

ORIGINAL ARTICLE 

Year : 2022  Volume
: 45
 Issue : 2  Page : 9498 


Estimation of evapotranspiration from measured meteorological parameters
Roopashree Shrivastava, Faby Sunny, Manish Chopra, Indumathi S Iyer, RB Oza
Radiation Safety Systems Division, Bhabha Atomic Research Centre, Mumbai, Maharashtra, India
Date of Submission  13Apr2022 
Date of Acceptance  07Oct2022 
Date of Web Publication  20Dec2022 
Correspondence Address: Roopashree Shrivastava Radiation Safety Systems Division, Bhabha Atomic Research Centre, Mumbai  400 085, Maharashtra India
Source of Support: None, Conflict of Interest: None  Check 
DOI: 10.4103/rpe.rpe_13_22
Five empirical equations, namely, FAO56 Penman–Monteith, Hargreaves–Samani, Makkink, Turc, and Priestley–Taylor are utilized in the estimation of evapotranspiration for the month of May in 2018 for Trombay site in Maharashtra. Evapotranspiration from a given surface is a function of incoming solar radiation, net radiation, ground heat flux, air temperature, relative humidity, and wind speed. Daily average measurements of these parameters are utilized in the empirical equations for the estimation of evapotranspiration. These estimated values are compared with the measured data from pan evaporimeter installed at Trombay. The measured data from the pan evaporimeter are corrected using the pan coefficient Kp which in turn is also estimated using empirical equations. The average value of the pan coefficient Kp is 0.8 for the site. The average measured value of evapotranspiration is 4.7 mm/d for May 2018, whereas the values estimated using the five empirical equations range from 3.3 mm/d to 12.7 mm/d. Among the five equations, the Turc equation was found to be in the best agreement with the measured values of evapotranspiration. Such studies are useful in the estimation of groundwater recharge, latent heat flux, and agriculture meteorology.
Keywords: Empirical equations, evapotranspiration, pan evaporimeter
How to cite this article: Shrivastava R, Sunny F, Chopra M, Iyer IS, Oza R B. Estimation of evapotranspiration from measured meteorological parameters. Radiat Prot Environ 2022;45:948 
How to cite this URL: Shrivastava R, Sunny F, Chopra M, Iyer IS, Oza R B. Estimation of evapotranspiration from measured meteorological parameters. Radiat Prot Environ [serial online] 2022 [cited 2023 Jan 28];45:948. Available from: https://www.rpe.org.in/text.asp?2022/45/2/94/364554 
Introduction   
The process by which water changes phase from liquid to gas is known as evaporation. Evaporation occurring from vegetation is given a more specific term – evapotranspiration or ET. This is defined as the loss of water from a vegetated surface through the combined process of soil evaporation and plant transpiration.^{[1]} Both evaporation and evapotranspiration are major parameters in the hydrological budget and play a key role in hydrological modeling. On a global scale, the evaporation/evapotranspiration has spatial and temporal variability. These parameters are necessary for groundwater recharge and rainfall runoff estimation. Both parameters depend on several meteorological parameters such as wind speed, atmospheric pressure, solar radiation, net radiation, and ambient temperature. Evaporation at a site can be directly estimated using pan evaporimeter and lysimeter. Measurements using the pan evaporimeter are corrected using the pan coefficient to estimate evapotranspiration. However, largescale field deployment of pan evaporimeters is cumbersome and is rarely carried out. Hence, generally, evaporation is estimated from its dependence on other more frequently observed meteorological parameters such as wind speed, air temperature, solar radiation, net radiation, vapor pressure, or a combination of these.^{[2]} Several empirical equations are suggested in literature which relate the evapotranspiration with these parameters. These empirical equations are classified as radiationbased or temperaturebased methods depending on the parameter on which evaporation is based. The performance of six equations for the Junagadh district located in Western India for the monsoon season during 1992–2012 has been analyzed.^{[2]} Similarly, in another study, a model to estimate evaporation based on air temperature was developed.^{[3]} The model was calibrated using daily air temperature and evaporation measurements for the period 1984–1991 and then validated for the period 1995–2000 for the Junagadh district in Gujarat. Similarly, the performance of Penman–Monteith and Priestley–Taylor methods using eddy correlation measurements for the estimation of evapotranspiration for September 2000–April 2002 at a nonirrigated pasture site in Florida, USA.^{[4]} Antonopoulos and Antonopoulos^{[5]} compared the performance of 13 empirical equations and one model based on an artificial neural network with the Penman–Monteith equation using data for the period 2009–2015. Their study concluded that recalibration of the constants used in some of the equations showed better results as compared to the original equations. The performance of eight empirical equations commonly used in the estimation of evapotranspiration for the Cumberland Plateau located in the humid South East United States has been intercompared.^{[6]} The performance of various evapotranspiration equations on a global basis for 1990–1995 was compared.^{[7]} The present study is carried out on an experimental basis at Trombay in Maharashtra state. The objective of the study is to evaluate the performance of the empirical equations available in the literature for the estimation of evapotranspiration and to identify the equation best suited for the site. Since this is an experimental study, data of 21 days from May 7, 2018, to May 27, 2018 are utilized. The methodology followed and results are described in the subsequent sections.
