RESEARCH

60 CHILEAN J. AGRIC. RES. - VOL. 69 - Nº 1 - 2009

METHODS TO ESTIMATE LETTUCE EVAPOTRANSPIRATION IN

GREENHOUSE CONDITIONS IN THE CENTRAL ZONE OF CHILE

Manuel Casanova P.1*, Ingmar Messing2, Abraham Joel2, and Alberto Cañete M.1

ABSTRACT

This study evaluates five methods to estimate crop evapotranspiration in greenhouse conditions. It compares their
performance in relation to the evapotranspiration directly determined from water balance measurements (ETlys) in
an irrigated lettuce (Lactuca sativa L.) crop during 9 weeks. Daily values of the reference evapotranspiration (ETo)
from Class A pan (CAP), Piche atmometers (ATM), Andersson evaporimeters (ANE), FAO-Radiation (FRE) and
FAO-Penman-Monteith (PME) equations were compared. The methods showed similar temporal variations but at
different ranges as follows: ANE < CAP < FRE < PME < ATM. Furthemore, ETo had a clear correlation with solar
radiation. Crop coefficients (Kc = ETlys/ETo) varied somewhat amongst the methods, but trends were identified
for two periods: in the first week, the overall mean Kc was 0.3 (± 0.1) and in weeks 2 to 9 on average 0.6 (± 0.3).
The greenhouse values of Kc were lower than those generally adopted for lettuce in field conditions. In terms of
irrigation design, crop evapotranspiration can be estimated by the methods in this study, on the condition that the
appropriate crop coefficients are applied. The fact that ANE showed values closest to those of ETlys, along with cost
and management convenience, makes it an advantageous alternative as compared to the other methods.

Key words: evapotranspiration, greenhouse, lettuce, lysimeter, Lactuca sativa.

INTRODUCTION Radin et al., 2004). Relative humidity, higher inside than

outside a greenhouse, has a similar impact but is balanced
Crop cultivation under plastic cover is an important by the high temperatures normally registered inside.
production system in Chile. Preliminary results of INE Lower wind velocity also tends to reduce the exchange
(2007) show that 1578 ha were cultivated in greenhouses. of water vapor between the canopy and the atmosphere
The need to provide fresh quality products during in a greenhouse. The control strategies for the greenhouse
prolonged periods of the year, along with the optimum climate, and the models behind them, assume that water
use of water under dry and hot climatic conditions have and nutrient uptake does not limit plant growth (Kläring,
led to the adoption of this technology. In these conditions, 2001). This assumption will be valid if irrigation and
the plastic cover of the greenhouses significantly changes fertilization management are controlled.
the internal radiation balance with respect to the external The most frequently used devices in Chile to estimate
environmental conditions, especially with regard to the evaporative demand in field conditions are the Class
absorption and reflection of incident solar radiation. As a A pan (CAP) and the Piche Atmometers (ATM). On
consequence, important effects in the evapotranspiration the other hand, Andersson evaporimeters (ANE) have
(ET) of the crops are observed. In fact, even the best been successfully used in spatial variability studies of
modern glasshouses reduce light input by at least 30%, evaporation in arid zones of Africa (Messing et al., 1998;
which should simultaneously cause a considerable Messing and Akrimi, 1998). The readings registered
reduction in ET (El Moujabber and Abi Zeid Daou, 1999; by all these devices are converted, using adequate
factors, into reference evapotranspiration values (ETo).
Parallel to these devices, empirical equations have been
1
Universidad de Chile, Facultad de Ciencias Agronómicas, Casilla developed that use climatic parameters to estimate ETo.
1004, Santiago, Chile. The Penman-Monteith (PME) equation has been recently
*
Corresponding author ([email protected]). recommended by Allen et al. (2006) and López-Urrea et
2
Swedish University of Agricultural Sciences, Box 7014. SE-750 07 al. (2006), as the sole method to determine ETo, although
Uppsala, Sweden.
Received: 29 November 2007. the FAO-Radiation formula (FRE) has been a widely
Accepted: 29 January 2008. accepted alternative (Nandagiri and Kovoor, 2006).

CHILEAN JOURNAL OF AGRICULTURAL RESEARCH 69 (1): 60-70 (JANUARY-MARCH 2009)

