Agricultural Water Management 142 (2014) 135–143

Contents lists available at ScienceDirect

Agricultural Water Management

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

Crop coefficient and evapotranspiration of grain maize modified

by planting density in an arid region of northwest China
Xuelian Jiang a , Shaozhong Kang a,∗ , Ling Tong a , Fusheng Li b , Donghao Li a , Risheng Ding a ,
Rangjian Qiu a
a
Center for Agricultural Water Research in China, China Agricultural University, Beijing 100083, China
b
College of Agriculture, Guangxi University, Nanning, Guangxi 530005, China

a r t i c l e i n f o a b s t r a c t

Article history: In order to investigate crop coefficient (Kc ) and evapotranspiration (ET) of grain maize modified by plant-
Received 9 December 2013 ing density, a field experiment was conducted from March to October in 2012 and 2013 in an arid region
Accepted 9 May 2014 of northwest China. Five planting densities, i.e. 67,500, 82,500, 97,500, 112,500 and 127,500 plants ha−1
Available online 2 June 2014
were conducted in 2012, and a higher planting density of 142,500 plants ha−1 was added in 2013. We
introduced a density ratio (Kdensity ) that is a function of leaf area index (LAI) to account for the effect
Keywords:
of planting density on Kc , and the daily Kc can be computed by Kdensity multiplying Kc at the reference
Crop coefficient
planting density (127,500 plants ha−1 ). The Allen method considering an adjustment coefficient (Acm ),
Evapotranspiration
Density coefficient
the single and dual Kc methods considering a density coefficient (Kd ), and Kdensity method were used
Density ratio to calculate Kc , and then the ET estimated by reference evapotranspiration and Kc was validated by the
measured ET using the eddy covariance system in 2012 and 2013. Results showed that higher planting
density increased ET and Kc and lowered soil evaporation and evaporation coefficient within the planting
densities of the experiments. Daily ET estimated by the Allen method performed very well after the end
of maize development stage, with mean bias error (MBE) of −0.06 and 0.12 mm d−1 , root mean square
error (RMSE) of 0.84 and 0.80 mm d−1 in 2012 and 2013 respectively. The single and dual Kc methods
can better simulate the daily ET when actual LAI was below the maximum LAI. Compared to the three
above-mentioned methods, the Kdensity method had higher accuracy in estimating daily ET over the whole
stage, with higher R2 and lower MBE and RMSE, indicating that Kdensity method had better performance
in calculating daily ET under different planting densities of grain maize.
© 2014 Elsevier B.V. All rights reserved.

1. Introduction (Papadopoulos, 1985; Papadopoulos and Pararajasingham, 1997).

Crop evapotranspiration (ET) is of importance in irrigation man-
agement and water allocation (Kang et al., 2002). It is affected
by many factors, e.g. weather parameters, crop characteristics,
irrigation scheduling and field management (Allen et al., 1998;
Kang et al., 2003). It is also affected by planting density (Allen et al.,
1998; Allen and Pereira, 2009). Higher planting density increases
the radiation intercepted by the plant canopy and reduces the radi-
ation at the soil surface (Papadopoulos and Pararajasingham, 1997),
but the increased radiation absorbed by plant canopy can lead to
intensify soil water consumption (Reicosky et al., 1985). The rate
of transpiration (T) tends to increase with the increase of plant-
ing density due to the well exposed leaf area at the top of plants
∗ Corresponding author. Tel.: +86 10 62737611; fax: +86 10 62737611.
E-mail addresses: kangsz@cau.edu.cn, kangshaozhong@tom.com (S. Kang).
Chen et al. (2010) found that higher plant density increases the T of
winter wheat but decreases soil evaporation (E) in the North China
Plain. Eberbach and Pala (2005) also reported that higher planting
density can result in higher ET and lower E.
The Priestley–Taylor method (Ding et al., 2013; Priestley and
Taylor, 1972), Penman–Monteith method (Agam et al., 2010;
Monteith, 1965; Utset et al., 2004), Shuttleworth–Wallace method
(Shuttleworth and Wallace, 1985; Teh et al., 2001), and Clumping
method (Brenner and Incoll, 1997; Domingo et al., 1999; Zhang
et al., 2008) are often used to estimate ET in the field. But ‘crop
coefficient (Kc ) × reference evapotranspiration (ET0 )’ is a useful and
convenient method to estimate ET (Allen et al., 1998; Doorenbos
and Pruitt, 1975; Kang et al., 2003). Kc can be calculated by dif-
ferent methods (i.e. the single and dual crop coefficient methods)
(Allen et al., 1998; Jensen et al., 1990). In the single crop coefficient
method, the effect of crop transpiration and soil evaporation are
combined into a single crop coefficient. The dual crop coefficient

http://dx.doi.org/10.1016/j.agwat.2014.05.006
0378-3774/© 2014 Elsevier B.V. All rights reserved.
136 X. Jiang et al. / Agricultural Water Management 142 (2014) 135–143

