Engr Optimizing ETo Model Selection: Spatial Insights for Climate-Smart Decision Making.

Abstract

This study investigates the performance evaluation of 17 empirical evapotranspiration (ETo) models against the FAO-PM reference using the ERA5 dataset with a 0.1⁰x 0.1⁰ grid resolution from 1959 to 2022 using two multi-criteria decision analysis (MCDA) methods, VIKOR and MMOORA. The study focuses on spatial performance metrics to assess the suitability of these models across diverse climatic regions in Nigeria. Results reveal substantial spatial variations in model performance. Temperature-based models by Linacre, Ivanor, and Papadakis show stronger performance in the northern region with a KGE value greater than or equal to 0.5 but perform worse across the south. Conversely, radiation-based models exhibit worse performance across all the grid cells, showing the need for adjustments to accurately simulate local evapotranspiration processes. Mass transfer‑based models Penman, Mahringer, Trabert, and WMO consistently display promising performance with a KGE value of greater than 0.5, with select models surpassing optimal KGE values across all the grid cells. Through the application of VIKOR and MMOORA, this study emphasizes the significance of spatial performance metrics in accurately selecting the suitability of ETo models for diverse climatic regions, advocating for the VIKOR method's superior efficacy in identifying appropriate models across varied geographical contexts.