Estimación de la demanda hídrica en el sector agrícola por medio de imágenes satelitales, caso de estudio cuenca del Río Prado, Tolima

Abstract

The present project shows the application of a methodology for the estimation of water demand for rice cultivation located in the AsoPrado irrigation district in Prado, Tolima. A literature review of existing methods for estimating the water demand of rice cultivation using satellite images was initially carried out, as a result of which the algorithm Simple Algorithm for Evapotranspiration Retrieving (SAFER) was selected, which makes estimates of crop evapotranspiration (Eta) from data provided by satellite imagery (Landsat, MODIS, Sentinel) and agrometeorological information. For the calculation of the Eta in the irrigation district AsoPrado were used the reflectance bands 1 and 2 of the product MOD13Q1, a compound of 16 days of the satellite mission MODIS, which has a greater availability of images for the study period from 2002 to 2012, with a total of 253 images. It was selected for the calculation of solar radiation and later reference evapotranspiration, a modification to the Hargreaves-Samani model, which involves only temperature data and extraterrestrial radiation, but considering that the difference between maximum and minimum temperature does not depend solely of extraterrestrial radiation, cloudiness and humidity, they are also subject to changes in topography. From these approximations the methodology was structured, and the results of its application were reflected in maps and time series of the different estimated variables of the energy balance for quantification of ETa through SAFER for the period of study. Reference evapotranspiration (ET0) was calculated simultaneously using the traditional Hargreaves-Samani method recommended by the Food and Agriculture Organization of the United Nations (FAO), which involves only temperature and extraterrestrial radiation data, which was the input to calculate the Eta considering the Kc reported by the FAO for rice cultivation. As a result, it was found that on a monthly basis and more specifically for dry months, the conventional methodology tends to overestimate ETa with respect to the SAFER algorithm between 30 % and 70 % per month, while in wet months, although the difference was smaller in the same way, an overestimation of 5 % to 30 % per month was presented. To conclude and taking into account that not all the precipitated water sheet reaches the plants due to loss due to foliage interception, deep percolation, evaporation and fast run-off, effective precipitation calculation was performed to more accurately estimate the water requirements of rice cultivation in the study area. To assess the scenarios where water stress may occur in the crop due to lack or excess of water; it was analyzed the capacity of the precipitation offered to satisfy the water losses that occur by evapotranspiration processes, putting on evidence the oversupply that occurs in the wet months that puts the health of the crop at risk due to the presence of diseases and the need for irrigation support in dry months so as not to impair the yield.