method has two coefficients, i.e. the basal crop coefficient (Kcb ) to
represent primarily the transpiration component of ET, and the soil
evaporation coefficient (Ke ) to describe evaporation from the soil
surface (Allen et al., 1998; Rosa et al., 2012; Zhao et al., 2013). As
soil evaporation may fluctuate daily as a result of irrigation or rain-
fall, applying Kc that expresses only the time-averaged effects on
ET, is more useful and convenient than computing a daily Kc based
on a dimensionless ‘stress’ coefficient (Ks ), Kcb , and Ke (Allen et al.,
1998; Allen and Pereira, 2009).
The Kc is chiefly affected by the amount, type, density and height
of vegetation under the assumption that the ET0 accounts for nearly
all variation caused by climate factors (Allen and Pereira, 2009). The
Kc tends to decrease with the decrease of leaf area or plant density
(Allen and Pereira, 2009; Qiu et al., 2013). Allen et al. (1998) found
that during the mid-season stage of crops, the vegetation nearly
covers the soil and varies with planting density, the Kc values should
be adjusted by a factor (Acm ) depending on the actual vegetation
development (The Allen method). Allen and Pereira (2009) pro-
posed a density coefficient (Kd ) to estimate both basal and average
Kc . In the dual Kc coefficient method, the basal crop coefficient (Kcb )
should be formalized using the estimated Kc during the mid-season
stage (at peak plant size or height) for full vegetation (Kc full ), the
minimum Kc for bare soil (Kc min ) and Kd (Kcb method). The single
crop coefficient (Kcm ) was similarly estimated and was adjusted
using a Ksoil representing background evaporation from wet soil
(Kcm method).
At present, there are many studies about the effects of planting
density on yield and water productivity (Chen et al., 2010; Eberbach
and Pala, 2005; Griesh and Yakout, 2001; Salah et al., 2008). But only
fewer studies have been conducted to assess the response of Kc and
ET to planting density (Allen et al., 1998; Allen and Pereira, 2009;
Daisuke et al., 2011; Qiu et al., 2013). The method proposed by Allen
et al. (1998) considered an adjustment for crops at the middle sea-
son stage that the soil is usually nearly completely covered by the
vegetation. The equations proposed by Allen and Pereira (2009) is
only applied to the mid and eventual late season stage. Whether
these formulas are applicable to the grain maize with different
planting densities in the arid region of northwest China should be
further studied.
As the planting area of grain maize in the arid area of northwest
China is rapidly developed, understanding the response of ET and
Kc of grain maize to planting density is important in improving the
irrigation management in the region with limited water resources.
Thus the objectives of this study were to (1) evaluate the effect
of planting density on crop coefficient and crop evapotranspira-
tion of grain maize; (2) develop a relatively simple and accurate
method with density ratio (Kdensity ) to calculate Kc and ET, and (3)
compare the accuracy of estimated ET using the Allen, Kcm , Kcb and
Kdensity methods, so as to validate whether Kdensity method had bet-
ter performance in calculating ET under different planting densities
or not.
2. Materials and methods
2.1. Experimental site
The experiment was conducted in two consecutive years during
April to September in 2012 (first season) and April to September in
2013 (second season) at Shiyanghe Experimental Station of China
Agricultural University, located in Wuwei City, Gansu Province
of northwest China (37◦ 52 N, 102◦ 50 E, altitude 1581 m). The
site is in a typical continental temperate climate zone with mean
annual precipitation of 164.4 mm, mean annual pan evaporation of
2000 mm. Average groundwater table is below 25 m, mean annual
sunshine duration is over 3000 h, frost-free days are 150 d and mean
annual temperature is 8.8 ◦ C.
The experimental soil is a light sandy loam, with an average
soil dry bulk density of 1.38 g cm−3 , average field water capacity
( FC ) of 0.29 cm3 cm−3 and wilting point ( WP ) of 0.12 cm3 cm−3 .
The electrical conductivity of irrigation water is 0.52 dS m−1 .
2.2. Experimental design and plant material
In the first season, the experiment had five planting densities, i.e.
67,500, 82,500, 97,500, 112,500 and 127,500 plants ha−1 , referred
as D1 , D2 , D3 , D4 and D5 . In the second season, a higher planting den-
sity of 142,500 plants ha−1 was added, referred as D6 . To achieve
different planting densities, different plant spacing in the row, i.e.,
37, 30, 25, 22, 20, 18 cm respectively, were for D1 , D2 , D3 , D4 , D5
and D6 , each planting density has the same row spacing of 40 cm.
Each treatment had three replicates and all plots were arranged in a
randomized complete block design. The plot size was 9.6 m × 5 m.
Grain maize (Zea mays L. cultivar Golden northwest No. 22) was
sown in one-line male plants and seven-line female plants. Female
plants were sowed on April 16 and 13 in the first and second sea-
sons, respectively, and two batches of male plants were planted on
April 23 and 26 in the first season, on April 19 and 22 in the sec-
ond season. The lengths of the four growth stages in each season
were divided according to Allen et al. (1998) (Table 1) and local
observations.
Before the sowing, the whole plots were fertilized with N
of 375 kg ha−1 , P2 O5 of 225 kg ha−1 and K2 O of 300 kg ha−1 as
a basal fertilizer. After fertilized, to ensure the emergence (the
temperature is low in April), each plot was covered with six
0.015 mm thick plastic films with each length of 5 m and width
of 1.2 m, the width of bare soil between two plastic films was
0.4 m. The plots were top-dressed with N of 600 kg ha−1 on June
10 and May 31 in the first and second seasons, respectively. Fer-
tilizers supply during the growing season was consistent with
the local management to ensure luxury nutrients conditions to
all planting density treatments. The irrigation method was bor-
der irrigation. During the whole growth stage, the experimental
plots were irrigated 5 times, i.e. on 6 June, 26 June, 13 July, 8

Table 1
Environmental variables at different growth stages of grain maize. Rs is solar radiation, Ta – air temperature, RH – relative humidity, P – precipitation and ET0 – reference
evapotranspiration.

Season Growth stage Rs (W m−2 ) Ta (◦ C) RH (%) P (mm) ET0 (mm)

2012 Initial (April 23–May 23) 224.6 15.6 21.5 9.8 99.1
Development (May 24–July 2) 267.5 20.6 20.3 18.7 166.9
Middle (July 3–August 22) 298.0 21.1 29.5 71.9 204.6
Late (August 23–September20) 248.5 17.5 31.3 29.0 124.6
Whole 259.7 18.7 25.6 129.4 595.2

