Desarrollo de un algoritmo de programación SIG para estimar la evapotranspiración en los cultivos mediante sensores remotos

Abstract

A study was conducted to evaluate the feasibility of GIS programming as a potential tool in the automation of image processing obtained by the LANDSAT 8 satellite to estimate evapotranspiration (ET) in crops. The objective was to develop an algorithm in python language implementing the SEBAL model. 35 complex equations were developed, which were automated for the processing of 13 scenes, thus covering the entire vegetative cycle of corn cultivation on the "Rancho Las Mercedes" property. For the validation of the results, the data measured by the INIFAP weather station network was consulted on the dates corresponding to the images. The results obtained show an RMSE ranging from 0.001 for the month of June to 0.013 for the month of April. The instantaneous ET values are between the ranges of 0.01 mm / hr and 0.80 mm / hr. Maps of their spatial distribution were generated, where it can be seen that the lowest ET values, those close to zero, correspond to human settlements and bare soils, while the highest values correspond to sites of vigorous vegetation and bodies of water.