M. CASANOVA et al. - METHODS TO ESTIMATE LETTUCE EVAPOTRANSPIRATION… 61

At the field level, these methods give reasonable Lettuce, which is a crop that is sensitive to
estimates of the crop water requirements, although meteorological variations and excess of rain (Radin et
they have advantages and disadvantages in specific al., 2004), represents an interesting case to analyze in
circumstances (Kassam and Smith, 2001; Turco et al., controlled greenhouse conditions regarding product
2008). PME is complex to use in greenhouses because it quality, mainly uniform plant size at harvest. Navarette
uses numerous climatic variables. CAP is a simple method et al. (1993) pointed this out as being a fundamental
to obtain ETo, although space limitations hinder its use quality parameter, very closely linked to environmental
in greenhouse conditions. ANE and ATM indubitably and management variables. Given the absence of national
show comparative advantages with respect to the pan and information concerning this species, the objective of
equations given their size, cost, and simple operation. this study was to compare the ETc values estimated by
A lot studies have been performed to compare these different methods (CAP, ATM, ANE, FRE and PME) with
methods. Araujo et al. (2005) compared lysimeters and the evapotranspiration measured by lysimeters (ETlys), for
CAP and found determination coefficients (R2) of 0.77 greenhouse-grown lettuce in the central zone of Chile.
and 0.65. López-Urrea et al. (2006) confronted PME and
FRE with lysimeters and observed R2 of 0.92 and 0.82, MATERIALS AND METHODS
respectively, whereas Vaughan et al. (2007) showed R2
around 0.74 (diurnal) and 0.23 (night) between PME and This study was carried out at the Antumapu
lysimeters. Papaioannou et al. (1996; 1998) compared Experimental Station (33º40’ S; 70º38’ W), in Santiago,
ATM with PME and obtained correlation coefficients (R) Chile, during the 2002 spring-summer period. The zone
ranging from 0.29 to 0.65 (yearly period) and 0.49 to 0.91 has a temperate mesothermal estenothermic semiarid
(over several years). In the case of ANE, Johansson (1969) Mediterranean climate (Santibáñez and Uribe, 1992).
reported a R value of 0.93 with CAP, whereas Messing The thermal regime is characterized by hot dry summers
(1998) compares it with ATM and CAP resulting in R2 and cold winters with temperatures varying between a
values of 0.78 and 0.85, respectively. maximum mean of 28.2 ºC in January and a minimum
El Moujabber and Abi Zeid Daou (1999) argued that the mean of 4.4 ºC in July; the frost-free period is 231 days
irrigation design based on climatic data is a very common with a mean of 11 frosts per year. Annually, an average
practice but most coefficients utilized to estimate the crop of 1621 degree-day and 1147 chill hours are registered.
evapotranspiration (ETc), whether lysimeters, devices, or Mean yearly precipitation is 419 mm with a water deficit
equations, are mainly valid for outdoor conditions, which of 997 mm and an eight-month dry period.
are contrasting to those of greenhouses. On the other A chapel-type greenhouse with a metallic structure
hand, it is necessary to further develop understanding facing north-south was used, with dimensions: height 4.0
of factors and processes that occur inside greenhouses, m, length 18.0 m, and width 11.5 m, and covered with
among which ETc, estimated from vegetative areas, is one 0.15 mm transparent polyethylene treated for ultraviolet
of the main factors to consider in the calculation of water radiation. Considering solar radiation (Rs) measured for
balances. This is necessary in order to get adequate and the same period at La Platina Regional Research Center
opportune estimations of water demand for the irrigation (Instituto de Investigaciones Agropecuarias, INIA)
system designs to integrate with other management weather station, the mean transmissivity of the cover
aspects (crop fertilization, pest and disease control, etc.) was 85%. On both sides walls had 0.20 mm polyethylene
in these controlled environments. Sarraf (2004) points out curtains which were raised or dropped according to
that, in the absence of sufficient local measurements on weather fluctuations.
crop water requirements in greenhouses, it is up till now A short-season lettuce hybrid was sown (August 20,
not clear whether greenhouses require more or less water 2002) var. capitata, cv. XP 6256 (Seminis Sudamérica
per unit of product than field cultures. Only a limited S.A., Santiago, Chile) and transplanted (September 30,
number of national publications specialized in this area 2002) at the three-leaf stage. The crop developed during
compare the ETc values for greenhouse irrigated crops 9 weeks in an artificial substrate composed of an Entic
using these methods with studies of water balance using Haploxeroll sandy loam (sand: 53.5%, silt: 32.4%, clay:
lysimeters. Although the use of ANE in greenhouses has 14.1%), organic residues, and fine sand in a volumetric
not been reported in international research, the other proportion of 2:1:1, respectively. Physical properties
methods in the present study have been evaluated under were characterized by extracting five samples of this
these conditions for several crops (Fernandes et al., 2003; substrate: bulk density with cylinder, particle density with
Blanco and Folegatti, 2003; 2004; Orgaz et al., 2005), picnometer, and water retention with pressure devices
although few with lettuce (Galvani et al., 1998). (Dane and Topp, 2002).
62 CHILEAN J. AGRIC. RES. - VOL. 69 - Nº 1 - 2009

