Comparative Study of Reference Evapotranspiration Estimation Methods Including Artificial Neural Network For Dry Sub-Humid Agro-Ecological Region

Download as pdf or txt
Download as pdf or txt
You are on page 1of 9

July-September 2016]

COMPARATIVE STUDY OF REFERENCE ET

233

Journal of Soil and Water Conservation 15(3): 233-241, July-September 2016


ISSN: 022-457X

Comparative study of reference evapotranspiration


estimation methods including Artificial Neural Network for
dry sub-humid agro-ecological region
SASWAT KUMAR KAR1, A.K. NEMA1, ABHISHEK SINGH1, B.L. SINHA2 and C.D. MISHRA1
Received: 14 May 2016; Accepted: 6 August 2016

ABSTRACT
In the present study, an attempt has been made to compare the reference evapotranspiration (ETo),
computed by eight different methods, namely, Penman-Monteith, Modified Penman-Monteith,
Hargreaves-Samani, Irmak, Hargreaves, Valiantzas, ANN and FAO(24) model for the dry subhumid agro-ecological region (Varanasi). An attempt was also made to find out utility of artificial
neural networks (ANN) for estimation of ET0 with minimum input. Feed forward network has
been used for prediction of ETo using resilient back-propagation method and the architecture 2-21(having parameters Tmean and solar radiation) was found to be the best one. The average annual
evapotranspiration (by PenmanMonteith method) for Varanasi was found to be 1447.4 mm. When
compared among the different methods for estimation of reference evapotranspiration with PenmanMonteith method, the FAO-24 and Hargraves-Samani (3) under estimate, however, Modified
Penman-Monteith, Hargreaves-Samani, Irmak, Hargreaves, Valiantzas over-estimate and ANN
closely estimates reference evapotranspiration.

Key words: Reference evapotranspiration, ANN, Trends, Penman-Monteith method

INTRODUCTION
Reference evapotranspiration (ET o ) is an
important parameter of the hydrologic cycle and
its accurate estimation is important for many
studies, such as hydrologic water balance,
irrigation scheduling, water resources planningmanagement and for hydrologic and climatic
studies. Evapotranspiration (ET) is the combined
process of evaporation from the soil and plant
surfaces and transpiration through the stomata of
the plant surface. The evapotranspiration rate from
a reference surface having no shortage of water is
called the reference crop evapotranspiration (ETo).
The reference surface is a hypothetical grass with
specific characteristics. The concept of the reference
evapotranspiration was introduced to study the
evaporative demand independent from crop type,
crop development and management practices. The
only factors affecting ETo are climatic parameters
and can be computed from weather data. Many
equations are used to estimate ETo and can be
divided into two main groups, those that are
physically based like Penman, Penman-Monteith

model, etc. and those that are empirical based like


Hargreaves, Hargreaves-Samani, Blaney-Criddle
model, etc. but the comparison between the results
of these methods reveals a wide divergence
(Doorenbos and Pruitt, 1977). ETo expresses the
evaporating power of the atmosphere at a specific
location and time of the year and does not consider
the crop characteristics and soil factors. The FAO
Penman Monteith method is the most reliable
method for determining ET o as it explicitly
incorporates both physiological and aerodynamic
parameters. It is a method with strong likelihood
of correctly predicting ET o in a wide range of
locations and climates and has provision for
application even when some data is missing.
Artificial neural network (ANN) technique has
drawn the attention of many researchers because
of the nature of being efficient and having
nonlinear modeling of complex systems. Many
studies have been done on the reliability of ANNs
for estimating ET o . Kisi (2007) compared the
potential of different ANN techniques and
empirical methods in reference evapotranspiration

1 Institute of Agricultural Sciences, Banaras Hindu University, Varanasi-221005 (U.P.); 2College of Agricultural Engineering &
Technology & Research (IJKVV) Mungali (C.G.); Email: anupamnema@ gmail.com

