Monitoramento do requerimento hídrico e rendimento de soja e milho via combinação de modelo de balanço hídrico do solo e sensoriamento remoto

• Ano de defesa: 2019
• Autor(a) principal: Chechi, Leonardo
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; Brasil; Engenharia Agrícola; UFSM; Programa de Pós-Graduação em Engenharia Agrícola; Centro de Ciências Rurais
• Palavras-chave em Português: SIMDualKc; Stewart; Evapotranspiration; Water deficit; Evapotranspiração; Déficit hídrico; CNPQ::CIENCIAS AGRARIAS::ENGENHARIA AGRICOLA
• Link de acesso: http://repositorio.ufsm.br/handle/1/19701

The increase in world population has resulted in increased demand for food, requiring increased crop productivity. Water stress is the major limiting factor for grain yield of spring-summer crops in Rio Grande do Sul, frequently requiring supplemental irrigation, especially in the most critical stages of water stress. In order to increase productivity and inputs use efficiency, current methodologies need to be improved and new alternatives sought to allow more accurate assessments of crop water requirement as well as crop response to water availability. In this way, the main objective of this work was to improve the estimation of the crop evapotranspiration (ETc) for soybean and corn under subtropical humid conditions by combining SIMDualKc soil water balance model and vegetation indices derived from remote sensing. In addition, the study sought to access soil evaporation (Es) and actual crop transpiration (Tc act), and relate Tc act to grain yield by applying empirical functions for corn and soybean crops in Rio Grande do Sul. The study methodology was composed of four steps: (i) calibration and validation of the SIMDualKc model; (ii) calibration and validation of the fraction of soil covered (fc) and leaf area index (LAI) estimated with NDVI; (iii) validation of SIMDualKc with fc and LAI derived from NDVI; and (iv) application S1 and S2 phases of the SIMDualKc-Stewart model to estimate grain yield. Steps i and ii consisted of field experiments carried out in irrigated and rainfed areas, in 2018/19 crop season, in the Depressão Central region of RS, with corn (first crop) and soybean (second crop). The areas used in step iii are farm areas and located in the major producing regions of Rio Grande do Sul state, including the Depressão Central, the Planalto Médio and Missões during the 2017/18 and 2018/19 crop seasons. In each farm a central pivot irrigated area and a rainfed area were used, when available. For stage iv, all study areas were used for calibration and validation of the SIMDualKc-Stewart model. The use of the SIMDualKc model combined with the NDVI vegetation index was efficient in simulating soil water balance by partitioning ETc act into Es and Tc act. The efficiency of the simulation was proved by the application of statistical indicators, with RMSE ranging from 2.96 to 6.52% of TAW among all soybean and corn crops areas. For the SIMDualKc-Stewart model the S2 phase was more accurate for grain yield estimation, where the RMSE was of 0.32 and 0.86 Mg ha-1, for soybean and corn, respectively. Thus, the methodology presented in the study proved to be efficient for the estimation of ETc act and soil water balance, as well as the crop response to water availability for soybean and maize grown in the main producing regions of Rio Grande do Sul state, in diversified conditions of soil and climate.