Data and methodology   
The present study utilizes empirical equations from literature in the estimation of evapotranspiration. Hourly average measurements available at Trombay are converted to daily averages for use in these equations. Some of the equations require the measurements of ground heat flux. At Trombay, ground heat flux is not routinely observed. It is computed from net radiation data following (Stensrud et al. 2007).^{[8]} In the following section, equations 1–5 are used to calculate average evapotranspiration in a day in mm/d. Similarly, equations 6–10 are used to estimate the pan coefficient Kp. Data of the Trombay meteorological station located in Maharashtra state for May 7–May 27, 2018, are used in this study. This is a complex topography site with undulating terrain due to the presence of hills and a water body. The height of the hills varies from 130 m to 300 m. In the summer months, the maximum temperature recorded is 44.5°C, whereas, in the winter months, the minimum temperature is 11.5°C. The annual rainfall at the site from June to September is 2400 mm. The annual average wind speed is 1.4 m/s and predominant wind directions are South South West and North North East. Here, hourly average data of air temperature, relative humidity, wind speed, wind direction, and solar and net radiation are recorded on a daily basis. As already mentioned, five empirical equations, namely, FAO56 Penman–Monteith, Hargreaves–Samani, Makkink, Turc, and Priestley–Taylor are utilized in the estimation of evapotranspiration for May 2018 for Trombay in Maharashtra and the estimated values are compared with those measured using pan evaporimeter. These equations are based on solar/net radiation, air temperature, or a combination of the two. Their details are as follows:
 FAO– 56 Penman Monteith
ET_{0} = reference crop evapotranspiration (mm/d)
R_{N} = net radiation (MJ/m^{2}/d)
G = soil heat flux (MJ/m^{2}/d)
e_{s} = daily mean saturation vapor pressure (kPa)
e_{a} = actual vapor pressure (kPa)
△ = slope of the vapor pressure curve (kPa°C^{ − 1})
γ = psychrometric constant (kPa°C^{ − 1})
u_{2} = mean daily wind speed at 2 m height (Wind speed at 20 m height scaled down to 2 m.^{[1]}
T_{mean} = mean daily air temperature (°C)
 HargreavesSamani
R_{a} = extraterrestrial solar radiation (MJ/m^{2}/d)
T_{min} = daily minimum air temperature (°C)
T_{max} = daily maximum air temperature (°C)
G_{SC} = solar constant (0.082 MJ/m^{2}/min)
d_{r} = eccentricity correction factor
 Makkink
R_{s} = solar radiation (MJ/m^{2}/d)
 Turc
λ = latent heat of evaporation (MJ/kg)
 Priestley Taylor
Estimation of evapotranspiration from pan evaporimeter measurements (E_{p}) requires the estimation of pan coefficient K_{p} (defined as ). Even though generic values of K_{p} from literature can be utilized, it is a good practice to estimate K_{p} from routinely observed meteorological parameters. The following five approaches were considered:^{[9]}
 Snyder
F is the windward side distance of green crop or a dry fallow in m (5 m in the present study).