234

KAR et al.

estimation. In the present study, an attempt has


been made to estimate the reference
evapotranspiration by different known methods
with the objective to reduce the input weather
parameters (due to no availability of various
weather data) without losing the qualitative result.
MATERIALS AND METHODS
Study area
Meteorological data from the observatory of
Institute of Agricultural sciences, BHU, Varanasi
(Latitude: 25.26 N, Longitude: 82.99E and Altitude:
80.71 m) were collected and analysed for the
purpose of this study. The observatory is situated
in the dry sub humid agro ecological region (BHU,
Varanasi). The meteorological data i.e. maximum
temperature (Tmax), minimum temperature (Tmin),
maximum relative humidity (RHmax), minimum
relative humidity (RHmin), sunshine hour, wind
speed, rainfall and pan evaporation of period of
29 years (1986-2014) were used for present study.
FAO has recommended the Penman-Monteith
method
for
estimation
of
reference
evapotranspiration as it is a close, simple
representation of the physical and physiological
factors governing the evapotranspiration process.
But, this method requires various meteorological
parameters that are usually not available in most
meteorological stations. Here, some other standard
methods like FAO-24 (Pan) model, Modified PM
method, Hargreaves method, Hargreaves Samani
method, Valiantzas method, Irmak method, ANN,
etc are tested for estimation of ET0. ANN was also
tested and input parameters were optimized. The
different methods of estimation of reference
evapotranspiration as mentioned above with the
required input parameters are discussed below.
FAO-56 (Penman Monteith) method
The Penman Monteith method is considered as
standard because it has ranked first for both humid
and arid regions (Allen et al., 1998). The equation
that is used for calculating the reference
evapotranspiration is given below:
(1)

where,
ETo = Reference evapotranspiration [mm day-1]
Rn = Net radiation at the crop surface [MJm-2 day-1]
G = Soil heat flux density [MJ m-2 day-1]

[Journal of Soil & Water Conservation 15(3)

T = Mean of daily air temperature [C]


u2 = Wind speed at 2 m height [m s-1]
es = Saturation vapour pressure [kPa]
ea = Actual vapour pressure [kPa]
= Slope vapour pressure curve [kPa C -1]
= Psychrometric constant [kPa C-1].
The parameters: air temperature, sunshine
duration, wind speed and relative humidity are
taken directly from the meteorological station and
are used to estimate other parameters such as the
net radiation, slope of saturation vapour pressure
curve, psychrometric constant, etc.
Modified Penman Monteith method
Due to difficulties in getting data, particularly
solar radiation, wind speed etc. Valiantzas (2006)
suggested the simplified algebraic formula for
Penmans equation, which requires commonly
available data of mean temperature, solar
radiation, mean relative humidity and wind speed.
(2)
where, T is the mean air temperature for the
examined time interval (0c), Q is the reflection
coefficient or albedo with typical value of the
albedo for a grass cover is 0.23, Ra is the
extraterrestrial radiation (MJm2d-1), RHmean is the
mean relative humidity for the examined time
interval(%) and fu=au+(bu*U), where au =1, bu=0.54.
FAO-24 (Pan) model
Evaporation from pan (USWB Class A Land
Pan) is due to combined effect of temperature,
relative humidity, wind speed, sunshine hour, etc.
It is used in case of limited data condition. The
reference evapotranspiration as given by FAO
(Pan) model is given by:
ETo = Epan * Kp

(3)

where, ETo = Reference evapotranspiration,


(mm day-1)
Epan = Pan evaporation, (mm day-1)
Kp = Pan coefficient (0.7)
Hargreaves method
Hargreaves (1975) suggested equation for
reference evapotranspiration based on solar
radiation, maximum temperature and minimum
temperature as follows:
(4)

July-September 2016]

