Assessment of reference evapotranspiration using ANN at Mulde, Maharashtra

Abstract

Evapotranspiration is a complex and non-linear phenomenon because it depends on several interacting climatological factors, such as temperature, humidity, wind speed, solar radiation. In the past decade, ANN intensively used in system modelling, rainfall-runoff modelling, reservoir operation; land drainage design, aquifer parameter estimation etc. shown that ANN is more accurate than conventional methods. Artificial Neural Networks (ANN) is effective tools to model nonlinear systems which may be difficult to present by conventional mathematical equations. The ANN models are well suited under relationships between the input and output variables are not explicit. The ANN models need less data than conventional methods when properly validated and numbers of variables remain constant from one simulation to another. The ANN approach estimates evapotranspiration

Estimation of reference evapotranspiration using artificial neural network models for semi-arid region of Haryana

“…is that these models are fully nonparametric and do not require a priori concept of the relations between the input variables and the output data (Kisi et al, 2016;Fahimi et al, 2017;Yama et al, 2020). The ANNs in particular have received extensive attention from researchers in estimation of ET 0 since 2000 (Bruton et al, 2000) and in subsequent studies (Traore et al, 2010;Sibale et al, 2016;Qasem et al, 2019;Ingle and Purohit, 2020). The ANN models needs to be fine tuned in terms of various hyper-parameters to get the minimum estimation error for a particular site.…”