2013 Initial (April 19–May 19) 239.2 17.4 33.2 6.0 114.3
Development (May 20–June 28) 221.5 20.3 49.8 18.2 165.1
Middle (June 29–August 18) 197.8 21.4 63.2 34.4 179.3
Late (August 19–September 12) 181.4 17.9 62.7 13.8 90.3
Whole 210.0 19.2 52.2 72.4 549.0
 X. Jiang et al. / Agricultural Water Management 142 (2014) 135–143
August and 27 August in the first season, on 6 June, 30 June,
14 July, 4 August and 26 August in the second season, and the
amount of each irrigation event was 100 mm.
2.3. Measurements and methods
2.3.1. Meteorological data
The measured meteorological data including precipitation (P),
solar radiation (Rs ), air temperature (Ta ), wind speed at 2 m above
ground (u2 ) and wind direction, and relative humidity (RH) during
the experimental period, were continuously observed by a standard
automatic weather station (Hobo, Onset Computer Corp., USA). The
data were sampled every 5 s interval, and 15 min averages were
calculated and recorded using a date logger.
2.3.2. Leaf area index
From 15 days after sowing, eight plants in each plot were
randomly selected to measure leaf length and maximum width
using a tape at intervals of 7–10 days during the whole growth
stage. Leaf area was calculated by summing rectangular area (leaf
length × maximum width) of each completely developed leaf mul-
tiplied by a factor of 0.74 (Li et al., 2008). Leaf area index (LAI) is
defined as the ratio of leaf area to land area.
2.3.3. Soil water balance
Soil water balance is an indirect method to estimate crop evapo-
transpiration (ET, mm) using the following equation (Rana and
Katerji, 2000):
ET = P + I + W − D − R − W
where P is effective precipitation (mm), I irrigation amount (mm),
W the capillary rise to the root zone (mm), R surface runoff (mm),
D the drainage from the root zone (mm) and W water content
change in the root zone and W can be calculated as:
W = Wt2 − Wt1
where Wt1 and Wt2 are the mean water content in the root zone at
time t1 and t2 , respectively.
The surface runoff (R) can be ignored since the experimental
sit is flat and the precipitation is not intensive. The capillary rise
(W) can also be negligible because the groundwater level is beyond
25 m. According to the observations of soil moisture, soil water
content at the 90–100 cm layer did not change before and after
irrigation, thus the drainage can be ignored. Eq. (1) can be simplified
as:
ET = P + I − W
Soil water content of each plot was measured by Diviner 2000
system (Sentek Pty Ltd., Australia). Calibration of the Diviner 2000
was conducted by determining the water contents of soil sam-
ples using gravimetric method as near as possible to the probe in
both seasons. Two PVC access tubes with the depth of 1.0 m were
installed below the plastic mulch and the bare soil in each plot.
Measurements were made at 0.1 m intervals in maximal soil depth
of 1.0 m every 5–7 days as well as before and after each irrigation
and rainfall event.
2.3.4. Soil evaporation
Micro-lysimeters were installed to measure soil evaporation (E)
in the both seasons. Two micro-lysimeters cylinders, which were
made from PVC tubes with a height of 20 cm and diameter of 10 cm,
were placed within bare soil between two plastic films in each plot.
The cylinders were taken out from soil and weighed daily by an
electronic scale with the precision of 0.1 g. The micro-lysimeters
were reinstalled after each irrigation and heavy rain.
137
2.3.5. Eddy covariance measurement
In the study, the EC system was installed in the middle of field
(length 400 m and width 200 m) that was 80 m away from the
experimental sit of planting density. Grain maize was the major
crop in the surrounding region, and has a large planting area to
provide adequate fetch length for EC system. Measured ET in 2012
and 2013 by an eddy covariance (EC) system was referenced to vali-
date ET model. The planting densities at the sites were 112,500 and
97,500 plants ha−1 , respectively, for the first and second seasons.
The maize varieties and crop field management were the same as
the planting density experiment in both seasons. The maximum
height of the maize is 1.60 m, and the sensor height was adjusted
weekly to keep the relative height of 1.0 m between sensors and
maize canopy constant. The minimum fetch length is 100 m.
The EC system consisted of a fast response 3-D sonic anemome-
ter/thermometer (model CSAT3), a Krypton hygrometer (model
KH20) and a temperature and humidity sensor (model HMP45C).
A data logger (model CR5000, Campbell Scientific Inc., USA) was
used to record all data of vertical fluctuations of wind, tempera-
ture and water vapor density at 0.1 s intervals, and temperature
and humidity at 30 min intervals.
Baldocchi (2003) proposed that temperature, humidity, fluctu-
ations of wind speed measured by eddy covariance system (EC) can
be used to calculate the latent and sensible heat fluxes (they can be
converted to water consumption), and its equations are as follows:
ET = a w q (4)
Hs = Cp a w T  (5)
(1) where ET and Hs are the latent and sensible heat fluxes (W m−2 ), a
air density (kg m−3 ), w the fluctuations of vertical wind, q the fluc-
tuations of vapor density, W  q the covariance between fluctuations
of vertical wind w and vapor density q ,  the latent heat of water
vaporization (J kg−1 ), Cp the specific heat of dry air at constant pres-
sure (J kg−1 K−1 ), T the fluctuations of sonic temperature (K), w T 
(2) the covariance between fluctuations of w and sonic temperature
T .
The latent and sensible heat fluxes were computed accord-
ing to the eddy covariance methodology using standardized
routines (Mauder and Foken, 2004) and the EdiRe software
(http://www.geos.ed.ac.uk/abs/research/micromet/EdiRe). The
steps for corrections of the eddy covariance measurements
included: (1) detection and elimination of raw peaks, (2) coor-
dinate rotation using the planar fit method (Paw et al., 2000;
Finnigan et al., 2003), (3) correction of oxygen following Tanner
and Greene (1989), (4) correction for density following Webb
et al. (1980) and (5) data gap filling using the MDV (mean diurnal
(3)
variation) method (Falge et al., 2001).
2.4. Evapotranspiration model
Under standard conditions that no limitations are placed on crop
growth from soil water and salinity stresses, crop density, pest and
diseases, weed infestation or fertility, crop evapotranspiration can
be calculated as follows (Allen et al., 1998):
ET = Kc ET0 (6)
In the dual crop coefficient approach, the Kc is divided into two
coefficients, one for crop transpiration, i.e. the basal crop coefficient
(Kcb ), and another for soil evaporation (Ke ) (Allen et al., 1998):
ET = (Kcb + Ke ) ET0 (7)
Reference crop evapotranspiration (ET0 ) is calculated according
to the FAO Penman–Monteith equation (Allen et al., 1998).
The Kc value tends to decrease when planting density or actual
leaf area index (LAI) is below the maximal LAI. Allen et al. (1998)
138 X. Jiang et al. / Agricultural Water Management 142 (2014) 135–143

