain of validity, or if it is not
performing to the expected level of accuracy or precision, its capacity can be extended by:
– adapting the model to new plants or cultivars (or re-parameterising existing plant
or cultivar adaptations),
– re-parameterising, modifying or adding formalisms, or
– coupling STICS with other models.
This chapter provides information and references on how to do this.
The first section focuses on model evaluation and offers guidelines so users can
assess the performance of the model for their particular application. The second and
third sections discuss how to adapt STICS to new crops or cultivars as well handle
model coupling.
Readers interested in the (re-)parameterisation of formalisms can refer to § 16.3,
which provides general guidelines and references on estimating model parameters.
Modifying or introducing new formalisms is a vast subject which depends heavily on
the type of formalism in question, and requires modifying the STICS code. Informa-
tion and advice about how to modify the STICS code will be the subject of a new
chapter in future versions of this book.
383
STICS soil-crop model
16.2.3 Methods
Various complementary model evaluation methods exist:
– Graphical plots are often used to compare observations and simulations, as well as
to qualitatively assess the agreement between model results and observed data points.
When represented in appropriate formats, these plots can yield significant insights
into model performance, pinpoint systematic bias or identify specific situations where
observations and simulations differ widely. Typical graphical plot examples are scatter
plots representing simulated values or residues as a function of observed values for a
given variable, together with their linear regressions (Bennett et al., 2013; Vezy et al.,
384
Model capacity extension methods
2021). Dynamic plots that display simulated and observed values over time for each
variable can also be very useful. Such plots can help identify time lags between simu-
lations and observations or periods that differ considerably.
– Goodness-of-fit criteria compute numerical metrics comparing model simulations
to corresponding observations. They are the most popular way of evaluating models
since they provide a quantitative assessment of model performance. Many different
metrics are used in the literature. Mean square error ( ) and its breakdown into
different terms, mean absolute error ( ), and model efficiency are among the most
popular. Readers can refer to D. Wallach et al. (2018) for a comprehensive description
of the main criteria used for crop models. Each individual goodness-of-fit criterion
provides an image of model performance from a particular point of view. Use of multiple
complementary criteria is thus advisable to ensure a good characterisation of the struc-
ture of the differences between model simulations and observations. Note also that
multiple criteria can be combined in an integrated indicator to rank the performance of
different simulated variables or different models (see e.g. Bellocchi et al., 2002).
– Behavioural analysis examines the response to changes in key variables or para
meters in the model to evaluate whether the model subcomponents behave, at least
qualitatively, according to current knowledge about the real physical system being
simulated. For example, Coucheney et al. (2015) used graphs to compare the relative
variations of a set of simulated and observed variables between classes of agro-
pedoclimatic indicators (global radiation, fertilisation rate, soil water content at field
capacity, etc.). Holzworth et al. (2011) referred to this evaluation process as conducting
‘sensibility tests’. They gave several examples, such as the plot of the predicted response
of lupin yield in function of in-season cumulative rainfall and its comparison with a
potential yield computed using a simple water use efficiency model. For this example,
they verified that the envelope curve of the predicted yield, which takes into account
various stresses, was consistent with the variation of the potential yield with respect to
the different rainfall conditions.
The behaviour analysis evaluation method is less frequently used and published than
the first two methods listed here. However, its importance is recognised because
demonstrating that model outputs more or less fit a dataset is a necessary but not
sufficient indication of validity. It is nearly impossible to put together a relevant, high-
quality dataset that can evaluate most of the model subcomponents on a representative
sample of the target population (Sinclair and Seligman, 2000). It is simply not feasible
to measure everything in the soil, crop and atmosphere for reasons of cost or techno
logical limitations. Moreover, the target population may also include rare or even future
events for which it is very difficult or impossible to have corresponding measured
data. This is why many authors advise systematically integrating behavioural analyses
into model evaluation to assess whether the model faithfully captures the under-
lying process driving the system (Bellocchi et al., 2010; Jakeman et al., 2006; Sinclair
and Seligman, 2000). We should mention the importance of scrutinising the results
obtained for variables for which there are no observations included in the evaluation
and calibration datasets used (e.g. harvest index, root characteristics) and the results
obtained for synthetic situations (e.g. non-limiting soil and weather), even if doing so
does not strictly fall within the behavioural analysis method according to the definition
given in the papers cited here. But since such data are not generally taken into account
in calibration procedures, they may reveal non-expected behaviours.
385
STICS soil-crop model
The CroPlotR R package (Vezy et al., 2021) can be used to facilitate the application of
all these methods on crop models.
386
Model capacity extension methods
Note, however, that if errors have large variances but are unbiased, a large dataset
can compensate for large variance (Montesino-San Martin et al., 2018). Quantity and
quality issues are especially important when dealing with remote-sensing data and
participatory science.
– Post questions and discussion topics on the STICS forum1: experts are there to try
to answer your questions!
Having a detailed knowledge of the system under study, the dataset and the model,
is of great help to correctly interpret discrepancies between measured data and
simulation outputs.
1. (https://w3.avignon.inra.fr/forge/projects/stics_main_projecu/boards)
387
STICS soil-crop model
estimation techniques. This typically consists in finding the parameter values which
minimise a distance between the results of model simulations obtained on an
experimental dataset and the corresponding observations for a given set of variables of
interest. We will call this process ‘model calibration’ in the remainder of the chapter.
It is difficult to propose a single process to adapt STICS to new crops or cultivars,
because it is bound to the user aim, relevant plant, experimental data, available biblio
graphy, and possible analogy between the crop/cultivar of interest and those already
parameterised in STICS. Moreover, the crop modelling community is nowhere near a
consensus on how to calibrate crop models (Wallach et al., 2021). Thus, this chapter puts
forward several methodological elements, guidelines, specific information about STICS,
and bibliographic and software references to help users determine their own methodo
logy according to their particular situation. Readers will find more detailed information,
especially on statistical methods, in the excellent book by Daniel Wallach et al. (2018).
388
Model capacity extension methods
same principles as outlined for model evaluation can also be applied for model cali-
bration (see § 16.2): more specifically, the selected dataset must be as representative
of the target population as possible, i.e. in terms of the variability of conditions and
variables to predict.
Should all available variables be used in the calibration process? Two opposite
approaches can be considered:
– Use all available observed variables: This would produce the best results if the model
were perfect – but it is not. Due to misspecifications in the model and errors in fixed
parameters, the parameter values that produce the best predictions for different vari-
ables are not the same. This means that using all observed variables for calibration
generally results in a sort of compromise rather than the best predictions.
– Use only the observations of the target variable(s): This may produce the best
predictions for these variable(s), but may not for others. It could even produce worse
predictions than the default parameter values (Guillaume et al., 2011).
When adapting the model to a new crop or cultivar, even if there is only one target variable,
we recommend using several observed variables to at least roughly calibrate the different
processes involved in the model. The tables given in § 16.3.3.4 describe the list of the
main parameters and associated variables for the main model processes. When resetting
the parameters to improve the model prediction performance for a particular objective,
i.e. if a set of parameter values for the crop and the cultivar considered is already avail-
able, then we would instead suggest keeping only observations of the target variable(s).
389
STICS soil-crop model
390
Figure 16.1. Proposed calibration patterns.
391
Model capacity extension methods
STICS soil-crop model
of fit and the simplicity of the model, i.e. between the risk of overfitting and underfit-
ting. If users have doubts about including a parameter or set of parameters in the list
of parameters to estimate in the calibration process, they can perform the calibration
with and without the (set of ) parameter(s) and compare the information criterion
values obtained to decide which list of parameters should be estimated (see Tremblay
and Wallach (2004) for an example of this technique in use).
Parameter values should be adjusted within plausible ranges. It is advisable to gather
prior information about the parameter values and to use bound-constraint minimisa-
tion or Bayesian methods to include this information in the calibration process.
392
Model capacity extension methods
Parameters Targeted
Step Forcing Process
to calibrate variable
+ activation of
standard Root
distribution: kdisrac, HR, resmes,
alloperirac azomes, msrac,
True density
Root + activation of or LRACH
1a (required to simulate
growth simulation of 2 Root and C&N stocks
soil C dynamic)
classes (recommended evolution (under
for perennial crops): perennial crops)
RTD , rapdia,
propracfmax
efcroijuv, efcroiveg,
In all situations efcroirepro, teopt, masec
teoptbis, sea, psisto
Biomass
2 Additional
growth
parameters propresP + Propres masecnp,
for simulation + tauxmortresp maperenne
of C reserves
LAI +
vmax2, inngrain1,
phenological In all situations QNplante
inngrain2, INNmin
stages
Additional
PropresPN
parameters
+ parazoper QNplantenp,
for simulation
Nitrogen + parazorac QNperenne,
3 of N partitioning
absorption + parazofmorte, QNrac
(structural N,
parazotmorte
reserves and Root)
Additional fixmaxveg, fixmaxgr,
parameters concNrac0, Qfix
for legumes concNrac100
vitircarb/vitircarbT,
4 Elaboration of yield IRmax, tmaxremp, mafruit
nbgrmax, pgrainmaxi
Vitirazo, stdrpdes,
tempdeshyd, CNgrain,
Elaboration of the quality
5 deshydbase, H2Orec, oil,
of harvested organs
vitprophuile, sugar
vitpropsucre
stlevamf, jvc, sensiphot IAMF
6 Development stamflax, stlevdrp, ILAX, IDRP,
stdrpmat IMAT
Emergence
date dlaimaxbrut/dlaimax,
tigefeuil, innturgmin,
7 Foliage growth LAI
innsen, rapsenturg,
psiturg, adens, durvieF
Remarks:
– We suggest repeating steps 2 and 3 after step 7 to ‘balance’ the effects: 1, 1a, 2, 3, 4,
5, 6, 7, 2, 3.
– Step 1a can be skipped knowing that the resmes and azomes values, from which the
root density growth settings are estimated, also depend on foliage growth (important
393
STICS soil-crop model
for calculating transpiration, which affects resmes) and aerial biomass growth (impor-
tant for calculating nitrogen which affects azomes).
– Step 2 could be divided into two separate steps: one to estimate 𝐞𝐟𝐜𝐫𝐨𝐢𝐣𝐮𝐯P,
𝐞𝐟𝐜𝐫𝐨𝐢𝐯𝐞𝐠P, 𝐞𝐟𝐜𝐫𝐨𝐢𝐫𝐞𝐩𝐫𝐨P, 𝐭𝐞𝐨𝐩𝐭P, 𝐭𝐞𝐨𝐩𝐭𝐛𝐢𝐬P, on masec with USMs in “almost
non-limiting” growth conditions, and one to estimate 𝐩𝐬𝐢𝐬𝐭𝐨P on the same variable
but with limited water conditions. The same approach could be applied for step 3 with
the estimation of 𝐕𝐦𝐚𝐱2P using USMs without N stress and then the estimation of
𝐢𝐧𝐧𝐠𝐫𝐚𝐢𝐧1P, 𝐢𝐧𝐧𝐠𝐫𝐚𝐢𝐧2P and 𝐈𝐍𝐍𝐦𝐢𝐧P using USMs with N stress.
Without *.lai files available but with the dates of observed stages (in italics: infor-
mation that differs from Table 16.1).
394
Model capacity extension methods
Parameters Targeted
Step Forcing Process
to calibrate variable
Additional
PropresPN
parameters
+ parazoper QNplantenp,
for simulation
+ parazorac QNperenne,
of N partitioning
+ parazofmorte, QNrac
Nitrogen (structural N,
3 parazotmorte
absorption reserves and Root)
fixmaxveg,
Additional
fixmaxgr,
parameters Qfix
Phenological concNrac0,
for legumes
stages concNrac100
vitircarb/vitircarbT,
4 Elaboration of yield IRmax, tmaxremp, mafruit
nbgrmax, pgrainmaxi
Vitirazo, stdrpdes,
tempdeshyd, CNgrain,
Elaboration of the quality
5 deshydbase, H2Orec, oil,
of harvested organs
vitprophuile, sugar
vitpropsucre
stlevamf, jvc,
IAMF
sensiphot
6 Development
stamflax, stlevdrp, ILAX, IDRP,
stdrpmat IMAT
Emergence
date dlaimaxbrut/dlaimax,
tigefeuil, innturgmin,
7 Foliage growth innsen, rapsenturg, LAI
psiturg, adens,
durvieF
Remarks:
– We can suggest an iterative process by following the steps in this order: 1, 2, 3, 6, 2,
3, 6, 2, 3, 4, 5, 6, 7 (to ‘balance’ the compensation effects).
– The same remarks as in Table 16.1 for steps 1a and 2.
With *.lai file available but without the dates of observed stages, except emer-
gence date (in italics: information that differs from Table 16.1 ).
395
STICS soil-crop model
stlevdrp, stdrpmat
396
Model capacity extension methods
Parameters Targeted
Step Forcing Process
to calibrate variable
Vitirazo, stdrpdes,
LAI + CNgrain,
Elaboration of the quality tempdeshyd, deshydbase,
5 emergence H2Orec, oil,
of harvested organs vitprophuile,
date sugar
vitpropsucre
dlaimaxbrut/dlaimax,
tigefeuil, innturgmin,
Emergence innsen, rapsenturg,
6 Foliage growth psiturg, adens, durvieF LAI
date
stlevamf, stamflax, jvc
(annual crops), sensiphot
Without *.lai files available and without the dates of observed stages (except
emergence date): we recommend starting with step 6 and then proceeding with steps
1, 2, 3, 4 and 5.
Without *.lai files available but with the dates of observed stages (in italics: infor-
mation that differs from the first cultivar calibration scheme Table (Table 16.4).
397
STICS soil-crop model
Remarks:
– We can suggest an iterative process by following the steps in this order: 1, 2, 6, 2, 6,
2, 3, 4, 5, 6, 7.
With *.lai files available and without the dates of observed stages, except emer-
gence date (or bud burst date) (in italics:: information that is different from the first
cultivar calibration scheme Table (Table 16.4).
Without *.lai files available and without the dates of observed stages, except
emergence date: adapt the process to this specific case.
398
Model capacity extension methods
399
STICS soil-crop model
400
Model capacity extension methods
Estimates from all these methods are based on the computation of a distance between
the selected model variables and corresponding observations on the given dataset.
Wallach et al. (2011) and He et al. (2010) selected different distances adapted for
crop models.
The CroptimizR R package (Buis et al., 2021) is dedicated to crop model calibration
and includes different types of methods and goodness-of-fit criteria. It has been inter-
faced with the STICS model through the SticsOnR (Lecharpentier et al., 2021a) and
SticsRFiles (Lecharpentier et al., 2021b) packages (see § 15.5.2). Examples of use with
STICS are described in the CroptimizR documentation.
401
STICS soil-crop model
One part of the original dataset is often used for the calibration process and another
for evaluation. The evaluation process is then similar to the one described in § 16.2.
There are, however, many ways to divide the dataset into two parts, and doing so means
that a significant part of the dataset will not be used for the parameter estimation step.
Cross-validation techniques can be used to avoid these drawbacks, but they are much
more computationally expensive (see e.g. Daniel Wallach et al. (2018) for more details).
402
Model capacity extension methods
403
STICS soil-crop model
The STICS software expects all its input files to be in the same working directory with
fixed names. As such, if users want to manage a parallel use of the model, they will
need to either manage separate directories for each individual simulation or tell the
model where to find those files with different locations or names.
A STICS simulation using consecutive USMs (§ 14.1) is computed as several separate
simulations (with the internal state stored and/or restored from a recup.tmp file when
changing the USM).
When using 𝐜𝐨𝐝𝐞𝐞𝐭𝐩C options involving calculations (options 2, 3 and 4), the calcula-
tions are performed during the input checking part of initialisation step. Accordingly,
to integrate climate external data directly into the computation loop, users must have
also performed those potential evapotranspiration (ETP) calculations externally and
provided ETP as forced Penman (option 1).
One aspect to consider carefully can be an ‘off by one’ error when using a variable
value (a large number of variables holds only a single value corresponding to the vari-
able calculation performed in the previous daily step or the current day depending on
when you try to access these values). Some variables correspond to daily state variables
and others are integrated on various time spans. The CorrespondanceVariablesDeS-
orties subroutine can be used to get access to the variable values. The SticsRFiles R
package can also be used to handle STICS output files extraction (§ 15.5.2).
Forcing parameters without modifying the main STICS input files can be done by
using the param.sti file read by the subroutine Lecture_Optimisation between the main
input files read and inputs checking steps. This param.sti file is a text file that must
contain the list and associated values of the STICS parameters to force (the name of
the parameter on one line and its value on the following line). Any type of parameters
(soil, plant, technical, option codes, etc.) can be forced using this file. The SticsRFiles R
package (§ 15.5.2) can also be used to handle STICS input files modifications.
404
References
Abbad H., El Jaafari S., Bort J., Araus J., 2004. Comparison of flag leaf and ear photosynthesis with
biomass and grain yield of durum wheat under various water conditions and genotypes. Agronomie
24, 19–28. https://doi.org/10.1051/agro:2003056
Abiven S., Recous S., Reyes V., Oliver R., 2005. Mineralisation of C and N from root, stem and leaf
residues in soil and role of their biochemical quality. Biology and Fertility of Soils 42, 119–128. https://
doi.org/10.1007/s00374-005-0006-0
Addiscott T.M., Wagenet R.J., 1985. Concepts of solute leaching in soils: A review of modelling
approaches. European Journal of Soil Science 36, 411–424. https://doi.org/10.1111/j.1365-2389.1985.
tb00347.x
Adiku S.G.K., Ozier-Lafontaine H., Bajazet T., 2001. Patterns of root growth and water uptake of a
maize-cowpea mixture grown under greenhouse conditions. Plant and Soil 235, 85–94. https://doi.
org/10.1023/A:1011847214706
Affholder F., Poeydebat C., Corbeels M., Scopel E., Tittonell P., 2013. The yield gap of major food
crops in family agriculture in the tropics: Assessment and analysis through field surveys and model-
ling. Field Crops Research 143, 106–118. https://doi.org/10.1016/j.fcr.2012.10.021
Affholder F., Scopel E., Neto J.M., Capillon A., 2003. Diagnosis of the productivity gap using a crop
model. Methodology and case study of small-scale maize production in central Brazil. Agronomie 23,
305–325. https://doi.org/10.1051/agro:2003004
Al Majou H., Bruand A., Duval O., Le Bas C., Vautier A., 2008. Prediction of soil water retention
properties after stratification by combining texture, bulk density and the type of horizon. Soil Use and
Management 24, 383–391. https://doi.org/10.1111/j.1475-2743.2008.00180.x
Allaire J., Xie Y., McPherson J., Luraschi J., Ushey K., Atkins A., Wickham H., Cheng J., Chang W.,
Iannone R., 2021. Rmarkdown: Dynamic documents for r.
Allen R.G., Pereira L.S., Raes D., Smith M., 1998. Crop evapotranspiration-Guidelines for computing
crop water requirements-. Fao, Rome FAO Irrigation and drainage paper, 300 pages.
Allen R.G., Smith M., Perrier A., Pereira L.S., 1994. An update for the definition of reference evapo-
transpiration. ICID bulletin 43, 1–34.
Allen W.A., Richardson A.J., 1968. Interaction of light with a plant canopy. Journal of the Optical
Society of America 58, 1023. https://doi.org/10.1364/JOSA.58.001023
Alm D.M., Stoller E.W., Wax L.M., 1993. An index model for predicting seed germination and emer-
gence rates. Weed Technology 7, 560–569. https://doi.org/10.1017/S0890037X00037349
Amir J., Sinclair T.R., 1991. A model of the temperature and solar-radiation effects on spring wheat
growth and yield. Field Crops Research 28, 47–58. https://doi.org/10.1016/0378-4290(91)90073-5
Angus J.F., Mackenzie D.H., Morton R., Schafer C.A., 1981. Phasic development in field crops II.
Thermal and photoperiodic responses of spring wheat. Field Crops Research 4, 269–283. https://doi.
org/10.1016/0378-4290(81)90078-2
Antonioletti R., 1986. Contribution à l’étude du climat du MontVentoux (Document interne INRA).
Station de Bioclimatologie (INRA), Montfavet (FRA).
Araus J.L., Brown H.R., Febrero A., Bort J., Serret M.D., 1993. Ear photosynthesis, carbon isotope
discrimination and the contribution of respiratory CO2 to differences in grain mass in durum wheat.