COMPARATIVE STUDY OF REFERENCE ET

Hargreaves Samani method


The extraterrestrial radiation based equation of
Hargreaves and Samani (1985) requiring only the
data for Ra, Tmax and Tmin is as follows:
(5)
Valiantzas method
A simplified form of modified PM equation
requiring only solar radiation (Rs), Tmean-, RHmean
takes the following form (Valiantzas, 2013).
(6)
Irmak method
Irmak et al. (2003) developed following
simplified equation for estimation of ETo using
solar radiation and mean temperature.
(7)
Artificial Neural Network
Artificial Neural Network (ANN) offers a
relatively quick and flexible means of modelling.
It gives complex nonlinear relationship among the
dependent and independent variables, due to
which it has received enormous interest by
researcher and widely used for calculation of
evaporation under limited weather data. An ANN
is a computer model, composed of individual
processing elements called units or neurons and
they are highly interconnected and operate in
parallel, these elements are inspired by biological
nervous systems. In this study, the neural networks
are adjusted or trained with a series of inputs so
that a particular input leads to a specific target
output from the training data set. In the ANN
analysis MLP type of feed-forward neural network
with back propagation learning algorithm is
applied with input layer, hidden layer and output
layer. Basically, it is a gradient descent technique
to minimize the squared error between the
calculated and predicted outputs. In this study,
since the purpose of this study was the estimation
of ETo, the ANN has only one output variable. The
output of a neuron is decided by an activation
function. There are a number of activation
functions that can be used in ANNs, such as step,
sigmoid, threshold, linear, etc. The sigmoid
activation function [f(x)] commonly used, can be
formulated mathematically as:

235

The database was divided into three subsets:


65% of data are used in the training phase, 20% in
the testing phase; 15% for validation. The first set
is for determining the weights and biases of the
network, the second one is for evaluating the
weights and biases and to decide when to stop
training and the last data set is for validating the
weights and biases to verify the effectiveness of
the stopping criterion and to estimate the expected
network operation on new data sets. The idea
behind this division is to overcome the over fitting
problem. The PM estimated daily ET0 values were
employed as substitute for measured ET0 data and
used as target output. Results showed that
increasing the number of neurons in hidden layer
did not improve the ANN results as reported by
Kisi (2006), but too many neurons may lead to the
over fitting. So the number of hidden nodes in the
ANN architecture was identified by trial and error
(Moghaddamnia et al., 2009). These computed ET0
values were used to train the ANN models. Various
architectures and input combinations of the models
were compared for crop evapotranspiration. The
input variables considered in the input layer
decrease gradually according to their reduced
sensitivity i.e. Tmean>VPD >RS > U2as suggested
by Anupriya et al. (2014) by calculating a sensitivity
coefficient of ETo to each climate variable (Irmak
et al., 2006). Input parameters decrease from four
to one and their effect on predicted ETo is analyzed.
Accuracy criteria
Accuracy tests were performed to evaluate the
reference evapotranspiration estimated by different
methods by the statistical analysis, as shown in
table in next page.
RESULTS AND DISCUSSION
Food and Agriculture Organization (FAO) has
suggested that the Penman-Monteith method is the
standard method for estimation of potential
evapotranspiration. The average annual
evapotranspiration for Varanasi is found to
be 1447.4 mm and the average daily
evapotranspiration is found to be 3.96 mm/day by
the above method. The daily average reference
evapotranspiration varies from minimum 1.65
mm/day to maximum 6.77 mm/day. Anupriya et
al. (2014) has suggested that Tmean, Vapour Pressure
Deficit (VPD), solar radiation (RS) and wind speed
(U2) are the parameters which are most sensitive
in the decreasing order towards reference
evapotranspiration. Artificial Neural Network
technique was utilised with the different

236
Performance Criteria
Mean Square Error
(MSE)

KAR et al.

[Journal of Soil & Water Conservation 15(3)

Formula

Remarks
An MSE of zero, means that the Pi predicts
observation of the parameter Ei with perfect
accuracy.