proposed an empirical formula (Allen method) to calculate Kc at ratio (Kdensity ) was defined in this study to describe the relation-
the mid-season stage under different planting densities: ship between relative ET and relative LAI under different planting
density, as follows:
Kc adj = Kc − Acm (9)
Kc d ETc d
 LAI ˇ
where Kc adj is adjusted Kc , Kc is Kc from the same crop under
Kdensity = = = (16)
normal standard crop management practice, and Acm is adjusted Kc dm ETc dm LAIdense
coefficient.
 LAI 0.5 where Kc d and ETc d is the adjusted Kc and ET under different plant-
Acm = 1 − (10) ing densities, Kc dm and ETc dm is the Kc and ET from the same crop
LAIdense
under well standard crop management practice, actively growing
where LAI is actual leaf area index, LAIdense is expected leaf area and completely shading the ground (In this study, planting density
index for the same crop under normal standard crop management of 127,500 plants ha−1 was selected as reference standard); ˇ is a
practices. adjustment coefficient.
Allen et al. (1998) also proposed that LAI is a function of planting ETc dadj at any growth stage under different plant densities is
density: obtained using the following equation:
LAI
 population ˛
= (15) ETc dadj = Kdensity × Kc dm ET0 (17)
LAIdense populationdense
where LAIdense is particular plant species under normal ‘dense’ or The estimated ET by the methods of Allen, Kcm and Kcb and
pristine growing condition; population is number of plants per unit Kdensity , marked as ETA , ETS , ETD and ETK respectively, was validated
area of soil surface under the actual growing condition [no. m−2 ]. by the measured ET using eddy covariance system (ETEC ) in both
Populationdense is number of plants per unit area of soil surface seasons.
under the ‘dense’ or pristine growing condition [no. m−2 ]. ˛ = 0.5
when population is formed from vigorous growing plants and ˛ = 1
2.5. Date analysis and evaluation of method performance
when plants are less vigorous.
Allen and Pereira (2009) introduced a density coefficient (Kd )
SPSS 13.0 version software (SPSS Inc., USA) was used to statisti-
to describe the effect of planting density on basal and single Kc . A
cally analyze the date. Analysis of variance (one-way ANOVA) was
basal crop coefficient (Kcb ) (Kcb method) for any period having LAI
performed and the significance of the means were performed using
below the maximal LAI is calculated as:
Duncan’s multiple range tests at significant level of 0.05.
Kcb adj = Kc min + Kd (Kcb full − Kc min ) (11) The performance of the adjusted ET was based on a linear
regression between estimated (Ei ) and observed (Qi ) values of ET.
where Kcb adj is the approximation for Kcb for the conditions rep-
Meanwhile, mean bias error (MBE) and root mean square error
resented by Kd , Kc full is the estimated Kc during the mid-season
(RMSE) were included. MBE and RMSE can be expressed as follows
stage (at peak plant size or height) for full vegetation, Kc min is the
(Poblete-Echeverría and Ortega-Farias, 2009):
minimum Kc for bare soil, Kd is density coefficient, which can be
estimated as a function of measured LAI by vegetation. The Kd is
1
n
calculated using the approaches proposed by Allen et al. (1998): MBE = (Ei − Qi ) (18)
n
Kd = 1 − e[−0.7 LAI] (12) i=1

Kcb full in Eq. (11) can be approximated as a function of mean crop

height and adjusted for climate according to Allen et al. (1998):
  h 0.3 
Kcb full = Fr min (1.0 + 0.1h, 1.20) + [0.04 (u2 − 2) − 0.004 (RHmin − 45)] (13)
3

where h is mean maximum crop height (m), u2 mean wind speed

at 2.0 m height during the mid-season (m s−1 ), RHmin the average
1/2
2
1
minimum daily relative humidity during the mid-season and Fr n

resistance correction factor. For the most annual agricultural crops,
the average resistance is approximately 100 s m−1 , the Fr can be
taken as 1 (Allen et al., 1996).
The method of Eq. (11) can be similarly applied to estimate the
single Kc (Kcm method), similar to the dual Kc approach:
  Kc full − Ksoil

Kcm = Ksoil + Kd max Kc full − Ksoil , (14)
2 maize
where Kcm is the average Kc considering the mean impact of evap-
oration from soil, Kc full represents Kc from a fully covered soil with
some background evaporation and can be estimated as equivalent
to Kcb full or equal to Kcb full + 0.05 (Allen et al., 1998; Wright, 1982),
Ksoil represents the average Kc from the non-vegetated (exposed)
portion of the soil surface during the same period as Kd and Kc full ,
and the Kd is defined in Eq. (12).
Previous studies have shown that crop evapotranspiration
increases linearly with the increasing of planting density (Liu et al.,
2012; Qiu et al., 2013), the LAI is a power function of populations
(Eq. (15)), and the function of LAI and Kc is similar, thus a density
RMSE = (Ei − Qi ) (19)
n
i=1
3. Results and discussion
3.1. Environmental variables at different growth stages of grain
Average daily environmental conditions at different growth
stages of grain maize is shown in Table 1. In the two growing sea-
sons, average daily Rs was 259.7 and 210.0 W m−2 , average daily Ta
was 18.7 and 19.2◦ C, average daily RH was 25.6 and 52.2%, average
daily u2 was 0.76 and 0.56 m s−1 , total P was 129.4 and 72.4 mm, and
total reference evapotranspiration (ET0 ) was 595.2 and 549.0 mm,
respectively. Environmental conditions varied in the two seasons,
daily averaged Rs , P and ET0 in the first season were 24, 78 and
8% higher while daily averaged RH and Ta were 3 and 51% lower
compared to those in the second season.
 X. Jiang et al. / Agricultural Water Management 142 (2014) 135–143 139

6 6
D1 D1
(a) D2 (b) D2
5 D3 5 D3
D4 D4
D5 D5
4 4

LAI (m2 m-2)

D6
LAI (m2 m-2)
3 3

2 2

1 1

0 0
5-1 5-31 6-30 7-30 8-29 9-28 5-1 5-31 6-30 7-30 8-29 9-28
Date (Month-Day) Date (Month-Day)

Fig. 1. Variations of leaf area index (LAI) for grain maize with different planting densities in 2012 (a) and 2013 (b). Inserted error bars denote standard error of the mean. D1 ,
D2 , D3 , D4 , D5 and D6 are 67,500, 82,500, 97,500, 112,500, 127,500 and 142,500 plants ha−1 , respectively.