The 0.15 m-thick substrate was placed on eight 16 x in standard conditions (Allen et al., 2006), was estimated
0.80 m metal tables, at a height of 0.70 m from the floor. on the basis of atmospheric demand measured with the
The six central tables were lined with polyethylene (0.20 studied devices and empirical equations. In total, one
mm) before placing the substrate on it and 10 cylinders standard Class A pan, 10 Piche atmometers, and two
connected to recipients were arranged under them to Andersson evaporimeters were used. Table 1 includes the
drain water to quantify the water percolation volume formulae used to define ETlys and ETo of each method in
(D) in test tubes (precision: 1 mL). The transplanted this study.
plants were placed in two rows spaced 0.25 x 0.27 m. Placement of the Class A pan (CAP) inside the
Fertilization prior to transplanting was comprised of 80 greenhouse, without considering standard reference
kg N ha-1 (urea), 100 kg P2O5 ha-1 (triple superphosphate), conditions, can be a first step for rough estimation of
and 120 kg K2O ha-1 (potassium nitrate). A system of two water demand when there are no other available methods
drip irrigation tapes per table was used over each row of or weather data are lacking (Baille, 1996). Nevertheless,
plants, with outlets every 0.20 m and a discharge of 2.20 L the use of CAP in these conditions is open to controversy
h-1 m-1. The irrigation frequency was daily and the applied and research on which pan coefficient (Kp) should be
water volume was measured in an external supply tank used is not conclusive (Fernandes et al., 2003). In this
(graduated at 1 cm), and corresponded to the evaporation study the CAP was placed at the center of the greenhouse
of the precedent day measured by means of a standard on a wooden structure at table height. A Kp of 0.65 or 0.75
Class A pan (inside greenhouse). With a calibrated FDR was used, depending on relative humidity (40-70% or >
sensor (Frequency Domain Reflectometry, repeatability 70%, respectively) (Doorenbos and Pruitt, 1976): case A
of ± 2% and operational range between 0 and 100%) was assumed, with a pan surrounded by a low green crop,
water content (W) of the substrate was recorded every windward distance equal to zero and low wind velocity
day before irrigation for each table (15 measurements, at (< 2 m s-1).
3 depths). The Piche atmometers were semi-protected from solar
ETc values, based on calculations of the water balance radiation; six of them were placed at a fixed height of 1.50
of the system, were estimated daily (∆t =1) using the tables m above the floor (ATMf), the other four were adjusted
as lysimeters (ETlys). The high porosity of the substrate weekly according to the growth of the lettuce plant and
and the gradient generated by a slight slope of the table were kept at 0.15 m above the plants (ATMm). ETo (mm
bottoms, described by Ben-Gal and Shani (2002), turned day-1) for these devices was calculated using the recorded
the tables into free-drainage lysimeters. In this way, water evaporation readings (Ep, mm d-1) multiplied by two
provided by irrigation (I), stored by the substrate (∆W), coefficients, both defined by Bouchet (1963): a with a
and drained (D) was calculated daily. The reference value of 0.27 which considers its semi-protection, and
evapotranspiration (ETo), which is normally determined ρ(σ) which is a temperature dependent factor.

Table 1. Devices and equations used to estimate evapotranspiration.

Method Formula/Equation Reference

I - (∆W + D)
Lysimeter ETlys = Ben-Gal and Shani (2002)
∆t

ATM ETo = α Epi ρ(σ) Bouchet (1963)

CAP ETo = Epan Kp Doorenbos and Pruitt (1976)
ANE ETo = Ea Eα Johansson (1969)

PME Allen et al. (1994)

Doorenbos and Pruitt (1976)

FRE
Frevert et al. (1983)

ETlys: lysimeter; CAP: Class A pan evaporimeter; ATM: Piche atmometer; ANE: Andersson evaporimeter; PME: Penmann Monteith equation; FRE:
FAO-Radiation equation.
M. CASANOVA et al. - METHODS TO ESTIMATE LETTUCE EVAPOTRANSPIRATION… 63

The Andersson evaporimeters (ANE) (Figure 1) average daily air temperature (T, ºC) and daily mean of
manufactured with transparent colorless plexiglass, wind speed (u2, m s-1) 2 m height, the saturation and actual
consist of a closed but ventilated cylindrical chamber with vapor pressure (es and ea respectively, kPa), the saturation
a 0.15 L capacity and an internal water height between vapor pressure deficit (es - ea), the slope of the saturation
25 and 40 mm; the evaporated water goes out through vapor pressure curve (∆, kPa ºC-1), and the psychrometric
circular holes located in the upper part of the vessel. Their constant (γ, 0.063 kPa ºC-1). Net radiation at the crop
design is directed to increase flow resistance to simulate surface (Rn, MJ m-2 d-1) was calculated as the difference
the aerodynamic conditions in the leaf-air interphase between the net shortwave radiation (Rns, [1-albedo]·Rs)
(Andersson, 1969). Furthermore, they react very quickly and net longwave radiation (Rnl) according to the detailed
to changes in the environmental evaporation demand procedure in Doorenbos and Pruitt (1976).
since they work with a small volume of water. They were The FAO-Radiation (FRE) equation (Table 1),
placed on 1.5 m-high pedestals inside the greenhouse and that links ETo with the solar radiation as evaporation
ETo (mm d-1) was obtained from the product between the equivalents (Rs, mm d-1) considered the corrections by
readings (determined with a micrometer with accuracy Frevert et al. (1983) that replace the original nomograms
0.01 mm) of registered evaporation (Ea), as well as an Ea (Nandagiri and Kovoor, 2006). Smith et al. (1998) inform
coefficient equal to 0.7 recommended by Johansson (1969). that this equation gives adequate ETo results in high
The daily ETo (mm d-1) for each method corresponded to humidity conditions where the aerodynamic element is
the difference in evaporation readings recorded at 08:30 h relatively small, as in greenhouse conditions, whereas
between two consecutive measurements. results become erratic in dry conditions.
In ETo (mm d-1) estimated through empirical equations In order to account for all the variables required for
(Table 1), the Penman-Monteith equation (PME) derived PME and FRE, both mechanical anemometer and weather
for 24 h (Allen et al., 1998) was utilized. It predicts ETo station, with radiation (piranometer) and temperature
for a hypothetical area of pasture used as reference with sensors, were installed at the center of the greenhouse to
a height of 0.12 m, resistance of 70 s m-1, and albedo of 1.6 and 2 m heigths respectively. Two hygrothermographs
23%. It has been used by Pollet and Bleyaert (2000) for (calibrated with an aspiration psychrometer) were also
greenhouse lettuce with 6% subestimates. Its structure utilized. The location of all the devices and instruments
includes the soil heat flux density (G, MJ m-2 d-1) with are shown in Figure 2 with more details in Table 2.
a zero value over the calculation time increment (24 h)

Figure 1. Sketch of Andersson evaporimeter (ANE).

