An Approach for Estimation of Evapotranspiration by Standardizing Parsimonious Method

Abstract

Evapotranspiration (ET) is one of the important components of the hydrological cycle which is essential for sustainable water resource management and ecohydrological studies. Accurate estimation of ET is a crucial task in datascarce regions due to limited meteorological variables. There exist a number of indirect methods among which the standard method for computing ET is FAO-56-Penman-Monteith (PM) method. However, due to paucity of flux data such as the components of net radiation, relative humidity, vapour pressure, and wind speed in many parts of the world, the use of standard benchmark method is limited. This limitation provides the widespread acceptance of the method which uses fewer variables and can give an accurate estimation of ET for water resource management. In this study, we have developed a framework to standardize the Hargreaves-based ET in the Kangsabati River basin. We utilize the weather datasets from six stations, namely Purulia, Bankura, Mohanpur, Jhargram, Kharagpur, and Midnapore to apply the ET standardization method. We have compared both the raw and corrected ET from Hargreaves with FAO-56-PM ET prior and after correction by using harmonization method. Performance evaluation of harmonization technique is done using statistical and graphical indicators for the duration of 2006-2010. It is observed that Purulia (r = 0.83 and d = 0.80) and Mohanpur (r = 0.85 and d = 0.87) stations are almost standardized appropriately on daily scale. Further, the highest r and R 2 was obtained for Mohanpur station (r = 0.972; d = 0.940), while least for Jhargram station (r = 0.961; d = 0.741) at monthly scale. Overall, this approach can be used to provide the utility in data-scarce conditions irrespective of agro-climatic conditions.

Prediction of Daily Streamflow Using Various Kernel Function Based Regression: A Case Study in India

“…In a recent study by Khosravi et al (2019) have compared the different methods based on the kernel and standalone tree based algorithms and found that the prediction power of kernel based functions are higher than non-kernel based functions. The findings of the study are corroborated with our results and support firmly the results with the error measures and graphical indicators obtained through model simulations (Kumari and Srivastava, 2020;Elbeltagi et al, 2021). Sattari et al, 2021 showed that poly kernel function provides the better results than the RBF kernel also they found that different scenarios yielded the same results of poly kernel functions.…”

A Review of Evapotranspiration Measurement Models, Techniques and Methods for Open and Closed Agricultural Field Applications

“…These constants limit the model to specific sites and can engender overestimations of ET rates which in turn can lead to excess irrigation. Hence, several studies investigated the validity of the Hargreaves model under various locations and suggested calibration parameters, which helped in reducing the overestimation of ET values [53]. For example, a calibration conducted by [54] decreased the overestimation of ET by 16.3%.…”

Spatio-temporal variability of vegetation and its relation to different hydroclimatic factors in Bangladesh

“…In recent decades, our understanding of vegetation modeling, water resource management, and hydrological and environmental evaluation has improved thanks to remote sensing-based vegetation indices [ 23 ]. Over the years, other datasets such as MeteoSat, MODIS and Landsat series were used for similar purposes at a variety of temporal and spatial scales [ 24 ].…”