Plant, Cell and Environment 16, 383–392. https://doi.org/10.1111/j.1365-3040.1993.tb00884.x
405
STICS soil-crop model
Araya S., Lyle G., Lewis M., Ostendorf B., 2016. Phenologic metrics derived from MODISNDVI as
indicators for PlantAvailableWater-holding Capacity. Ecological Indicators 60, 1263–1272. https://
doi.org/10.1016/j.ecolind.2015.09.012
Asseng S., Martre P., Maiorano A., Rötter R.P., O’Leary G.J., Fitzgerald G.J., Girousse C., Motzo R.,
Giunta F., Babar M.A., Reynolds M.P., Kheir A.M.S., Thorburn P.J., Waha K., Ruane A.C.,
Aggarwal P.K., Ahmed M., Balkovič J., Basso B., Biernath C., Bindi M., Cammarano D., Challinor A.J.,
Sanctis G.D., Dumont B., Rezaei E.E., Fereres E., Ferrise R., Garcia-Vila M., Gayler S., Gao Y., Horan H.,
Hoogenboom G., Izaurralde R.C., Jabloun M., Jones C.D., Kassie B.T., Kersebaum K.-C., Klein C.,
Koehler A.-K., Liu B., Minoli S., Martin M.M.S., Müller C., Kumar S.N., Nendel C., Olesen J.E.,
Palosuo T., Porter J.R., Priesack E., Ripoche D., Semenov M.A., Stöckle C., Stratonovitch P., Streck T.,
Supit I., Tao F., Velde M.V. der, Wallach D., Wang E., Webber H., Wolf J., Xiao L., Zhang Z., Zhao Z.,
Zhu Y., Ewert F., 2019. Climate change impact and adaptation for wheat protein. Global Change
Biology 25, 155–173. https://doi.org/10.1111/gcb.14481
Aura E., 1996. Modelling non-uniform soil water uptake by a single plant root. Plant and Soil 186,
237–243. https://doi.org/10.1007/BF02415519
Autret B., Mary B., Strullu L., Chlebowski F., Mäder P., Mayer J., Olesen J.E., Beaudoin N., 2020. Long-
term modelling of crop yield, nitrogen losses and GHG balance in organic cropping systems. Science
of The Total Environment 710. https://doi.org/10.1016/j.scitotenv.2019.134597
Avice J.C., Ourry A., Lemaire G., Boucaud J., 1996. Nitrogen and carbon flows estimated by 15N
and 13C pulse-chase labeling during regrowth of alfalfa. Plant Physiology 112, 281–290. https://doi.
org/10.1104/pp.112.1.281
Baier W., 1969. Concepts of soil moisture availability and their effect on soil moisture esti-
mates from a meteorological budget. Agricultural Meteorology 6, 165–178. https://doi.
org/10.1016/0002-1571(69)90002-8
Baldocchi D.D., Luxmoore R.J., Hatfield J.L., 1991. Discerning the forest from the trees: An essay on
scaling canopy stomatal conductance. Agricultural and Forest Meteorology 54, 197–226. https://doi.
org/10.1016/0168-1923(91)90006-C
Balesdent J., Recous S., 1997. Les temps de résidence du carbone et le potentiel de stockage de
carbone dans quelques sols cultivés français. Canadian Journal of Soil Science 77, 187–193. https://
doi.org/10.4141/S96-109
Barber S.A., Silberbush M., 2015. Plant root morphology and nutrient uptake, in: Barber, S.A.,
Bouldin, D.R. (Eds.), Roots, Nutrient and Water Influx, and Plant Growth, ASASpecialPublications.
American Society of Agronomy, Crop Science Society of America; Soil Science Society of America,
Madison, WI, USA, pp. 65–87. https://doi.org/10.2134/asaspecpub49.c4
Barbottin A., Lecomte C., Bouchard C., Jeuffroy M.-H., 2005. Nitrogen remobilization during grain
filling in wheat: Genotypic and environmental effects. Crop Science 45, 1141–1150. https://doi.
org/10.2135/cropsci2003.0361
Baret F., 1986. Contribution au suivi radiométrique de cultures de céréales (Thèse de doctorat).
Université de Paris XI.
Baret F., Andrieu B., Folmer J.C., Hanocq J.F., Sarrouy C., 1993. Gap fraction measurement from hemis-
pherical infrared photography and its use to evaluate PAR interception efficiency, in: Varlet-Grancher,
C., Bonhomme, R., Sinoquet, H. (Eds.), Crop Structure and Light Microclimate. Characterization and
Applications, Science Update. INRA Editions, Versaille, France, pp. 359–372.
Baret F., Houles V., Guérif M., 2007. Quantification of plant stress using remote sensing observations
and crop models: The case of nitrogen management. Journal of Experimental Botany 58, 869–880.
https://doi.org/10.1093/jxb/erl231
Baret F., Olioso A., Luciani J.L., 1992. Root biomass fraction as a function of growth degree days in
wheat. Plant and Soil 140, 137–144. https://doi.org/10.1007/BF00012815
Beaudoin N., 2006. Experimental analysis and modelling of the effects of cropping practices on
nonpoint nitrate pollution of a deep aquifer in an arable farming area (Thèse de doctorat). Institut
National Agronomique Paris-Grignon.
406
References
Beaudoin N., Gallois N., Viennot P., Le Bas C., Puech T., Schott C., Buis S., Mary B., 2018. Evaluation
of a spatialized agronomic model in predicting yield and N leaching at the scale of the Seine-
NormandieBasin. Environmental Science and Pollution Research 25, 23529–23558. https://doi.
org/10.1007/s11356-016-7478-3
Beaudoin N., Launay M., Ripoche D., Buis S., Justes E., Ruget F., Jégo G., García de Cortázar Atauri I.,
Léonard J., Raynal H., de Noblet-Ducoudre N., Seguin B., Lescourret F., Aubertot J.-N., Habets F.,
Jayet P.A., Gate P., Mary B., 2019. Modélisation du fonctionnement des agro-écosystèmes: L’épopée
STICS, in: Richard, G., Stengel, P., Lemaire, G., Cellier, P., Valceschini, E. (Eds.), Une Agronomie Pour Le
XXIe Siècle. Editions Quae, pp. 138–149.
Beaudoin N., Launay M., Sauboua E., Ponsardin G., Mary B., 2008. Evaluation of the soil crop model
STICS over 8 years against the "on farm" database of Bruyeres catchment. European Journal of Agronomy
29, 46–57. https://doi.org/10.1016/j.eja.2008.03.001
Beaudoin N., Venet E., Maucorps J., Vandenberghe C., Pugeaux N., Viennot P., Gourcy L., Brayer C.,
Machet J.-M., Couturier A., Billy C., Vigour N., Hulin G., Dorel G., Mary B., 2021. Long term response
of water and nitrogen fluxes to GoodAgriculturalPractices at field and catchment scales. Science of The
Total Environment 776. https://doi.org/10.1016/j.scitotenv.2021.145954
Beaujouan V., Durand P., Ruiz L., Aurousseau P., Cotteret G., 2002. A hydrological model dedicated to
topography-based simulation of nitrogen transfer and transformation: Rationale and application to the
geomorphology-denitrification relationship. Hydrological Processes 16, 493–507.
Beblik A., Kersebaum K.C., 2001. Performance of a nitrogen dynamics model applied to evaluate agricul-
tural management practices, in: Shaffer, M.J., Ma, L., Hansen, S. (Eds.), Modeling Carbon and Nitrogen
Dynamics for Soil Management. CRC Press, Boca Raton, USA, pp. 550–569.
Bedoussac L., Tribouillois H., Plaza Bonilla D., Journet E.-P., Justes E., 2017. Designing and evaluating
arable cropping systems with cash and cover crop legumes in sole crop and intercrop to improve nitrogen
use efficiency. Agricultural Research & Technology Open Access Journal 12. https://doi.org/10.19080/
ARTOAJ.2017.12.555842
Bélanger G., Rochette P., Castonguay Y., Bootsma A., Mongrain D., Ryan D.A.J., 2002. Climate change
and winter survival of perennial forage crops in EasternCanada. Agronomy Journal 94, 1120–1130.
https://doi.org/10.2134/agronj2002.1120
Bellocchi G., Acutis M., Fila G., Donatelli M., 2002. An indicator of solar radiation model perfor-
mance based on a fuzzy expert system. Agronomy Journal 94, 1222–1233. https://doi.org/10.2134/
agronj2002.1222
Bellocchi G., Rivington M., Donatelli M., Matthews K., 2011. Validation of biophysical models: Issues and
methodologies, in: Lichtfouse, E., Hamelin, M., Navarrete, M., Debaeke, P. (Eds.), Sustainable Agriculture.
Springer, Dordrecht, The Netherlands, pp. 577–603. https://doi.org/10.1007/978-94-007-0394-0_26
Bellocchi G., Rivington M., Donatelli M., Matthews K., 2010. Validation of biophysical models: Issues and
methodologies. A review. Agronomy for Sustainable Development 30, 109–130. https://doi.org/10.1051/
agro/2009001
Bennett N.D., Croke B.F.W., Guariso G., Guillaume J.H.A., Hamilton S.H., Jakeman A.J., Marsili-
Libelli S., Newham L.T.H., Norton J.P., Perrin C., Pierce S.A., Robson B., Seppelt R., Voinov A.A.,
Fath B.D., Andreassian V., 2013. Characterising performance of environmental models. Environmental
Modelling & Software 40, 1–20. https://doi.org/10.1016/j.envsoft.2012.09.011
Benoit M., Garnier J., Beaudoin N., Billen G., 2016. A participative network of organic and conventional
crop farms in the SeineBasin (France) for evaluating nitrate leaching and yield performance. Agricultural
Systems 148, 105–113. https://doi.org/10.1016/j.agsy.2016.07.005
Bergez J.E., Chabrier P., Gary C., Jeuffroy M.N., Makowski D., Quesnel G., Ramat E., Raynal H., Rousse N.,
Wallach D., Debaeke P., Durand P., Duru M., Dury J., Faverdin P., Gascuel-Odoux C., Garcia F., 2013.
An open platform to build, evaluate and simulate integrated models of farming and agro-ecosystems.
Environmental Modelling & Software 39, 39–49. https://doi.org/10.1016/j.envsoft.2012.03.011
Bergez J.E., Charron M.H., Leenhardt D., Poupa J.C., 2012. MOUSTICS: A generic dynamic plot-based
biodecisional model. Computers and Electronics in Agriculture 82, 8–14. https://doi.org/10.1016/j.
compag.2011.12.006
407
STICS soil-crop model
Bergez J.E., Raynal H., Launay M., Beaudoin N., Casellas E., Caubel J., Chabrier P., Coucheney E.,
Dury J., García de Cortázar Atauri I., Justes E., Mary B., Ripoche D., Ruget F., 2014. Evolution of
the STICS crop model to tackle new environmental issues: New formalisms and integration in the
modelling and simulation platform RECORD. Environmental Modelling & Software. https://doi.
org/10.1016/j.envsoft.2014.07.010
Bertin N., Gary C., 1993. Évaluation d’un modèle dynamique de croissance et de développement de la
tomate (Lycopersicon esculentumMill), TOMGRO, pour différents niveaux d’offre et de demande en
assimilats. Agronomie 13, 395–405. https://doi.org/10.1051/agro:19930504
Bessou C., Mary B., Leonard J., Roussel M., Grehan E., Gabrielle B., 2010. Modelling soil compaction
impacts on nitrous oxide emissions in arable fields. European Journal of Soil Science 61, 348–363.
Bethlenfalvay G.J., Abu-Shakra S.S., Fishbeck K., Phillips D.A., 1978. The effect of source-
sink manipulations on nitrogen fixation in peas. Physiologia Plantarum 43, 31–34. https://doi.
org/10.1111/j.1399-3054.1978.tb01562.x
Bethlenfalvay G.J., Phillips D.A., 1977. Ontogenetic interactions between photosynthesis and
symbiotic nitrogen fixation in legumes. Plant Physiology 60, 419–421. https://doi.org/10.1104/
pp.60.3.419
Beutier D., Renon H., 1978. Gas solubilities near the solvent critical point. AIChE Journal 24, 1122–
1125. https://doi.org/10.1002/aic.690240628
Bidabe B., 1965. L’action des températures sur l’évolution des bourgeons de l’entrée en dormance à la
floraison. Congr. Pom.(96 sess) Paris.
Bindi M., Sinclair T.R., Harrison J., 1999. Analysis of SeedGrowth by LinearIncrease in HarvestIndex.
Crop Science 39, 486–493. https://doi.org/10.2135/cropsci1999.0011183X0039000200031x
Bingham I.J., 1995. A comparison of the dynamics of root growth and biomass partitioning in wild oat
(Avena fatua L.) And spring wheat. Weed Research 35, 57–66. https://doi.org/10.1111/j.1365-3180.1995.
tb02017.x
Blum A., 1996. Crop responses to drought and the interpretation of adaptation, in: Belhassen, E.
(Ed.), Drought Tolerance in Higher Plants: Genetical, Physiological and Molecular Biological Analysis.
Springer, Dordrecht, The Netherlands, pp. 57–70. https://doi.org/10.1007/978-94-017-1299-6_8
Boesten J.J.T.I., Stroosnijder L., 1986. Simple model for daily evaporation from fallow tilled soil under
spring conditions in a temperate climate. Netherlands Journal of Agricultural Science 34, 75–90.
https://doi.org/10.18174/njas.v34i1.16818
Bonachela S., 1996. Root growth of triticale and barley grown for grain or for forage-plus-grain in a
Mediterranean climate. Plant and Soil 183, 239–251. https://doi.org/10.1007/BF00011439
Bonhomme R., Derieux M., Edmeades G.O., 1994. Flowering of DiverseMaizeCultivars in Relation
to Temperature and Photoperiod in MultilocationFieldTrials. Crop Science 34, 156–164. https://doi.
org/10.2135/cropsci1994.0011183X003400010028x
Bonhomme R., Ruget F., Derieux M., Vincourt P., 1982. Relations entre production de matière sèche
aérienne et énergie interceptée chez différents génotypes de maïs. Comptes rendus de l’Académie des
sciences. Série III, Sciences de la vie 294, 393–398.
Boote K.J., Jones J.W., 1987. Equations to define canopy photosynthesis from quantum efficiency,
maximum leaf rate, light extinction, leaf area index, and photon flux density, in: Biggins, J. (Ed.),
Progress in Photosynthesis Research. Springer, Dordrecht, The Netherlands, pp. 415–418. https://doi.
org/10.1007/978-94-017-0519-6_85
Boote K.J., Jones J.W., Hoogenboom G., 2018. Simulation of crop growth: CROPGRO model, in: Peart,
R.M., Curry, R.B. (Eds.), Agricultural Systems Modeling and Simulation. CRC Press, pp. 651–692.
https://doi.org/10.1201/9781482269765-18
Boote K.J., Pickering N.B., 1994. Modeling photosynthesis of row crop canopies. HortScience 29,
1423–1434. https://doi.org/10.21273/HORTSCI.29.12.1423
Bouaziz A., Bruckler L., 1989. Modeling wheat seedling growth and emergence: I. Seedling growth
affected by soil water potential. Soil Science Society of America Journal 53, 1832–1838. https://doi.
org/10.2136/sssaj1989.03615995005300060036x
408
References
Boulard T., Wang S., 2000. Greenhouse crop transpiration model from external climate conditions.
Acta Horticulturae 235–244. https://doi.org/10.17660/ActaHortic.2000.534.27
Bouma J., 1989. Using soil survey data for quantitative land evaluation, in: Stewart, B.A. (Ed.), Advances
in Soil Science. Springer, New York, NY, pp. 177–213. https://doi.org/10.1007/978-1-4612-3532-3_4
Boyer J.S., 1970. Leaf enlargement and metabolic rates in corn, soybean, and sunflower at various leaf
water potentials. Plant Physiology 46, 233–235. https://doi.org/10.1104/pp.46.2.233
Bradford K.J., 2002. Applications of hydrothermal time to quantifying and modeling seed germination
and dormancy. Weed Science 50, 248–260. https://doi.org/10.1614/0043-1745(2002)050[0248:AOHT
TQ]2.0.CO;2
Bradford K.J., 1990. A Water relations analysis of seed germination rates. Plant Physiology 94,
840–849. https://doi.org/10.1104/pp.94.2.840
Bradford K.J., Hsiao T.C., 1982. Physiological responses to moderate water stress, in: Lange,
O.L., Nobel, P.S., Osmond, C.B., Ziegler, H. (Eds.), Physiological Plant Ecology II, Encyclopedia
of PlantPhysiology (NewSeries). Springer, Berlin, Heidelberg, pp. 263–324. https://doi.
org/10.1007/978-3-642-68150-9_10
Brilli L., Bechini L., Bindi M., Carozzi M., Cavalli D., Conant R., Dorich C.D., Doro L., Ehrhardt F.,
Farina R., Ferrise R., Fitton N., Francaviglia R., Grace P., Iocola I., Klumpp K., Léonard J., Martin R.,
Massad R.S., Recous S., Seddaiu G., Sharp J., Smith P., Smith W.N., Soussana J.-F., Bellocchi G., 2017.
Review and analysis of strengths and weaknesses of agro-ecosystem models for simulating C and N
fluxes. Science of The Total Environment 598, 445–470. https://doi.org/10.1016/j.scitotenv.2017.03.208
Brisson N., Bonhomme R., Ameglio T., Gautier H., Olioso A., Drouet J.-L., 1996. Modèle de culture :
Simulation sous contraintes à partir d’un potentiel de production, in: Cruiziat, P., Lagouarde, J.-P.
(Eds.), Actes. Tome 2 : Du Couvert Végétal à La Région. INRA Editions, Le Croisic, France, pp. 39–51.
Brisson N., Bussière F., Ozier-Lafontaine H., Tournebize R., Sinoquet H., 2004. Adaptation of the
crop model STICS to intercropping. Theoretical basis and parameterisation. Agronomie 24, 409–421.
https://doi.org/10.1051/agro:2004031
Brisson N., Delécolle R., 1992. Développement et modèles de simulation de cultures. Agronomie 12,
253–263. https://doi.org/10.1051/agro:19920305
Brisson N., Gary C., Justes E., Roche R., Mary B., Ripoche D., Zimmer D., Sierra J., Bertuzzi P.,
Burger P., Bussière F., Cabidoche Y.M., Cellier P., Debaeke P., Gaudillère J.P., Hénault C., Maraux F.,
Seguin B., Sinoquet H., 2003. An overview of the crop model STICS. European Journal of Agronomy
18, 309–332. https://doi.org/10.1016/S1161-0301(02)00110-7
Brisson N., Gate P., Gouache D., Charmet G., Oury F.-X., Huard F., 2010. Why are wheat yields stagna-
ting in Europe? A comprehensive data analysis for France. Field Crops Research 119, 201–212. https://
doi.org/10.1016/j.fcr.2010.07.012
Brisson N., Itier B., L’Hotel J.C., Lorendeau J.Y., 1998a. Parameterisation of the Shuttleworth-Wallace
model to estimate daily maximum transpiration for use in crop models. Ecological Modelling 107,
159–169. https://doi.org/10.1016/S0304-3800(97)00215-9
Brisson N., Launay M., Mary B., Beaudoin N. (Eds.), 2009. Conceptual basis, formalisations and para-
meterization of the Stics crop model. Editions Quae.
Brisson N., Levrault F., 2010. Livre vert du projet Climator 2007-2010. Changement climatique, agri-
culture et forêt en France : Simulations d’impacts sur les principales espèces. ADEME Editions.
Brisson N., Mary B., Ripoche D., Jeuffroy M.H., Ruget F., Nicoullaud B., Gate P., Devienne-Barret F.,
Antonioletti R., Durr C., Richard G., Beaudoin N., Recous S., Tayot X., Plenet D., Cellier P.,
Machet J.-M., Meynard J.M., Delécolle R., 1998b. STICS: A generic model for the simulation of crops
and their water and nitrogen balances. I. Theory and parameterization applied to wheat and corn.
Agronomie 18, 311–346. https://doi.org/10.1051/agro:19980501
Brisson N., Olioso A., Clastre P., 1993. Daily transpiration of field soybeans as related to hydraulic
conductance, root distribution, soil potential and midday leaf potential. Plant and Soil 154, 227–237.
https://doi.org/10.1007/BF00012528
Brisson N., Perrier A., 1991. A semiempirical model of bare soil evaporation for crop simulation
models. Water Resources Research 27, 719–727. https://doi.org/10.1029/91WR00075
409
STICS soil-crop model
Brisson N., Rebière B., Zimmer D., Renault P., 2002. Response of the root system of a winter wheat
crop to waterlogging. Plant and Soil 243, 43–55. https://doi.org/10.1023/A:1019947903041
Brisson Nadine, Ruget F., Gate P., Lorgeou J., Nicoullaud B., Tayot X., Plenet D., Jeuffroy M.-H.,
Bouthier A., Ripoche D., Mary B., Justes E., 2002. STICS: A generic model for simulating crops and
their water and nitrogen balances. II. Model validation for wheat and maize. Agronomie 22, 69–92.
https://doi.org/10.1051/agro:2001005
Brisson N., Seguin B., Bertuzzi P., 1992. Agrometeorological soil water balance for crop simulation models.
Agricultural and Forest Meteorology 59, 267–287. https://doi.org/10.1016/0168-1923(92)90097-N
Brisson N., Wery J., Boote K.J., 2006. Fundamental concepts of crop models illustrated by a compara-
tive approach, in: Wallach, D., Makowski, D., Jones, J.W. (Eds.), Working with Dynamic Crop Models:
Evaluation, Analysis, Parameterization, and Applications. Elsevier, pp. 257–277.
Brochet P., Gerbier N., 1968. L’évapotranspiration. Aspect agrométéorologique. Evaluation pratique
de l’évapotranspiration potentielle. Direction de la météorologie nationale.
Bronson K.F., Touchton J.T., Hauck R.D., Kelley K.R., 1991. Nitrogen-15 recovery in winter wheat
as affected by application timing and dicyandiamide. Soil Science Society of America Journal 55,
130–135. https://doi.org/10.2136/sssaj1991.03615995005500010023x
Brougham R.W., 1956. Effect of intensity of defoliation on regrowth of pasture. Australian Journal of
Agricultural Research 7, 377–387. https://doi.org/10.1071/AR9560377
Bruand A., Duval O., Cousin I., 2004. Estimation des propriétés de rétention en eau des sols à partir de
la base de données SOLHYDRO: Une première proposition combinant le type d’horizon, sa texture et
sa densité apparente. Etude et Gestion des Sols 11, 323–334.
Brunt D., 2007. Notes on radiation in the atmosphere. I. Quarterly Journal of the Royal Meteorological
Society 58, 389–420. https://doi.org/10.1002/qj.49705824704
Brutsaert W., 1982. Evaporation into the atmosphere. Theory, history and applications, Environmental
FluidMechanics. Springer, Dordrecht, The Netherlands. https://doi.org/10.1007/978-94-017-1497-6
Buchi L., Valsangiacomo A., Burel E., Charles R., 2016. Integrating simulation data from a crop model
in the development of an agri-environmental indicator for soil cover in Switzerland. European Journal
of Agronomy 76, 149–159. https://doi.org/10.1016/j.eja.2015.11.004
Buis S., Coucheney E., Launay M., Lecharpentier P., Mary B., Ripoche D., Beaudoin N., Ruget F.,
García de Cortázar Atauri I., Justes E., Constantin J., Andrianarisoa K.-S., Le Bas C., Léonard J., 2016.