RH = relative humidity at 2 m
 Cuenca
 Orang
 Perriera
 Allen and Pruit
Results and Discussion   
[Table 1] indicates the daily variation of meteorological parameters at Trombay during the period of study. The minimum temperature recorded was 24.5°C, whereas the maximum was 36.0°C. The difference in daily maximum and minimum temperature varied from 6.5°C to 11.5°C during the study period. On an average basis, over the period of study, the air temperature was 30.4°C ± 0.6°C, relative humidity was 77.7% ± 3.3%, the wind speed was 1.6 m/s ± 0.4 m/s, solar radiation was 21.3 MJ/m^{2}/d ± 2.3 MJ/m^{2}/d, and net radiation was 10.0 MJ/m^{2}/d ± 1.3 MJ/m^{2}/d. [Figure 1] shows the time series of evapotranspiration (mm/d) from pan evaporimeter and the five empirical equations, namely, FAO56 Penman–Monteith, Hargreaves–Samani, Makkink, Turc, and Priestley–Taylor. From the figure, it is clear that the observed evapotranspiration values during the period of study ranged from 1.6 mm/d to 6.4 mm/d, whereas those estimated using the empirical equations range from 2.1 mm/d to 14.6 mm/d. The average measured value of evapotranspiration is 4.7 mm/d for May 2018, whereas the values estimated using the five empirical equations range from 3.3 mm/d to 12.7 mm/d. All the empirical equations used in the study except Hargreaves–Samani have indicated a reasonable agreement with the observed evapotranspiration values. The use of Hargreaves–Samani empirical equation has resulted in an overprediction of the evapotranspiration by a factor of 3–4. This is primarily because the Hargreaves–Samani equation is based on extraterrestrial solar radiation which is actually the solar flux at the top of the atmosphere on a given day and is different from the actual solar radiation. In the equation suggested by Hargreaves–Samani, usage of solar radiation in place of extraterrestrial radiation leads to a better agreement with the observed data which is evident from [Figure 1]. This study was carried out on a preliminary basis to test the suitability of five empirical equations for the estimation of evapotranspiration for May 2018 at Trombay in Maharashtra. Results indicate that the average value of the pan coefficient Kp is 0.8 for the site. Among the five equations, the Turc equation was found to be in the best agreement with the measured values of evapotranspiration. Such studies are useful in the estimation of groundwater recharge, latent heat flux, and agriculture meteorology.  Table 1: Daily meteorological parameters during May 2018 at Trombay site
Click here to view 
 Figure 1: Bar plot of evapotranspiration (mm/d) from pan evaporimeter measurements and empirical equations
Click here to view 
Five empirical equations were utilized for the estimation of daily evapotranspiration at a tropical site location in Mumbai, India, for May 2018. Results were compared with measurements of evapotranspiration being carried out at the site. The average measured value of evapotranspiration was found to be 4.7 mm/d, whereas the values estimated using the five empirical equations ranged from 3.3 mm/d to 12.7 mm/d. Among the five equations, the Turc equation was found to be in the best agreement with the measured values of evapotranspiration. Such studies are useful in the estimation of groundwater recharge, latent heat flux, and agriculture meteorology.
Acknowledgments
The authors express their gratitude to Dr. D. K. Aswal, Director, Health, Safety and Environment Group, BARC and Shri Probal Chaudhury, Head of the Radiation Safety Systems Division, BARC for constant encouragement and fruitful discussions during the period of study. The authors also thank Shri B. B. Ghorpade of Radiation Safety Systems Division for assistance in the measurements of meteorological variables.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References   
1.  
2.  Gundalia MJ, Dholakia MB. Estimation of pan evaporation using mean air temperature and radiation for monsoon season in Junagadh region. Int J Eng Res Appl 2013a; 3:6470. 
3.  Gundalia MJ, Dholakia MB. Modeling pan evaporation using mean air temperature and mean pan evaporation relationship in middle south Saurashtra region. Int J Water Resour Environ Eng 2013b; 5:6229. 
4.  Summer DM, Jacobs JM. Utility of PenmanMonteith, PriestleyTaylor, reference evapotranspiration, and pan evaporation methods to estimate pasture evapotranspiration. J Hydrol 2005;308:81104. 
5.  Antonopoulos VZ, Antonopoulos AV. Evaluation of different methods to estimate reference evapotranspiration in a Mediterranenan area. Water Utility J 2018;18:6177. 
6.  Yoder RE, Odhiambo LO, Wright WC. Evaluation of methods for estimating daily reference crop evapotranspiration at a site in the humid southeast United States. Appl Eng Agric 2005;21:197202. 
7.  Weib M, Menzel L. A global comparison of four potential evapotranspiration equations and their relevance to stream flow modeling in semiarid environments. Adv Geosci 2008;18:1523. 
8.  Stensrud DJ. Parameterization Schemes Keys to Understanding Numerical Weather Prediction Models. Cambridge: Cambridge University Press; 2007. p. 460. 
9.  Pradhan S, Sehgal VK, Das DK, Bandopadhyay KK, Singh R. Evaluation of pan coefficient methods for estimating FAO56 reference crop evapotranspiration in a semi – Arid environment. J Agrometeorol 2013;15:903. 
[Figure 1]
[Table 1]