Root Mean square error


(RMSE)
Coefficient of
correlation (r)

Flemming (1975)
It always lies between -1 to +1, if r =1, the
correlation coefficient is perfect and positive
and if r = -1, then it is perfect and negative.
Coefficient of determination (r2) is a statistical
measure of how well the regression line
approximates the real data points. An r2 of 1
indicates that the line of regression is of best fit.

Coefficient of
determination (r2)

where, Pi = Predicted ETo, Ei = Estimated ETo, N = Number of observation

combination of these four sensitive parameters


as input for estimation of reference
evapotranspiration. To find out the best ANN
architecture, providing near accurate estimation
of evapotranspiration with limited climatic data,
first of all the optimum number of nodes in hidden
layer for each ANN architecture were decided by
running the ANN analysis to a number of times
until a significantly least relative error is found in
the holdout. The best ANN architecture was chosen
and evaluated among various ANN architectures
with different number of neurons in hidden layer.
Table 1, represent the results obtained from ANN
of feed forward with learning algorithm of back
propagation (BP) type. For statistical comparison
of performance of these ANN architectures, the

predicted evapotranspiration by these neural


networks were compared with the estimates
provided by the FAO-56(PM) evaporation model.
The detail performance evaluation is briefed in the
Table 1.
The ANN architecture 4-3-1 estimated
evapotranspiration with maximum accuracy, i.e.
relative error (Holdout) = 0.001, MSE = 0.003 mm
day-1, RMSE = 0.053 mm day-1, r = 0.992 and r2 =
0.985. But, this architecture requires four input
parameters. While the ANN architecture 2-2-1
requires only two input parameters i.e., Tmean, solar
radiation (R s) and estimate evapotranspiration
quite accurately with less error i.e., relative error
(Holdout)= 0.005, MSE =0.017 mm day-1, RMSE =
0.130 mm day-1, r = 0.988 and r2= 0.976. Therefore,

Table 1. Comparison of various neural network predicted ETo data with FAO-56 (PM) estimated ETo
ANN architecture
Inputs

No of neurons
Input
Hidden
Output
Nodes in hidden layer
MSE
RMSE
Coefficient of
correlation (r)
Coefficient of
determination (r2)
Relative Error
(Holdout)

4-3-1
Tmean,
Solar
radiation,
VPD, U2

3-3-1
Tmean,
Solar
radiation
(Rs), VPD

2-2-1
Tmean,
Solar
radiation
(Rs)

1-9-1
Tmean

3-5-1
Solar
radiation
(Rs),
( VPD, U2)

2-6-1
Tmean,
U2

2-3-1
Solar
radiation
(Rs), VPD

4
1
1
3
0.003
0.053
0.992

3
1
1
3
0.015
0.124
0.989

2
1
1
2
0.017
0.130
0.988

1
1
1
9
0.624
0.790
0.920

3
1
1
5
0.024
0.156
0.985

2
1
1
6
0.129
0.359
0.959

2
1
1
3
0.185
0.430
0.944

0.985

0.978

0.976

0.846

0.970

0.920

0.892

0.001

0.006

0.005

0.089

0.009

0.034

0.061

where, Tmean= Daily mean temperature (C), Solar radiation (Rs) on daily basis, Vapour Pressure Deficit (VPD) = es-ea,
U2= Wind speed daily basis

July-September 2016]

COMPARATIVE STUDY OF REFERENCE ET

ANN architecture 2-2-1 was chosen for comparison


with other empirical methods of evapotranspiration
estimation. Fig. 1 is a general architecture (2-2-1)
of a feed forward neural network consisting of one
input layer, one hidden layers and one output layer
and found to be the best one among other
architectures under consideration. The comparison
of estimated ETo values by different 9 methods