Multicriteria evaluation of the stics soil-crop model and implementation of an automated evaluation
system, in: Ewert, F., Boote, K.J., Rötter, R.P., Thorburn, P., Nendel, C. (Eds.), International Crop Mo
delling Symposium "Crop Modelling for Agriculture and Food Security Under Global Change" 2016,
15-17/03/2016. Berlin, Germany, pp. 214–215.
Buis S., Lecharpentier P., Vezy R., Giner M., 2021. CroptimizR: A package to estimate parameters of
crop models. https://doi.org/10.5281/ZENODO.4066451
Buis S., Wallach D., Guillaume S., Varella H., Lecharpentier P., Launay M., Guérif M., Bergez J.E.,
Justes E., 2011. The STICS crop model and associated software for analysis, parameterization, and
evaluation, in: Ahuja, L.R., Ma, L. (Eds.), Methods of Introducing System Models into Agricultural
Research, Advances in AgriculturalSystemsModeling. John Wiley & Sons, Ltd, pp. 395–426. https://
doi.org/10.2134/advagricsystmodel2.c14
Burch G.J., Smith R.C.G., Mason W.K., 1978. Agronomic and PhysiologicalResponses of Soybean
and SorghumCrops to WaterDeficitsII. CropEvaporation, SoilWaterDepletion and RootDistribution.
Functional Plant Biology 5, 169–177. https://doi.org/10.1071/PP9780169
Burger P., 2001. Analyse de la variabilité de la teneur en protéines de la graine de soja : Approche par
voie d’enquête et étude expérimentale de conduites de cultures visant de hautes teneurs en protéines
(Thèse de doctorat). Institut National Agronomique Paris-Grignon.
Bussière F., 1995. Rainfall interception by plant canopies, consequences for water partitioning in
intercropping systems, in: Sinoquet, H., Cruz, P. (Eds.), Ecophysiology of Tropical Intercropping,
Science Update. INRA Editions, Versailles, pp. 163–174.
410
References
Buvaneshwari S., Riotte J., Sekhar M., Sharma A.K., Helliwell R., Kumar M.S.M., Braun J.J., Ruiz L.,
2020. Potash fertilizer promotes incipient salinization in groundwater irrigated semi-arid agriculture.
Scientific Reports 10, 3691. https://doi.org/10.1038/s41598-020-60365-z
Buvaneshwari S., Sekhar M., Riotte J., Raynal H., Casellas E., Mohan Kumar M.S., Troldborg M., Ruiz L.,
2018. Modelling the impact of nutrient recycling on groundwater resource quantity and quality in an
irrigated semi-arid tropical catchment, in: Geophysical Research Abstracts, EGU General Assembly
2018, 8-13/04/2018. p. 17226.
Caldwell R.M., 1995. Simulation models for intercropping systems, in: Sinoquet, H., Cruz, P. (Eds.),
Ecophysiology of Tropical Intercropping. INRA Editions, Pointe à Pitre, France, pp. 353–358.
Caldwell R.M., Hansen J.W., 1993. Simulation of multiple cropping systems with CropSys, in: de
Vries, F.P., Teng, P., Metselaar, K. (Eds.), Systems Approaches for Agricultural Development. Springer,
Dordrecht, The Netherlands, pp. 397–412. https://doi.org/10.1007/978-94-011-2842-1_24
Casals M.L., 1996. Introduction des mécanismes de résistance à la sécheresse dans un modèle
dynamique de croissance et de développement, du blé dur (Thèse de doctorat). Institut National
Agronomique Paris-Grignon.
Caubel J., Launay M., García de Cortázar Atauri I., Ripoche D., Huard F., Buis S., Brisson N., 2014. A
new integrated approach to assess the impacts of climate change on grapevine fungal diseases: The
coupled MILA-STICS model. Journal International des Sciences de la Vigne et du Vin 45–54.
Caubel J., Launay M., Lannou C., Brisson N., 2012. Generic response functions to simulate climate-
based processes in models for the development of airborne fungal crop pathogens. Ecological
Modelling 242, 92–104. https://doi.org/10.1016/j.ecolmodel.2012.05.012
Caubel J., Launay M., Ripoche D., Gouache D., Buis S., Huard F., Huber L., Brun F., Bancal M.O.,
2017. Climate change effects on leaf rust of wheat: Implementing a coupled crop-disease model in
a French regional application. European Journal of Agronomy 90, 53–66. https://doi.org/10.1016/j.
eja.2017.07.004
Cellier P., Richard G., Robin P., 1996. Partition of sensible heat fluxes into bare soil and the atmosphere.
Agricultural and Forest Meteorology 82, 245–265. https://doi.org/10.1016/0168-1923(95)02328-3
Cellier P., Ruget F., Chartier M., Bonhomme R., 1993. Estimating the temperature of a maize
apex during early growth stages. Agricultural and Forest Meteorology 63, 35–54. https://doi.
org/10.1016/0168-1923(93)90021-9
Chantigny M.H., Rochette P., Angers D.A., Massé D., Côté D., 2004. Ammonia Volatilization and
SelectedSoilCharacteristicsFollowingApplication of AnaerobicallyDigestedPigSlurry. Soil Science
Society of America Journal 68, 306–312. https://doi.org/10.2136/sssaj2004.3060
Chapman D.C., Lake D.W., 2003. Computing runoff, in: New York Standards and Specifications for
Erosion and Sediment Control. New York State Soil; Water Conservation Committee, New York,
USA, pp. 16 p.
Chapman S.C., Hammer G.L., Palta J.A., 1993. Predicting leaf area development of sunflower. Field
Crops Research 34, 101–112. https://doi.org/10.1016/0378-4290(93)90114-3
Chuine I., 2010. Why does phenology drive species distribution? Philosophical Transactions of the
Royal Society of London B: Biological Sciences 365, 3149–3160. https://doi.org/10.1098/rstb.2010.0142
Clapp R.B., Hornberger G.M., 1978. Empirical equations for some soil hydraulic properties. Water
Resources Research 14, 601–604. https://doi.org/10.1029/WR014i004p00601
Clivot H., Mary B., Valé M., Cohan J.-P., Champolivier L., Piraux F., Laurent F., Justes E., 2017.
Quantifying in situ and modeling net nitrogen mineralization from soil organic matter in arable crop-
ping systems. Soil Biology and Biochemistry 111, 44–59. https://doi.org/10.1016/j.soilbio.2017.03.010
Clivot H., Mouny J.-C., Duparque A., Dinh J.-L., Denoroy P., Houot S., Vertès F., Trochard R.,
Bouthier A., Sagot S., Mary B., 2019. Modeling soil organic carbon evolution in long-term arable
experiments with AMG model. Environmental Modelling & Software 118, 99–113. https://doi.
org/10.1016/j.envsoft.2019.04.004
Cockborne A.-M. de, Jauzein M., Stengel P., Guennelon R., 1988. Variation du coefficient de diffusion
de NO3– dans les sols : Influence de la teneur en eau et de la porosité. Agronomie 8, 905–914. https://
doi.org/10.1051/agro:19881010
411
STICS soil-crop model
Coleman K., Jenkinson D.S., 1996. RothC-26.3 - AModel for the turnover of carbon in soil, in:
Powlson, D.S., Smith, P., Smith, J.U. (Eds.), Evaluation of Soil Organic Matter Models. Springer, Berlin,
Heidelberg, pp. 237–246. https://doi.org/10.1007/978-3-642-61094-3_17
Confalonieri R., Orlando F., Paleari L., Stella T., Gilardelli C., Movedi E., Pagani V., Cappelli G.,
Vertemara A., Alberti L., Alberti P., Atanassiu S., Bonaiti M., Cappelletti G., Ceruti M., Confalonieri A.,
Corgatelli G., Corti P., Dell’Oro M., Ghidoni A., Lamarta A., Maghini A., Mambretti M., Manchia A.,
Massoni G., Mutti P., Pariani S., Pasini D., Pesenti A., Pizzamiglio G., Ravasio A., Rea A.,
Santorsola D., Serafini G., Slavazza M., Acutis M., 2016. Uncertainty in crop model predictions: What
is the role of users? Environmental Modelling & Software 81, 165–173. https://doi.org/10.1016/j.
envsoft.2016.04.009
Constantin J., Beaudoin N., Launay M., Duval J., Mary B., 2012. Long-term nitrogen dynamics in
various catch crop scenarios: Test and simulations with STICS model in a temperate climate.
Agriculture Ecosystems & Environment 147, 36–46. https://doi.org/10.1016/j.agee.2011.06.006
Constantin J., Durr C., Tribouillois H., Justes E., 2015a. Catch crop emergence success depends on
weather and soil seedbed conditions in interaction with sowing date: A simulation study using the
SIMPLE emergence model. Field Crops Research 176, 22–33. https://doi.org/10.1016/j.fcr.2015.02.017
Constantin J., Le Bas C., Justes E., 2015b. Large-scale assessment of optimal emergence and destruc-
tion dates for cover crops to reduce nitrate leaching in temperate conditions using the STICS soil-crop
model. European Journal of Agronomy 69, 75–87. https://doi.org/10.1016/j.eja.2015.06.002
Constantin J., Raynal H., Casellas E., Hoffmann H., Bindi M., Doro L., Eckersten H., Gaiser T., Grosz B.,
Haas E., Kersebaum K.-C., Klatt S., Kuhnert M., Lewan E., Maharjan G.R., Moriondo M., Nendel C.,
Roggero P.P., Specka X., Trombi G., Villa A., Wang E., Weihermüller L., Yeluripati J., Zhao Z., Ewert F.,
Bergez J.-E., 2019. Management and spatial resolution effects on yield and water balance at regional
scale in crop models. Agricultural and Forest Meteorology 275, 184–195. https://doi.org/10.1016/j.
agrformet.2019.05.013
Constantin Julie, Willaume M., Murgue C., Lacroix B., Therond O., 2015. The soil-crop models STICS
and AqYield predict yield and soil water content for irrigated crops equally well with limited data.
Agricultural and Forest Meteorology 206, 55–68. https://doi.org/10.1016/j.agrformet.2015.02.011
Cordery I., Graham A., 1989. Forecasting wheat yields using a water budgeting model. Australian
Journal of Agricultural Research 40, 715–728. https://doi.org/10.1071/AR9890715
Corre-Hellou G., Brisson N., Launay M., Fustec J., Crozat Y., 2007. Effect of root depth penetra-
tion on soil nitrogen competitive interactions and dry matter production in pea-barley intercrops
given different soil nitrogen supplies. Field Crops Research 103, 76–85. https://doi.org/10.1016/j.
fcr.2007.04.008
Corre-Hellou G., Faure M., Launay M., Brisson N., Crozat Y., 2009. Adaptation of the STICS intercrop
model to simulate crop growth and N accumulation in pea-barley intercrops. Field Crops Research
113, 72–81. https://doi.org/10.1016/j.fcr.2009.04.007
Corwin D.L., Wagenet R.J., 1996. Applications of GIS to the Modeling of NonPointSourcePollutants
in the VadoseZone: AConferenceOverview. Journal of Environmental Quality 25, 403–411. https://
doi.org/10.2134/jeq1996.00472425002500030004x
Coucheney E., Buis S., Launay M., Constantin J., Mary B., García de Cortázar Atauri I., Ripoche D.,
Beaudoin N., Ruget F., Andrianarisoa K.S., Le Bas C., Justes E., Léonard J., 2015. Accuracy, robustness
and behavior of the STICS soil–crop model for plant, water and nitrogen outputs: Evaluation over
a wide range of agro-environmental conditions in France. Environmental Modelling & Software 64,
177–190. https://doi.org/10.1016/j.envsoft.2014.11.024
Courault D., Hadria R., Ruget F., Olioso A., Duchemin B., Hagolle O., Dedieu G., 2010. Combined
use of FORMOSAT-2 images with a crop model for biomass and water monitoring of permanent
grassland in Mediterranean region. Hydrology and Earth System Sciences 14, 1731–1744. https://doi.
org/10.5194/hess-14-1731-2010
Cousin R., 1997. Peas (Pisum sativumL.). Field Crops Research 53, 111–130. https://doi.org/10.1016/
S0378-4290(97)00026-9
Cowan I.R., 1968. The Interception and Absorption of Radiation in PlantStands. The Journal of
Applied Ecology 5, 367. https://doi.org/10.2307/2401567
412
References
Cruz P., Duru M., Therond O., Theau J.-P., Ducourtieux C., Jouany C., Al Haj Khaled R., Ansquer P.,
2002. Une nouvelle approche pour caractériser les prairies naturelles et leur valeur d’usage. Fourrages
172, 335–354.
Cruz P., Theau J.-P., Lecloux E., Jouany C., Duru M., 2010. Typologie fonctionnelle de graminées four-
ragères pérennes : Une classification multitraits. Fourrages 201, 11–17.
Dale R.F., Coelho D.T., Gallo K.P., 1980. Prediction of DailyGreenLeafAreaIndex for Corn. Agronomy
Journal 72, 999–1005. https://doi.org/10.2134/agronj1980.00021962007200060032x
de Willigen P., Heinen M., Mollier A., Van Noordwijk M., 2002. Two-dimensional growth of a root
system modelled as a diffusion process. I. Analytical solutions. Plant and Soil 240, 225–234. https://
doi.org/10.1023/A:1015744529454
de Wit C.T. (Ed.), 1978. Simulation of assimilation, respiration and transpiration of crops, Centre
for Agricultural Publishing and Documentation., Simulation monographs. Pudoc, Wageningen, The
Netherlands.
de Wit C.T., Brouwer R., Penning de Vries F.W.T., 1970. The simulation of photosynthetic systems,
in: Setlik, I. (Ed.), Prediction and Measurement of Photosynthetic Productivity, Proceeding IBP/
PPTechnicalMeetingTrebon 1969. Pudoc, Wageningen, The Netherlands, pp. 47–70.
De Notaris C., Rasmussen J., Sorensen P., Olesen J.E., 2018. Nitrogen leaching: A crop rotation pers-
pective on the effect of N surplus, field management and use of catch crops. Agriculture Ecosystems &
Environment 255, 1–11. https://doi.org/10.1016/j.agee.2017.12.009
De Tourdonnet S., 1999. Control of quality and N pollution in greenhouse lettuce production: A
modelling study. Acta Horticulturae 263–270. https://doi.org/10.17660/ActaHortic.1999.507.30
De Villele O., 1974. Besoins en eau des cultures sous serre - Essai de conduite des arrosages en fonction
de l’ensoleillement. Acta Horticulturae 123–135. https://doi.org/10.17660/ActaHortic.1974.35.16
Debaeke P., Aboudrare A., 2004. Adaptation of crop management to water-limited environments.
European Journal of Agronomy 21, 433–446. https://doi.org/10.1016/j.eja.2004.07.006
Debaeke P., Burger P., Brisson N., 2001. A simple model to simulate N2 symbiotic fixation of annual
grain legumes: Application to soybean, in: Proceedings of the Second International Symposium on
Modelling Cropping System. Florence (Italy), pp. 17–18.
Defossez P., Richard G., Boizard H., O’Sullivan M.F., 2003. Modeling change in soil compaction due to
agricultural traffic as function of soil water content. Geoderma 116, 89–105. https://doi.org/10.1016/
S0016-7061(03)00096-X
Deleens E., Gregory N., Bourdu R., 1984. Transition between seed reserve use and photosynthetic
supply during development of maize seedlings. Plant Science Letters 37, 35–39.
Demestihas C., Plenet D., Genard M., García de Cortázar Atauri I., Launay M., Ripoche D.,
Beaudoin N., Simon S., Charreyron M., Raynal C., Lescourret F., 2018. Analyzing ecosystem services
in apple orchards using the STICS model. European Journal of Agronomy 94, 108–119. https://doi.
org/10.1016/j.eja.2018.01.009
Denmead O.T., 1973. Relative significance of soil plant evaporation in estimating evapotranspiration, in:
Slatyer, R.O. (Ed.), Plant Response to Climatic Factors. Proceedings of the UNESCO Uppsala Symposium
1970. UNESCO, Uppsalla, Sweden, pp. 505–511.
Denmead O.T., Shaw R.H., 1962. Availability of SoilWater to Plants as Affected by SoilMoistureContent
and MeteorologicalConditions. Agronomy Journal 54, 385–390. https://doi.org/10.2134/agronj1962.
00021962005400050005x
Desclaux D., Roumet P., 1996. Impact of drought stress on the phenology of two soybean (Glycine
maxL. Merr) cultivars. Field Crops Research 46, 61–70. https://doi.org/10.1016/0378-4290(95)00086-0
Devienne-Barret F., Justes E., Machet J.M., Mary B., 2000. Integrated Control of NitrateUptake by
CropGrowthRate and SoilNitrateAvailability under FieldConditions. Annals of Botany 86, 995–1005.
https://doi.org/10.1006/anbo.2000.1264
Diaz S., Cabido M., 1997. Plant functional types and ecosystem functions in relation to global change.
Journal of Vegetation Science 8, 463–474. https://doi.org/10.2307/3237198
413
STICS soil-crop model
Dimassi B., Cohan J.-P., Labreuche J., Mary B., 2013. Changes in soil carbon and nitrogen following
tillage conversion in a long-term experiment in NorthernFrance. Agriculture Ecosystems & Environment
169, 12–20. https://doi.org/10.1016/j.agee.2013.01.012
Dimassi B., Mary B., Wylleman R., Labreuche J., Couture D., Piraux F., Cohan J.-P., 2014. Long-term
effect of contrasted tillage and crop management on soil carbon dynamics during 41 years. Agriculture
Ecosystems & Environment 188, 134–146. https://doi.org/10.1016/j.agee.2014.02.014
Dinesh H., Pearce J.M., 2016. The potential of agrivoltaic systems. Renewable & Sustainable Energy
Reviews 54, 299–308. https://doi.org/10.1016/j.rser.2015.10.024
Donohue R.J., Lawes R.A., Mata G., Gobbett D., Ouzman J., 2018. Towards a national, remote-
sensing-based model for predicting field-scale crop yield. Field Crops Research 227, 79–90. https://doi.
org/10.1016/j.fcr.2018.08.005
Doorenbos K., Kassam A.H., Bentvelsen C., Uittenbogaard G., 1980. Yield response to water, in: Johl,
S.S. (Ed.), Irrigation and Agricultural Development. Pergamon, pp. 257–280. https://doi.org/10.1016/
B978-0-08-025675-7.50021-2
Dorsainvil F., 2002. Evaluation, par modélisation, de l’impact environnemental des modes de conduite
des cultures intermédiaires sur les bilans d’eau et d’azote dans les systèmes de culture (Thèse de doctorat).
Institut National Agronomique Paris-Grignon.
Douguedroit A., 1986. Les topoclimats thermiques de moyenne montagne. Agrométéorologie des régions
de moyenne montagne, Les colloques de l’INRA, Toulouse 198–213.
Drouet J.-L., Pagès L., 2003. GRAAL: A model of GRowth, Architecture and carbon ALlocation
during the vegetative phase of the whole maize plant. Ecological Modelling 165, 147–173. https://doi.
org/10.1016/S0304-3800(03)00072-3
Ducharne A., Baubion C., Beaudoin N., Benoit M., Billen G., Brisson N., Garnier J., Kieken H.,
Lebonvallet S., Ledoux E., Mary B., Mignolet C., Poux X., Sauboua E., Schott C., Théry S., Viennot P.,
2007. Long term prospective of the SeineRiver system: Confronting climatic and direct anthropogenic
changes. Science of The Total Environment 375, 292–311. https://doi.org/10.1016/j.scitotenv.2006.12.011
Dufour L., Metay A., Talbot G., Dupraz C., 2013. Assessing LightCompetition for CerealProduction in
TemperateAgroforestrySystems using Experimentation and CropModelling. Journal of Agronomy and
Crop Science 199, 217–227. https://doi.org/10.1111/jac.12008
Dumont B., Leemans V., Mansouri M., Bodson B., Destain J.-P., Destain M.-F., 2014. Parameter iden-
tification of the STICS crop model, using an accelerated formal MCMC approach. Environmental
Modelling & Software 52, 121–135. https://doi.org/10.1016/j.envsoft.2013.10.022
Duncan W.G., 1971. Leaf Angles, LeafArea, and CanopyPhotosynthesis. Crop Science 11, 482–485.
https://doi.org/10.2135/cropsci1971.0011183X001100040006x
Dupraz C., Wolz K.J., Lecomte I., Talbot G., Vincent G., Mulia R., Bussière F., Ozier-Lafontaine H.,
Andrianarisoa S., Jackson N., Lawson G., Dones N., Sinoquet H., Lusiana B., Harja D.,
Domenicano S., Reyes F., Gosme M., Van Noordwijk M., 2019. Hi-sAFe: A3DAgroforestryModel for
IntegratingDynamicTree–CropInteractions. Sustainability 11. https://doi.org/10.3390/su11082293
Durand J.-L., Delusca K., Boote K., Lizaso J., Manderscheid R., Weigel H.J., Ruane A.C., Rosenzweig C.,
Jones J., Ahuja L., Anapalli S., Basso B., Baron C., Bertuzzi P., Biernath C., Deryng D., Ewert F., Gaiser T.,
Gayler S., Heinlein F., Kersebaum K.C., Kim S.-H., Müller C., Nendel C., Olioso A., Priesack E.,
Villegas J.R., Ripoche D., Rötter R.P., Seidel S.I., Srivastava A., Tao F., Timlin D., Twine T., Wang E.,
Webber H., Zhao Z., 2018. How accurately do maize crop models simulate the interactions of atmos-
pheric CO2 concentration levels with limited water supply on water use and yield? European Journal of
Agronomy, Recent advances in crop modelling to support sustainable agricultural production and food
security under global change 100, 67–75. https://doi.org/10.1016/j.eja.2017.01.002
Durand R., 1967. Action de la temperature et du rayonnement sur la croissance. Annales de Physiologie
Végétale 9, 5.
