Agricultural Water Management 218 (2019) 1–7

Contents lists available at ScienceDirect

Agricultural Water Management

journal homepage: www.elsevier.com/locate/agwat

Soybean irrigation requirements and canopy-atmosphere coupling in T

Southern Brazil
Evandro H.F.M. da Silvaa, Alexandre O. Gonçalvesa,b, Rodolfo A. Pereiraa,
⁎
Izael M. Fattori Júniora, Luiz R. Sobenkoa, Fábio R. Marina,
a
University of São Paulo (USP), College of Agriculture “Luiz de Queiroz” (ESALQ), Piracicaba, SP, Brazil
b
Embrapa Soils, Rio de Janeiro, SP, Brazil

A R T I C LE I N FO A B S T R A C T

Keywords: The lack of base information on crop water requirements under tropical conditions is one of the causes for
Crop coefficient inefficient water use and inadequate irrigation management. The crop coefficient (Kc) approach is universally
Water use adopted for scheduling and quantifying the water management and, recently, its relation with the evaporative
Leaf conductance atmosphere demand has been included in this approach to improve the decision-making on irrigation man-
Canopy-atmosphere coupling
agement. In this paper, we measured crop evapotranspiration (ETc) in two experiments as evaporative heat flux
from a centre pivot-irrigated soybean plantations in tropical Brazil to quantify the crop water needs. Irrigation
requirements were determined by comparing ETc with reference evapotranspiration (ETo), derived from FAO-56
(Allen et al., 1998). In this study, the average Kc value obtained was 0.92, which is nearly 20% lower than those
recommended by the FAO-56 (Allen et al., 1998) and varied inversely as a function of ETo. Based on aero-
dynamic and leaf diffusive resistance (rs) data, Ω was computed and presented values relatively high compared
to other crops, which means a relatively poor coupling of soybean canopies to the atmosphere. Throughout both
experiments, the Kc value decreased by up to 44.8% when the ETo increased from 2 to 4 mm d−1, depending on
the phenological phase, indicating the need for taking into account the ETo values when selection the Kc values
for improving the irrigation management in tropical soybean.

1. Introduction inadequate irrigation management. In this sense, crop water con-

sumption mensuration can be obtained by the product of the reference
Together, Brazil and the USA, currently account for about 75% of evapotranspiration (ETo) and the crop coefficient (Kc), resulting is es-
the global soybean crop (Glycine max L.) production. Nearly 34 million timated crop evapotranspiration (ETc) (Allen et al., 1998). This ap-
hectares of land are cultivated with soybean, spread out on 14 Brazilian proach has been universally adopted for scheduling and quantifying the
states, from south to north, with an average yield of 3.4 Mg ha−1. Only water amount to be applied to the crops and has been supported by
11% of the Brazilian soybean crops are currently irrigated. long-term data. In Brazil, the increase in irrigated soybean crops urges
Recently, the high international ratings for soybean and the climatic farmers to improve their irrigation management.
variability in major producing areas in Brazil have increased the de- This approach has been proven useful for practical purposes, and
mand for permissions to establish new irrigated soybean fields, thereby several studies have been performed to provide Kc values for global
increasing soybean yields in the tropics (Kuss et al., 2008; Garcia y soybean production (Ortega-Farias et al., 2004; Karam et al., 2005;
Garcia et al., 2010; Zhang et al., 2016). The Brazilian National Water Suyker and Verma, 2009; Payero and Irmak, 2013; Kamble et al.,
Agency (ANA - Agência Nacional das Águas (2014)) has reported that in 2013). However, information about Brazilian soybean crops in terms of
2014, new permissions were given to irrigate more than 28,000 land irrigation management improvement is still scarce.
hectares of soybean crops under central pivot systems. Previous studies on coffee, citrus orchards, sugarcane and maize
Suitable irrigation practices must be established on the basis of how crops (Marin et al., 2005; Marin and Angelocci, 2011; Nassif et al.,
to use the available water resources more efficiently to achieve higher 2014; Marin et al., 2016; Sobenko et al., 2018; Nassif et al., 2019) have
yields. The lack of basic information on crop water needs under tropical shown that, besides common crop variables influence Kc (e.g. crop
conditions is one of the causes for inefficient agricultural water use and phenology and arrangements, ecophysiology characteristics), the level