64 CHILEAN J. AGRIC. RES. - VOL. 69 - Nº 1 - 2009

Figure 2. Spatial distribution of devices and instruments inside the greenhouse. 1: Atmometer; 2: Thermohigrograph;
3: Class A pan; 4: Andersson evaporimeter; 5: Meteorological station; 6: Anemometer.

Table 2. Origin of instruments and devices used inside the greenhouse.

Instrument/device Brand-Model Location
FDR (Frequency domain reflectometry) Aquaterr® (M-200). Fremont, California, USA
Piche Atmometer Casella (CEL) Bedford, England
Andersson Evaporimeter Swedish University of Agricultural Sciences, Uppsala, Sweden
Department of Soil Science, Hydrotechnics
Division
Weather Station LI-1200 (Li-Cor Inc.) Lincoln, Nebraska, USA
Mechanical Anemometer R-Fuess Berlin-Steglitz, Germany
Higrothermograph Bendix Corporation, Friez Instruments Baltimore, Maryland, USA
Division (Model 594)
Aspiration Psychrometer Fisher® (Assmann Model) Drebach, Germany

The combined effects of crop transpiration and of the period. Large fluctuations in all measured climatic
soil evaporation can be integrated into a unic mean variables except wind speed were observed inside the
crop coefficient (Kc) that can be considered for normal greenhouse (Table 3).
irrigation planning and management purposes, for the The ETo (mm d-1) values determined using the
development of basic irrigation schedules, and for most different methods and the values obtained with a
hydrologic water balance studies (Allen et al., 1998). lysimeter (ETlys) showed a high daily variability during
Kc is consequently estimated for lettuce in greenhouse the assessment period. This degree of oscillation follows
conditions in the present study. the natural fluctuations of the weather variables with
direct effects on the measurement devices (Blanco and
RESULTS AND DISCUSSION Folegatti, 2003). However, when analyzing weekly
means, the trends are more obvious (Table 4) and closely
Bulk density and particle density of the substrate correlated with variations in solar radiation measured in
were 0.94 and 2.89 Mg m-3 respectively, giving a total evapotranspiration equivalents (Figure 4). The values
porosity of 67.5%. During post-transplant period (weeks of ETlys and ETo for the different methods showed the
1 to 9), water contents (W) close to 33 kPa tension were following relative levels: ETlys ≈ ANE < CAP ≈ FRE ≈
observed, with a mean of 19.2% ± 1.4 during the first PME < ATMf ≈ ATMm.
three weeks (Figure 3) and 21.3% ± 0.8 for the remainder
M. CASANOVA et al. - METHODS TO ESTIMATE LETTUCE EVAPOTRANSPIRATION… 65

Figure 3. Substrate water content during assay and its water retention curve.

Table 3. Range of climatic variables in the greenhouse during measurement period.

Value
Variable Minimum Maximum Mean ± SD
Temperature, ºC 4.0 36.8 20.0 ± 2.4
Solar radiation, Mj m-2 d-1 3.7 23.4 17.5 ± 5.5
Net radiation, Mj m-2 d-1 4.1 15.1 11.3 ± 3.0
Relative humidity, % 22.5 100.0 66.3 ± 8.6
Vapor pressure deficit, kPa 0.2 2.4 1.3 ± 0.5
Wind speed, m s-1 0.0 0.1 0.0 ± 0.0
SD: standard deviation.

Pan coefficients (Kp) for CAP calculated by Galvani

et al. (1998) with the Snyder (1992) empirical formula
for greenhouse conditions differ from the Kp values given
by Doorenbos and Pruit (1976) that were used in the
present investigation (Table 1). Nevertheless, applying
the variables measured here with that formula, the results
obtained do not differ significantly from those reported
in the present study. Moreover, Gundekar et al. (2008)
recently reported that Kp obtained by empirical formulae
are not different from those found by Doorenbos and Pruitt
(1976), and concluded that the Snyder (1992) formula has
the advantage of simplifying the calculation.
A simple correlation analysis of the daily ETo among
the different devices and equations, revealed high
correlation (R) coefficients (Table 5). Even though the
correlation between ETlys and ETo based in daily values
CAP: Class A pan evaporimeter; ATM: fixed (f) and movable (m) Piche did not exceed 0.35, R increased to values between 0.72
atmometer; ANE: Andersson evaporimeter; FRE: FAO-Radiation
equation; PME: Penmann Monteith equation. and 0.86 when weekly values were analyzed.
The results of the ETlys/ETo, being an approximation
Figure 4. Weekly mean reference evapotranspiration of the crop coefficient (Kc) for lettuce, show the lowest
for different methods and solar radiation inside the values during the first week (post-transplant) with a mean
greenhouse. of 0.3 ± 0.1 for the bulk of methods (Table 6). Between
weeks 2 and 9, a mean of 0.6 ± 0.3 is observed. In this
66 CHILEAN J. AGRIC. RES. - VOL. 69 - Nº 1 - 2009

Table 4. Daily evapotranspiration, weekly means for different methods.