Dürr C., Aubertot J.-N., Richard G., Dubrulle P., Duval Y., Boiffin J., 2001. Simple: A model for simulation
of plant emergence predicting the effects of soil tillage and sowing operations. Soil Science Society of
America Journal 65, 414–423. https://doi.org/10.2136/sssaj2001.652414x
414
References
Duru M., Ducrocq H., Tirilly V., 1995. Modeling growth of cocksfoot (Dactylis glomerataL.) And
tall fescue (Festuca arundinacea schreb.) At the end of spring in relation to herbage nitrogen status.
Journal of Plant Nutrition 18, 2033–2047. https://doi.org/10.1080/01904169509365042
Duru M., Justes E., Langlet A., Tirilly V., Rouziès S., Sos L., Viard R., 1993. Comparaison des dyna-
miques d’apparition et de mortalité des organes de fétuque élevée, dactyle et luzerne (feuilles, talles
et tiges). Agronomie 13, 237–252.
Duru M., Tallowin J., Cruz P., 2005. Functional diversity in low-input grassland farming systems:
Characterisation, effect and management. Agronomy research 3, 125–138.
Duval Y., Boiffin J., 1990. Influence of soil crusting on the emergence of suger beet seedlings, in:
Proceedings of the First Congress of the European Society of Agronomy, 5-7/12/1990. ESA, Colmar,
France, p. 311.
Dwyer L.M., Stewart D.W., 1986. Leaf AreaDevelopment in Field-GrownMaize. Agronomy Journal 78,
334–343. https://doi.org/10.2134/agronj1986.00021962007800020024x
Eagleman J.R., 1971. An experimentally derived model for actual evapotranspiration. Agricultural
Meteorology 8, 385–394. https://doi.org/10.1016/0002-1571(71)90124-5
Ehrhardt F., Soussana J.-F., Bellocchi G., Grace P., McAuliffe R., Recous S., Sándor R., Smith P., Snow V.,
de Antoni Migliorati M., Basso B., Bhatia A., Brilli L., Doltra J., Dorich C.D., Doro L., Fitton N.,
Giacomini S.J., Grant B., Harrison M.T., Jones S.K., Kirschbaum M.U.F., Klumpp K., Laville P.,
Léonard J., Liebig M., Lieffering M., Martin R., Massad R.S., Meier E., Merbold L., Moore A.D.,
Myrgiotis V., Newton P., Pattey E., Rolinski S., Sharp J., Smith W.N., Wu L., Zhang Q., 2018. Assessing
uncertainties in crop and pasture ensemble model simulations of productivity and N2O emissions.
Global Change Biology 24, e603–e616. https://doi.org/10.1111/gcb.13965
Ewert F., van Ittersum M.K., Bezlepkina I., Therond O., Andersen E., Belhouchette H., Bockstaller C.,
Brouwer F., Heckelei T., Janssen S., Knapen R., Kuiper M., Louhichi K., Olsson J.A., Turpin N.,
Wery J., Wien J.E., Wolf J., 2009. A methodology for enhanced flexibility of integrated assessment in
agriculture. Environmental Science & Policy 12, 546–561. https://doi.org/10.1016/j.envsci.2009.02.005
Ewert F., van Ittersum M.K., Heckelei T., Therond O., Bezlepkina I., Andersen E., 2011. Scale changes
and model linking methods for integrated assessment of agri-environmental systems. Agriculture
Ecosystems & Environment 142, 6–17. https://doi.org/10.1016/j.agee.2011.05.016
Faivre R., Leenhardt D., Voltz M., Benoit M., Papy F., Dedieu G., Wallach D.D., 2004. Spatialising crop
models. Agronomie 24, 205–217. https://doi.org/10.1051/agro:2004016
Falconnier G.N., Corbeels M., Boote K.J., Affholder F., Adam M., MacCarthy D.S., Ruane A.C.,
Nendel C., Whitbread A.M., Justes E., Ahuja L.R., Akinseye F.M., Alou I.N., Amouzou K.A.,
Anapalli S.S., Baron C., Basso B., Baudron F., Bertuzzi P., Challinor A.J., Chen Y., Deryng D.,
Elsayed M.L., Faye B., Gaiser T., Galdos M., Gayler S., Gerardeaux E., Giner M., Grant B.,
Hoogenboom G., Ibrahim E.S., Kamali B., Kersebaum K.C., Kim S.-H., van der Laan M., Leroux L.,
Lizaso J.I., Maestrini B., Meier E.A., Mequanint F., Ndoli A., Porter C.H., Priesack E., Ripoche D.,
Sida T.S., Singh U., Smith W.N., Srivastava A., Sinha S., Tao F., Thorburn P.J., Timlin D., Traore B.,
Twine T., Webber H., 2020. Modelling climate change impacts on maize yields under low nitrogen
input conditions in sub-SaharanAfrica. Global Change Biology 26, 5942–5964. https://doi.
org/10.1111/gcb.15261
Falconnier G.N., Journet E.P., Bedoussac L., Vermue A., Chlebowski F., Beaudoin N., Justes E., 2019.
Calibration and evaluation of the STICS soil-crop model for faba bean to explain variability in yield
and N-2 fixation. European Journal of Agronomy 104, 63–77. https://doi.org/10.1016/j.eja.2019.01.001
Fan J., McConkey B., Wang H., Janzen H., 2016. Root distribution by depth for temperate agricultural
crops. Field Crops Research 189, 68–74. https://doi.org/10.1016/j.fcr.2016.02.013
Farquhar G.D., von Caemmerer S., Berry J.A., 1980. A biochemical model of photosynthetic CO2
assimilation in leaves of C3 species. Planta 149, 78–90. https://doi.org/10.1007/BF00386231
Feddes R.A., Hooghart J.C., 1987. Crop factors in relation to Makkink reference-crop evapo-
transpiration, in: Evaporation and Weather. Proceedings and Informations / Committee for
Hydrological Research T.N.O. The Hague, The Netherlands, pp. 33–47.
415
STICS soil-crop model
Ferchaud F., Mary B., 2016. Drainage and NitrateLeachingAssessedDuring 7 YearsUnderPerennial and
AnnualBioenergyCrops. Bioenergy Research 9, 656–670. https://doi.org/10.1007/s12155-015-9710-2
Ferrant S., Gascoin S., Veloso A., Salmon-Monviola J., Claverie M., Rivalland V., Dedieu G.,
Demarez V., Ceschia E., Probst J.L., Durand P., Bustillo V., 2014. Agro-hydrology and multi-temporal
high-resolution remote sensing: Toward an explicit spatial processes calibration. Hydrology and Earth
System Sciences 18, 5219–5237. https://doi.org/10.5194/hess-18-5219-2014
Finn G.A., Brun W.A., 1982. Effect of Atmospheric CO2 Enrichment on Growth, Nonstructural
Carbohydrate Content, and RootNoduleActivity in Soybean. Plant Physiology 69, 327–331. https://
doi.org/10.1104/pp.69.2.327
Fisher D.K., Elliot R.L., 1996. Modelling rangeland/pasture evapotranspiration using the Shuttleworth-
Wallace approach, in: Camp, C.R., Sadler, E.J., Yoder, R.E. (Eds.), Evapotranspiration and Irrigation S
cheduling Proceedings of the International Conference, 3-6/11/1996. American Society of Agricultural
Engineers, San Antonio, USA, pp. 664–672.
Flood J., 2010. The importance of plant health to food security. Food Security 2, 215–231. https://doi.
org/10.1007/s12571-010-0072-5
Fontaine S., Barot S., Barré P., Bdioui N., Mary B., Rumpel C., 2007. Stability of organic carbon in
deep soil layers controlled by fresh carbon supply. Nature 450, 277–280. https://doi.org/10.1038/
nature06275
Friend A.D., 1991. Use of a model of photosynthesis and leaf microenvironment to predict optimal
stomatal conductance and leaf nitrogen partitioning. Plant, Cell and Environment 14, 895–905.
https://doi.org/10.1111/j.1365-3040.1991.tb00958.x
Gabrielle B., Mary B., Roche R., Smith P., Gosse G., 2002. Simulation of carbon and nitrogen dynamics
in arable soils: A comparison of approaches. European Journal of Agronomy 18, 107–120. https://doi.
org/10.1016/S1161-0301(02)00099-0
García de Cortázar Atauri I., 2006. Adaptation du modèle STICS à la vigne (Vitis viniferaL.) :
Utilisation dans le cadre d’une étude d’impact du changement climatique à l’échelle de la France
(Thèse de doctorat). École nationale supérieure agronomique de Montpellier.
García de Cortázar Atauri I., Brisson N., Gaudillere J.P., 2009a. Performance of several models for
predicting budburst date of grapevine (Vitis viniferaL.). International Journal of Biometeorology 53,
317–326. https://doi.org/10.1007/s00484-009-0217-4
García de Cortázar Atauri I., Brisson N., Ollat N., Jacquet O., Payan J.C., 2009b. Asynchronous
dynamics of grapevine ("Vitis vinifera") maturation: Experimental study for a modelling approach.
Journal International des Sciences de la Vigne et du Vin 43, 83–97. https://doi.org/10.20870/
oeno-one.2009.43.2.801
García de Cortázar Atauri I., Brisson N., Seguin B., Gaudillère J.P., Baculat B., 2005. Simulation of
budbreak date for vine. TheBRIN model. Some applications in climate change study, in: Schultz,
H.R. (Ed.), Proceedings of the XIV International GESCO Viticulture Congress 2005, 23-27/08/2005.
Geisenheim, Germany, pp. 485–490.
Gardner W.R., 1991. Modeling water uptake by roots. Irrigation Science 12, 109–114. https://doi.
org/10.1007/BF00192281
Gary C.C., Daudet F.A., Cruiziat P., Brisson N., Ozier Lafontaine H., Bréda N., 1995. Le concept de
facteur limitant, in: Actes. Tome 1: De La Plante Au Couvert Végétal. INRA Editions, Le Croisic,
France, pp. 121–128.
Gary C., Jones J.W., Longuenesse J.J., 1993. Modelling daily changes in specific leaf area of tomato:
The contribution of the leaf assimilate pool. Acta Horticulturae 205–210. https://doi.org/10.17660/
ActaHortic.1993.328.19
Gateau C., Granger S., Kockmann F., Ruget F., Novak S., 2006. Evaluation du potentiel herbager et de
sa variabilité en élevage allaitant. Régionalisation de la démarche ISOP en Saône-et-Loire. Fourrages
186, 257–269.
Génermont S., Cellier P., 1997. A mechanistic model for estimating ammonia volatilization from
slurry applied to bare soil. Agricultural and Forest Meteorology 88, 145–167. https://doi.org/10.1016/
S0168-1923(97)00044-0
416
References
Gervois S., Noblet-Ducoudre N. de, Viovy N., Ciais P., Brisson N., Seguin B., Perrier A., 2004.
Including Croplands in a GlobalBiosphereModel: Methodology and Evaluation at SpecificSites. Earth
Interactions 8, 1–25. https://doi.org/10.1175/1087-3562(2004)8<1:ICIAGB>2.0.CO;2
Gerwitz A., Page E.R., 1974. An EmpiricalMathematicalModel to DescribePlantRootSystems. Journal
of Applied Ecology 11, 773–781. https://doi.org/10.2307/2402227
Giacomini S.J., Recous S., Mary B., Aita C., 2007. Simulating the effects of N availability, straw
particle size and location in soil on C and N mineralization. Plant and Soil 301, 289–301. https://doi.
org/10.1007/s11104-007-9448-5
Giauffret C., Derieux M., 1991. Variabilité génétique de la croissance de la jeune racine en fonction
de la température, in: Picard, D. (Ed.), Physiologie Et Production Du Maïs. Pau (France), 13-15 Nov.
1990. INRA Editions, pp. 69–74.
Girard M.L., 1997. Modélisation de l’accumulation de biomasse et d’azote dans les grains de blé tendre
d’hiver (Triticum aestivumL.) ; simulation de leur teneur en protéines à la récolte (Thèse de doctorat).
Institut National Agronomique Paris-Grignon.
Gonzalez-Camacho J.M., Mailhol J.C., Ruget F., 2008. Impacts locaux attendus de l’augmentation
du CO2 dans l’atmosphère sur la productivité de l’eau du maïs dans la Drôme, France. Irrigation and
Drainage 57, 229–243. https://doi.org/10.1002/ird.332
Gonzalez-Dugo V., Durand J.-L., Gastal F., 2010. Water deficit and nitrogen nutrition of crops. A
review. Agronomy for Sustainable Development 30, 529–544. https://doi.org/10.1051/agro/2009059
Gordon A.J., Ryle G.J.A., Mitchell D.F., Powell D.C.E., 1985. The Flux of 14C-LabelledPhotosynthate
through SoyabeanRootNodules during N2 Fixation. Journal of Experimental Botany 36, 756–769.
https://doi.org/10.1093/jxb/36.5.756
Goudriaan J., Gurney W.S.C., Nisbet R.M., Blythe S.P., 1986. Numerical approaches, in: Metz, J.A.J.,
Diekmann, O. (Eds.), The Dynamics of Physiologically Structured Populations, Lecture Notes in
Biomathematics. Springer, Berlin, Heidelberg, pp. 453–473.
Gras R., 1994. Sols caillouteux et production végétale, Mieux comprendre. INRA Editions, Versailles,
France.
Graux A.-I., 2005. Etude des mécanismes de remobilisation d’assimilats chez la betterave sucrière en
réponse aux stress abiotiques : Expérimentation et modélisation (Mémoire de DEA). ENSAIA.
Graux A.I., Resmond R., Casellas E., Delaby L., Faverdin P., Le Bas C., Ripoche D., Ruget F., Therond O.,
Vertes F., Peyraud J.L., 2020. High-resolution assessment of French grassland dry matter and nitrogen
yields. European Journal of Agronomy 112, 125952. https://doi.org/10.1016/j.eja.2019.125952
Grebet P., 1993. Functions, subroutines and a program for easy computation of commonly required
data related to the sun, in: Varlet-Grancher, C., Bonhomme, R., Sinoquet, H. (Eds.), Crop Structure
and Light Microclimate. pp. 501–511.
Greenwood D.J., Gastal F., Lemaire G., Draycott A., Millard P., Neeteson J.J., 1991. Growth Rate
and % N of FieldGrownCrops: Theory and Experiments. Annals of Botany 67, 181–190. https://doi.
org/10.1093/oxfordjournals.aob.a088118
Grosz B., Dechow R., Gebbert S., Hoffmann H., Zhao G., Constantin J., Raynal H., Wallach D.,
Coucheney E., Lewan E., Eckersten H., Specka X., Kersebaum K.-C., Nendel C., Kuhnert M.,
Yeluripati J., Haas E., Teixeira E., Bindi M., Trombi G., Moriondo M., Doro L., Roggero P.P., Zhao Z.,
Wang E., Tao F., Rötter R., Kassie B., Cammarano D., Asseng S., Weihermüller L., Siebert S., Gaiser T.,
Ewert F., 2017. The implication of input data aggregation on up-scaling soil organic carbon changes.
Environmental Modelling & Software 96, 361–377. https://doi.org/10.1016/j.envsoft.2017.06.046
Guérif M., Beaudoin N., Durr C., Houles V., Machet J.-M., Mary B., Moulin S., Richard G., 2001.
Designing a field experiment for assessing soil and crop spatial variability and defining site-specific
management strategies, in: Proceedings of the 3rd European Conference on Precision Agriculture,
18-20/06/2001. ENSAM, Montpellier, France, pp. 677–682.
Guérif M., Houlès V., Mary B., Beaudoin N., Machet J.-M., Moulin S., Nicoullaud B., 2007. Modulation
intra-parcellaire de la fertilisation azotée du blé fondée sur le modèle de culture Stics. Intérêt de
la démarche et méthodes de spatialisation, in: Guérif, M., King, D. (Eds.), Agriculture de Précision,
Update Sciences and Technologies. Editions Quae, pp. 225–248.
417
STICS soil-crop model
Guérin V., Pladys D., Trinchant J.-C., Rigaud J., 1991. Proteolysis and nitrogen fixation in faba-
bean (Vicia faba) nodules under water stress. Physiologia Plantarum 82, 360–366. https://doi.
org/10.1111/j.1399-3054.1991.tb02918.x
Guillaume S., Bergez J.E., Wallach D., Justes E., 2011. Methodological comparison of calibration
procedures for durum wheat parameters in the STICS model. European Journal of Agronomy 35,
115–126. https://doi.org/10.1016/j.eja.2011.05.003
Guyot G., 1997. Climatologie de l’environnement. De la plante aux écosystèmes. Masson.
Hallaire M., 1964. Le potentiel efficace de l’eau dans le sol en régime de dessèchement, in: L’eau Et La
Production Végétale. INRA, Paris, France, pp. 27–62.
Hammer G.L., Muchow R.C., 1994. Assessing climatic risk to sorghum production in water-limited
subtropical environments I. Development and testing of a simulation model. Field Crops Research 36,
221–234. https://doi.org/10.1016/0378-4290(94)90114-7
Hansen S., Jensen H.E., Nielsen N.E., Svendsen H., 1990. DAISY: ASoilPlantSystemModel. Danish
simulation model for transformation and transport of energy and matter in the soil plant atmosphere
system (NPO No. A10). Copenhagen.
He D., Wang E., 2019. The potential of using LAI time series to predict plant available water capacity
(PAWC) of soils, in: Agronomy Australia Conference. Agronomy Australia, Wagga Wagga.
He D., Wang E., Wang J., Robertson M.J., 2017. Data requirement for effective calibration of process-
based crop models. Agricultural and Forest Meteorology 234-235, 136–148. https://doi.org/10.1016/j.
agrformet.2016.12.015
He J., Dukes M.D., Jones J.W., Graham W.D., Judge J., 2009. Applying GLUE for EstimatingCERES-
MaizeGenetic and SoilParameters for SweetCornProduction. Transactions of the ASABE 52,
1907–1921. https://doi.org/10.13031/2013.29218
He J., Jones J.W., Graham W.D., Dukes M.D., 2010. Influence of likelihood function choice for esti-
mating crop model parameters using the generalized likelihood uncertainty estimation method.
Agricultural Systems 103, 256–264. https://doi.org/10.1016/j.agsy.2010.01.006
Heinen M., Mollier A., De Willigen P., 2003. Growth of a root system described as diffusion. II. Numerical
model and application. Plant and Soil 252, 251–265. https://doi.org/10.1023/A:1024749022761
Hidayat T., Koesmaryono Y., Impron, Ghulamahdi M., 2019. Intensifying of reflected radiation under
oil palm canopy and its effect on growth and production of soybean. IOP Conference Series: Earth and
Environmental Science 365, 012001. https://doi.org/10.1088/1755-1315/365/1/012001
Hoffmann H., Enders A., Siebert S., Gaiser T., Ewert F., 2016a. Climate and soil input data aggregation
effects in crop models. https://doi.org/10.7910/DVN/C0J5BB
Hoffmann H., Zhao G., Asseng S., Bindi M., Biernath C., Constantin J., Coucheney E., Dechow R.,
Doro L., Eckersten H., Gaiser T., Grosz B., Heinlein F., Kassie B.T., Kersebaum K.-C., Klein C.,
Kuhnert M., Lewan E., Moriondo M., Nendel C., Priesack E., Raynal H., Roggero P.P., Rötter R.P.,
Siebert S., Specka X., Tao F., Teixeira E., Trombi G., Wallach D., Weihermüller L., Yeluripati J., Ewert F.,
2016b. Impact of SpatialSoil and ClimateInputDataAggregation on RegionalYieldSimulations. PLOS
ONE 11, e0151782. https://doi.org/10.1371/journal.pone.0151782
Hoffmann H., Zhao G., van Bussel L.G.J., Enders A., Specka X., Sosa C., Yeluripati J., Tao F.,
Constantin J., Raynal H., Teixeira E., Grosz B., Doro L., Zhao Z., Wang E., Nendel C., Kersebaum K.C.,
Haas E., Kiese R., Klatt S., Eckersten H., Vanuytrecht E., Kuhnert M., Lewan E., Rotter R., Roggero P.P.,
Wallach D., Cammarano D., Asseng S., Krauss G., Siebert S., Gaiser T., Ewert F., 2015. Variability
of effects of spatial climate data aggregation on regional yield simulation by crop models. Climate
Research 65, 53–69. https://doi.org/10.3354/cr01326
Holzworth D.P., Huth N.I., deVoil P.G., 2011. Simple software processes and tests improve the relia-
bility and usefulness of a model. Environmental Modelling & Software 26, 510–516. https://doi.
org/10.1016/j.envsoft.2010.10.014
Holzworth D.P., Snow V., Janssen S., Athanasiadis I.N., Donatelli M., Hoogenboom G., White J.W.,
Thorburn P., 2015. Agricultural production systems modelling and software: Current status and
future prospects. Environmental Modelling & Software 72, 276–286. https://doi.org/10.1016/j.
envsoft.2014.12.013
418
References
Hooghoudt S.B., 1940. General consideration of the problem of field drainage by parallel drains,
ditches, watercourses and channels, in: Contribution to the Knowledge of Some Physical Parameters of
the Soil. Bodemkundig Instituut, Groningen, The Netherlands.