⁎
Corresponding author at: Biosystems Engineering Department, Av. Pádua Dias, 11, Piracicaba, SP, 13418-900, Brazil.
E-mail address: [email protected] (F.R. Marin).

https://doi.org/10.1016/j.agwat.2019.03.003
Received 22 October 2018; Received in revised form 26 February 2019; Accepted 1 March 2019
0378-3774/ © 2019 Elsevier B.V. All rights reserved.
E.H.F.M. da Silva, et al. Agricultural Water Management 218 (2019) 1–7

of ETo during the days prior to irrigation should also be taken into two heights, respectively, ETc is the crop evapotranspiration (mm d−1),
account. Rn is the surface radiation balance (MJ m-2 d−1), G is the soil heat flux
In this paper, we measured ETc of an irrigated tropical soybean crop (MJ m-2 d−1) and λ is the latent heat of vaporisation (MJ m-2 d−1).
throughout two seasons to a) provide Kc values based on current soy- Daily reference evapotranspiration ETo was estimated based on the
bean crops for improving irrigation management, and b) to evaluate to Penman-Monteith equation as parameterised by Allen et al. (1998),
what degree the Kc values would be affected by atmospheric evapora- using weather data collected from an automatic standard weather sta-
tive demand and to determine the effects of the crop coefficient-re- tion (Campbell Scientific, Inc., model CR1000 and such sensors) located
ference evapotranspiration approach. near the experimental field.
To evaluate the ecophysiological responses of the crop to the
2. Material and methods weather, we measured the diurnal courses of leaf diffusive resistance to
water vapour (rs), which is the inverse of leaf diffusive conductance
Two soybean crop experiments were conducted at the experimental (gs = 1/rs), with a well-calibrated cycling porometer (Delta-T Inc.,
area of the University of São Paulo (USP), “Luiz de Queiroz” College of model AP4). This was determined during 2 days of the vegetative stage
Agriculture (ESALQ), Piracicaba, São Paulo State, Brazil (latitude (V2 and V5) and 3 days of the reproductive stage (R1, R3 and R6) for
22°42′S; longitude 47°30′W; 546 m a.s.l). For experiments 1 (Exp. 1) Exp. 1 and during 1 day of the vegetative stage (V5) and 2 days of the
and 2 (Exp. 2) were grown from October 2016 to March 2017 and from reproductive stage (R1 and R2) for Exp. 2. The rs was measured in three
December 2017 to April 2018, respectively. According to Köppen exposed leaves from five different plants from 0800 h to 1800 h (local
(1931) climate classification, the local climate is Cwa (high-altitude time). Boundary lines and regression analyses were made using
tropical climate), with rainy summers and dry winters. Annual average SigmaPlot (Systat Software Inc., version 12.0) to obtain the relation-
temperature and precipitation are 22 °C and 1280 mm, respectively. ships between the upper limit of the scattering in the diagrams and to
Both experiments were irrigated by centre pivot, sprinklers i-Wob® indicate the hypothetical response of each independent variable when
UP3® (Senninger Irrigation, Inc.). Irrigation frequency and water the other independent ones are not limiting (Jarvis, 1976; Jones, 1994).
amount were scheduled based on an agrometeorological water balance Mean rs values were also used to compute the decoupling factor (Ω),
model ensuring that soil moisture content exceeded 80% of field ca- which was defined by Eq. (4) and described by McNaughton and Jarvis
pacity and preventing the formation of a shallow root system that (1983). The Ω is a useful indicator on how the canopy interacts with the
would lead to a water supply insufficient to match the atmospheric atmosphere and hence controls the plant transpiration as responding to
demand. Effective root depth was assumed to be 0.50 m. For both ex- weather. Conceptually, Ω can be interpreted by analysis of its extreme
periments, the 3-ha experimental field was planted with the soybean values. As Ω tends to unity (Ω→1 as rs/ra →0), this means that the net
cultivar BRS399-RR at a row spacing of 0.45 m and with 18 seeds per radiation is the main contributor to the evapotranspiration process and
linear metre, resulting in a plant density of 35.5 plants m−2. The soil the crop is considered to be decoupled from the atmospheric conditions.
was classified as an Eutric RHodic Ferralic Nitisol with 3.9% organic If Ω tends to zero (Ω→0 as rs/ra →∞), it indicates the complete cou-
matter content, bulk density of 1480 kg m-3, clay texture (sand, 286 g pling of the crop to the atmosphere, that is, the evapotranspiration
kg-1; silt, 114 g kg-1; clay, 600 g kg-1) and water contents at field ca- process is greatly influenced by the vapour pressure deficit and the
pacity and permanent wilting point of 0.338 and 0.249 m3 m-3, re- wind speed (McNaughton and Jarvis, 1983):
spectively. 1
Crop evapotranspiration ETc was determined during the growing Ω= ,
γ r
seasons by the surface energy balance using the Bowen ratio (β ) 1 + ⎡ (s + γ ) rS ⎤
⎣ a⎦ (4)
method, based on vertical differences in air temperature ( ΔT ) and va-
pour pressure ( Δea ). The reliability of the method was tested following where Ω is the decoupling factor (dimensionless), s is the slope of the
the algorithm presented by Perez et al. (1999), as this method can show saturation vapour pressure curve (kPa), rs is the stomatal resistance to
some variation. When the variability exceeded 2 h, these periods were vapour diffusion measured by the cycling porometer (s m−1) and ra is
discarded, and interpolations for gap fillings were conducted. Also, data the bulk aerodynamic resistance of the soybean (s m-1). The bulk
for the entire day were discarded when periods of such undesirable aerodynamic resistance ra was calculated by Eq. (5) (Jones, 1992). The
variability exceed 2 h. parameters displacement of the zero plane (d) and surface roughness
The Bowen ratio systems were installed in the centre of each ex- length (zo) of Eq. (5) were estimated as a function of crop height (h),
perimental plot, measuring the following weather variables: surface using Eqs. (6) and (7), as proposed by Stanhill (1969) and Rosenberg
radiation balance (Rn ) (Kipp & Zone Inc., model NR-Lite2), soil heat et al. (1983), respectively:
flux (G ) (Hukeseflux Inc., model HFP01), measured at two points 2

