Stochastic calibration of the SEBAL algorithm for variable rate irrigation prescription in subtropical climate

• Grant number: 17/19398-5
• Support Opportunities: Research Grants - Research Partnership for Technological Innovation - PITE
• Start date: April 01, 2019
• End date: March 31, 2021
• Field of knowledge: Agronomical Sciences - Agricultural Engineering - Soil and Water Engineering
• Agreement: IBM Brasil
• Principal Investigator: Marcos Vinícius Folegatti
• Grantee: Marcos Vinícius Folegatti
• Host Institution: Escola Superior de Agricultura Luiz de Queiroz (ESALQ). Universidade de São Paulo (USP). Piracicaba , SP, Brazil
• Company: IBM Brasil - Indústria, Máquinas e Serviços Ltda
• City: PiracicabaRio de Janeiro
• Partner institutions: EMBRAPA AGROPECUARIA OESTE/EMBRAPA
• Associated researchers: Danilton Luiz Flumignan ; Fabio Ricardo Marin ; Jéfferson de Oliveira Costa ; João Paulo Francisco ; Luciano Sobral Fraga Junior ; Rubmara Ketzer Oliveira ; Wagner Wolff

Abstract

The spatio-temporal variation of evapotranspiration (ET) is the key to improve the water management over irrigated agricultural areas. Many remote sensing methods have been developed for mapping ET, among them, the most widely used is the Surface Energy Balance Algorithm for Land (SEBAL). However, SEBAL was developed for particular satellite sensors and images resolutions, in which functions and empirical parameters within its algorithm are corresponding to these. Therefore, the aim of this study is going to be the stochastic calibration of the SEBAL algorithm for images from unmanned aircraft vehicle (UAV) under subtropical climate. The study will be carried out in the city of Piracicaba in São Paulo state. UAV's images, meteorological data (eddy covariance, Bowen ration, and weather station), and lysimetric data will match for the UAV's campaigns, being measured in real-time. Utilizing the SEBAL algorithm is going to be estimated all the components of the energy balance and consequently the ET. Thus, the calibration is going to be done using the least squares method for residuals between the ET, observed (lysimeters, eddy covariance, and Bowen ratio) and ET, estimated by the SEBAL algorithm. The stochastic optimization method Particle Swarm Optimization (PSO) is going to be used to minimize the sum of squared deviations. The new parameters empirical of the SEBAL algorithm are going to be obtained in uncertainty levels and may be used to compose the variable rate irrigation (VRI) prescription and to understand the spatial variability of water stress, helping the farmers on decision-making and yield increase. (AU)