Hucl P., 1993. Effects of temperature and moisture stress on the germination of diverse common bean
genotypes. Canadian Journal of Plant Science 73, 697–702. https://doi.org/10.4141/cjps93-091
Humblot P., Leconte-Demarsy D., Clerino P., Szopa S., Castell J.-F., Jayet P.A., 2013. Assessment of
ozone impacts on farming systems: A bio-economic modeling approach applied to the widely diverse
French case. Ecological Economics 85, 50–58. https://doi.org/10.1016/j.ecolecon.2012.10.012
Hungria M., Vargas M.A.T., 2000. Environmental factors affecting N2 fixation in grain legumes in
the tropics, with an emphasis on Brazil. Field Crops Research 65, 151–164. https://doi.org/10.1016/
S0378-4290(99)00084-2
Hunt L.A., Pararajasingham S., 1995. CROPSIM — WHEAT: A model describing the growth and deve-
lopment of wheat. Canadian Journal of Plant Science 75, 619–632. https://doi.org/10.4141/cjps95-107
Huth N.I., Carberry P.S., Poulton P.L., Brennan L.E., Keating B.A., 2002. A framework for simula-
ting agroforestry options for the low rainfall areas of Australia using APSIM. European Journal of
Agronomy 18, 171–185. https://doi.org/10.1016/S1161-0301(02)00103-X
Hyvönen R., Ågren G., Linder S., Persson T., Cotrufo M., Ekblad A., Freeman M., Grelle A., Janssens I.,
Jarvis P., Kellomäki S., Lindroth A., Lundmark T., Norby R., Oren R., Pilegaard K., Ryan M.,
Sigurdsson B., Strömgren M., van Oijen M., Wallin G., 2007. The likely impact of elevated [CO2],
nitrogen deposition, increased temperature and management on carbon sequestration in tempe-
rate and boreal forest ecosystems: A literature review. New Phytologist 173, 463–480. https://doi.
org/10.1111/j.1469-8137.2007.01967.x
Idso S.B., 1991. A general relationship between CO2-induced increases in net photosynthesis and
concomitant reductions in stomatal conductance. Environmental and Experimental Botany 31,
381–383. https://doi.org/10.1016/0098-8472(91)90035-M
Idso S.B., Jackson R.D., Reginato R.J., 1978. Extending the "DegreeDay"Concept of
PlantPhenologicalDevelopment to IncludeWaterStressEffects. Ecology 59, 431–433. https://doi.
org/10.2307/1936570
INRA, 2018. Base de DonnéesGéographique des Sols de France à 1/1 000 000 version 3.2.8.0,
10/09/1998. https://doi.org/10.15454/BPN57S
IPCC, 2007. Climate Change 2007: The physical science basis. Contribution of working group I to the
fourth assessment report of the intergovernmental panel on climate change. Cambridge University
Press, Cambridge, United Kingdom ; New York, USA.
Itabari J.K., Gregory P.J., Jones R.K., 1993. Effects of Temperature, SoilWaterStatus and Depth of
Planting on Germination and Emergence of Maize (Zea mays) Adapted to Semi-AridEasternKenya.
Experimental Agriculture 29, 351–364. https://doi.org/10.1017/S0014479700020913
Itier B., Brisson N., Doussan C., Tournebize R., 1997. Bilan hydrique en agrométéorologie, in: Cruiziat,
P. (Ed.), Actes. Tome 2 : Du Couvert Végétal à La Région. INRA Editions, Le Croisic, France, pp. 383–397.
Jacquemoud S., Baret F., Hanocq J.F., 1992. Modeling spectral and bidirectional soil reflectance. Remote
Sensing of Environment 41, 123–132. https://doi.org/10.1016/0034-4257(92)90072-R
Jakeman A.J., Letcher R.A., Norton J.P., 2006. Ten iterative steps in development and evaluation of
environmental models. Environmental Modelling & Software 21, 602–614. https://doi.org/10.1016/j.
envsoft.2006.01.004
Jakku E., Thorburn P.J., 2010. A conceptual framework for guiding the participatory development of
agricultural decision support systems. Agricultural Systems 103, 675–682. https://doi.org/10.1016/j.
agsy.2010.08.007
Jamieson P.D., Brooking I.R., Porter J.R., Wilson D.R., 1995. Prediction of leaf appearance in wheat: A
question of temperature. Field Crops Research 41, 35–44. https://doi.org/10.1016/0378-4290(94)00102-I
Jayasundara H.P.S., Thomson B.D., Tang C., 1997. Responses of cool season grain legumes to soil
abiotic stresses, in: Sparks, D.L. (Ed.), Advances in Agronomy. Academic Press, pp. 77–151. https://doi.
org/10.1016/S0065-2113(08)60242-4
419
STICS soil-crop model
Jayet P.A., Petsakos A., 2013. Evaluating the Efficiency of a UniformN-InputTax under
DifferentPolicyScenarios at DifferentScales. Environmental Modeling & Assessment 18, 57–72.
https://doi.org/10.1007/s10666-012-9331-5
Jégo G., Chantigny M., Pattey E., Bélanger G., Rochette P., Vanasse A., Goyer C., 2014. Improved
snow-cover model for multi-annual simulations with the STICS crop model under cold, humid
continental climates. Agricultural and Forest Meteorology 195-196, 38–51. https://doi.org/10.1016/j.
agrformet.2014.05.002
Jégo G., Pattey E., Liu J., 2012a. Using LeafAreaIndex, retrieved from optical imagery, in the STICS
crop model for predicting yield and biomass of field crops. Field Crops Research 131, 63–74. https://
doi.org/10.1016/j.fcr.2012.02.012
Jégo G., Sánchez-Pérez J.M., Justes E., 2012b. Predicting soil water and mineral nitrogen contents
with the STICS model for estimating nitrate leaching under agricultural fields. Agricultural Water
Management 107, 54–65. https://doi.org/10.1016/j.agwat.2012.01.007
Jensen E.S., 1987. Seasonal patterns of growth and nitrogen fixation in field-grown pea. Plant and Soil
101, 29–37. https://doi.org/10.1007/BF02371027
Jeuffroy M.H., Barre C., Bouchard C., Demotes-Mainard S., Devienne-Barret F., Girard M.L.,
Recous S., 2000. Fonctionnement d’un peuplement de blé en conditions de nutrition azotée sub-
optimale, in: Bonhomme, R., Maillard, P. (Eds.), Fonctionnement Des Peuplements Végétaux Sous
Contraintes Environnementales, Colloques de l’INRA. INRA Editions, Paris, France, pp. 289–304.
Jeuffroy M.-H., Recous S., 1999. Azodyn: A simple model simulating the date of nitrogen deficiency
for decision support in wheat fertilization. European Journal of Agronomy 10, 129–144. https://doi.
org/10.1016/S1161-0301(98)00059-8
Jeuffroy M.-H., Warembourg F.R., 1991. Carbon Transfer and Partitioning between Vegetative and
ReproductiveOrgans in Pisum sativumL. Plant Physiology 97, 440–448. https://doi.org/10.1104/
pp.97.1.440
Jing Q., Huffman T., Shang J., Liu J., Pattey E., Morrison M., Jégo G., Qian B., 2017. Modelling soybean
yield responses to seeding date under projected climate change scenarios. Canadian Journal of Plant
Science 97, 1152–1164. https://doi.org/10.1139/cjps-2017-0065
Johnson I., Clark D., Clothier B., Giltrap D., Rissman C., Roskruge N., Thorburn P., White R., 2021.
Overseer whole-model review- Assessment of the model approach (MPI Technical Paper No.
2021/12). Ministry for Primary Industries; the Ministry for the Environment, Wellington New
Zealand.
Jones H.G., 2013. Plants and microclimate: A quantitative approach to environmental plant physiology,
3rd edition. Cambridge University Press, Cambridge. https://doi.org/10.1017/CBO9780511845727
Jones J.W., Antle J.M., Basso B., Boote K.J., Conant R.T., Foster I., Godfray H.C.J., Herrero M.,
Howitt R.E., Janssen S., Keating B.A., Munoz-Carpena R., Porter C.H., Rosenzweig C., Wheeler T.R.,
2017. Brief history of agricultural systems modeling. Agricultural Systems 155, 240–254. https://doi.
org/10.1016/j.agsy.2016.05.014
Jones J.W., Dayan E., Allen L.H., van Keulen H., Challa H., 1991. A dynamic tomato growth and yield
model (TOMGRO). Transactions of the ASAE 34, 0663–0672. https://doi.org/10.13031/2013.31715
Jones J.W., Hoogenboom G., Porter C.H., Boote K.J., Batchelor W.D., Hunt L.A., Wilkens P.W.,
Singh U., Gijsman A.J., Ritchie J.T., 2003. The DSSAT cropping system model. European Journal of
Agronomy 18, 235–265. https://doi.org/10.1016/S1161-0301(02)00107-7
Justes E., Beaudoin N., Bertuzzi P., Charles R., Constantin J., Durr C., Hermon C., Joannon A.,
Le Bas C., Mary B., Mignolet C., Montfort F., Ruiz L., Sarthou J.-P., Souchere V., Tournebize J.,
Rechauchère O., Savini I., Barbier M., Leiser H., 2013. Les cultures intermédiaires pour une produc-
tion agricole durable, Matière à débattre et décider. Editions Quae.
Justes E., Beaudoin N., Bertuzzi P., Charles R., Constantin J., Dürr C., Hermon C., Joannon A., Le
Bas C., Mary B., Mignolet C., Montfort F., Ruiz L., Sarthou J.-P., Souchère V., Tournebize J., Savini I.,
Réchauchère O., 2012. The use of cover crops to reduce nitrate leaching. Effect on the water and
nitrogen balance and other ecosystem services, in: Proceedings n°719, 7-8/12/2012. The International
Fertiliser Society, Leek, United Kingdom, pp. 27 p.
420
References
Justes E., Constantin J., Beaudoin N., Dürr C., Mary B., Ruiz L., 2017. Main lessons drawn from the
simulation study, in: Justes, E. (Ed.), Cover Crops for Sustainable Farming. Springer, Dordrecht, The
Netherlands. https://doi.org/10.1007/978-94-024-0986-4_3
Justes E., Jeuffroy M.H., Mary B., 1997. Wheat, barley, and durum wheat, in: Lemaire, G. (Ed.),
Diagnosis of the Nitrogen Status in Crops. Springer, Berlin, Heidelberg, pp. 73–91. https://doi.
org/10.1007/978-3-642-60684-7_4
Justes E., Mary B., Meynard J.M., Machet J.M., Thelier-Huche L., 1994. Determination of a
CriticalNitrogenDilutionCurve for WinterWheatCrops. Annals of Botany 74, 397–407. https://doi.
org/10.1006/anbo.1994.1133
Justes E., Mary B., Nicolardot B., 2009. Quantifying and modelling C and N mineralization kine-
tics of catch crop residues in soil: Parameterization of the residue decomposition module of STICS
model for mature and non mature residues. Plant and Soil 325, 171–185. https://doi.org/10.1007/
s11104-009-9966-4
Kage H., Ehlers W., 1996. Does transport of water to roots limit water uptake of field crops? Zeitschrift
für Pflanzenernährung und Bodenkunde 159, 583–590. https://doi.org/10.1002/jpln.1996.3581590609
Keating B.A., Thorburn P.J., 2018. Modelling crops and cropping systems—Evolving purpose, practice
and prospects. Recent advances in crop modelling to support sustainable agricultural production and
food security under global change 100, 163–176. https://doi.org/10.1016/j.eja.2018.04.007
Kersebaum K.C., Richter J., 1991. Modelling nitrogen dynamics in a plant-soil system with a simple
model for advisory purposes, in: Groot, J.J.R., De Willigen, P., Verberne, E.L.J. (Eds.), Nitrogen
Turnover in the Soil-Crop System, Developments in Plant and SoilSciences. Springer, Dordrecht, The
Netherlands, pp. 273–281. https://doi.org/10.1007/978-94-011-3434-7_12
Khalil K., Mary B., Renault P., 2004. Nitrous oxide production by nitrification and denitrification in
soil aggregates as affected by O2 concentration. Soil Biology and Biochemistry 36, 687–699. https://
doi.org/10.1016/j.soilbio.2004.01.004
Kiniry J.R., Williams J.R., Gassman P.W., Debaeke P., 1992. A general, process-oriented model for two
competing plant species. Transactions of the ASAE 35, 801–810. https://doi.org/10.13031/2013.28665
Kirschbaum M., 2006. The temperature dependence of organic-matter decomposition—still a topic
of debate. Soil Biology and Biochemistry 38, 2510–2518. https://doi.org/10.1016/j.soilbio.2006.01.030
Kizza T., Sekhar M., Rao S.M., Ruiz L., 2016. Model based estimation of turmeric yield response to
saline groundwater irrigation. International Journal of Scientific Research and Engineering Studies
(IJSRES) 3, 8–12.
Kollas C., Kersebaum K.C., Nendel C., Manevski K., Müller C., Palosuo T., Armas-Herrera C.M.,
Beaudoin N., Bindi M., Charfeddine M., Conradt T., Constantin J., Eitzinger J., Ewert F., Ferrise R.,
Gaiser T., García de Cortázar Atauri I., Giglio L., Hlavinka P., Hoffmann H., Hoffmann M.P., Launay M.,
Manderscheid R., Mary B., Mirschel W., Moriondo M., Olesen J.E., Öztürk I., Pacholski A., Ripoche-
Wachter D., Roggero P.P., Roncossek S., Rötter R.P., Ruget F., Sharif B., Trnka M., Ventrella D., Waha K.,
Wegehenkel M., Weigel H.-J., Wu L., 2015. Crop rotation modelling-AEuropean model intercompa-
rison. European Journal of Agronomy 70, 98–111. https://doi.org/10.1016/j.eja.2015.06.007
Kouchi H., Nakaji K., 1985. Utilization and Metabolism of Photoassimilated13CINSoybeanRoots and
Nodules. Soil Science and Plant Nutrition 31, 323–334. https://doi.org/10.1080/00380768.1985.1055
7440
Kuhnert M., Yeluripati J., Smith P., Hoffmann H., van Oijen M., Constantin J., Coucheney E.,
Dechow R., Eckersten H., Gaiser T., Grosz B., Haas E., Kersebaum K.-C., Kiese R., Klatt S., Lewan E.,
Nendel C., Raynal H., Sosa C., Specka X., Teixeira E., Wang E., Weihermüller L., Zhao G., Zhao Z.,
Ogle S., Ewert F., 2017. Impact analysis of climate data aggregation at different spatial scales on simu-
lated net primary productivity for croplands. European Journal of Agronomy 88, 41–52. https://doi.
org/10.1016/j.eja.2016.06.005
Kvaki’c M., Pellerin S., Ciais P., Achat D.L., Augusto L., Denoroy P., Gerber J.S., Goll D.,
Mollier A., Mueller N.D., Wang X., Ringeval B., 2018. Quantifying the Limitation to
WorldCerealProductionDueToSoilPhosphorusStatus. Global Biogeochemical Cycles 32, 143–157.
https://doi.org/10.1002/2017GB005754
421
STICS soil-crop model
Lacape M.J., Wery J., Annerose D.J.M., 1998. Relationships between plant and soil water status in five
field-grown cotton (Gossypium hirsutumL.) cultivars. Field Crops Research 57, 29–43. https://doi.
org/10.1016/S0378-4290(97)00111-1
Lagarias J.C., Reeds J.A., Wright M.H., Wright P.E., 1998. Convergence Properties of the Nelder–
MeadSimplexMethod in LowDimensions. SIAM Journal on Optimization 9, 112–147. https://doi.
org/10.1137/S1052623496303470
Lammoglia S.-K., Makowski D., Moeys J., Justes E., Barriuso E., Mamy L., 2017. Sensitivity analysis
of the STICS-MACRO model to identify cropping practices reducing pesticides losses. Science of The
Total Environment 580, 117–129. https://doi.org/10.1016/j.scitotenv.2016.10.010
Launay C., Constantin J., Chlebowski F., Houot S., Graux A.-I., Klumpp K., Martin R., Mary B.,
Pellerin S., Therond O., 2021. Estimating the carbon storage potential and greenhouse gas emissions
of French arable cropland using high-resolution modeling. Global Change Biology 27, 1645–1661.
https://doi.org/10.1111/gcb.15512
Launay M., Brisson N., 2004. STICS’ adaptability to a novel crop as an application of modularity in
crop modelling: Example of sugar beet, in: Jacobsen, S.E., Jensen, C.R., Porter, J.R. (Eds.), European
Agriculture in a Global Context, 8th ESA Congress, 11-15/07/2004. Royal Veterinary; Agricultural
University, Copenhagen, Denmark, pp. 283–284.
Launay M., Brisson N., Satger S., Hauggaard-Nielsen H., Corre-Hellou G., Kasynova E., Ruske R.,
Jensen E.S., Gooding M.J., 2009. Exploring options for managing strategies for pea-barley intercrop-
ping using a modeling approach. European Journal of Agronomy 31, 85–98. https://doi.org/10.1016/j.
eja.2009.04.002
Lawrie A.C., Wheeler C.T., 1974. The effects of flowering and fruit formation on the supply of photo-
synthetic assimilates to the nodules of Pisum sativum l. In relation to the fixation of nitrogen. New
Phytologist 73, 1119–1127. https://doi.org/10.1111/j.1469-8137.1974.tb02141.x
Le Bas C., 2016. Impact des incertitudes des paramètres sols estimés à partir de la Base de
DonnéesGéographique des Sols de France sur les sorties du modèle de culture Stics. Approche
méthodologique (Thèse de doctorat). AgroParisTech.
Lecharpentier P., Vezy Rémi, Buis S., Giner M., 2021b. SticsRFiles: Read and modify STICS input/
output files. https://doi.org/10.5281/ZENODO.4443206
Lecharpentier P., Vezy Rémi, Buis S., Giner M., 2021a. SticsOnR: Manage STICS simulations running
the executable orJavaStics. https://doi.org/10.5281/ZENODO.4443130
Lecharpentier P., Vezy Remi, SticsRPacks, 2021c. SticsRFiles: Read and modify STICS input/output
files.
Leclère D., Jayet P.A., Noblet-Ducoudré N. de, 2013. Farm-level AutonomousAdaptation of
EuropeanAgriculturalSupply to ClimateChange. Ecological Economics 87, 1–14. https://doi.
org/10.1016/j.ecolecon.2012.11.010
Lecoeur J., Sinclair T.R., 2001. Analysis of nitrogen partitioning in field pea resulting in linear
increase in nitrogen harvest index. Field Crops Research 71, 151–158. https://doi.org/10.1016/
S0378-4290(01)00151-4
Ledoux E., Gomez E., Monget J.M., Viavattene C., Viennot P., Ducharne A., Benoit M., Mignolet C.,
Schott C., Mary B., 2007. Agriculture and groundwater nitrate contamination in the Seine basin.
TheSTICS-MODCOU modelling chain. Science of The Total Environment 375, 33–47. https://doi.
org/10.1016/j.scitotenv.2006.12.002
Leenhardt D., Voltz M., Rambal S., 1995. A survey of several agroclimatic soil water balance models
with reference to their spatial application. European Journal of Agronomy 4, 1–14. https://doi.
org/10.1016/S1161-0301(14)80012-9
Lefebvre M.-P., 2010. Spatialisation de modèles de fonctionnement hydromécanique des sols appli-
quée à la prévision des risques de tassement à l’échelle de la France (Thèse de doctorat). Université
d’Orléans.
Lemaire G., Gastal F., 1997. N uptake and distribution in plant canopies, in: Lemaire, G. (Ed.),
Diagnosis of the Nitrogen Status in Crops. Springer, Berlin, Heidelberg, pp. 3–43. https://doi.
org/10.1007/978-3-642-60684-7_1
422
References
Lemaire G., Salette J., Sigogne M., Terrasson J.-P., 1984. Relation entre dynamique de croissance et
dynamique de prélèvement d’azote pour un peuplement de graminées fourragères. I. — Etude de l’effet
du milieu. Agronomie 4, 423–430. https://doi.org/10.1051/agro:19840503
Léonard J., 2016. Nitrification, denitrification and N2O emissions in STICS.
Levavasseur F., Mary B., Houot S., 2021. C and N dynamics with repeated organic amendments can
be simulated with the STICS model. Nutrient Cycling in Agroecosystems. https://doi.org/10.1007/
s10705-020-10106-5
Limaux F., Recous S., Meynard J.-M., Guckert A., 1999. Relationship between rate of crop growth
at date of fertiliser N application and fate of fertiliser N applied to winter wheat. Plant and Soil 214,
49–59. https://doi.org/10.1023/A:1004629511235
Loague K., Corwin D.L., 1996. Uncertainty in regional-scale assessments of non-point source pollutants,
in: Corwin, D.L., Loague, K. (Eds.), Applications of GIS to the Modeling of Non-Point Source Pollutants
in the Vadose Zone, SSSASpecialPublication. pp. 131–152. https://doi.org/10.2136/sssaspecpub48.c7
Loague K., Green R., 1991. Statistical and graphical methods for evaluating solute transport
models: Overview and application. Journal of Contaminant Hydrology 7, 51–73. https://doi.
org/10.1016/0169-7722(91)90038-3
Ludwig B., John B., Ellerbrock R., Kaiser M., Flessa H., 2003. Stabilization of carbon from maize in
a sandy soil in a long-term experiment. European Journal of Soil Science 54, 117–126. https://doi.
org/10.1046/j.1365-2389.2003.00496.x
Ludwig B., Schulz E., Rethemeyer J., Merbach I., Flessa H., 2007. Predictive modelling of C dyna-
mics in the long-term fertilization experiment at BadLauchstädt with the RothamstedCarbonModel.