0.05 m below the ground, and vertical gradients of air temperature and
⎡ln
ra = ⎣
( ) ⎤⎦
z−d
z0
partial vapour pressure (Vaisala Inc., model HMP155), measured uk (5)
0.80 m between the sensors and 0.20 m above the canopy. Data were
recorded daily at 10-s intervals and stored as 15-min averages by a h0.979
d=
data-logger (Campbell Scientific, Inc., model CR1000). Eqs. (1)–(3) 1.425 (6)
were used to estimate ETc as follows:
h0.997
z0 =
(7)
es = ( 17.3 T
0.6108 e 237.3 T ) RH (1)
7.638
−1
where u is the wind speed (m s ), z is the measurement height of u
ΔT (m), d is the displacement of the zero plane (m), zo is the surface
β=γ
Δea (2) roughness length (m), k is the von Kármán constant (0.40) and h is the
crop height (m).
Rn − G
ETc =
λ (1 + β ) (3)
3. Results
where ea is the current water vapour pressure (kPa), T is the air tem-
perature (°C), RH is the relative humidity (decimals), β is the Bowen Fig. 1 shows the daily variations of ETc and ETo throughout the two
ratio, γ is the psychrometric constant (kPa °C−1), ΔT and Δea are air well-irrigated experiments and the duration period of the phenological
temperature (°C) and partial vapour pressure difference (kPa) between stages. In both experiments, irrigation was managed to meet the crop

2
E.H.F.M. da Silva, et al. Agricultural Water Management 218 (2019) 1–7

Fig. 1. Daily variation in soybean evapotranspiration (ETc), reference evapotranspiration (ETo) and measured rainfall throughout the experimental period.

