Avaliação do uso de dados meteorológicos de satélite (NasaPower) para manejo de irrigação no Brasil

• Autor(a) principal: Silva, Otávio Coscrato Cardoso da
• Data de Publicação: 2021
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFG
• dARK ID: ark:/38995/001300000fb29
• Texto Completo: http://repositorio.bc.ufg.br/tede/handle/tede/11341
• topic: Simulação de cultivo; Base de dados; Projeto Power; INMET; Cultivation simulation; Data base; Power Project; INMET; CIENCIAS AGRARIAS::AGRONOMIA

The use of information derived from satellites has been gaining space in areas where previously there was no use. This is incrising due to the technological advances in sensors and components that are installed in equipment launched in orbit. In agriculture, it was no different, this segment of the productive chain so important for the Brazilian and world economy has been taking advantage of these to improve its techniques and practices of crop management in several ways, where one is crop irrigation management. One of the ways of carrying out irrigation management is through the water balance, whose main input component is the use of meteorological data to calculate reference evapotranspiration (ETo). The weather data are obtained through institutions in the most cases, which have conventional weather stations (EMC) or automatic stations (EMA) installed in rural properties. The problem is that these methods, despite being reliable, are economically costly and depends on trained personnel to collect this information in the case of EMC and can still not be close to the agricultural application area. As an alternative to this mode, the NASAPOWER platform has been gaining prominence in some studies due to the availability of daily meteorological data derived from satellite for any region of the globe. Thus, this work aimed to evaluate the behavior of meteorological data provided by NASAPOWER in relation to the data provided by INMET, collected through EMC, in the application of irrigation management in different cultures and regions in Brazil, using the Penman-Moteith equation - FAO to obtain ETo. The study considered simulations of the main crops of economic value used in different regions of the country, having input information of cultivation coefficient throughout the cycle, determination of the local available water to the crop, cultivation period, and two bases of meteorological data with a period of 31 years. The results of irrigation requirement based on the NASAPOWER platform proved to be adjusted in most simulations with used EMC, and its use in cotton culture in the Luís Eduardo Magalhães region, soybean in Cruz Alta, corn in Balsas and orange in Araraquara are recommended. Its use was not recommended for sugar cane in Uberaba and beans in Primavera do Leste. While for sweet corn in Unaí, rice in Uruguaiana, sugar cane in Rio Verde and coffee in Araguari, there is a recommendation with reservations. In general, the evaluation of irrigation for all evaluated crops obtained a correlation coefficient (r) of 0.95, determination coefficient (R2) of 0.95, square root of the average error (RMSE) of 115 mm, mean absolute error (MAE) of 85 mm and a Willmott coefficient (d) of 0.96.