European Journal of Soil Science 58, 1155–1163. https://doi.org/10.1111/j.1365-2389.2007.00907.x
Luo Z., Wang E., Sun O.J., 2010. Can no-tillage stimulate carbon sequestration in agricultural soils? A
meta-analysis of paired experiments. Agriculture Ecosystems & Environment 139, 224–231. https://doi.
org/10.1016/j.agee.2010.08.006
Maas S.J., 1993. Parameterized Model of GramineousCropGrowth: I. LeafArea and DryMassSimulation.
Agronomy Journal 85, 348–353. https://doi.org/10.2134/agronj1993.00021962008500020034x
MacDuff J.H., Jarvis S.C., Davidson I.A., 1996. Inhibition of N2 fixation by white clover (Trifolium
repensL.) At low concentrations of NO3– in flowing solution culture. Plant and Soil 180, 287–295.
https://doi.org/10.1007/BF00015312
Magid J., De Neergaard A., Brandt M., 2006. Heterogeneous distribution may substantially decrease initial
decomposition, long-term microbial growth and N-immobilization from high C-to-N ratio resources.
European Journal of Soil Science 57, 517–529. https://doi.org/10.1111/j.1365-2389.2006.00805.x
Maharjan G.R., Hoffmann H., Webber H., Srivastava A.K., Weihermueller L., Villa A., Coucheney E.,
Lewan E., Trombi G., Moriondo M., Bindi M., Grosz B., Dechow R., Kuhnert M., Doro L.,
Kersebaum K.-C., Stella T., Specka X., Nendel C., Constantin J., Raynal H., Ewert F., Gaiser T., 2019.
Effects of input data aggregation on simulated crop yields in temperate and Mediterranean climates.
European Journal of Agronomy 103, 32–46. https://doi.org/10.1016/j.eja.2018.11.001
Mailhol J.C., Ruelle P., Revol P., Delage L., Lescot J.M., 1996. Operative modelling for actual evapo-
transpiration assessment: Calibration methodology, in: Proceedings of the International Conference on
Evaporation and Irrigation Sheduling , 3-6/11/1996/. American Society of Agricultural Engineers, San
Antonio, Texas, USA, pp. 474–479.
Makowski D., Wallach D., Tremblay M., 2002. Using a Bayesian approach to parameter estimation;
comparison of the GLUE and MCMC methods. Agronomie 22, 191–203. https://doi.org/10.1051/
agro:2002007
Malone R.W., Meek D.W., Ma L., Jaynes D.B., Nolan B.T., Karlen D.L., Ahuja L.R., Ma L., 2011. Quality
assurance of weather data for agricultural system model input: A case study using the walnut creek
watershed in central Iowa, in: Methods of Introducing System Models into Agricultural Research. John
Wiley & Sons, Ltd, pp. 283–295. https://doi.org/10.2134/advagricsystmodel2.c10
Mary B., Beaudoin N., Justes E., Machet J.M., 1999. Calculation of nitrogen mineralization and leaching
in fallow soil using a simple dynamic model. European Journal of Soil Science 50, 549–566. https://doi.
org/10.1046/j.1365-2389.1999.00264.x
423
STICS soil-crop model
Mary B., Clivot H., Blaszczyk N., Labreuche J., Ferchaud F., 2020. Soil carbon storage and minerali-
zation rates are affected by carbon inputs rather than physical disturbance: Evidence from a 47-year
tillage experiment. Agriculture Ecosystems & Environment 299, 106972. https://doi.org/10.1016/j.
agee.2020.106972
Mary B., Guérif M., 2005. Effet du stress azoté sur la plante : Définition d’un indice de nutrition
instantané. Carry-le-Rouet, pp. 23–27.
McBratney A.B., Minasny B., Cattle S.R., Vervoort R.W., 2002. From pedotransfer functions to soil
inference. Geoderma 109, 41–73. https://doi.org/10.1016/S0016-7061(02)00139-8
McCann R., McFarland M.J., Witz J.A., 1991. Near-surface bare soil temperature model for biophy-
sical models. Transactions of the ASAE 34, 0748–0755. https://doi.org/10.13031/2013.31726
McMaster G.S., Klepper B., Rickman R.W., Wilhelm W.W., Willis W.O., 1991. Simulation of shoot
vegetative development and growth of unstressed winter wheat. Ecological Modelling 53, 189–204.
https://doi.org/10.1016/0304-3800(91)90156-U
McMillan W.D., Burgy R.H., 1960. Interception loss from grass. Journal of Geophysical Research 65,
2389–2394. https://doi.org/10.1029/JZ065i008p02389
Mesbah M., Pattey E., Jégo G., 2017. A model-based methodology to derive optimum nitrogen
rates for rainfed crops - a case study for corn using STICS in Canada. Computers and Electronics in
Agriculture 142, 572–584. https://doi.org/10.1016/j.compag.2017.11.011
Meurer K.H.E., Haddaway N.R., Bolinder M.A., Katterer T., 2018. Tillage intensity affects total SOC
stocks in boreo-temperate regions only in the topsoil-A systematic review using an ESM approach.
Earth-Science Reviews 177, 613–622. https://doi.org/10.1016/j.earscirev.2017.12.015
Meyer W.S., Green G.C., 1981. Plant indicators of wheat and soybean crop water stress. Irrigation
Science 2, 167–176. https://doi.org/10.1007/BF00257978
Mignolet C., Schott C., Benoit M., 2004. Spatial dynamics of agricultural practices on a basin terri-
tory: A retrospective study to implement models simulating nitrate flow. The case of the Seine basin.
Agronomie 24, 219–236. https://doi.org/10.1051/agro:2004015
Mignolet C., Schott C., Benoît M., 2007. Spatial dynamics of farming practices in the Seine basin:
Methods for agronomic approaches on a regional scale. Science of The Total Environment 375, 13–32.
https://doi.org/10.1016/j.scitotenv.2006.12.004
Milroy S.P., Goyne P.J., 1995. Leaf area development in barley—model construction and response to
soil moisture status. Australian Journal of Agricultural Research 46, 845–860. https://doi.org/10.1071/
AR9950845
Mondelaers K., Aertsens J., Van Huylenbroeck G., 2009. A meta-analysis of the differences in envi-
ronmental impacts between organic and conventional farming. British Food Journal 111, 1098–1119.
https://doi.org/10.1108/00070700910992925
Monteith J.L., 1972. Solar Radiation and Productivity in TropicalEcosystems. The Journal of Applied
Ecology 9, 747–766. https://doi.org/10.2307/2401901
Monteith J.L., 1965. Evaporation and environment. Cambridge University Press, pp. 205–234.
Montesino-San Martin M., Wallach D., Olesen J.E., Challinor A.J., Hoffman M.P., Koehler A.K.,
Rötter R.P., Porter J.R., 2018. Data requirements for crop modelling—Applying the learning curve
approach to the simulation of winter wheat flowering time under climate change. European Journal of
Agronomy 95, 33–44. https://doi.org/10.1016/j.eja.2018.02.003
Moore A.D., Eckard R.J., Thorburn P.J., Grace P.R., Wang E., Chen D., 2014. Mathematical modeling
for improved greenhouse gas balances, agro-ecosystems, and policy development: Lessons from the
Australian experience. WIREs Climate Change 5, 735–752. https://doi.org/10.1002/wcc.304
Moreau D., Pointurier O., Perthame L., Beaudoin N., Villerd J., Colbach N., 2021. Integrating plant-
plant competition for nitrogen into a 3D individual-based model simulating the effects of cropping
systems on weed dynamics. Field Crops Research 268. https://doi.org/10.1016/j.fcr.2021.108166
Moreau P., Ruiz L., Mabon F., Raimbault T., Durand P., Delaby L., Devienne S., Vertes F., 2012.
Reconciling technical, economic and environmental efficiency of farming systems in vulnerable areas.
Agriculture Ecosystems & Environment 147, 89–99. https://doi.org/10.1016/j.agee.2011.06.005
424
References
Moreau P., Viaud V., Parnaudeau V., Salmon-Monviola J., Durand P., 2013. An approach for global
sensitivity analysis of a complex environmental model to spatial inputs and parameters: A case study of
an agro-hydrological model. Environmental Modelling & Software 47, 74–87. https://doi.org/10.1016/j.
envsoft.2013.04.006
Morvan T., 1999. Quantification et modélisation des flux d’azote résultant de l’épandage de lisier
(Thèse de doctorat). Université de Paris VI.
Morvan T., Leterme P., 2001. Vers une prévision opérationnelle des flux d’azote résultant de l’épandage
de lisier : Paramétrage d’un modèle dynamique de Simulation des transformations de l’azote des lisiers
(STAL). Ingénieries eau-agriculture-territoires 26, 17–26.
Muchow R.C., Carberry P.S., 1990. Phenology and leaf-area development in a tropical grain sorghum.
Field Crops Research 23, 221–237. https://doi.org/10.1016/0378-4290(90)90056-H
Muchow R.C., Sinclair T.R., Bennett J.M., 1990. Temperature and SolarRadiationEffects on
PotentialMaizeYield across Locations. Agronomy Journal 82, 338–343. https://doi.org/10.2134/agronj
1990.00021962008200020033x
Mulder V., Lacoste M., Richer-De-Forges A., Arrouays D., 2016. GlobalSoilMapFrance: High-resolution
spatial modelling the soils of France up to two meter depth. Science of The Total Environment 573,
1352–1369. https://doi.org/10.1016/j.scitotenv.2016.07.066
Mumen M., 2006. Caractérisation du fonctionnement hydrique des sols à l’aide d’un modèle mécaniste
de transferts d’eau et de chaleur mis en œuvre en fonction des informations disponibles sur le sol
(Thèse de doctorat). Université d’Avignon et des Pays de Vaucluse.
Munier-Jolain N.G., Munier-Jolain N.M., Roche R., Ney B., Duthion C., 1998. Seed growth rate in grain
legumes I. Effect of photoassimilate availability on seed growth rate. Journal of Experimental Botany
49, 1963–1969. https://doi.org/10.1093/jxb/49.329.1963
Nelder J.A., 1961. The Fitting of a Generalization of the LogisticCurve. Biometrics 17, 89–110. https://
doi.org/10.2307/2527498
Nesme T., Brisson N., Lescourret F., Bellon S., Crete X., Plenet D., Habib R., 2006. Epistics: A dynamic
model to generate nitrogen fertilisation and irrigation schedules in apple orchards, with special atten-
tion to qualitative evaluation of the model. Agricultural Systems 90, 202–225. https://doi.org/10.1016/j.
agsy.2005.12.006
Nicolardot B., Parnaudeau V., Chabalier P., 2006. Effet de la température sur la minéralisation de diffé-
rents produits organiques apportés dans les sols, in: Findeling, A., Marlet, S., Montange, D., Saint
Macary, H. (Eds.), Impacts Du Recyclage Des Déchets Agro-Industriels : Actes de l’atelier de Clôture de
l’ATP, 1-2/07/2004. Cirad, Montpellier, France.
Nicolardot B., Recous S., Mary B., 2001. Simulation of C and N mineralisation during crop residue
decomposition: A simple dynamic model based on the C:N ratio of the residues. Plant and Soil 228,
83–103. https://doi.org/10.1023/A:1004813801728
Nicoullaud B., Couturier A., Beaudoin N., Mary B., Coutadeur C., King D., 2004. Modélisation spatiale
à l’échelle parcellaire des effets de la variabilité des sols et des pratiques culturales sur la pollution
nitrique agricole, in: Monestiez, P., Lardon, S., Seguin, B. (Eds.), Organisation Spatiale Des Activités
Agricoles Et Processus Environnementaux, Science Update. INRA Editions, Paris, France, pp. 143–161.
Nicoullaud B., King D., Tardieu F., 1994. Vertical distribution of maize roots in relation to permanent
soil characteristics. Plant and Soil 159, 245–254. https://doi.org/10.1007/BF00009287
Novak V., 1989. Determination of critical moisture contents of soils in the course of evapotranspira-
tion. Sov. Soil Sci. 122–126.
Ojeda J.J., Huth N., Holzworth D., Raymundo R., Zyskowski R.F., Sinton S.M., Michel A.J., Brown H.E.,
2021. Assessing errors during simulation configuration in crop models – A global case study using
APSIM-Potato. Ecological Modelling 458, 109703. https://doi.org/10.1016/j.ecolmodel.2021.109703
Ong C.K., 1983. Response to Temperature in a Stand of PearlMillet ( Pennisetum typhoidesS. & H.).
Journal of Experimental Botany 34, 322–336. https://doi.org/10.1093/jxb/34.3.322
Oorts K., Laurent F., Mary B., Thiébeau P., Labreuche J., Nicolardot B., 2007. Experimental and simu-
lated soil mineral N dynamics for long-term tillage systems in northern France. Soil and Tillage
Research 94, 441–456. https://doi.org/10.1016/j.still.2006.09.004
425
STICS soil-crop model
Otegui M.E., Nicolini M.G., Ruiz R.A., Dodds P.A., 1995. Sowing DateEffects on GrainYieldComponents
for DifferentMaizeGenotypes. Agronomy Journal 87, 29–33. https://doi.org/10.2134/agronj1995.000
21962008700010006x
Ozier-Lafontaine H., Lecompte F., Sillon J.F., 1999. Fractal analysis of the root architecture of
Gliricidia sepium for the spatial prediction of root branching, size and mass: Model development and
evaluation in agroforestry. Plant and Soil 209, 167–179. https://doi.org/10.1023/A:1004461130561
Pachepsky Y., Rawls W.J., 2004. Preface: Status of pedotransfer functions, in: Pachepsky, Y., Rawls,
W.J. (Eds.), Development of Pedotransfer Functions in Soil Hydrology, Developments in Soil Science.
Elsevier, Amsterdam, The Netherlands, p. 512.
Pagès L., Bécel C., Boukcim H., Moreau D., Nguyen C., Voisin A.-S., 2014. Calibration and evaluation
of ArchiSimple, a simple model of root system architecture. Ecological Modelling 290, 76–84. https://
doi.org/10.1016/j.ecolmodel.2013.11.014
Palacios E., Quevedo A.N., 1996. Irrigation scheduling using small evaporation pan and computer
program for developing countries, in: Proceedings of the International Conference on Evapotranspiration
and Irrigation Scheduling, 3-6/11/1996. American Society of Agricultural Engineers, San Antonio,
Texas, USA, pp. 1029–1034.
Pararajasingham S., Hunt L.A., 1991. Wheat SpikeTemperature in Relation to BaseTemperature
for GrainFillingDuration. Canadian Journal of Plant Science 71, 63–69. https://doi.org/10.4141/
cjps91-007
Parton W.J., Schimel D.S., Cole C.V., Ojima D.S., 1987. Analysis of
FactorsControllingSoilOrganicMatterLevels in GreatPlainsGrasslands. Soil Science Society of America
Journal 51, 1173–1179. https://doi.org/10.2136/sssaj1987.03615995005100050015x
Passioura J.B., 1996. Simulation Models: Science, SnakeOil, Education, or Engineering? Agronomy
Journal 88, 690–694. https://doi.org/10.2134/agronj1996.00021962008800050002x
Peart R.M., Jones J.W., Curry R.B., Boote K., Allen L.H., 1989. Impact of climate change on crop yield
in the southwestern USA: A simulation study, in: Smith, J.B., Tirpak, D.A. (Eds.), The Potential Effects
of Global Climate Change on the United States. Report to Congress. United States Environmental
Protection Agency, Washington, DC, USA, pp. 1–54.
Pellegrino A., Wery J., Lebon E., 2002. Crop management adaptation to water-limited environments,
in: Villalobos, F.J., Testi, L. (Eds.), Proceedings of the VII ESA Congress, 15-18/07/2002. Junta de
Andalucía, Consejería de Agricultura y Pesca, Cordoba, Spain, pp. 313–314.
Pellerin S., Bamière L., Launay C., Martin R., Schiavo M., Angers D., Augusto L., Balesdent J., Basile-
Doelsch I., Bellassen V., Cardinael R., Cécillon L., Ceschia E., Chenu C., Constantin J., Daroussin J.,
Delacote P., Delame N., Gastal F., Gilbert D., Graux A.-I., Guenet B., Houot S., Klumpp K., Letort E.,
Litrico I., Martin M., Menasseri-Aubry S., Meziere D., Morvan T., Mosnier C., Roger-Estrade J., Saint-
André L., Sierra J., Therond O., Viaud V., Grateau R., Le Perchec S., Savini I., Rechauchère O., 2020.
Stocker du carbone dans les sols français. Quel potentiel au regard de l’objectif 4 pour 1000 et à quel
coût ? (Other). INRA.
Peña-Cabriales J.J., Castellanos J.Z., 1993. Effects of water stress on N2 fixation and grain yield of
Phaseolus vulgarisL. Plant and Soil 152, 151–155. https://doi.org/10.1007/BF00016345
Penman H.L., 1948. Natural evaporation from open water, bare soil and grass. Proceedings of the
Royal Society of London. Series A. Mathematical and Physical Sciences 193, 120–145. https://doi.
org/10.1098/rspa.1948.0037
Perarnaud V., Seguin B., Malezieux E., Deque M., Loustau D., 2005. Agrometeorological research
and applications needed to prepare agriculture and forestry to 21st century climate change. Climatic
Change 70, 319–340. https://doi.org/10.1007/s10584-005-5953-9
Perfect E., Sukop M.C., Haszler G.R., 2002. Prediction of Dispersivity for UndisturbedSoilColumns
from WaterRetentionParameters. Soil Science Society of America Journal 66, 696–701. https://doi.
org/10.2136/sssaj2002.6960
Perrin A., Basset-Mens C., Huat J., Gabrielle B., 2017. The variability of field emissions is critical to
assessing the environmental impacts of vegetables: ABenin case-study. Journal of Cleaner Production
153, 104–113. https://doi.org/10.1016/j.jclepro.2017.03.159
426
References
Peyrard C., Ferchaud F., Mary B., Grehan E., Leonard J., 2017. Management Practices of miscanthus
x giganteus Strongly Influence Soil Properties and N2O Emissions Over the Long Term. Bioenergy
Research 10, 208–224. https://doi.org/10.1007/s12155-016-9796-1
Plaza-Bonilla D., Nolot J.-M., Raffaillac D., Justes E., 2015. Cover crops mitigate nitrate leaching in crop-
ping systems including grain legumes: Field evidence and model simulations. Agriculture Ecosystems &
Environment 212, 1–12. https://doi.org/10.1016/j.agee.2015.06.014
Porter C.H., Villalobos C., Holzworth D., Nelson R., White J.W., Athanasiadis I.N., Janssen S., Ripoche D.,
Cufi J., Raes D., Zhang M., Knapen R., Sahajpal R., Boote K., Jones J.W., 2014. Harmonization and
translation of crop modeling data to ensure interoperability. Environmental Modelling & Software 62,
495–508. https://doi.org/10.1016/j.envsoft.2014.09.004
Post D.F., Fimbres A., Matthias A.D., Sano E.E., Accioly L., Batchily A.K., Ferreira L.G., 2000. Predicting
soil albedo from soil color and spectral reflectance data. Soil Science Society of America Journal 64,
1027–1034. https://doi.org/10.2136/sssaj2000.6431027x
Powlson D.S., Hart P.B.S., Poulton P.R., Johnston A.E., Jenkinson D.S., 1992. Influence of soil type, crop
management and weather on the recovery of 15N-labelled fertilizer applied to winter wheat in spring.
The Journal of Agricultural Science 118, 83–100. https://doi.org/10.1017/S0021859600068040
Powlson D.S., Pruden T.L.G., Johnston A.E., Jenkinson D.S., 1986. The nitrogen cycle in the
BroadbalkWheatExperiment: Recovery and losses of 15N-labelled fertilizer applied in spring and
inputs of nitrogen from the atmosphere. The Journal of Agricultural Science 107, 591–609. https://doi.
org/10.1017/S0021859600069768
Priestley C.H.B., Taylor R.J., 1972. On the assessment of surface heat flux and evaporation using large-scale
parameters. Monthly Weather Review 100, 81–92. https://doi.org/10.1175/1520-0493(1972)100%3C008
1:OTAOSH%3E2.3.CO;2
Puech T., Schott C., Mignolet C., 2015. Modélisation de la pollution nitrique d’origine agricole des grands
aquifères du bassin de Seine-Normandie à l’échelle des masses d’eau. Vol. 1. Évolution des systèmes de
culture sur le bassin Seine- Normandie depuis les années 2000 : Construction d’une base de données
spatialisée sur les pratiques agricoles (Contract). Agence de l’Eau Seine Normandie (AESN).
Puntel L.A., Sawyer J.E., Barker D.W., Thorburn P.J., Castellano M.J., Moore K.J., VanLoocke A., Heaton E.A.,
Archontoulis S.V., 2018. A SystemsModelingApproach to ForecastCornEconomicOptimumNitrogenRate.
Frontiers in Plant Science 9, 436. https://doi.org/10.3389/fpls.2018.00436
Queyrel W., Habets F., Blanchoud H., Ripoche D., Launay M., 2016. Pesticide fate modeling in soils
with the crop model STICS: Feasibility for assessment of agricultural practices. Science of The Total
Environment 542, Part A, 787–802. https://doi.org/10.1016/j.scitotenv.2015.10.066
Quintana Segui P., Martin E., Habets F., Noilhan J., 2009. Improvement, calibration and validation of a
distributed hydrological model over France. Hydrology and Earth System Sciences 13, 163–181. https://
doi.org/10.5194/hess-13-163-2009
R Core Team, 2020. R: A language and environment for statistical computing. R Foundation for Statistical
Computing, Vienna, Austria.
Rakotovololona L., 2018. Quantification expérimentale et modélisation de la production, des flux d’eau
et d’azote en systèmes de culture biologiques (Thèse de doctorat). Institut agronomique, vétérinaire et
forestier de France.