3.2. Effect of planting density on leaf area index
Fig. 1 shows dynamics of LAI in both seasons. Variation of LAI
showed a single peak curve under different planting densities. LAI
started to rise at the development season, reached to the maxi-
mum at the mid-season, and declined at the late season due to leaf
yellowing or wilting. LAI increased with the increasing of planting
density, and the LAI difference among different planting densities
was significant (P < 0.05) except at the initial stage and the begin-
ning of crop development stage. Compared to low planting density
(D1 ), the highest planting density (D5 and D6 in the first and sec-
ond seasons, respectively) increased LAI by about 60% and 70% at
the mid-season stage, respectively, in the first (Fig. 1a) and second
seasons (Fig. 1b).
3.3. Effect of planting density on evapotranspiration of grain
maize
As shown in Table 2, total ET and E in the first season were
close to those in the second season for all treatments due to sim-
ilar planting time, length of the growth season and varieties of
crop. According to water content in the root zone, the root zone
depletion (Dr ) was larger than readily available water (RAW) for all
treatments at the different growth stage, all planting density treat-
ments were under optimal growing conditions. Higher planting
density increased the total ET but decreased the total E signifi-
cantly (P < 0.05) over the entire season. The differences of ET and
E among different planting densities occurred mainly at the crop
development and middle stages in both seasons, and the ET differ-
ences among different planting densities may result from the LAI
differences. Increased planting density had higher LAI (Fig. 1), which
will reduce soil evaporation but increase total radiation absorbed
by plant canopy significantly, which will intensify crop water con-
sumption (Reicosky et al., 1985).
3.4. Seasonal variation of crop coefficient with different planting
densities
Table 3 shows the crop coefficient (Kc ) calculated from ET/ET0
and soil evaporation coefficient (Ke ) calculated from E/ET0 at differ-
ent growth stages of grain maize with different planting densities in
the first and second seasons. On average, the seasonal Kc was about
0.80 and 0.85, and Ke was about 0.12 and 0.11, respectively, for both
seasons. Kc had the minimal value of 0.17–0.24 and 0.15–0.21 at the
initial season stage, reached to the maximal values of 1.01–1.17
and 1.09–1.36, respectively, at the mid-season for both seasons.
The Kc from Allen et al. (1998) (corrected for climatic conditions)
was 0.16 in the initial stage, 1.11–1.16 and 0.36–0.61 in the middle
and late stage respectively. Kang et al. (2003) reported that a 10-
year average spring maize Kc was 0.45 ± 0.08 in June (initial stage),
1.43 ± 0.37 in August (middle stage) and 0.93 ± 0.21 in October (late
stage), with an average of 1.04 over the entire growth season in
northwest China. Li et al. (2008) indicated that the average Kc of
maize was 0.44 in May (initial stage), 1.46 in July (middle stage)
and 1.22 in September (late stage) in northwest China. Our results
are generally similar to Allen et al. (1998), but the values at the late
stage are higher than the FAO value. The reason of smaller Kc is that
maize is harvested early, substantial green leaf existed on the plants
and crop transpiration was still large. The Kc in our study is smaller
than Kang et al. (2003) and Li et al. (2008) at the whole growth stage.
This was caused by different crop varieties. Compared to the spring
maize, the grain maize has smaller crop height, LAI and ET, so the Kc
of grain maize is also small. With the increase of planting density,
the Kc tended to increase significantly but Ke increased slightly at
different growth stages, indicating that the differences in ET with
different planting densities can be attributed to the differences of
crop transpiration (T).
As shown in Fig. 2, the relationship between LAI and Kc , Kcb
were power functions that Kc increased rapidly when LAI was below
3 m2 m−2 and slowly beyond this threshold. Previous studies have
also shown similar relationship between Kc and LAI (Al-Kaisi et al.,
1989; Dunchemin et al., 2006; Kang et al., 2003). The Kc increased
in a linear or exponential limit fashion until approximately a LAI
of 3 m2 m−2 , but Kc increased slowly after LAI increased above
3 m2 m−2 (Kang et al., 2003).
3.5. Comparison of different methods in estimating
evapotranspiration
3.5.1. Adjusting crop coefficient under different plant densities
In this study, the adjusted mean Kc at different growth stages
under different planting densities was calculated using the
methods of Allen, Kcm , Kcb and Kdensity (Fig. 3). In this study, the
adjustment coefficient ˇ of Eq. (16) can be taken as 0.45, which
was fitted with the measured LAI and Kc at the intervals of 5–7
days under different planting densities in 2012 and 2013, ˇ is 0.45
(R2 = 0.71, n = 144, RMSE = 0.06). Comparing to the Kc calculated
from ET/ET0 under different planting densities, the adjusted Kc
calculated using the Allen method was underestimated by 8%
(Fig. 3a), with a coefficient of determination (R2 ) of 0.90, mean
bias error (MBE) of −0.099 mm d−1 , root mean square (RMSE)
of 0.016 mm d−1 . Adjusted Kc at the initial season stage was
significantly lower than the measured values, with the maximum
deviation of 0.35, but the method performed well after the initial
140 X. Jiang et al. / Agricultural Water Management 142 (2014) 135–143

1.5
(a)
1.2

0.9

Estimated Kc
0.6

0.3 y = 0.92x
R² = 0.90
0

-0.3
0 0.3 0.6 0.9 1.2 1.5
Observed K c

1.5
(b)
1.2

Estimated Kc
0.9
y = 1.06x
0.6 R² = 0.86

0.3

0
0 0.3 0.6 0.9 1.2 1.5
Observed K c
1.5
(c)
1.2
Estimated Kc

0.9
Fig. 2. Relationship between crop coefficient (Kc d ) calculated from ET/ET0 , basal
crop coefficient (Kcb d ) calculated from T/ET0 and leaf area index (LAI) with different 0.6
planting densities in 2012 and 2013, inserted error bars denote standard error of
the mean. y = 0.97x
0.3 R² = 0.88
season stages, which are consistent with the studies of Allen et al.
0
(1998). As clearly stated by Allen et al. (1998), the method is the
0 0.3 0.6 0.9 1.2 1.5
adjustment for the middle and late season that nearly completely
cover the soil when LAI is reduced by planting density. The Kcb Observed K c
method overestimated the measured Kc by 6%, with R2 of 0.86,
1.5
MBE of −0.076 mm d−1 and RMSE of 0.018 mm d−1 (Fig. 3b). The
adjusted Kc by the Kcm method was underestimated by 3%, with R2 , (d)
MBE and RMSE of 0.88, −0.005 mm d−1 and 0.014 mm d−1 (Fig. 3c). 1.2
Ringersma and Sikking (2001) applied the Kcb method to estimate
Estimated Kc

ET of Sahelian vegetation barriers, they found the method overesti- 0.9

mate Kcb full , even with an adjustment of resistance correction factor
(Fr ). The Kc of beans, cantaloupe, strawberry and tomato in Cali- 0.6
fornia estimated by the Kcm method was lower than the reported y = 0.98x
R² = 0.99
Kc (Allen and Pereira, 2009). However, the adjusted Kc by Kdensity 0.3
method was only underestimated by 2%, with higher R2 of 0.99,
lower MBE of −0.016 mm d−1 and RMSE of 0.001 mm d−1 (Fig. 3d).
0
0 0.3 0.6 0.9 1.2 1.5
3.5.2. Comparison of estimated ET by different methods
Observed K c
In 2012, compared to daily ET measured by the EC system (ETEC )
(Fig. 4a–d), daily ET estimated by the Allen method (ETA ) was Fig. 3. Relationship between adjusted Kc by the Allen (a), Kcm (b), Kcb (c) and Kdensity
underestimated by 2%, with MBE, RMSE and R2 of −0.06 mm d−1 , (d) methods and measured Kc calculated by ET/ET0 with different planting densities
0.84 mm d−1 and 0.80, respectively (Table 4). The Kcb method (ETD ) in 2012 and 2013.
overestimated the daily ET by 9%, with MBE of 0.48 mm d−1 , RMSE of
0.97 mm d−1 and R2 of 0.73, respectively. Daily ET estimated by the
Kcm method (ETS ) was overestimated by 8%, with R2 , MBE and RMSE
of 0.81, 0.35 mm d−1 and 0.86 mm d−1 , respectively. However, daily
ET estimated by Kdensity method (ETK ) was underestimated by only
 X. Jiang et al. / Agricultural Water Management 142 (2014) 135–143 141

Table 2
Evapotranspiration (ETd ) calculated by soil water balance and soil evaporation (Ed ) determined by microlysimeters of grain maize with different planting densities at different
growth stages in 2012 and 2013. D1 , D2 , D3 , D4 , D5 and D6 are 67,500, 82,500, 97,500, 112,500, 127,500 and 142,500 plants ha−1 , respectively. For total ET and E comparison
over the growing season, different small letters are significantly different (P < 0.05), and the same letters are not significantly different (P > 0.05).