Mean evapotranspiration
Post-transplant time (weeks)
Method 1 2 3 4 5 6 7 8 9 1-9
mm d ± SD
-1

ETlys 0.6 ± 0.5 1.2 ± 0.7 1.0 ± 0.8 1.5 ± 0.6 1.5 ± 0.8 1.8 ± 0.7 1.7 ± 0.6 2.2 ± 1.1 2.1 ± 0.3 1.5 ± 0.9
ETo
CAP 1.6 ± 0.7 1.6 ± 0.7 1.9 ± 1.0 1.4 ± 0.8 2.4 ± 0.8 2.5 ± 0.7 2.9 ± 0.7 3.2 ± 0.6 2.8 ± 0.6 2.3 ± 1.0
ATMf 3.0 ± 0.7 3.3 ± 1.5 3.5 ± 1.5 2.6 ± 1.2 4.1 ± 1.2 4.6 ± 1.5 5.1 ± 1.2 5.8 ± 1.5 4.3 ± 0.8 4.0 ± 1.6
ATMm 3.2 ± 0.7 3.3 ± 1.5 3.5 ± 1.4 2.6 ± 1.2 4.0 ± 1.1 4.6 ± 1.5 5.1 ± 1.1 5.4 ± 1.4 4.1 ± 0.9 4.0 ± 1.5
ANE 1.0 ± 0.4 1.2 ± 0.5 1.3 ± 0.6 0.9 ± 0.6 1.6 ± 0.5 1.8 ± 0.5 2.0 ± 0.4 2.2 ± 0.4 1.8 ± 0.4 1.5 ± 0.7
PME 1.9 ± 0.6 2.2 ± 0.7 2.3 ± 0.5 2.0 ± 0.7 2.6 ± 0.6 2.8 ± 0.6 3.0 ± 0.4 3.2 ± 0.3 2.9 ± 0.4 2.5 ± 0.7
FRE 2.0 ± 1.0 2.3 ± 1.3 2.6 ± 1.1 1.8 ± 1.2 3.0 ± 1.1 3.1 ± 1.0 3.5 ± 0.7 3.9 ± 0.7 3.3 ± 0.7 2.8 ± 1.2
ETlys: lysimeter; CAP: Class A pan evaporimeter; ATM: fixed (f) and movable (m) Piche atmometer; ANE: Andersson evaporimeter; PME: Penmann
Monteith equation; FRE: FAO-Radiation equation; SD: standard deviation.

Table 5. Correlation matrix for reference evapotranspiration (ETo) and lysimeter evapotranspiration (ETlys) (mm d-1),
daily and weekly values (in parentheses).

Method ETlys CAP ATMf ATMm ANE PME FRE

ETlys -
CAP 0.34
(0.80) -
ATMf 0.28 0.85
(0.78) (0.97) -
ATMm 0.25 0.83 0.99
(0.72) (0.96) (0.99) -

ETo
ANE 0.31 0.89 0.96 0.95
(0.80) (0.99) (0.99) (0.98) -
PME 0.31 0.85 0.88 0.87 0.91
(0.79) (0.99) (0.98) (0.97) (0.99) -
FRE 0.35 0.84 0.89 0.87 0.90 0.97
(0.86) (0.99) (0.97) (0.95) (0.99) (0.98) -
ETlys: lysimeter; ETo: Reference evapotranspiration; CAP: Class A pan evaporimeter; ATM: fixed (f) and movable (m) Piche atmometer; ANE: Andersson
evaporimeter; PME: Penmann monteith equation; FRE: FAO-Radiation equation.

context, Allen et al. (1998) recommended a single Kc agronomic management factors like distance between
value for greenhouse lettuce, whereas Katerji and Rana rows and irrigation system. Furthermore, in greenhouse
(2006) reported that it is possible to find differences of conditions, the differences in Kc can also be attributed to
± 40% between the Kc reported by Allen et al. (1998) the size of the greenhouse and the substrate used.
and those obtained experimentally, especially in the If the ETo values reported in Table 4 were multiplied
middle of the growth cycle. These differences are due to by Kc values indicated for each method in Table 6, it
the complexity of the coefficient which integrates various would be possible to estimate ETc for greenhouse lettuce.
functions (Katerji et al., 1991; Testi et al., 2004) such Thus, a linear regression analysis of the paired daily values
as aerodynamic factors linked to crop height, biological ETc-ETlys showed a fairly adequate degree of explanation
factors related to leaf growth and senescence, physical (R2) between the methods (Table 7), and furthermore
factors linked to soil evaporation, physiological factors emphazised close to 1:1 relationships between them. Figure
of stomatal response to the air vapor pressure deficit, and 5 graphically shows the estimated ETc for lettuce compared
M. CASANOVA et al. - METHODS TO ESTIMATE LETTUCE EVAPOTRANSPIRATION… 67

with the lysimeter (ETlys) measurement. Consequently, the

estimated Kc values could be appropriate to determine
lettuce water requirements in controlled environments
and design suitable irrigation schedules to satisfy the
water demand of this crop. Movable atmometers (ATMm)
were excluded from this analysis since their behavior is
almost identical to that of fixed atmometers (ATMf).
Few studies have been conducted in closed
environments such as greenhouses (Fernandes et al., 2003;
Blanco and Folegatti, 2003; 2004; Orgaz et al., 2005) and
very few for lettuce (Galvani et al., 1998; Radin et al.,
2004). Some studies corroborate the results in the present
study, whereas others give contrasting results. Arruda
et al. (2006), for example, suggested ETo estimation
using the following methods, in order of decreasing CAP: Class A pan evaporimeter; ATM: Piche atmometer; ANE: Andersson
evaporimeter; PME: Penmann Monteith equation; FRE: FAO-radiation
importance: Penman-Monteith equation (hourly or daily equation; ETlys: lysimeter.
measurements) inside greenhouse, measurements based
on solar radiation also inside greenhouse, measurements Figure 5. Weekly mean lettuce evapotranspiration for
based on solar radiation outside greenhouse, and ETo different methods and solar radiation inside the
outside greenhouse with any method. greenhouse.