Rao Kanneganti V., Fick G.W., 1991. A warm-season annual grass growth model parameterized for maize
and sudangrass. Agricultural Systems 36, 439–470. https://doi.org/10.1016/0308-521X(91)90071-H
Ratto M., Tarantola S., Saltelli A., 2001. Sensitivity analysis in model calibration: GSA-GLUE approach.
Computer Physics Communications 136, 212–224. https://doi.org/10.1016/S0010-4655(01)00159-X
Rebière B., 1996. Effet d’un excès d’eau sur la croissance d’une culture de blé d’hiver. Identification d’une
variable de stress et analyse de la réponse de la culture (Thèse de doctorat). Université Louis Pasteur de
Strasbourg, Strasbourg.
Recous S., Fresneau C., Faurie G., Mary B., 1988. The fate of labelled15N urea and ammonium nitrate
applied to a winter wheat crop: I. Nitrogen transformations in the soil. Plant and Soil 112, 205–214.
https://doi.org/10.1007/BF02139997
427
STICS soil-crop model
Recous S., Machet J.-M., 1999. Short-term immobilisation and crop uptake of fertiliser nitrogen
applied to winter wheat: Effect of date of application in spring. Plant and Soil 206, 137–149. https://
doi.org/10.1023/A:1004377006602
Recous S., Machet J.M., Mary B., 1992. The partitioning of fertilizer-N between soil and crop:
Comparison of ammonium and nitrate applications. Plant and Soil 144, 101–111. https://doi.
org/10.1007/BF00018850
Recous S., Robin D., Darwis D., Mary B., 1995. Soil inorganic N availability: Effect on
maize residue decomposition. Soil Biology and Biochemistry 27, 1529–1538. https://doi.
org/10.1016/0038-0717(95)00096-W
Refsgaard J.C., Henriksen H.J., Harrar W.G., Scholten H., Kassahun A., 2005. Quality assurance
in model based water management - review of existing practice and outline of new approaches.
Environmental Modelling & Software 20, 1201–1215. https://doi.org/10.1016/j.envsoft.2004.07.006
Rennie R.J., Kemp G.A., 1981. Dinitrogen fixation in pea beans (Phaseolus vulgaris) as affected by
growth stage and temperature regime. Canadian Journal of Botany 59, 1181–1188. https://doi.
org/10.1139/b81-161
Richard G., Brisson N., Lebonvallet S., Ripoche D., Boizard H., Défossez P., Chanzy A., Roger-
Estrade J., 2007. Eléments de description de la structure du sol introduits dans Stics, in: Séminaire
Stics. Annales 2007. 20-22/03/2007. INRA, Reims, France, pp. 152–155.
Richard G., Cellier P., 1998. Effect of tillage on bare soil energy balance and thermal regime: An expe-
rimental study. Agronomie 18, 163–181. https://doi.org/10.1051/agro:19980301
Richardson E.A., Seeley S.D., Walker D.R., 1974. A model for estimating the completion of rest for
’Redhaven’ and ’Elberta’ peach trees. HortScience 9, 331–332.
Richardson E.A., Seeley S.D., Walker D.R., Anderson J.L., Ashcroft G.L., 1975. Pheno-climatography
of spring peach bud development. HortScience 10, 236–237.
Rickman R.W., Waldman S.E., Klepper B., 1996. MODWht3: ADevelopment-
DrivenWheatGrowthSimulation. Agronomy Journal 88, 176–185. https://doi.org/10.2134/agronj199
6.00021962008800020010x
Riou C., Itier B., Seguin B., 1988. The influence of surface roughness on the simplified relationship
between daily evaporation and surface temperature. International Journal of Remote Sensing 9, 1529–
1533. https://doi.org/10.1080/01431168808954957
Ripoche D., Weiss M., Prevot L., 2001. Driving the STICS crop model by exogeneous values of leaf
area index. Application of remote sensing, in: Proceedings of the Second International Symposium on
Modelling Cropping System. Florence, Italy, pp. 169–170.
Ritchie J.T., 1985. A user-orientated model of the soil water balance in wheat, in: Day, W., Atkin, R.K.
(Eds.), Wheat Growth and Modelling, NATOASIScience (SeriesA: LifeSciences). Springer, Boston,
USA, pp. 293–305. https://doi.org/10.1007/978-1-4899-3665-3_27
Ritchie J.T., 1972. Model for predicting evaporation from a row crop with incomplete cover. Water
Resources Research 8, 1204–1213. https://doi.org/10.1029/WR008i005p01204
Ritchie J.T., Otter S., 1985. Description and performance of CERESWheat: A user-oriented wheat
yield model, in: ARS Wheat Yield Project, ARS. Natural Technology Information Service, Springfield,
USA, pp. 159–175.
Robert M., Thomas A., Sekhar M., Raynal H., Casellas E., Casel P., Chabrier P., Joannon A., Bergez J.E.,
2018. A dynamic model for water management at the farm level integrating strategic, tactical and
operational decisions. Environmental Modelling & Software 100, 123–135. https://doi.org/10.1016/j.
envsoft.2017.11.013
Robertson M.J., Fukai S., Ludlow M.M., Hammer G.L., 1993. Water extraction by grain sorghum in a
sub-humid environment. II. Extraction in relation to root growth. Field Crops Research 33, 99–112.
https://doi.org/10.1016/0378-4290(93)90096-6
Rodrigo A., Recous S., Neel C., Mary B., 1997. Modelling temperature and moisture effects on C–N
transformations in soils: Comparison of nine models. Ecological Modelling 102, 325–339. https://doi.
org/10.1016/S0304-3800(97)00067-7
428
References
Román Dobarco M., Cousin I., Le Bas C., Martin M.P., 2019. Pedotransfer functions for predicting avai-
lable water capacity in French soils, their applicability domain and associated uncertainty. Geoderma
336, 81–95. https://doi.org/10.1016/j.geoderma.2018.08.022
Rosenthal W.D., Arkin G.F., Shouse P.J., 1985. Water deficit effects on sorghum transpiration, in:
Proceedings of National Conference on Advances in Evapotranspiration. 16-17/12/1985. ASAE publica-
tion, Chicago, USA, pp. 159–169.
Rotter R.P., Palosuo T., Kersebaum K.C., Angulo C., Bindi M., Ewert F., Ferrise R., Hlavinka P.,
Moriondo M., Nendel C., Olesen J.E., Patil R.H., Ruget F., Takac J., Trnka M., 2012. Simulation of spring
barley yield in different climatic zones of Northern and CentralEurope: A comparison of nine crop
models. Field Crops Research 133, 23–36. https://doi.org/10.1016/j.fcr.2012.03.016
RStudio Team, 2021. RStudio: Integrated development environment for r. RStudio, PBC, Boston, MA.
Ruget F., Bethenod O., Combe L., 1996. Repercussions of increased atmospheric CO2 on maize
morphogenesis and growth for various temperature and radiation levels. Maydica 41, 181–191.
Ruget F., Brisson N., 2007. Modéliser l’impact de la sécheresse sur la production fourragère par le
modèle STICS, in: Journées AFPF 2007 – Productions Fourragères Et Adaptations 48 à La Sécheresse,
27-28/03/2007. Association Francophone pour les Prairies et les Fourrages, Rennes, France.
Ruget F., Clastre P., Moreau J.-C., Cloppet E., Souverain F., Lacroix B., Lorgeou J., 2012. Conséquences
possibles des changements climatiques sur la production fourragère en France. II. Exemples de quelques
systèmes d’élevage. Fourrages 211, 243–251.
Ruget F., Novak S., Granger S., 2006. Du modèle STICS au système ISOP pour estimer la production
fourragère. Adaptation à la prairie, application spatialisée. Fourrages 186, 241–256.
Saffih-Hdadi K., Mary B., 2008. Modeling consequences of straw residues export on soil organic carbon.
Soil Biology and Biochemistry 40, 594–607. https://doi.org/10.1016/j.soilbio.2007.08.022
Saltelli A., Aleksankina K., Becker W., Fennell P., Ferretti F., Holst N., Li S., Wu Q., 2019. Why so
many published sensitivity analyses are false: A systematic review of sensitivity analysis practices.
Environmental Modelling & Software 114, 29–39. https://doi.org/10.1016/j.envsoft.2019.01.012
Sau F., Boote K.J., McNair Bostick W., Jones J.W., Inés Mínguez M., 2004. Testing and
ImprovingEvapotranspiration and SoilWaterBalance of the DSSATCropModels. Agronomy Journal 96,
1243–1257. https://doi.org/10.2134/agronj2004.1243
Savary S., Ficke A., Aubertot J.-N., Hollier C., 2012. Crop losses due to diseases and their implications
for global food production losses and food security. Food Security 4, 519–537. https://doi.org/10.1007/
s12571-012-0200-5
Sawyer J., Nafziger E., Randall G., Bundy L., Rehm G., Joern B., 2006. Concepts and Rationale for
RegionalNitrogenRateGuidelines for Corn (Departmental Report). Iowa State University Extension,
Ames.
Schnebelen N., Nicoullaud B., Bourennane H., Couturier A., Verbeque B., Revalier C., Bruand A.,
Ledoux E., 2004. The STICS model to predict nitrate leaching following agricultural practices.
Agronomie 24, 423–435. https://doi.org/10.1051/agro:2004039
Schott C., Mimet A., Mignolet C.C., 2010. Création d’une base de données sur les pratiques agricoles
de Basse-Normandie de 1970 à 2006 (Rapport de contrat). INRA.
Schwartz M.D. (Ed.), 2013. Phenology: An integrative environmental science. Springer, Dordrecht, The
Netherlands.
Scopel E., Muller B., Arreola Tostado J.M., Chavez Guerra E., Maraux F., 1998. Quantifying and model-
ling the effects of a light crop residue mulch on the water balance: An application to rainfed maize in
western Mexico, in: Proceedings Og XVI World Congress of Soil Science, 20-26/08/1998. Montpellier,
France.
Seghieri J., Floret Ch., Pontanier R., 1995. Plant phenology in relation to water availability: Herbaceous
and woody species in the savannas of northern Cameroon. Journal of Tropical Ecology 11, 237–254.
https://doi.org/10.1017/S0266467400008713
Seguin B., Antonioletti R., 1988. Quelques éléments sur le climat du MontVentoux. Bulletin
Climatologique et Agroclimatologique du Vaucluse 17, 26–34.
429
STICS soil-crop model
Seguin B., Itier B., 1983. Using midday surface temperature to estimate daily evaporation
from satellite thermal IR data. International Journal of Remote Sensing 4, 371–383. https://doi.
org/10.1080/01431168308948554
Seidel S.J., Palosuo T., Thorburn P., Wallach D., 2018. Towards improved calibration of crop models
– Where are we now and where should we go? European Journal of Agronomy 94, 25–35. https://doi.
org/10.1016/j.eja.2018.01.006
Sellers W.D., 1965. Physical climatology. University of Chicago Press.
Sene K., 1994. Parameterisations for energy transfers from a sparse vine crop. Agricultural and Forest
Meteorology 71, 1–18. https://doi.org/10.1016/0168-1923(94)90097-3
Sheng M., Liu J., Zhu A.-X., Rossiter D.G., Liu H., Liu Z., Zhu L., 2019. Comparison of GLUE and
DREAM for the estimation of cultivar parameters in the APSIM-maize model. Agricultural and Forest
Meteorology 278, 107659. https://doi.org/10.1016/j.agrformet.2019.107659
Shili-Touzi I., Tourdonnet S. de, Launay M., Doré T., 2010. Does intercropping winter wheat (Triticum
aestivum) with red fescue (Festuca rubra) as a cover crop improve agronomic and environmental
performance? A modeling approach. Field Crops Research 116, 218–229. https://doi.org/10.1016/j.
fcr.2009.11.007
Shuttleworth W.J., Wallace J.S., 1985. Evaporation from sparse crops-an energy combination theory.
Quarterly Journal of the Royal Meteorological Society 111, 839–855. https://doi.org/10.1002/
qj.49711146910
Sierra J., Brisson N., Ripoche D., Deque M., 2010. Modelling the impact of thermal adaptation of soil
microorganisms and crop system on the dynamics of organic matter in a tropical soil under a climate
change scenario. Ecological Modelling 221, 2850–2858. https://doi.org/10.1016/j.ecolmodel.2010.08.031
Sierra J., Brisson N., Ripoche D., Noël C., 2003. Application of the STICS crop model to predict nitrogen
availability and nitrate transport in a tropical acid soil cropped with maize. Plant and Soil 256, 333–345.
https://doi.org/10.1023/A:1026106208320
Silva F.A.M., Naudin K., Corbeels M., Scopel E., Affholder F., 2019. Impact of conservation agriculture
on the agronomic and environmental performances of maize cropping under contrasting climatic condi-
tions of the BrazilianCerrado. Field Crops Research 230, 72–83. https://doi.org/10.1016/j.fcr.2018.10.009
Silva J.V., Giller K.E., 2020. Grand challenges for the 21st century: What crop models can and can’t (yet)
do. The Journal of Agricultural Science 158, 794–805. https://doi.org/10.1017/S0021859621000150
Sinclair T.R., 1986. Water and nitrogen limitations in soybean grain production I. Model development.
Field Crops Research 15, 125–141. https://doi.org/10.1016/0378-4290(86)90082-1
Sinclair T.R., de Wit C.T., 1976. Analysis of the Carbon and NitrogenLimitations to SoybeanYield.
Agronomy Journal 68, 319–324. https://doi.org/10.2134/agronj1976.00021962006800020021x
Sinclair T.R., Mosca G., Bona S., 1993. Simulation Analysis of Variation among Seasons in
WinterWheatYields in NorthernItaly. Journal of Agronomy and Crop Science 170, 202–207. https://doi.
org/10.1111/j.1439-037X.1993.tb01076.x
Sinclair T.R., Seligman N., 2000. Criteria for publishing papers on crop modeling. Field Crops Research
68, 165–172. https://doi.org/10.1016/S0378-4290(00)00105-2
Sinclair T.R., Seligman N.G., 1996. Crop Modeling: FromInfancy to Maturity. Agronomy Journal 88,
698–704. https://doi.org/10.2134/agronj1996.00021962008800050004x
Singels A., de Jager J.M., 1991. Refinement and validation of the PUTU wheat crop growth model 3.
Grain growth. South African Journal of Plant and Soil 8, 73–77. https://doi.org/10.1080/02571862.199
1.10634583
Smith M., Allen R., Pereira L., 1998. Revised FAO methodology for crop-water requirements, in:
Consultants Meeting on Management of Nutrients and Water in Rainfed Arid and Semi-Arid Areas,
26-29/05/1997, IAEA-TECDOC. International Atomic Energy Agency (IAEA), Vienne, Austria, pp.
51–58.
Sonohat G., Sinoquet H., Varlet-Grancher C., Rakocevic M., Jacquet A., Simon J.-C., Adam B., 2002.
Leaf dispersion and light partitioning in three-dimensionally digitized tall fescue-white clover mixtures.
Plant, Cell and Environment 25, 529–538. https://doi.org/10.1046/j.1365-3040.2002.00833.x
430
References
Spaeth S.C., Sinclair T.R., 1985. Linear Increase in SoybeanHarvestIndex during Seed-Filling.
Agronomy Journal 77, 207–211. https://doi.org/10.2134/agronj1985.00021962007700020008x
Spitters C.J.T., Toussaint H.A.J.M., Goudriaan J., 1986. Separating the diffuse and direct
component of global radiation and its implications for modeling canopy photosynthesis PartI.
Components of incoming radiation. Agricultural and Forest Meteorology 38, 217–229. https://doi.
org/10.1016/0168-1923(86)90060-2
Sprent J.I., Stephens J.H., Rupela O.P., 1988. Environmental effects on nitrogen fixation, in: Bliss,
F.A., Summerfield, R.J. (Eds.), World Crops: Cool Season Food Legumes, Current PlantScience and
Biotechnology in Agriculture. Springer, Dordrecht, The Netherlands, pp. 801–810. https://doi.
org/10.1007/978-94-009-2764-3_64
Sreelash K., Buis S., Sekhar M., Ruiz L., Tomer S.K., Guerif M., 2017. Estimation of available water
capacity components of two-layered soils using crop model inversion: Effect of crop type and water
regime. Journal of Hydrology 546, 166–178. https://doi.org/10.1016/j.jhydrol.2016.12.049
Sreelash K., Sekhar M., Ruiz L., Tomer S.K., Guerif M., Buis S., Durand P., Gascuel-Odoux C.,
2012. Parameter estimation of a two-horizon soil profile by combining crop canopy and surface
soil moisture observations using GLUE. Journal of Hydrology 456, 57–67. https://doi.org/10.1016/j.
jhydrol.2012.06.012
Stockle C.O., Kjelgaard J., 1996. Parameterizing Penman-Monteith surface resistance for estimating
daily crop ET, in: Camp, C.R., Sadler, E.J., Yoder, R.E. (Eds.), Proceedings of the International Conference
on Evapotranspiration and Irrigation Sheduling, 3-6/11/1996. American Society of Agricultural
Engineers, San Antonio, Texas, USA, pp. 632–638.
Stockle C.O., Williams J.R., Rosenberg N.J., Jones C.A., 1992. A method for estimating the direct
and climatic effects of rising atmospheric carbon dioxide on growth and yield of crops: PartI—
Modification of the EPIC model for climate change analysis. Agricultural Systems 38, 225–238.
https://doi.org/10.1016/0308-521X(92)90067-X
Stöckle C.O., Donatelli M., Nelson R., 2003. CropSyst, a cropping systems simulation model. European
Journal of Agronomy 18, 289–307. https://doi.org/10.1016/S1161-0301(02)00109-0
Strullu L., Beaudoin N., García de Cortázar Atauri I., Mary B., 2014. Simulation of Biomass and
NitrogenDynamics in PerennialOrgans and Shoots of Miscanthus × GiganteusUsing the STICSModel.
BioEnergy Research 7, 1253–1269. https://doi.org/10.1007/s12155-014-9462-4
Strullu L., Beaudoin N., Thiébeau P., Julier B., Mary B., Ruget F., Ripoche D., Rakotovololona L.,
Louarn G., 2020. Simulation using the STICS model of C&N dynamics in alfalfa from sowing to crop
destruction. European Journal of Agronomy 112. https://doi.org/10.1016/j.eja.2019.125948
Strullu L., Ferchaud F., Yates N., Shield I., Beaudoin N., García de Cortázar Atauri I., Besnard A.,
Mary B., 2015. Multisite YieldGapAnalysis of Miscanthus × giganteusUsing the STICSModel.
BioEnergy Research 8, 1735–1749. https://doi.org/10.1007/s12155-015-9625-y
Tao F., Rötter R.P., Palosuo T., DíazNANAAmbrona C.G.H., Mínguez M.I., Semenov M.A.,
Kersebaum K.C., Nendel C., Specka X., Hoffmann H., Ewert F., Dambreville A., Martre P.,
Rodríguez L., Ruiz-Ramos M., Gaiser T., Höhn J.G., Salo T., Ferrise R., Bindi M., Cammarano D.,
Schulma A.H., 2018. Contribution of crop model structure, parameters and climate projections to
uncertainty in climate change impact assessments. Global Change Biology 24, 1291–1307. https://
doi.org/10.1111/gcb.14019
Tardieu F., Granier C., Muller B., 1999. Modelling leaf expansion in a fluctuating environment: Are
changes in specific leaf area a consequence of changes in expansion rate. New Phytologist 143, 33–43.
https://doi.org/10.1046/j.1469-8137.1999.00433.x
Tavakoly A.A., Habets F., Saleh F., Yang Z.-L., Bourgeois C., Maidment D.R., 2019. An integrated
framework to model nitrate contaminants with interactions of agriculture, groundwater, and surface
water at regional scales: TheSTICS–EauDyssée coupled models applied over the SeineRiverBasin.
Journal of Hydrology 568, 943–958. https://doi.org/10.1016/j.jhydrol.2018.11.061
Teittinen M., Karvonen T., Peltonen J., 1994. A DynamicModel for Water and NitrogenLimitedGrowth
in SpringWheat to PredictYield and Quality. Journal of Agronomy and Crop Science 172, 90–103.
https://doi.org/10.1111/j.1439-037X.1994.tb00534.x
431
STICS soil-crop model
Tetegan M., Nicoullaud B., Baize D., Bouthier A., Cousin I., 2011. The contribution of rock fragments
to the available water content of stony soils: Proposition of new pedotransfer functions. Geoderma
165, 40–49. https://doi.org/10.1016/j.geoderma.2011.07.001
Texeira J.L., Liu Y., Zhang H.J., Pereira L.S., 1996. Evaluation of the ISAREG irrigation schedu-
ling model in north China plain, in: Camp, C.R., Sadler, E.J., Yoder, R.E. (Eds.), Proceedings of the
International Conference on Evapotranspiration and Irrigation Sheduling, 3-6/11/1996. American
Society of Agricultural Engineers, San Antonio, Texas, USA, pp. 632–638.
Theau J.-P., Coléno F.C., Duru M., Rauzy Y., 1998. L’utilisation de l’herbe paturée et fauchée en réfé-
rence au potentiel de production des prairies. Fourrages 156, 589–601.
Therond O., Tichit M., Tibi A., Accatino F., Biju-Duval L., Bockstaller C., Bohan D., Bonaudo T.,
Boval M., Cahuzac E., Casellas E., Chauvel B., Choler P., Constantin J., Cousin I., Daroussin J.,
David M., Delacote P., Derocles S., Sousa L. de, Domingues J.P., Dross C., Duru M., Eugène M.,
Fontaine C., Garcia B., Geijzendorffer I.R., Girardin A., Graux A.-I., Jouven M., Langlois B., Le Bas C.,
Le Bissonnais Y., Lelievre V., Lifran R., Maigne E., Martin G., Märtin R., Martin-Laurent F., Martinet V.,
Mclaughlin O., Meillet A., Mignolet C., Mouchet M., Nozieres-Petit M.-O., Ostermann O.P.,
Paracchini M.L., Pellerin S., Peyraud J.-L., Michaut S.P., Picaud C., Plantureux S., Poméon T., Porcher E.,
Puech T., Puillet L., Rambonilaza T., Raynal H., Resmond R., Ripoche D., Ruget F., Rulleau B., Rush A.,
Salles J.-M., Sauvant D., Schott C., Tardieu L., 2017. Volet "écosystèmes agricoles" de l’EvaluationFran-
çaise des Ecosystèmes et des ServicesEcosystémiques (Rapport d'expertise). INRA.