Season Growth stage ETd (mm) Ed (mm)

D1 D2 D3 D4 D5 D6 D1 D2 D3 D4 D5 D6

2012 Initial 17.5 19.3 20.9 22.6 24.5 – 11.4 10.5 10.7 10.4 10.3 –
Development 137.7 142.0 153.9 158.8 162.0 – 13.2 12.3 11.7 11.4 10.8 –
Middle 215.5 220.4 228.1 236.9 247.8 – 33.0 30.3 28.7 28.2 26.5 –
Late 66.6 73.6 74.5 80.9 85.4 – 18.1 17.6 16.8 15.3 15.0 –
Whole 437.4e 455.3d 477.3c 499.2b 519.7a – 75.7a 70.7b 68.0c 65.2d 62.5e –

2013 Initial 17.6 18.1 18.5 20.6 21.4 24.0 12.6 12.6 12.1 12.0 11.5 11.3
Development 127.4 129.4 138.0 141.7 149.5 160.5 15.5 14.7 14.1 13.8 13.2 12.8
Middle 196.0 206.1 213.7 227.3 236.7 244.1 28.5 27.2 25.5 24.6 23.5 23.2
Late 72.2 77.3 82.6 85.5 87.9 91.7 11.7 11.1 10.5 10.1 9.5 9.0
Whole 413.2f 430.9e 452.8d 475.2c 495.5b 520.2a 68.3a 65.7b 62.3c 60.4d 57.7e 56.3e

Table 3
Crop coefficient (Kc d ) calculated from ET/ET0 and soil evaporation coefficient (Ke d ) calculated from (E/ET0 ) at different growth stages of grain maize with different planting
densities in 2012 and 2013. D1 , D2 , D3 , D4 , D5 and D6 are 67,500, 82,500, 97,500, 112,500, 127,500 and 142,500 plants ha−1 , respectively. For Kc d and Ke d comparison in entire
season, different small letters are significantly different (P < 0.05), and the same letters are not significantly different (P > 0.05).

Season Growth stage Kc d Ke d

D1 D2 D3 D4 D5 D6 D1 D2 D3 D4 D5 D6

2012 Initial 0.17 0.19 0.21 0.22 0.24 – 0.11 0.11 0.11 0.10 0.10 –
Development 0.73 0.76 0.82 0.85 0.86 – 0.07 0.07 0.06 0.06 0.06 –
Middle 1.01 1.04 1.07 1.11 1.17 – 0.16 0.16 0.14 0.13 0.12 –
Late 0.71 0.79 0.80 0.87 0.92 – 0.19 0.19 0.18 0.16 0.16 –
Whole 0.73e 0.76d 0.80c 0.83b 0.87a – 0.13a 0.13a 0.11b 0.11b 0.11b –

2013 Initial 0.15 0.16 0.16 0.18 0.19 0.21 0.11 0.11 0.11 0.10 0.10 0.10
Development 0.77 0.78 0.84 0.86 0.91 0.97 0.09 0.09 0.09 0.08 0.08 0.08
Middle 1.09 1.15 1.19 1.27 1.32 1.36 0.16 0.15 0.14 0.14 0.13 0.13
Late 0.80 0.86 0.91 0.95 0.97 1.02 0.13 0.12 0.12 0.11 0.11 0.10
Whole 0.75f 0.78e 0.82d 0.87c 0.90b 0.95a 0.12a 0.12a 0.11b 0.11b 0.11b 0.10c

1%, with MBE, RMSE and R2 of 0.05, 0.67 mm d−1 and 0.85, respec-
tively.
In 2013, compared to daily ET measured by ETEC (Fig. 4e, 4f, 4 g,
4 h), daily ETA was underestimated by 7%, with R2 , MBE and RMSE
of 0.65, 0.12 mm d−1 and 0.80 mm d−1 , respectively (Table 4). Daily
ETS was underestimated by 9%, with R2 , MBE and RMSE of 0.79,
−0.04 mm d−1 and 0.73 mm d−1 . Daily ETD was underestimated by
4%, with R2 , MBE and RMSE of 0.75, 0.14 mm d−1 and 0.78 mm d−1 .
But daily ETK was close to daily ETEC , which had higher R2 of 0.85
and lower RMSE of 0.67 mm d−1 . Thus Kdensity method had higher
accuracy in estimating daily ET over the whole growth stage.
As seen from Fig. 5, when LAI was less than 3 m2 m−2 , daily ETA
was lower than daily ETEC in 2012 (Fig. 5a), especially at the initial
stage, but daily ETA was higher than daily ETEC in 2013 (Fig. 5b).
Compared to ETEC , daily ETS and ETD overestimated the measured
ET significantly, especially the ETD , for the entire season in 2012,

Fig. 4. Comparison of daily ET estimated the Allen (ETA ), Kcm (ETS ), Kcb (ETD ), Kdensity (ETK ) methods and measured ET by eddy covariance in 2012 (a, b, c, d) and 2013 (e, f, g,
h).
142 X. Jiang et al. / Agricultural Water Management 142 (2014) 135–143

Table 4
Statistical results of daily ET estimated using the Allen method (ETA ), Kcm (ETS ), Kcb (ETD ) and Kdensity (ETK ) methods and in comparison with the measured ET by eddy
covariance (ETEC ) in 2012 and 2013, respectively. ET is crop evapotranspiration, R2 determination coefficient, n sample numbers, MBE mean bias error (mm d−1 ) and RMSE
root mean square error (mm d−1 ).

