Daily reference evapotranspiration models in the rice zone of Valle del Cauca, Colombia

Abstract

The opportune calculations of the daily reference evapotranspiration (ETo) quantify and schedule the use of the water availability and consequently determine a daily consumption, mostly in the conditions where the lack of climatic information interrupts the performance of diverse models to estimate the daily reference evapotranspiration in a certain region. Therefore, the objective of this work is to evaluate the behavior of the equations of Hargreaves-Samani, Priestley and Taylor, García and López, and Linacre in the calculation of daily reference evapotranspiration (ETo) for the equation of FAO Penman-Monteith (PM) using daily information from five stations near the rice production area of Valle del Cauca, Colombia. The daily evapotranspiration of 28 consecutive years was estimated, the results were confirmed by the statistical indicators: Determination Coefficient (R²); the square root of the mean square of the error (RMSE); Nash-Sutcliffe Efficiency (NSE), and box diagrams. The Priestley-Taylor (PT) and Linacre (LN) models have shown the closest results to the standard model. Bengala and Ciat Quilichao stations are shown as the areas with the best behavior of all the models with the coefficient of determination index (R²); all the stations presented acceptable results with the square root of the mean square of the error (RMSE) for the different comparisons made but not with the Nash-Sutcliffe Efficiency (Nash-Sutcliffe Efficiency). However, there is limited evidence of the daily use of the Garcia and Lopez model, being this method and that of the Hargreaves-Samani the least ideal for the Independence station.