water requirements, which were 277 and 378 mm for Exp. 1 and Exp. 2, 1.55 (Fig. 2E). Indeed, the overall mean value of Kc was 0.92 ± 0.02
respectively. Based on the water balance and the possibility of water with a statistically significant (P < 0.001) inverse relationship be-
uptake from deeper layers, evapotranspiration in both crops was non- tween Kc and ETo in both experiments and all phenological phases
limited, ensuring maximum ETc rates. analized (Fig. 2B, D and F).
Along the soybean crop of Exp. 1 (Fig. 1A), ETc was systematically Angular coefficients of linear regression analysis between gs and
lower than ETo, with average values of 2.57 ± 0.07 and 3.87 ± 0.09 mm weather variables as well as for ETo values were negative for all of
d−1, respectively. For Exp. 1, relative air humidity (RH ) was higher them, but none were statistically significant at 10%. However, gs in
than 76%, while average air temperature (T ) was 23 °C. Under these turm responded in a nonlinear manner to air temperature, wind speed,
conditions, average ETc was 4.0 mm d−1, reaching a peak of 5.03 mm VPD and solar radiation, as showed by boundary lines in Fig. 3 and
d−1, and stayed above 4.0 mm d−1 over a period of 13 days (12% of respective equations shown in Table 1.
total length) in Exp. 1. The Exp. 2 period was relatively warmer, with We found statistically significant (P < 0.001) inverse relationships
average T and RH values of 23.7 °C and 78.6%, respectively; average between Kc values and ETo (Fig. 2B–D), and based on these we pro-
values of ETc and ETo were 3.97 ± 0.10 and 3.50 ± 0.07 mm d−1, re- posed Kc values for different phenological stages and ETo ranges for the
spectively (Fig. 1B). The ETc in Exp. 2 was generally higher because of two experiments (Table 2). By comparing the Kc values ranging from
the availability of water associated with a high atmospheric demand ETo < 2 and ETo > 4 mm d−1, crop water requirements decreased by
from the prevailing dry air mass, with temperatures ranging from 25 to 44.8, 18.4 and 24.2% for vegetative, reproductive R1 to R4 and R5 to
35 °C and a mean air relative humidity below 73%. The ETc reached a R6 stages, respectively (Table 2).
peak of 6.49 mm d−1 and remained above 4.0 mm d−1 over a period of The average Ω value was 0.59 at vegetative stage V1. At the vege-
55 days (51% of total length) for Exp. 2. tative stage V2, average Ω values of 0.88 and 0.70 were obtained for
The ratio between ETc and ETo represents the Kc value and is de- Exp. 1 and Exp. 2, respectively. For later stages, Ω decreased to 0.55 in
termined from the slope of the straight line forced to pass through the average, but still showing a relatively weak canopy-atmosphere cou-
origin. When considering the whole data across the entire period of the pling level (Table 3).
two experiments, and separating them by phenological stages, the mean Fig. 4 shows patterns of hourly variation of gs throughout the two
Kc value of 0.79 ± 0.03, ranging from 0.26 to 1.78 (Fig. 2A). During the experimental period and the duration period of the phenological stages.
initial stages of the reproductive development, including blooming and In both experiments, there was a sharp drop in gs over the last hours of
pod development (R1 to R4) the mean Kc value was 1.12 ± 0.03, ran- the day and gs data was remarkable variable along the daylength
ging from 0.63 to 2.06 (Fig. 2C). From the beginning to the end of grain mainly in Exp 1 (Fig. 4A). Exp. 1 showed relatively higher average gs
filling stages (R5 to R6), ETc/ETo ratio was 0.82 ranging from 0.32 to values along the measurements days, excepting for the phenological

3
E.H.F.M. da Silva, et al. Agricultural Water Management 218 (2019) 1–7

Fig. 2. Soybean evapotranspiration (ETc , A, C and E) and crop coefficient (Kc, B, D and F), separated by phonological stages, as a function of reference evapo-
transpiration (ETo ).

stage V5. growers, as well as in unnecessary use of natural water reservoirs.

Due to the increase in aerodynamic resistance or stomatal closure, Ω
tends to be higher in vegetation with less roughness because of an in-
4. Discussion crease in the aerodynamic resistance of the canopy. In general, Ω values
showed soybean relatively decoupled to the atmosphere (Table 2)
The soybean crop Kc values agreed with those reported for the same compared to coffee (Marin et al., 2005), citrus (Marin et al., 2011),
climatic conditions such as the Brazilian Southern summer (Doorenbos sugarcane (Nassif et al., 2014; Marin et al., 2016; Nassif et al., 2019)
and Pruitt, 1977; Allen et al., 1998; Suyker and Verma, 2009; Payero and maize (Sobenko et al., 2018). In those previous papers, we hy-
and Irmak, 2013). However, Kc values found here were nearly 20% pothesized that one possible explanation for the decline of Kc with ETo
lower than those recommended by the FAO-56 (Allen et al., 1998). The increases is the high level of canopy-atmosphere coupling found for
Kc decreased as ETo increased for both experiments and all phenolo- those crops. Under such conditions there would be a feed-back between
gical stages at (P < 0.001) suggesting that keeping the same Kc levels plants and their immediate aerial environment which may diminish the
for irrigation management would imply in water and energy waste for