Table 6. Weekly mean crop coefficients (Kc) of lettuce using different greenhouse methods.
Crop coefficienta
Week CAP ATMf ATMm ANE PME FRE Meanh ± SD
1 0.4 0.2 0.2 0.6 0.3 0.3 0.3 ± 0.1
2 0.7 0.4 0.4 1.0 0.5 0.5
3 0.5 0.3 0.3 0.8 0.5 0.4
4 1.1 0.6 0.4 1.6 0.8 0.8
5 0.6 0.4 0.3 0.9 0.6 0.5
0.6 ± 0.3
6 0.7 0.4 0.3 1.0 0.6 0.6
7 0.6 0.3 0.2 0.9 0.6 0.5
8 0.7 0.4 0.2 1.0 0.7 0.6
9 0.7 0.5 0.3 1.2 0.7 0.6
2 to 9 Mean ± SD 0.7 ± 0.1 0.4 ± 0.1 0.3 ± 0.1 1.0 ± 0.2 0.6 ± 0.1 0.6 ± 0.1
Kc = ETlys/ETo; ETlys: evapotranspiration using lysimeter; ETo: reference evapotranspiration;
a

CAP: Class A pan evaporimeter; ATM: fixed (f) and movable (m) Piche atmometer; ANE: Andersson evaporimeter; PME: Penmann Monteith equation;
FRE: FAO-radiation equation; SD: standard deviation.

Table 7. Linear regression of daily lysimeter evapotranspiration of lettuce (ETlys) versus estimated values (ETc) for five
methods (devices and formulae) during the study period.
Method Equation Determination coefficient (R2)
CAP ETlys = 0.98 ETc 0.76
ATM ETlys = 0.97 ETc 0.72
ANE ETlys = 1.00 ETc 0.72
FRE ETlys = 0.92 ETc 0.73
PME ETlys = 1.05 ETc 0.80
CAP: Class A pan evaporimeter; ATM: Piche atmometer; ANE: Andersson evaporimeter; FRE: FAO-Radiation equation; PME: Penmann Monteith
equation.
68 CHILEAN J. AGRIC. RES. - VOL. 69 - Nº 1 - 2009

