Simulação do crescimento, desenvolvimento e produtividade de milho em clima presente e futuro

Abstract

Corn is one of the most important summer crops around the world and this grain plays an important role in the sustainability and food security of the world's population. Agricultural modeling is an important tool in planning agricultural activities. Of the existing maize models, the CSM-Ceres-Maize and Hybrid-Maize models are easy-to-use process-based models that can simulate maize growth, development and yield. The objectives of this dissertation were (a) to compare different methods of estimating genetic parameters in the CSM-Ceres-Maize model, (b) to compare the capacity of the CSM-Ceres-Maize and Hybrid-Maize models to simulate growth, development and productivity of maize with different genetic variability in a subtropical environment and (c) to simulate maize productivity in the Rio Grande do Sul State under future climate change scenarios using the Hybrid-Maize model. For the calibration of the models, field experiments were carried out during the 2013/14 and 2014/15 growing seasons, and for the evaluation of these models, data were collected in field experiment in the 2015/16 and 2017/18 growing seasons. Two improved maize cultivars, one of open pollination variety 'BRS Planalto' and one simple hybrid 'AS 1573PRO', and two 'Bico de Ouro' and 'Cinquentinha' were used. To simulate maize yields with different genetic variability in relation to future climatic scenarios, the scenarios RCP 2.6, RCP 4.5 and RCP 8.5 of the fifth IPCC report using the Hybrid-Maize model, were used. Simulations showed a decrease in maize yield in the northern half of the state, and up to 5.5 Mg ha-1, while in the southern half showed an increase in maize productivity in the period 2070-2098 in relation to the period 1975-2005.