4
E.H.F.M. da Silva, et al. Agricultural Water Management 218 (2019) 1–7

Fig. 3. Relationships between hourly soybean leaf diffusion conductance (gs) and air temperature (A), vapour pressure deficit (VPD, B), wind speed (C) and solar
radiation (D), with hypothetical boundary line fitted for each environmental variable.

dependence of transpiration on stomatal conductance (Jarvis and Table 2

McNaughton, 1986; Steduto et al., 2009). Crop coefficient (Kc) as a function of phenological stage and ranges of reference
However, for soybean, the overall average values of Ω = 0.59, but evapotranspiration (ETo) for soybean crop. Values in parentheses represent
the inverse ETo-Kc relationship was also observed, casting doubt on our standard deviation.
previous hypothesis on gs controlling the crop transpiration as the in- Phenological stage ETo range Kc
verse ETo-Kc relationship occurred regardless of coupling level. For
canopies completely decoupled from the atmosphere, the values of gs Vegetative < 2.0 mm d−1 1.27 [0.47]
2.0 – 4.0 mm d−1 0.78 [0.27]
were higher than the canopy boundary layer conductance (i.e. the in-
> 4.0 mm d−1 0.70 [0.24]
verse of bulk aerodynamic resistance – ra), which means that changes in R1-R4 < 2.0 mm d−1 1.25 [0.41]
gs will have only a slight effect on transpiration rate, and incoming solar 2.0 – 4.0 mm d−1 1.22 [0.25]
radiation would be the primary driver of transpiration. In this case, > 4.0 mm d−1 1.02 [0.26]
R5-R6 < 2.0 mm d−1 0.91 [0.11]
despite stomatal control of transpiration is weaker than form highly
2.0 – 4.0 mm d−1 0.81 [0.12]
coupled canopies, for Ω = 0.59 as found herein, it is reasonable to > 4.0 mm d−1 0.69 [0.17]
accept gs playing some role on the crop transpiration control (Meinzer
et al., 1997).
Meyer and Green (1981) stated that soybean gs at approximately the however, reached peak values with some optimum combination values
same time each day are subject to a considerable amount of variability for these weather variables, rapidly decreasing thereafter as the atmo-
even for very high level of available soil water. This agrees with gs sphere evaporative demand increases (Teare and Kanemasu, 1972;
measuments made for other crops (e.g. Rochette et al., 1991; Angelocci Sionit and Kramer, 1976; Choudhury, 1983; OosteRHuis and Walker,
et al., 2004), as well as with our dataset. The maximum gs values, 1987; Buttery et al. (1993)). Denmead and Shaw (1962) and Grantz and

Table 1
Regression equations among soybean leaf diffusion conductance (gs, cm s−1) and air temperature (T, ºC), vapour pressure deficit (VPD, kPa), wind speed (u,
m s−1) and solar radiation (Rn, W m-2).
Variable Equation R2 P value

T ⎡−0.5 T − 29.3918
2 0.867 < 0.001
⎢
gS = 0.7493 + 2.0975e⎣
( 2.8571 ) ⎤⎥⎦
VPD gS = −0.1873 + 6.5046VPD − 3.7922VPD 2 + 0.5745VPD3 0.897 < 0.001
u gS = −0.6729 + 3.9858u − 1.2171u2 + 0.096u3 0.977 < 0.001
Rn gS = 0.758 + 0.0094Rn − 1.023x10−5Rn2 + 2.6794x10−9Rn3 0.821 < 0.001

5
E.H.F.M. da Silva, et al. Agricultural Water Management 218 (2019) 1–7

Table 3 Jarvis, 1983).