237

including ANN (2-2-1) is represented in Table 2


on ten days basis for providing clear view of
reference evapotranspiration by different methods.
It is found that mean daily ET o estimated by
methods like PM, Hargreaves, Irmak, HS-5 and
ANN lies in the close proximity. However, annual
average ETo obtained from Hargreaves and Irmak
method are almost same. Graphical representation

Table 2. Comparison of the results obtained from various ETo estimation methods
Month
JAN

FEB

MAR

APR

MAY

JUN

JUL

AUG

SEPT

OCT

NOV

DEC

ETo(mm)

No. of days
I (10)
II(10)
III(11)
I(10)
II(10)
III(8)
I(10)
II(10)
III(11)
I(10)
II(10)
III(10)
I(10)
II(10)
III(11)
I(10)
II(10)
III(10)
I(10)
II(10)
III(11)
I(10)
II(10)
III(11)
I(10)
II(10)
III(10)
I(10)
II(10)
III(11)
I(10)
II(10)
III(10)
I(10)
II(10)
III(11)
Annual
avg.
Daily
average

Pan
1.04
1.26
1.43
1.62
1.90
2.22
2.56
3.03
3.86
4.59
5.13
5.59
5.42
5.64
5.48
5.80
5.35
4.23
3.73
3.21
3.19
3.08
2.68
2.93
3.02
2.73
2.61
2.49
2.25
2.02
1.92
1.84
1.54
1.37
1.24
1.16
1107.10
3.03

Average ETo (mm/day) estimated on ten days basis by different methods


P-M
Hargreaves
Irmak
HS
Modified PM
Valiantzas
ANN
1.69
1.97
2.29
2.42
2.00
2.30
1.83
1.97
2.22
2.54
2.63
2.30
2.65
1.99
2.28
2.55
2.87
2.94
2.65
3.06
2.28
2.54
2.95
3.18
3.24
2.92
3.36
2.57
2.92
3.22
3.43
3.58
3.30
3.71
2.87
3.30
3.64
3.77
3.98
3.76
4.23
3.37
3.76
4.04
4.09
4.49
4.33
4.85
3.82
4.20
4.41
4.35
5.11
4.78
5.24
4.10
4.83
4.90
4.68
5.85
5.51
6.01
4.63
5.39
5.46
5.05
6.54
6.09
6.79
5.22
5.69
5.58
5.15
6.98
6.54
7.08
5.60
6.06
5.98
5.40
7.10
6.89
7.36
6.08
6.29
6.14
5.49
6.99
7.05
7.36
6.26
6.36
6.19
5.51
7.15
7.08
7.33
6.39
6.39
6.14
5.50
7.14
7.01
7.18
6.44
6.45
6.22
5.53
7.13
7.09
7.23
6.55
5.91
5.67
5.19
6.20
6.45
6.34
5.97
4.89
4.60
4.55
5.60
5.16
4.93
4.74
4.66
4.39
4.46
5.19
4.86
4.62
4.54
4.33
4.14
4.28
4.77
4.43
4.12
4.20
4.49
4.39
4.42
4.55
4.66
4.34
4.38
4.42
4.40
4.45
4.46
4.57
4.34
4.43
4.00
3.99
4.18
4.22
4.10
3.88
4.00
4.28
4.32
4.41
4.28
4.38
4.25
4.35
4.17
4.34
4.33
4.14
4.25
4.15
4.19
3.93
4.07
4.19
3.97
4.05
4.00
3.96
3.98
4.27
4.30
4.00
4.13
4.23
4.08
3.78
4.25
4.30
4.00
4.06
4.39
4.03
3.47
4.01
4.10
4.03
3.82
4.30
3.62
3.07
3.74
3.91
3.95
3.55
4.24
3.24
2.74
3.34
3.63
3.79
3.25
3.97
2.70
2.45
3.06
3.38
3.45
2.99
3.72
2.37
2.25
2.85
3.17
3.22
2.76
3.47
2.23
2.06
2.59
2.93
3.00
2.55
3.23
2.07
1.92
2.33
2.72
2.76
2.35
2.90
1.89
1.75
2.08
2.45
2.58
2.12
2.55
1.79
1447.41
1505.51
1502.38 1677.61
1600.15
1700.01
1448.27
3.97

4.12

4.12

4.6

4.38

4.66

3.97

238

KAR et al.

