Evaluation of evapotranspiration estimates from observed and reanalysis data sets over Indian region

In this study, we have computed the evapotranspiration (ET) from the input variables of India Meteorological Department (IMD) for different stations in Monsoon Core Region (MCR) of India and Indian Peninsular Region (IPR) and compared with the ERA Interim (ERA‐I) and CRU ET data sets. While studying the discrepancies among the data sets, rainfall (source: IMD gridded), relative humidity (source: ERA Interim gridded), air temperature (source: IMD gridded) and soil moisture (source: TRMM/LPRM/TMI‐Model) were made use to illustrate the ET variations. When compared with IMD ET, our results show the CRU ET is underestimated but maintained the close pattern over MCR and IPR during South West (SW) monsoon (June–September) and North East (NE) monsoon (October–December) period, respectively. ERA‐I ET bounded to have mixed response over MCR and are higher than the IMD ET over IPR. Daily comparison of the IMD and ERA‐I ET data sets shows a large bias during the beginning of SW monsoon (June month) compared to other months. Site wise correlations show the substantial positive correlations between IMD and CRU ET over MCR than IPR. Overall analysis shows the monsoon features were better explained by the variations in IMD ET compared to CRU and ERA‐I ET data sets. The reported disparities among the data sets play an important role in the choice of selection for different applications such as water resource assessments, crop water requirements, monitoring of droughts etc. Comparison of evapotranspiration (ET) data sets obtained from IMD inputs, ERA‐I and CRU sources show substantial discrepancies. ET of IMD could explain the monsoon features over India compared to other data sets. The reported disparities play a key role in selecting the data sets for different applications. Mean seasonal ET over (a) MCR (b) IPR during SW and NE monsoon seasons of India from 1979 to 2014 obtained from IMD, ERS‐I and CRU data sets.