Stomatal resistance (rs), bulk aerodynamic resistance (ra) and decoupling factor Based on the relationships among gs and weather variables (Fig. 3),
(Ω ) for soybean crop. we selected 30.3 °C, 1.1 kPa, 2.12 m s−1 and 644 W m-2, respectively,
Exp. 1 Exp. 2 for air temperature, VPD, wind speed and solar radiation as corre-
sponding values for the highest gs. Estimating ETo (Allen et al., 1998)
DOYa Hour rs ra Ω DOY Hour rs ra Ω from these values, a daily rate of 5.17 mm d−1 was found. These esti-
(s cm−1) (s cm−1)
mations were based on the assumption that net radiation was 50% of
332 08h00 0.14 1.45 0.97 25 10h00 0.97 0.47 0.68 incoming solar radiation and that the photoperiod was 12 h. Boundary
09h00 0.29 0.70 0.90 12h00 0.87 0.50 0.74 functions showed gs decreases when the atmospheric demand is high,
10h00 0.82 0.72 0.79 14h00 0.73 0.76 0.84 corroborating the hypothesis that soybean plants partially control
11h00 0.85 0.76 0.81 16h00 0.59 0.30 0.73 transpiration rates as ETo increases.
12h00 0.46 0.76 0.88 18h00 2.20 0.55 0.52
14h00 0.34 0.47 0.85 40 08h00 2.04 0.53 0.55
Allen et al. (1998) have claimed that Kc values must be used under
15h00 0.14 1.38 0.96 09h00 1.97 0.37 0.58 standard climatic conditions (sub-humid climate, minimum relative
16h00 0.33 0.64 0.86 11h00 2.70 0.38 0.48 humidity of 45% and wind speeds averaging 2 m s−1) and that varia-
349 08h00 1.03 0.22 0.40 14h00 4.01 0.33 0.37 tions in wind speed may alter aerodynamic resistance and, hence, Kc.
10h00 1.66 0.19 0.29 17h00 6.86 0.41 0.50
They also pointed out that Kc tends to increase under conditions of high
11h00 1.03 0.23 0.45 53 09h00 1.18 0.52 0.65
12h00 1.10 0.19 0.37 10h00 1.27 0.35 0.58 wind speed and air temperatures, and low relative humidity. Our data
14h00 1.26 0.20 0.38 12h00 1.74 0.26 0.52 showed that high air temperature, VPD and wind speed tended to re-
15h00 1.11 0.19 0.39 14h00 1.23 0.26 0.55 strict soybean transpiration, along with changing Kc values as ETo
16h00 1.20 0.20 0.39 16h00 3.11 0.17 0.35 varied. Yet, we measured ETc throughout two irrigated soybean cycles
17h00 2.00 0.21 0.28 18h00 6.08 0.23 0.22
to update Kc values specifically for Brazilian cultivars and those dataset
18h00 2.73 0.19 0.19
4 09h00 0.73 0.60 0.73 suggested that the irrigation management procedures suggested by
10h00 0.66 0.39 0.68 Allen et al. (1998) should be updated, as the use of conventional Kc
11h00 0.68 0.27 0.60 values proposed by Allen et al. (1998) might lead to a crop water re-
12h00 0.46 0.23 0.66
quirement higher than necessary.
13 09h00 0.63 0.19 0.51
10h00 0.78 0.23 0.50
11h00 0.77 0.34 0.62
12h00 0.50 0.34 0.71 5. Conclusions
14h00 0.36 0.32 0.79
15h00 0.41 0.24 0.70
We measured ETc along two irrigated soybean crop plots to update
16h00 0.70 0.55 0.73
17h00 0.97 0.43 0.61 Kc values for irrigation management in accordance with the current
37 11h00 0.42 1.29 0.92 technologies and practices in Southern Brazil and discussed the para-
12h00 0.38 2.91 0.92 digm involving the crop coefficient-reference evapotranspiration ap-
13h00 0.30 5.19 0.98
proach. The average Kc value obtained was 0.92, and the maximum
15h00 0.37 2.04 0.95
16h00 0.43 0.38 0.78
average value of 1.12 was found between the reproductive stages R1
17h00 1.36 2.70 0.62 and R4, representing a value 20% lower than those suggested by Allen
18h00 1.69 0.31 0.38 et al. (1998).
a
The Kc values decreased as ETo at statiscally significant values
DOY = day of year with porometric measurements. (P < 0.001) for both experiment and all phenological stages. Soybean
leaves reduced the stomatal conductance under high air temperature,
Zeiger (1986) reported that with high atmospheric water demand, even wind speed, VPD and, mainly, solar radiation, even with high soil water
at a high soil water content, the relationship between gs and meteor- availability. Based on these results, we proposed that the Kc re-
ological variables demonstrated that leaves restrict water loss mainly commendation for practical purposes should take into account the ETo
through stomatal closure. Therefore, the relationship between ETc and values of the previous days before irrigation, besides the aspects already
ETo seems to be due to an increase in inner resistance to water transport suggested by Allen et al. (1998) for estimating the soybean water re-
in plants when subjected to conditions of high atmospheric demand for queriments.
water due to an opposing tendency for transpiration and stomatal
movement in relation to increased free-air VPD (McNaughton and

Fig. 4. Hourly variation of measured leaf diffusion conductance (gs) in soybean experiments 1 (A) and 2 (B).