Season Regression equation n R2 MBE (mm d−1 ) RMSE (mm d−1 )

2012 ETA = 0.98 ETEC 142 0.80 −0.06 0.84

ETS = 1.08ETEC 142 0.81 0.35 0.86
ETD = 1.09 ETEC 142 0.73 0.48 0.97
ETK = 0.99 ETEC 142 0.85 0.05 0.67

2013 ETA = 0.93 ETEC 142 0.65 0.12 0.80

ETS = 0.91ETEC 142 0.79 −0.04 0.73
ETD = 0.96 ETEC 142 0.75 0.14 0.78
ETK = 0.98 ETEC 142 0.85 0.07 0.67

9
ETA
(a)
ETK
7
ETS
ETD
ET(mm d-1)

5 ETEC

-1
4-20 5-10 5-30 6-19 7-9 7-29 8-18 9-7 9-27
Date(Month-Day)

9
ETA
(b)
7 ETK
ETS
ETD
ET(mm d-1)

5
ETEC

-1
4-20 5-10 5-30 6-19 7-9 7-29 8-18 9-7 9-27
Date(Month-Day)

Fig. 5. Seasonal variations of daily ET estimated by the Allen (ETA ), Kcm (ETS ), Kcb (ETD ) Kdensity (ETK ) methods and measured ET by eddy covariance in 2012 (a) and 2013 (b).