CONCLUSIONS
métodos muestran similares variaciones en el tiempo,
pero en los diferentes rangos siguientes: ANE < CAP
Under the conditions of this experiment in greenhouse
< FRE < PME < ATM. Además, ETo muestra una clara
in the central zone of Chile, the estimation of the
correlación con la radiación solar. Los coeficientes de
reference evapotranspiration (ETo) with the studied
cultivo (Kc = ETlys/ETo) variaron entre métodos pero se
devices and equations and without taking into account
identificaron tendencias para dos períodos: en la primera
standard reference conditions, can constitute an approach
semana el Kc promedio fue de 0,3 (± 0,1) y dentro de
to estimate crop water requirements.
las semanas 2 a 9 estos valores fueron del orden de 0,6
The variations of ETlys and ETo observed over time
(± 0,3). Los valores de Kc encontrados en invernadero
were large when analyzing on a daily basis whereas on a
fueron inferiores a los adoptados para lechuga en
weekly basis, the variability was reduced and trends more
condiciones de campo. En términos de diseño de riego,
revealed. Thus, the correlations of ETo values between
la evapotranspiración del cultivo puede ser estimada si
the different devices and equations were high with a R
se emplean los métodos de este estudio, con la salvedad
coefficient fluctuating between 0.83 and 0.99 on a daily
que se apliquen los coeficientes de cultivo apropiados.
basis and from 0.95 to 0.99 on a weekly basis. On the
El hecho que ANE muestre los valores más cercanos a
other hand, R between ETlys and ETo fluctuated between
los de ETlys, combinado con su conveniencia en cuanto a
0.25 and 0.35 on a daily basis but reached values between
costo y manejo, lo convierte en una alternativa ventajosa
0.72 and 0.86 for weekly means.
respecto a los otros métodos.
The relative levels of ETlys and ETo values followed the
order: ETlys ≈ ANE < CAP ≈ FRE ≈ PME < ATMf ≈ ATMm,
Palabras clave: evapotranspiración, invernadero,
and showed a clear correlation with solar radiation.
lechuga, lisímetro, Lactuca sativa.
The use of an Andersson evaporimeter in greenhouses
is recommended and adopting a device factor of
0.7 generated ETo almost equal to the actual crop
evapotranspiration measured in the lysimeters (ETlys). LITERATURE CITED
Estimated crop coefficient (Kc) based on the ETlys/
ETo ratio, varied amongst the methods but overall mean Allen, R.G., L.S. Pereira, D. Raes, and M. Smith. 1998.
values were identified for two growth stages of the crop: Crop evapotranspiration guidelines for computing
first week post-transplant (0.3 ± 0.1) and second week crop water requirements. Irrigation and Drainage
onwards (0.6 ± 0.3). Paper N° 56. 300 p. FAO, Rome, Italy.
The lettuce evapotranspiration values (ETc) indirectly Allen, R.G., L.S. Pereira, D. Raes, y M. Smith. 2006.
obtained using devices and equations (ETo·Kc) were Evapotranspiración del cultivo. Guías para la
similar to the values measured directly by lettuce determinación de los requerimientos de agua de los
lysimeter water balance (ETlys), and showed determination cultivos. Cuadernos de Riego y Drenaje N° 56. 322 p.
coefficients (R2) between 0.72 and 0.80. FAO, Roma, Italia.
Andersson, S. 1969. A new evaporimeter. Grundförbättring
(Journal of Agricultural Land Improvement) 22:59-
RESUMEN 66.
Araujo, G., F. Bussmeyer, and E. Sakai. 2005. Crop
Métodos de estimación de la evapotranspiración de
coefficient for irrigated beans derived using three
un cultivo de lechuga en condiciones de invernadero,
reference evaporation methods. Agric. For. Meteorol.
zona central de Chile. Este estudio evalúa cinco
135:135-143.
métodos de estimación de la evapotranspiración de
Arruda, F.B., R. Pires, E. Sakai, P. Furlani, A. Torre Neto,
cultivos en condiciones de invernadero. Sus desempeños
E. Silva, e R. Calheiros. 2006. Evapotranspiração
se evalúan en base a la evapotranspiración determinada
de referência estimada em estufa com cobertura
directamente a partir de mediciones de balance
em polietileno em Campinas. In Proceeding VII
hídrico (ETlys), para un cultivo de lechuga (Lactuca
Congreso Latinoamericano y del Caribe de Ingeniería
sativa L.) bajo riego y un período de 9 semanas. Se
Agrícola, V Congreso Internacional de Ingeniería
compararon los valores diarios de evapotranspiración
Agrícola, Chillán, Chile. 10-12 de mayo (CD-Rom).
de referencia (ETo) provenientes de una bandeja clase
Universidad de Concepción, Facultad de Ingeniería
A (CAP), atmómetros de Piche (ATM), evaporímetros
Agrícola e Instituto de Investigaciones Agropecuarias,
Andersson (ANE), la ecuación FAO-Radiación (FRE)
Centro Regional de Investigación Quilamapu, Chillán,
y la ecuación FAO-Penman-Monteith (PME). Los
Chile.
M. CASANOVA et al. - METHODS TO ESTIMATE LETTUCE EVAPOTRANSPIRATION… 69