6
E.H.F.M. da Silva, et al. Agricultural Water Management 218 (2019) 1–7

Acknowledgements the atmosphere of a citrus orchard in Southern Brazil. Agric. Water Manag. 98,
1091–1096.
Marin, F.R., Angelocci, L.R., Righi, E.Z., Sentelhas, P.C., 2005. Evapotranspiration and
We thank the Research Foundation of the State of São Paulo irrigation requirements of a coffee plantation in Southern Brazil. Exp. Agric. 41 (02),
(FAPESP grants 2017/23468-9, 2015/25702-3, 2017/20925-0), the 187–197.
Brazilian Research Council (CNPq grant 301424/2015, 401662/2016, Marin, F.R., Angelocci, L.R., Nassif, D.S., Costa, L.G., Vianna, M.S., Carvalho, K.S., 2016.
Crop coefficient changes with reference evapotranspiration for highly canopy-at-
425174/2018, 300916/2018) and the Coordenação de mosphere coupled crops. Agric. Water Manag. 163, 139–145.
Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) for the McNaughton, K.G., Jarvis, P.G., 1983. Predicting effects of vegetation changes on tran-
scholarships. spiration and evaporation. In: In: Koslowski, T.T. (Ed.), Water Deficit and Plant
Growth, vol7. Academic Press, New York, pp. 1–47.
Meinzer, F., Andrade, J., Goldstein, G., Holbrook, N., Cavelier, J., Jackson, P., 1997.
References Control of transpiration from the upper canopy of a tropical forest: the role of sto-
matal, boundary layer and hydraulic architecture components. Plant Cell Environ. 20
(10), 1242–1252.
Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop Evapotranspiration—Guidelines
Meyer, W.S., Green, G.C., 1981. Plant indicators of wheat and soybean crop water stress.
for Computing Crop Water Requirements. FAO, Rome pp. 300 (Irrigation and drai-
Irrig. Sci. 2.3, 167–176.
nage paper, n. 56).
Nassif, D.S.P., Marin, F.R., Costa, L.G., 2014. Evapotranspiration and transpiration cou-
Angelocci, L.R., Marin, F.R., Oliveira, R.F., Righi, E.Z., 2004. Transpiration, leaf diffusive
pling to the atmosphere of sugarcane in Southern Brazil: scaling up from leaf to field.
conductance, and atmospheric water demand relationship in an irrigated acid lime
Sugar Technol. 1–5.
orchard. Braz. J. Plant Phys. CN-0000 16, 53–64.
Nassif, D.S.P., da Costa, L.G., Vianna, M., dos, S., dos Santos Carvalho, K., Marin, F.R.,
ANA - Agência Nacional das Águas, 2014. Levantamento da Agricultura Irrigada por Pivôs
2019. The role of decoupling factor on sugarcane crop water use under tropical
Centrais no Brasil, 33.
conditions. Exp. Agric. 1–11.
Buttery, B.R., Tan, C.S., Buzzell, R.I., Gaynor, J.D., MacTavish, D.C., 1993. Plant Soil 149,
Ortega-Farias, S., Olioso, A., Antonioletti, R., Brisson, N., 2004. Evaluation of the
283–288.
Penman-Monteith model for estimating soybean evapotranspiration. Irrig. Sci.
Choudhury, B., 1983. Simulating the effects of weather variables and soil water potential
23, 1–9.
on a corn canopy temperature. Agric. For. Meteorol. 29, 169–182.
Payero, J.O., Irmak, S., 2013. Daily energy fluxes, evapotranspiration and crop coefficient
Denmead, O.T., Shaw, R.H., 1962. Availability of soil water to plants as affected by soil
of soybean. Agric. Water Manag. 129, 31–43.
moisture content and meteorological conditions. Agric. J. 45, 385–390.
Perez, P.J., Castellvi, F., Iban¸ ez, M., Rosell, J.I., 1999. Assessment of reliability of Bowen
Doorenbos, J., Pruitt, W.O., 1977. Background and development of methods to predict
ratio method for partitioning fluxes. Agric. For. Meteorol 97, 141–150.
reference crop evapotranspiration (ETo). Crop Water Requirements. FAO Irrigation
RHuis, D.M., Walker, Sue, 1987. Stomatal resistance measurement as an indicator of