and underestimated slightly the measured ET at the mid-season
stage in 2013. Daily ETK was close to the measured daily ETEC over
the whole growth stage in both seasons.
4. Conclusions
Higher planting density increased crop evapotranspiration (ET)
and crop coefficient (Kc ) significantly and decreased soil evapora-
tion (E) and evaporation coefficient (Ke ) slightly, and the differences
of ET, E and Kc among different planting densities were mainly
caused by the difference of leaf area index (LAI). The basal crop
coefficient (Kcb ) and Kc increased significantly until approximately
a LAI of 3 m2 m−2 , but it increased slightly when LAI was larger
than 3 m2 m−2 . The Allen method performed very well after the
development stage of grain maize. The Kcm and Kcb method can
well simulate the daily ET when the LAI was below the maximal
LAI. Comparing to the three above-mentioned methods, the Kdensity
method had higher accuracy in estimating daily ET over the whole
growth stage, with higher coefficient of determination and lower
root mean square error. Thus Kdensity method had better perfor-
mance in simulating daily ET under different planting densities of
grain maize.
Acknowledgements
We are grateful for the research grants from NSFC (51321001),
the National High Technology Research and Development Program
of China (863 Program, 2011AA100502), and The Discipline Inno-
vative Engineering Plan (111 Program, B14002).
 X. Jiang et al. / Agricultural Water Management 142 (2014) 135–143
References
Agam, N., Kustas, W.P., Anderson, M.C., Norman, J.M., Colaizzi, P.D., Howell, T.A.,
Prueger, J.H., Meyers, T.P., Wilson, T.B., 2010. Application of the Priestley–Taylor
approach in a two-source surface energy balance model. J. Hydrometeorol. 11
(1), 185–198.
Al-Kaisi, M., Brun, L.J., Enz, J.W., 1989. Transpiration and evapotranspiration from
maize as related to leaf area index. Agric. For. Meteorol. 48 (1–2), 111–116.
Allen, R.G., Pereira, L.S., 2009. Estimating crop coefficients from fraction of ground
cover and height. Irrig. Sci. 28, 17–34.
Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop Evaporation: Guidelines for
Computing Crop Water Requirements. Irrigation and Drainage Paper No. 56,
FAO, Rome, pp. 300.
Allen, R.G., Pruitt, W.O., Businger, J.A., Fritschen, L.J., Jensen, M.E., Quinn, F.H., 1996.
Evaporation and transpiration. In: Wootton, et al. (Task Com.) (Eds.), ASCE Hand-
book of Hydrology, 2nd ed., American Society of Civil Engineers, New York, pp.
125–252 (784, Chapter 4).
Baldocchi, D.D., 2003. Assessing the eddy covariance technique for evaluating carbon
dioxide exchange rates of ecosystems: past, present and future. Global Change
Biol. 9, 479–492.
Brenner, A.J., Incoll, L.D., 1997. The effect of clumping and stomatal response on evap-
oration from sparsely vegetation shrublands. Agric. For. Meteorol. 84, 187–205.
Chen, S.Y., Zhang, X.Y., Sun, H.Y., Ren, T.S., Wang, Y.M., 2010. Effects of winter wheat
row spacing on evapotranspiration, grain yield and water use efficiency. Agric.
Water Manage. 97 (8), 1126–1132.
Daisuke, Y., Chiyo, K., Keisuke, K., Kenta, I., Makito, M., Shinzo, Y., Morihiro, M.,
Hideaki, H., Taku, F., 2011. Analyzing evapotranspiration components and crop
coefficients for catch crop field with small area at different planting densities in
a greenhouse. Environ. Control Biol. 49 (4), 217–225.
Ding, R.S., Kang, S.Z., Li, F.S., Zhang, Y.Q., Tong, L., 2013. Evapotranspiration mea-
surement and estimation using modified Priestley–Taylor model in an irrigated
maize field with mulching. Agric. For. Meteorol. 168, 140–148.
Domingo, F., Villagarcía, L., Brennerb, A.J., Puigdefaá Bregas, J., 1999. Evapotranspi-
ration model for semi-arid shrub-lands tested against data from SE Spain. Agric.
For. Meteorol. 95, 67–84.
Doorenbos, J., Pruitt, W.O., 1975. Crop Water Requirements. FAO Irrigation and
Drainage Paper 24, United Nations, Rome.
Dunchemin, B., Hadria, R., Erraki, S., Boulet, G., Maisongrande, P., Chehbouni,
A., Escadafal, R.J., Ezzaharb, J., Hoedjesa, J.C.B., Kharroud, M.H., Khabbab, S.,
Mougenota, B., Oliosoe, A., Rodriguezf, J.C., Simonneauxa, V., 2006. Monitoring
wheat phenology and irrigation in Central Morocco: on the use of relationships
between evapotranspiration, crops coefficients, leaf area index and remotely-
sensed vegetation indices. Agric. Water Manage. 79, 1–27.
Eberbach, P., Pala, M., 2005. Crop row spacing and its influence on partitioning of
evapotranspiration by winter-grown wheat in Northern Syria. Plant Soil 268 (1),
195–208.
Falge, E., Baldocchi, D.D., Olson, R., Anthoni, P., Aubinet, M., Bernhofer, C., Burba,
G., Ceulemans, R., Clement, R., Dolman, H., Granier, A., Gross, P., Grunwald, T.,
Hollinger, D., Jensen, N.-O., Katul, G., Keronen, P., Kowalski, A., Ta Lai, C., Law,
B.E., Meyers, T., Moncrieff, J., Moors, E., Munger, J.W., Pilegaard, K., Rannik, U.,
Rebmann, C., Suyker, A., Tenhunen, J., Tu, K., Verma, S., Vesala, T., Wilson, K.,
Wofsy, S., 2001. Gap filling strategies for long term energy flux data sets. Agric.
For. Meteorol. 107, 71–77.
Finnigan, J.J., Clement, R., Malhi, Y., Leuning, R., Cleugh, H.A., 2003. A re-evaluation
of long-term flux measurement techniques. Part I. Averaging and coordinate
rotation. Bound-Lay. Meteorol. 107, 1–48.
Griesh, M.H., Yakout, G.M., 2001. Effect of plant population density and nitro-
gen fertilization on yield and yield components of some white and yellow
maize hybrids under drip irrigation system in sandy soil. In: Proceedings of the
International Conference on Plant Nutrition-Food Security and Sustainability of
Agro-Ecosystems, Madrid, Spain., pp. 810–811.
Jensen, M.E., Burman, R.D., Allen, R.G., 1990. Evapotranspiration and Irrigation Water
Requirements. ASCE Manuals and Reports on Engineering Practice No. 70, pp.
332.
Kang, S.Z., Gu, B.J., Du, T.S., Zhang, J.H., 2003. Crop coefficient and ratio of transpira-
tion to evapotranspiration of winter wheat and maize in a semi-humid region.
Agric. Water Manage. 59, 239–254.
Kang, S.Z., Zhang, L., Liang, Y.L., Hu, X.T., Cai, H.J., Gu, B.J., 2002. Effect of limited
irrigation on yield and water use efficiency of winter wheat in the Loess Plateau
of China. Agric. Water Manage. 55, 203–216.
143
Li, S.E., Kang, S.Z., Li, F.S., Zhang, L., 2008. Evapotranspiration and crop coefficient
of spring maize with plastic mulch using eddy covariance in northwest China.
Agric. Water Manage. 95, 1214–1222.
Liu, Z.D., Sao, Y.F., Yu, J.C., Liu, Z.G., Nan, J.Q., 2012. Effects of varieties and planting
density on plant traits and water consumption characteristics of spring maize.
Trans. Chin. Soc. Agric. Eng. 28 (11), 125–131 (in Chinese).
Mauder, M., Foken, T., 2004. Documentation and Instruction Manual of the Eddy
Covariance Software Package TK2. Work Report University of Bayreuth, Depart-
ment of Micrometeorology, vol. 26, no. 44., pp. 1614–8916.
Monteith, J.L., 1965. Evaporation and environment. Symp. Soc. Exp. Biol. 19,
205–234.
Papadopoulos, A.P., (Ph.D. thesis) 1985. Plant Spacing Effects on Photo Syntheti-
cally Active Radiation Interception and Utilization by Greenhouse Tomatoes.
University of Guelph, Guelph, ON.
Papadopoulos, A.P., Pararajasingham, S., 1997. The influence of plant spacing on light
interception and use in greenhouse tomato (Lycopersicon esculentum Mill.): a
review. Sci. Hortic. 69 (1–2), 1–29.
Paw, U.K.T., Baldocchi, D.D., Meyers, T.P., Wilson, K.B., 2000. Correction of eddy
covariance measurements incorporating both advective effects and density
fluxes. Bound-Lay. Meteorol. 97, 487–511.
Poblete-Echeverría, C., Ortega-Farias, S., 2009. Estimation of actual evapotranspira-
tion for a drip-irrigated Merlot vineyard using a three-source model. Irrig. Sci.
28, 65–78.
Priestley, C., Taylor, R.J., 1972. On the assessment of surface heat flux and evaporation
using large-scale parameters. Mon. Weather Rev. 100 (2), 81–92.
Qiu, R.J., Song, J.J., Du, T.S., Kang, S.Z., Tong, L., Chen, R.Q., Wu, L.S., 2013. Response
of evapotranspiration and yield to planting density of solar greenhouse grown
tomato in northwest China. Agric. Water Manage. 130, 44–51.
Rana, G., Katerji, N., 2000. Measurement and estimation of actual evapotranspi-
ration in the field under Mediterranean climate: a review. Eur. J. Agron. 13,
125–153.
Reicosky, D.C., Warnes, D.D., Evans, S.D., 1985. Soybean evapotranspiration, leaf
water potential and foliage temperature as affected by row spacing and
irrigation. Field Crop Res. 10, 37–48.
Ringersma, J., Sikking, A.F.S., 2001. Determining transpiration coefficients of Sahelian
vegetation barriers. Agrofor. Syst. 51, 1–9.
Rosa, R.D., Paredes, P., Rodrigues, G.C., Alves, I., Fernando, R.M., Pereira, L.S., Allen,
R.G., 2012. Implementing the dual crop coefficient approach in interactive soft-
ware. 1. Background and computational strategy. Agric. Water Manage. 103,
8–24.
Salah, E.E., Essam, A.A.E., Mohamed, S.A., Urs, S., 2008. Irrigation rate and planting
density effects on yield and water use efficiency of drip-irrigated corn. Agric.
Water Manage. 95, 836–844.
Shuttleworth, W.J., Wallace, J.S., 1985. Evaporation from sparse crops – an energy
combination theory. Q. J. Roy. Meteor. Soc. 111, 839–855.
Tanner, B.D., Greene, J.P., 1989. Measurement of Sensible Heat and Water-Vapor
Fluxes Using Eddy-Correlation Methods. Final Report Prepared for US Army
Dugway Proving Grounds, US Army, Dugway, UT, pp. 17.
Teh, C.B.S., Simmonds, L.P., Wheeler, T.R., 2001. Modelling the partitioning of
solar radiation capture and evapotranspiration intercropping systems. In:
Proceedings of the 2nd International Conference on Tropical Climatology, Mete-
orology and Hydrology TCMH-2001, Brussels, Belgium.
Utset, A., Farre, I., Martinez-Cob, A., Cavero, J., 2004. Comparing Penman–Monteith
and Priestley–Taylor approaches as reference-evapotranspiration inputs for
modeling maize water-use under Mediterranean conditions. Agric. Water Man-
age. 66 (3), 205–219.
Webb, E.K., Pearman, G.I., Leuning, R., 1980. Correction of flux measurements
for density effects due to heat and water vapor. Q. J. Roy. Meteor. Soc. 106,
85–100.
Wright, J.L., 1982. New evapotranspiration crop coefficients. J. Irrig. Drain. E: ASCE
108, 57–74.
Zhang, B.Z., Kang, S.Z., Li, F.S., Zhang, L., 2008. Comparison of three evapotrans-
piration models to Bowen ratio-energy balance method for a vineyard in
an arid desert region of northwest China. Agric. For. Meteorol. 148 (10),
1629–1640.
Zhao, N.N., Liu, Y., Cai, J.B., Paredes, P., Rosa, R.D., Pereira, L.S., 2013. Dual crop coef-
ficient modelling applied to the winter wheat–summer maize crop sequence
in North China plain: basal crop coefficients and soil evaporation component.
Agric. Water Manage. 117, 93–105.