Baille, A. 1996. Principle and methods for predicting Johansson, W. 1969. The influence of meteorological
crop water requirement in greenhouse environments. elements on the evaporation from the Andersson
Cahier Options Méditerranéennes 31:177-187. evaporimeter. Grundförbättring (Journal of
Ben-Gal, A., and U. Shani. 2002. A highly conductive Agricultural Land Improvement) 22:83-105.
drainage extension to control the lower boundary Kassam, A., and M. Smith. 2001. FAO methodologies
condition of lysimeters. Plant Soil 239:9-17. on crop water use and crop water productivity. Paper
Blanco, F., and M. Folegatti. 2003. Evapotranspiration CWP-M07. 18 p. FAO, Rome, Italy.
and crop coefficient of cucumber in greenhouse. Rev. Katerji, N., A. Hamdy, A. Raad, et M. Mastrorilli. 1991.
Bras. Eng. Agríc. Ambient. 7:285-291. Conséquence d’une contrainte hydrique appliquée à
Blanco, F., and M. Folegatti. 2004. Evaluation of différents stades phénologiques sur le rendement des
evaporation-measuring equipment for estimating plantes de poivron. Agronomie 11:679-687.
evapotranspiration within a greenhouse. Rev. Bras. Katerji, N., and G. Rana. 2006. Modelling evapotranspira-
Eng. Agríc. Ambient. 8:184-188. tion of six irrigated crops under Mediterranean climate
Bouchet, R.J. 1963. Evapotranspiration réelle, conditions. Agric. For. Meteorol. 138:142-155.
évapotranspiration potentielle, et production agricole. Kläring, H.P. 2001. Strategies to control water and nutrient
Ann. Agron. 14:743-824. supplies to greenhouse crops. A review. Agronomie
Dane, J.H., and G.C. Topp (eds.) 2002. Methods of soil 21:311-321.
analysis. Part 4. Physical methods (V1). 1.692 p. López-Urrea, F., F. Martín de Santa, C. Fabeiro, and A.
American Society of Agronomy, Soil Science Society Moratalla. 2006. An evaluation of two hourly reference
of America, Madison, Wisconsin, USA. evapotranspiration equations for semiarid conditions.
De Tourdonnet, S., J. Meynard, F. Lafolie, J. Roger- Agric. Water Manage. 86:277-282.
Estrade, J. Lagier, and M. Sebillotte. 2001. Non- Messing, I. 1998. A simple practical evaporimeter:
uniformity of environmental conditions in greenhouse comparison of Andersson evaporimeter with Class A
lettuce production increases the risk of N pollution pan, Piche atmometer, and Penman evaporation. Arid
and lower product quality. Agronomie 21:297-309. Soil Res. Rehab. 12:275-290.
Doorenbos, J., y W. Pruitt. 1976. Las necesidades de agua Messing, I., M. Afors, K. Radkvist, and E. Lewan. 1998.
de los cultivos. Estudio Riego y Drenaje N° 24. 194 p. Influence of shelterbelt type on potential evaporation
FAO, Roma, Italia. in an arid environment. Arid Soil Res. Rehab. 12:123-
El Moujabber, M., et R. Abi Zeid Daou. 1999. Gestion 138.
durable de l’eau pour les cultures protégées au Liban. Messing, I., and K. Akrimi. 1998. Evaporative demand,
Cahiers Agricultures 8:309-313. water use, and crop yield at sheltered and unsheltered
Fernandes, C., J.E. Cora, and J.A.C. de Araujo. 2003. sites in an arid environment. Arid Soil Res. Rehab.
Reference evapotranspiration inside greenhouses. Sci. 12:223-235.
Agríc. (Piracicaba, Braz.) 60:591-594. Nandagiri, L., and G. Kovoor. 2006. Performance
Frevert, D., R. Hill, and B. Braaten. 1983. Estimate of evaluation of reference evapotranspiration equations
FAO evapotranspiration coefficients. J. Irrig. Drain. across a range of Indian climates. J. Irrig. Drain. Eng.
109:265-270. 132:238-249.
Galvani, E., R.T. Dantas, J.F. Escobedo, e E.S. Klosowski. Navarette, M., M. Sebillotte, B. Jeannequin, and J.
1998. Parámetros meteorológicos em cultura de Alface Longuenesse. 1993. Inter-plant variability: a daily
(Lactuca sativa L.) cultivada em casas de vegetação criterion for the grower, an open question for the
com orientações leste-oeste, norte-sul e condições modeler. Acta Hort. (ISHS) 328:211-218.
externas. Rev. Brás. Agrometeorol. 6:157-163. Orgaz, F., M.D. Fernández, S. Bonachela, M. Gallardo, and
Gundekar, H.G., U.M. Khodke, S. Sarkar, and R.K. Rai. E. Federes. 2005. Evapotranspiration of horticultural
2008. Evaluation of pan coefficient for reference crop crops in an unheated plastic greenhouse. Agric. Water
evapotranspiration for semi-arid region. Irrig. Sci. Manage. 72:81-96.
26:169-175. Papaioannou, G., S. Kaloudis, and P. Kerkides. 1998.
INE. 2007. Resultados Preliminares de VII Censo On the proper employment of Piche evaporimeters
Agropecuario y Forestal de Chile. Temporada 2006- in estimating evapotranspiration. Int. J. Climatol.
2007. Cuadro 9. Instituto Nacional de Estadísticas 18:1247-1260.
(INE), Santiago, Chile. Disponible en http://www. Papaioannou, G., K. Vouraki, and P. Kerkides. 1996.
censoagropecuario.cl/noticias/08/6/10062008.html Piche evaporimeter data as a substitute for Penman
(Accessed July 2008). equation’s aerodynamic term. Agric. For. Meteorol.
82(1-4):83-92.
70 CHILEAN J. AGRIC. RES. - VOL. 69 - Nº 1 - 2009

Pollet, S., and P. Bleyaert. 2000. Application of the Penman- Snyder, R.L. 1992. Equation for evaporation pan to
Monteith model to calculate the evapotranspiration evapotranspiration conversions. J. Irrig. Drain. Eng.
of head lettuce (Lactuca sativa L. var. capitata) in 118:977-980.
glasshouse conditions. Acta Hort. (ISHS) 519:151- Testi, L., F.J. Villalobos, and F. Orgaz. 2004.
161. Evapotranspiration of a young irrigated olive orchard
Radin, B., C. Reisser Júnior, R. Matzenauer, e H. in southern Spain. Agric. For. Meteorol. 121:1-18.
Bergamaschi. 2004. Crescimento de cultivares de Turco, J.E.P., D. Perecin, e D.L. Pinto Jr. 2008. Influência
alface conduzidas em estufa e a campo. Hort. Bras. da acurácia de instrumentos de medidas na comparação
22(2):178-181. de métodos de estimativa da evapotranspiração de
Santibañez, F., y J.M. Uribe. 1992. Atlas agroclimático referência. Irriga (Botucatu-Brasil) 13(1):63-80.
de Chile. 65 p. Universidad de Chile, Facultad de Vaughan, P.J., T.J. Trout, and J.E. Ayars. 2007. A
Ciencias Agrarias y Forestales. Editorial Universitaria, processing method for weighing lysimeter data and
Santiago, Chile. comparison to micrometeorological ETo predictions.
Sarraf, S. 2004. Irrigation management and maintenance in Agric. Water Manage. 88:141-146.
greenhouse crops in Lebanon. p. 83-93. In Abdelhaq,
H. (ed.) Integrated production and protection in
greenhouse vegetable crops. Technical Booklet. FAO,
Rome, Italy.
Smith, M., R. Allen, and L. Pereira. 1998. Revised FAO
methodology for crop-water requirements. p. 51-
58. In Proceedings of management of nutrients and
water in rainfed arid and semi-arid areas, Vienna,
Austria. 26-29 May 1997. A Consultants Meeting,
FAO/IAEA (Food and Agriculture Organization of
the United Nations/International Atomic Energy
Agency), Division of Nuclear Techniques in Food and
Agriculture, Vienna, Austria.