Evaluation of reference evapotranspiration methods for the northeastern region of India

• Radiation-based IRMAK3 and TURC models performed best among the selected models.
• FAO24 Blaney-Criddle is the only suitable temperature based method for the region.
• Performance of particular model is more subjected to it structure rather than type.
• In particular, FAO24BC and TURC performed better as compared to 1957MAKK model.
• Radiation and temperature extreme is the dominant factors to drive the ET0 process.

The study planed to identify a suitable alternative to the FAO-56 Penman-Monteith (FAO56PM) equation for calculating reference evapotranspiration (ET0) from chosen temperature and radiation based models utilizing monthly meteorological data from 30 destinations in diverse agro-ecological regions of the Northeast (NE) India i.e., Assam Bengal Plain (ABP), eastern Himalaya (EH), and the northeastern hilly (NEH) region. Radiation-based IRMAK3 most appropriate in the ABP (weighted root mean square deviation, WRMSD=0.17 mm d−1, r2=0.98, for Nagrakata), and TURC model being in the first three rank of most of the sites, with the lowest error and highest correlation in NEH (WRMSD=0.10 mm d−1, r2=0.92, for Shillong), and EH (WRMSD=0.23 mm d−1, r2=0.95, for Gangtok). Findings reveal that IRMAK3 and TURC models performed equally well and were observed to be the best among selected models for the majority of stations followed by FAO24 Blaney-Criddle (FAO24BC), and 1957MAKK. Pair-wise regression equations were developed for preferred FAO56PM ET0 estimates to ET0 estimates by alternative methods. Cross-correlation of eighteen chose methods demonstrated that the five equations (i.e. four radiation- and one temperature-based) performed exceptionally well when contrasted with the FAO56PM model, thus being advised for assessing ET0 under limiting data conditions as have yielded a better estimate of ET0 with a small error.