and Drainage Paper 24. pp. 108–119.
water deficit stress in wheat and soybeans. South Afr. J. Plant Soil 4.3, 113–120.
Garcia y Garcia, A., Persson, T., Guerra, L., Hoogenboom, G., 2010. Response of soybean
Rochette, P., et al., 1991. Estimation of maize (Zea mays L.) canopy conductance by
genotypes to different irrigation regimes in a humid region of the southeastern USA.
scaling up leaf stomatal conductance. Agric. For. Meteorol 54, 241–261.
Agric. Water Manag. 97 (7), 981–987.
Rosenberg, N.J., Blad, B.L., Verma, S.B., 1983. Microclimate: The Biological Environment.
Grantz, D.A., Zeiger, E., 1986. Stomatal response to light and leaf-air water vapor pres-
Wiley, USA, pp. 495.
sure difference show similar kinetics in sugarcane and soybean. Plant Physiol. 81,
Sionit, N., Kramer, P.J., 1976. Water potential and stomatal resistance of sunflower and
865–868.
soybean subjected to water stress during various growth stages. Plant Physiol. 58,
Jarvis, P.G., 1976. The Interpretation of Variations in the Leaf Water Potential and
537–540.
Stomatal Condutance Found in Canopies in the Field. Trans. Royal Soc, Philosoph.
Sobenko, L.R., Souza, T.T., Gonçalves, A.O., Bianchini, V.J., Silva, E.H.F.M., Souza, L.,
London, UK.
Marin, F.R., 2018. Irrigation requirements are lower than those usually prescribed for
Jarvis, P.G., McNaughton, K.J., 1986. Stomatal control of transpiration: scaling up from
a maize crop in southern brazil. Exp. Agric. 1–10.
leaf to region. Adv. Ecol. Res. CN -1056 15 (1), 49.
Stanhill, G., 1969. A simple instrument for the field measurement of turbulent diffusion
Jones, H.G., 1992. Plants and Microclimate: A Quantitative Approach to Environmental
flux. J. Appl. Meteorol. Climatol. 8, 509–513.
Plant Physiology. Cambridge University Press, New York, USA.
Steduto, P., Raes, D., Hsiao, T.C., Fereres, E., Heng, L.K., Howell, T.A., Geerts, S., 2009.
Jones, H.G., 1994. Plants and microclimate: a quantitative approach to environmental
Concepts and Applications of AquaCrop: The FAO Crop Water Productivity Model. In:
plant physiology, 2nd edn. Cambridge University Press, Cambridge.
Cao, W., White, J.W., Wang, E. (Eds.), Crop Modeling and Decision Support. Springer,
Kamble, B., Kilic, A., Hubbard, K., 2013. Estimating crop coefficients using remote sen-
Berlin Heidelberg, pp. 175–191. Retrieved from http://link.springer.com/chapter/
sing-based vegetation index. Remote Sens. (Basel) 5 (4), 1588–1602.
10.1007/978-3-642-01132-0_19.
Karam, F., Masaad, R., Sfeir, T., Mounzer, O., Rouphael, Y., 2005. Evapotranspiration and
Suyker, A.E., Verma, S.B., 2009. Evapotranspiration of irrigated and rainfed maize–soy-
seed yield of field grown soybean under deficit irrigation conditions. Agric. Water
bean cropping systems. Agric. For. Meteorol. 149 (3), 443–452.
Manag. 75 (3), 226–244.
Teare, I.D., Kanemasu, E.T., 1972. Stomatal-diffusion resistance and water potential of
Köppen, W., 1931. Grundriss Der Klimakunde: Outline of Climate Science. Walter de
soybean and sorghum leaves. New Phytol. 71, 805–810.
Gruyter, Berlin.
Zhang, B., Feng, G., Kong, X., Lal, R., Ouyang, Y., Adeli, A., Jenkins, J.N., 2016.
Kuss, R.C.R., König, O., Dutra, L.M.C., Bellé, R.A., Roggia, S., Sturmer, G.R., 2008. Plant
Simulating yield potential by irrigation and yield gap of rainfed soybean using APEX
population and irrigation management strategies in the soybean tillage. Cienc. Rural
model in a humid region. Agric. Water Manag. 177, 440–453.
38, 1133–1137.
Marin, F.R., Angelocci, L.R., 2011. Irrigation requirements and transpiration coupling to