Thornley J., 1996. Modelling Water in Crops and PlantEcosystems. Annals of Botany 77, 261–275.
https://doi.org/10.1006/anbo.1996.0030
Thorsen S.M., Roer A.-G., Oijen M. van, 2010. Modelling the dynamics of snow cover, soil frost and
surface ice in Norwegian grasslands. Polar Research 29, 110–126. https://doi.org/10.3402/polar.
v29i1.6057
Tolk J.A., Howel T.A., Steiner J.L., 1996. Corn canopy resistance determined from whole plant trans-
piration, in: Camp, C.R., Sadler, E.J., Yoder, R.E. (Eds.), Proceedings of the International Conference
on Evapotranspiration and Irrigation Scheduling, 3-6/11/1996. American Society of Agricultural
Engineers, San Antonio, Texas, USA, pp. 664–672.
Toreti A., Deryng D., Tubiello F.N., Müller C., Kimball B.A., Moser G., Boote K., Asseng S., Pugh T.A.M.,
Vanuytrecht E., Pleijel H., Webber H., Durand J.-L., Dentener F., Ceglar A., Wang X., Badeck F., Lecerf R.,
Wall G.W., van den Berg M., Hoegy P., Lopez-Lozano R., Zampieri M., Galmarini S., O’NALeary G.J.,
Manderscheid R., Mencos Contreras E., Rosenzwei C., 2020. Narrowing uncertainties in the effects of
elevated CO2 on crops. Nature Food 1, 775–782. https://doi.org/10.1038/s43016-020-00195-4
Tóth B., Weynants M., Nemes A., Makó A., Bilas G., Tóth G., 2015. New generation of hydraulic pedo-
transfer functions for Europe. European Journal of Soil Science 66, 226–238. https://doi.org/10.1111/
ejss.12192
Tournebize J., Kao C., Nikolic N., Zimmer D., 2004. Adaptation of the STICS model to subsurface
drained soils. Agronomie 24, 305–313. https://doi.org/10.1051/agro:2004030
Trapani N., Hall A.J., Sadras V.O., Vilella F., 1992. Ontogenetic changes in radiation use effi-
ciency of sunflower (Helianthus annuusL.) crops. Field Crops Research 29, 301–316. https://doi.
org/10.1016/0378-4290(92)90032-5
Tremblay M., Wallach D., 2004. Comparison of parameter estimation methods for crop models.
Agronomie 24, 351–365. https://doi.org/10.1051/agro:2004033
Tribouillois H., Constantin J., Justes E., 2018b. Cover crops mitigate direct greenhouse gases balance
but reduce drainage under climate change scenarios in temperate climate with dry summers. Global
Change Biology 24, 2513–2529. https://doi.org/10.1111/gcb.14091
Tribouillois H., Constantin J., Justes E., 2018a. Analysis and modeling of cover crop emergence:
Accuracy of a static model and the dynamic STICS soil-crop model. European Journal of Agronomy 93,
73–81. https://doi.org/10.1016/j.eja.2017.12.004
Tribouillois H., Cruz P., Cohan J.-P., Justes E., 2015. Modelling agroecosystem nitrogen functions
provided by cover crop species in bispecific mixtures using functional traits and environmental factors.
Agriculture, Ecosystems & Environment 207, 218–228. https://doi.org/10.1016/j.agee.2015.04.016
432
References
Trnka M., Kocmánková E., Balek J., Eitzinger J., Ruget F., Formayer H., Hlavinka P., Schaumberger A.,
Horáková V., Možný M., Žalud Z., 2010. Simple snow cover model for agrometeorological applications.
Agricultural and Forest Meteorology 150, 1115–1127. https://doi.org/10.1016/j.agrformet.2010.04.012
Valé M., 2006. Quantification et prédiction de la minéralisation nette de l’azote du sol in situ, sous
divers pédoclimats et systèmes de culture français (Thèse de doctorat). Institut National Polytechnique
de Toulouse.
van Bavel C.H.M., 1953. A drought criterion and its application in evaluating drought incidence and
hazard. Agronomy Journal 45, 167–172. https://doi.org/10.2134/agronj1953.00021962004500040009x
van Keulen H., Seligman N.G., 1987. Simulation of water use, nitrogen nutrition and growth of a spring
wheat crop, Simulation monographs. Pudoc, Wageningen, The Netherlands.
Van Der Ploeg R.R., Ringe H., Machulla G., 1995. Late FallSite‐SpecificSoilNitrateUpperLimits
for GroundwaterProtectionPurposes. Journal of Environmental Quality 24, 725–733. https://doi.
org/10.2134/jeq1995.00472425002400040025x
Van Ittersum M.K., Leffelaar P.A., van Keulen H., Kropff M.J., Bastiaans L., Goudriaan J., 2003. On
approaches and applications of the Wageningen crop models. European Journal of Agronomy 18,
201–234. https://doi.org/10.1016/S1161-0301(02)00106-5
Varella H., Guérif M., Buis S., 2010a. Global sensitivity analysis measures the quality of parameter
estimation: The case of soil parameters and a crop model. Environmental Modelling & Software 25,
310–319. https://doi.org/10.1016/j.envsoft.2009.09.012
Varella H., Guérif M., Buis S., Beaudoin N., 2010b. Soil properties estimation by inversion of a crop
model and observations on crops improves the prediction of agro-environmental variables. European
Journal of Agronomy 33, 139–147. https://doi.org/10.1016/j.eja.2010.04.005
Varlet-Grancher C., Bonhomme R., 1979. Application aux couverts végétaux des lois de rayonnement
en milieu diffusant. II - Interception de l’énergie solaire par une culture. Annales Agronomiques 30, 1.
Varlet-Grancher C., Bonhomme R., Chartier M., Artis P., 1982. Efficience de la conversion de l’énergie
solaire par un couvert végétal. Acta Oecologica Oecologia Plantarum 3, 3–26.
Varlet-Grancher C., Bonhomme R., Chartier M., Artis P., 1981. Évolution de la réponse photosyn-
thétique des feuilles et efficience théorique de la photosynthèse brute d’une culture de canne à sucre
(Saccharum officinarumL.). Agronomie 1, 473–481. https://doi.org/10.1051/agro:19810607
Varlet-Grancher C., Bonhomme R., Sinoquet H. (Eds.), 1993. Crop structure and light microclimate.
Characterization and applications, Science Update. INRA Editions.
Varlet-Grancher C., Gosse G., Chartier M., Sinoquet H., Bonhomme R., Allirand J.M., 1989. Mise au
point : Rayonnement solaire absorbé ou intercepté par un couvert végétal. Agronomie 9, 419–439.
https://doi.org/10.1051/agro:19890501
Vehtari A., Gelman A., Simpson D., Carpenter B., Bürkner P.-C., 2021. Rank-Normalization, Folding,
and Localization: AnImprovedR for AssessingConvergence of MCMC. Bayesian Analysis 16, 667–718.
https://doi.org/10.1214/20-BA1221
Vereecken H., Weynants M., Javaux M., Pachepsky Ya.A., Schaap M.G., van Genuchten M.Th., 2010.
Using PedotransferFunctions to Estimate the van Genuchten-MualemSoilHydraulicProperties:
AReview. Vadose Zone Journal 9, 795–820. https://doi.org/10.2136/vzj2010.0045
Vezy R., Buis S., Lecharpentier P., Giner M., 2021. CroPlotR: A package to analyse crop model simula-
tions outputs with plots and statistics.
Vinten A.J.A., Redman M.H., 1990. Calibration and validation of a model of non-interac-
tive solute leaching in a clay-loam arable soil. Journal of Soil Science 41, 199–214. https://doi.
org/10.1111/j.1365-2389.1990.tb00057.x
Voisin A.S., 2003. Symbiotic N2 fixation activity in relation to C economy of Pisum sativumL. As a
function of plant phenology. Journal of Experimental Botany 54, 2733–2744. https://doi.org/10.1093/
jxb/erg290
Voisin A.-S., Salon C., Munier-Jolain N.G., Ney B., 2002. Quantitative effects of soil nitrate, growth
potential and phenology on symbiotic nitrogen fixation of pea (Pisum sativumL.). Plant and Soil 243,
31–42. https://doi.org/10.1023/A:1019966207970
433
STICS soil-crop model
Vrugt J.A., 2016. Markov chain MonteCarlo simulation using the DREAM software package: Theory,
concepts, and MATLAB implementation. Environmental Modelling & Software 75, 273–316. https://
doi.org/10.1016/j.envsoft.2015.08.013
Wagenet R.J., Hutson J.L., 1996. Scale-dependency of solute transport modeling/GIS applications.
Journal of Environmental Quality 25, 499–510. https://doi.org/10.2134/jeq1996.00472425002500030
015x
Wallace J.S., 1995. Towards a coupled light partitioning and transpiration model for use in intercrops
and agroforestry.
Wallach D., Buis S., Lecharpentier P., Bourges J., Clastre P., Launay M., Bergez J.E., Guérif M.,
Soudais J., Justes E., 2011. A package of parameter estimation methods and implementation for the
STICS crop-soil model. Environmental Modelling & Software 26, 386–394. https://doi.org/10.1016/j.
envsoft.2010.09.004
Wallach D., Makowski D., Jones J.W., Brun F., 2018. Model evaluation, in: Wallach, D., Makowski, D.,
Jones, J.W., Brun, F. (Eds.), Working with Dynamic Crop Models: Methods, Tools and Examples for
Agriculture and Environment. Academic Press, San Diego, CA, USA, pp. 311–373.
Wallach Daniel, Makowski D., Jones J.W., Brun F., 2018. Working with dynamic crop models: Methods,
tools and examples for agriculture and environment, Third edition. Academic Press, London San
Diego, CA Cambridge, MA Oxford.
Wallach D., Palosuo T., Thorburn P., Hochman Z., Gourdain E., Andrianasolo F., Asseng S., Basso B.,
Buis S., Crout N., Dibari C., Dumont B., Ferrise R., Gaiser T., Garcia C., Gayler S., Ghahramani A.,
Hiremath S., Hoek S., Horan H., Hoogenboom G., Huang M., Jabloun M., Jansson P.-E., Jing Q.,
Justes E., Kersebaum K.C., Klosterhalfen A., Launay M., Lewan E., Luo Q., Maestrini B., Mielenz H.,
Moriondo M., Nariman Zadeh H., Padovan G., Olesen J.E., Poyda A., Priesack E., Pullens J.W.M.,
Qian B., Schütze N., Shelia V., Souissi A., Specka X., Srivastava A.K., Stella T., Streck T., Trombi G.,
Wallor E., Wang J., Weber T.K.D., Weihermüller L., de Wit A., Wöhling T., Xiao L., Zhao C., Zhu Y.,
Seidel S.J., 2021. The chaos in calibrating crop models: Lessons learned from a multi-model cali-
bration exercise. Environmental Modelling & Software 145, 105206. https://doi.org/10.1016/j.
envsoft.2021.105206
Warembourg F.R., 1983. Estimating the true cost of dinitrogen fixation by nodulated plants in undis-
turbed conditions. Canadian Journal of Microbiology 29, 930–937. https://doi.org/10.1139/m83-149
Warren-Wilson J., 1972. Control of crop processes, in: Rees, A.R., Cockshull, K.E., Hand, D.W., Hurd,
R.G. (Eds.), Crop Processes in Controlled Environments. Academic Press, London; New York, pp. 7–30.
Waterer J.G., Vessey J.K., 1993. Effect of low static nitrate concentrations on mineral nitrogen uptake,
nodulation, and nitrogen fixation in field pea. Journal of Plant Nutrition 16, 1775–1789. https://doi.
org/10.1080/01904169309364649
Weaich K., Cass A., Bristow K.L., 1996. Pre-emergent shoot growth of maize (Zea mays, L.) As a function
of soil strength. Soil and Tillage Research 40, 3–23. https://doi.org/10.1016/S0167-1987(96)80003-1
Weir A.H., Bragg P.L., Porter J.R., Rayner J.H., 1984. A winter wheat crop simulation model without
water or nutrient limitations. The Journal of Agricultural Science 102, 371–382. https://doi.org/10.1017/
S0021859600042702
Whisler F.D., Acock B., Baker D.N., Fye R.E., Hodges H.F., Lambert J.R., Lemmon H.E., McKinion J.M.,
Reddy V.R., 1986. Crop simulation models in agronomic systems, in: Brady, N.C. (Ed.), Advances in
Agronomy. Academic Press, pp. 141–208. https://doi.org/10.1016/S0065-2113(08)60282-5
Williams J., C. A. Jones, P. T. Dyke, 1984. A modeling approach to determining the relation-
ship between erosion and soil productivity. Transactions of the ASAE 27, 0129–0144. https://doi.
org/10.13031/2013.32748
Williams J.R., Jones C.A., Kiniry J.R., Spanel D.A., 1989. The EPIC crop growth model. Transactions of
the ASAE 32, 0497–0511. https://doi.org/10.13031/2013.31032
Wilson J.B., 1988. A Review of Evidence on the Control of Shoot: RootRatio, in Relation to Models.
Annals of Botany 61, 433–449. https://doi.org/10.1093/oxfordjournals.aob.a087575
Wösten J.H.M., Lilly A., Nemes A., Le Bas C., 1999. Development and use of a database of hydraulic
properties of European soils. Geoderma 90, 169–185. https://doi.org/10.1016/S0016-7061(98)00132-3
434
References
Wösten J.H.M., Pachepsky Ya.A., Rawls W.J., 2001. Pedotransfer functions: Brifging the gap between
available basic soil data and missing soil hydraulic characteristics. Journal of Hydrology 251, 123–150.
https://doi.org/10.1016/S0022-1694(01)00464-4
Wu X., Vuichard N., Ciais P., Viovy N., Noblet-Ducoudre N. de, Wang X., Magliulo V., Wattenbach M.,
Vitale L., Tommasi P.D., Moors E.J., Jans W., Elbers J., Ceschia E., Tallec T., Bernhofer C., Gruenwald T.,
Moureaux C., Manise T., Ligne A., Cellier P., Loubet B., Larmanou E., Ripoche D., 2016. ORCHIDEE-
CROP (v0), a new process-based agro-land surface model: Model description and evaluation over
Europe. Geoscientific Model Development 9, 857–873. https://doi.org/10.5194/gmd-9-857-2016
Xie Y., 2021. Bookdown: Authoring books and technical documents with r markdown.
Xie Y., 2016. Bookdown: Authoring books and technical documents with R markdown. Chapman;
Hall/CRC, Boca Raton, Florida.
Yin X., Beaudoin N., Ferchaud F., Mary B., Strullu L., Chlébowski F., Clivot H., Herre C., Duval J.,
Louarn G., 2020. Long-term modelling of soil N mineralization and N fate using STICS in a 34-year
crop rotation experiment. Geoderma 357, 113956. https://doi.org/10.1016/j.geoderma.2019.113956
Yin X.G., Kersebaum K.C., Beaudoin N., Constantin J., Chen F., Louarn G., Manevski K., Hoffmann M.,
Kollas C., Armas-Herrera C.M., Baby S., Bindi M., Dibari C., Ferchaud F., Ferrise R., García de Cortázar
Atauri I., Launay M., Mary B., Moriondo M., Ozturk I., Ruget F., Sharif B., Wachter-Ripoche D.,
Olesen J.E., 2020. Uncertainties in simulating N uptake, net N mineralization, soil mineral N and N
leaching in European crop rotations using process-based models. Field Crops Research 255. https://
doi.org/10.1016/j.fcr.2020.107863
Zhao G., Hoffmann H., van Bussel L.G.J., Enders A., Specka X., Sosa C., Yeluripati J., Tao F.,
Constantin J., Raynal H., Teixeira E., Grosz B., Doro L., Zhao Z., Nendel C., Kiese R., Eckersten H.,
Haas E., Vanuytrecht E., Wang E., Kuhnert M., Trombi G., Moriondo M., Bindi M., Lewan E.,
Bach M., Kersebaum K.-C., Roetter R., Roggero P.P., Wallach D., Cammarano D., Asseng S., Krauss G.,
Siebert S., Gaiser T., Ewert F., 2015. Effect of weather data aggregation on regional crop simulation
for different crops, production conditions, and response variables. Climate Research 65, 141–157.
https://doi.org/10.3354/cr01301
Zhao G., Hoffmann H., Yeluripati J., Xenia S., Nendel C., Coucheney E., Kuhnert M., Tao F.,
Constantin J., Raynal H., Teixeira E., Grosz B., Doro L., Kiese R., Eckersten H., Haas E., Cammarano D.,
Kassie B., Moriondo M., Trombi G., Bindi M., Biernath C., Heinlein F., Klein C., Priesack E., Lewan E.,
Kersebaum K.-C., Rotter R., Roggero P.P., Wallach D., Asseng S., Siebert S., Gaiser T., Ewert F., 2016.
Evaluating the precision of eight spatial sampling schemes in estimating regional means of simu-
lated yield for two crops. Environmental Modelling & Software 80, 100–112. https://doi.org/10.1016/j.
envsoft.2016.02.022
435
Chapter 17
Definition of symbols
437
STICS soil-crop model
438
Definition of symbols
439
STICS soil-crop model
440
Definition of symbols
441
STICS soil-crop model
442
Definition of symbols
443
STICS soil-crop model
444
Definition of symbols
445
STICS soil-crop model
446
Definition of symbols
447
STICS soil-crop model
Concentration of mineral N
𝐜𝐨𝐧𝐜𝐫𝐫C kg ha−1 mm−1
(NH4+NO3-N) in the rain
Minimum concentration of NO3-N in soil
𝐜𝐨𝐧𝐜𝐬𝐞𝐮𝐢𝐥S kg ha−1 mm−1
(unavailable for leaching and for uptake)
Maximal reduction factor applied to root
𝐜𝐨𝐧𝐭𝐫𝐝𝐚𝐦𝐚𝐱P growth rate due to soil strengthness (high ND
bulk density)
Temperature to substract to Tmin to
𝐜𝐨𝐫𝐞𝐜𝐓𝐫𝐨𝐬𝐞𝐞C estimate dew point temperature (in case of °C
missing air humidity data)
448
Definition of symbols
449
STICS soil-crop model
450
Definition of symbols
451
STICS soil-crop model
452
Definition of symbols
453
STICS soil-crop model
454
Definition of symbols
455
STICS soil-crop model
456
Definition of symbols
457
STICS soil-crop model
458
Definition of symbols
459
STICS soil-crop model
460
Definition of symbols
461
STICS soil-crop model
462
Definition of symbols
463
STICS soil-crop model
464
Definition of symbols
465
STICS soil-crop model
466
Definition of symbols
467
STICS soil-crop model
468
Definition of symbols
469
STICS soil-crop model
470
Definition of symbols
471
STICS soil-crop model
472
Definition of symbols
473
STICS soil-crop model
474
Definition of symbols
475
STICS soil-crop model
476
Definition of symbols
477
STICS soil-crop model
478
Definition of symbols
479
STICS soil-crop model
480
Definition of symbols
481
STICS soil-crop model
482
Definition of symbols
483
STICS soil-crop model
484
Definition of symbols
485
STICS soil-crop model
486
Definition of symbols
487
STICS soil-crop model
488
Definition of symbols
489
STICS soil-crop model
490
Definition of symbols
491
STICS soil-crop model
492
Definition of symbols
493
STICS soil-crop model
494
Definition of symbols
495
STICS soil-crop model
496
Definition of symbols
497
STICS soil-crop model
498
Definition of symbols
499
STICS soil-crop model
500
Definition of symbols
501
STICS soil-crop model
502
Definition of symbols
503
STICS soil-crop model
504
Definition of symbols
505
STICS soil-crop model
506
Definition of symbols
507
STICS soil-crop model
508
Definition of symbols
509
STICS soil-crop model
510
Definition of symbols
511
STICS soil-crop model
512
Definition of symbols
513
STICS soil-crop model
514
Authors list
François Affholder Benjamin Dumont
CIRAD, AIDA Liege University, Gembloux Agro-Bio Tech
34398, Montpellier cedex 5, NA, Gembloux
FRANCE BELGIUM
francois.affholder@cirad.fr benjamin.dumont@ulg.ac.be
515
STICS soil-crop model
516
The STICS soil-crop model simulates the production and environmental
impacts of cropping systems under current and changing climatic conditions.
It synthesises a significant share of agronomic knowledge about field and
cropping system behaviour. The model algorithms presented in this book
mainly draw from literature references in the bioclimatology, soil and crop
sciences; the model also considers most management practices. The book
is structured to reflect how the crop-soil system is modelled according to
weather conditions and farming practices. Each chapter focuses on a set
of important functions, such as plant functioning (development initiation,
growth, yield formation, water uptake and transpiration) and soil processes
of the water cycle and organic matter transformation and associated fluxes.
This new edition integrates all the STICS updates that have been developed,
evaluated and published over the last decade. Thanks to the innovative
editorial design of this digital format, modellers can develop a research
version of the model and users can track changes in the current version.
While this book is primarily intended for soil-crop modellers and agronomists
using STICS, its many illustrative examples of the processes involved and case
studies of crops and cropping systems may be of interest to a wider audience.
59 €
ISBN : 978-2-7592-3678-7
Réf. : 02881