[Journal of Soil & Water Conservation 15(3)

Fig. 1. Architecture (2-2-1) of feed forward neural network with one hidden layer

of average ETo by different methods on monthly


basis is shown in Fig 2. It shows maximum ETo
during month of May and lowest during month
of Dec-Jan. When compared among the
different methods for estimation of reference
evapotranspiration with standard PenmanMonteith method, it is found that Pan (FAO-24)
under estimate and Modified Penman-Monteith,
Hargreaves-Samani, Irmak, Hargreaves, Valiantzas
overestimate however ANN closely estimates
reference evapotranspiration.
Long term changes in evaporation and potential
evapotranspiration can have profound
implications for hydrological process as well as for
agricultural crop performance. Large changes in
the evapotranspiration demand would change the
water requirement and water balance in the crop
growing seasons. Trend analysis as shown in Fig.
3 depicts the decreasing trend. It has also observed
that annual reference crop evapotranspiration

varies from 1553.02 mm (year 1993) to 1383.58 mm


(year 2013). Similar trends were also observed by
Rao and Wani (2011) for semi-arid climate of
ICRISAT, Hyderabad.
Pan evaporation data is available at almost all
metrological observatories. An attempt was made
to correlate these Pan data with other estimation
methods. The results of statistical analysis (Table
3) shows that Penman- Monteith method have a
close relationship with Pan method having MSE=
0.968, RMSE= 0.984 and coefficient of correlation
(r) = 0.977 followed by ANN having MSE= 1.002,
RMSE= 1.001 and coefficient of correlation (r) =
0.971. Similar pattern was also found by Kale et al.
(2013). A single linear regression (y = mx + c) for
each estimation was also accomplished by plotting
the scatter plot for the test data with respect to data
obtained from FAO-24 estimations and is depicted
in Fig. 4 to 10.

Table 3. Statistical comparison of ETo estimated by different methods with ETo estimated by FAO-24 (Pan) model
ETo estimation
methods
PenmanMonteith
Hargreaves
Irmak
HargreavesSamani
Mod PM
Valiantzas
ANN

MSE

RMSE
0.984

Coefficient of
correlation (r)
0.977

Coefficient of
determination (r2)
0.955

Conversion Equation from


Pan model to others
Y1 = 0.9591X + 1.056

C
(y intercept)
1.056

0.968
1.397
1.649
2.534

1.182
1.284
1.592

0.960
0.940
0.979

0.922
0.884
0.959

Y2 = 0.8051X + 1.683
Y3 = 0.5877 X + 2.334
Y4 = 0.973 X + 1.645

1.683
2.334
1.645

1.899
2.852
1.002

1.378
1.689
1.001

0.985
0.952
0.971

0.970
0.907
0.942

Y5 = 1.0309 X + 1.257
Y6 = 0.9663 X + 1.727
Y7 = 0.9457 X + 1.100

1.257
1.727
1.100

where, X = Estimated ETo by Pan method and Y1, Y2, Y3, Y4, Y5, Y6 and Y7 are estimated ETo by Penman- Monteith, Hargreaves,
Irmak, Hargreaves samani, Modified PM, Valiantzas and ANN methods, respectively.

July-September 2016]

COMPARATIVE STUDY OF REFERENCE ET

Fig. 2. Monthly average reference evapotranspiration (ETo) by different 8 methods

Fig. 3. Trend in reference crop evapotranspiration (ETo) at Varanasi, UP

Fig. 4. Reference evapotranspiration by Pan Vs P-M

Fig. 5. Reference evapotranspiration by Pan Vs Hargreaves method

239

240

KAR et al.

Fig. 6. Reference evapotranspiration by PAN Vs IRMAK

Fig. 7. Reference evapotranspiration by Pan Vs Hargreaves-Samani

Fig. 8. Reference evapotranspiration by Pan Vs Modified PM

Fig.9. Reference evapotranspiration by Pan Vs Valiantzas

Fig. 10. Reference evapotranspiration by Pan Vs ANN

[Journal of Soil & Water Conservation 15(3)

July-September 2016]

COMPARATIVE STUDY OF REFERENCE ET

CONCLUSION
Estimation of the crop evapotranspiration is
very useful for crop planning in dry Sub-humid
Agro-ecological region. In case of limited weather
data, Artificial Neural Network (architecture: 2-21) with input parameters mean temperature and
solar radiation (R s ) estimate the reference
evapotranspiration very accurately. The Pan data
can be correlated with penman Monteith method
high degree of correlation.
REFERENCES
Allen, R.G., Pereira, L.S., Raes, D. and Smith, M. 1998.
Crop evapotranspiration- Guidelines for computing
crop water requirements. Journal of irrigation and
drainage engineering. Paper No. 56, FAO, Rome.
Doorenbos, J. and Pruitt, W. O. 1977. Guidelines for
computing crop water requirements. (FAO). Journal
of irrigation and drainage engineering. Paper No.
24 (rev). FAO, Rome.
Hargreaves, G. H. and Samani, Z. A. 1985. Reference
crop evapotranspiration from temperature. Applied
Engineering in Agriculture 1(2): 96-99.
Hargreaves, G. H. 1975. Moisture availability and crop
production. Trans. American Society of Agricultural
Engineering 18(5): 980-984.
Irmak, S., Payero, J. O., Martin, D. L., Irmak, A. and
Howell, T. A. 2006. Sensitivity analyses and
sensitivity coefficients of standardized daily ASCE
Penman-Monteith equation. Journal of Irrigation and
Drainage Engineering 132(6): 564-578.
Irmak, S., Allen, R. G. and Whitty, E. B. 2003. Daily grass

241

and Alfalfa- reference evapotranspiration estimates


and Alfalfa to grass evapotranspiration ratios in
Florida. Journal of Irrigation and Drainage Engineering
129(5): 360-370.
Kale, M. U., Nagdeve, M. B. and Bagade, S. J. 2013.
Estimation of evapotranspiration with ANN
technique. Journal of Indian Water Resources Society
33(1): 23-29.
Kisi, O. 2006. Daily pan evaporation modelling using a
neuro-fuzzy computing technique. J. Hydrol.
(Amsterdam). 329: 636646.
Kisi, O. 2007. Evapotranspiration modeling from
climatic data using a neural computing technique.
Hydrol. Processes 21(14): 1925-1934.
Moghaddamnia, A., Ghafari Gousheh, M., Piri, J., Amin,
S. and Han, D. 2009. Evaporation estimation using
artificial neural networks and adaptive neuro-fuzzy
inference system techniques. Adv. Water Resource
32(1): 88-97.
Priya, A., Nema, A. K. and Islam, Adlul 2014. Effect of
climate change and elevated CO 2 on reference
evapotranspiration in Varanasi, India - A case study.
Journal of Agrometeorology 16(1): 1-8.
Rao, A. and Wani, S. 2011. Evapotranspiration paradox
at a semi-arid location in India. Journal of
Agrometeorology 13(1): 3-8.
Valiantzas, J. D. 2006. Simplified versions for the
Penman evaporation equation using routine
weather data. J. Hydrol. 331(3-4): 690-702.
Valiantzas, J. D. 2013. Simple ETo forms of Penmans
equation without wind and/or humidity data. I:
Theoretical development. Journal of Irrigation and
Drainage Engineering 139(1): 1-8.

You might also like