Risco de ocorrência de excesso hídrico para a cultura do girassol na região central do Rio Grande do Sul

• Autor(a) principal: Lucas, Dioneia Daiane Pitol
• Data de Publicação: 2014
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Manancial - Repositório Digital da UFSM
• Texto Completo: http://repositorio.ufsm.br/handle/1/3225
• topic: Helianthus annuus L.; Datas de semeadura; Risco agroclimático; Análise numérica; Série histórica; Ajuste de funções; Sowing dates; Climate risk; Numerical analysis; Historical series; Adjustment functions; CNPQ::CIENCIAS AGRARIAS::AGRONOMIA

The approach to knowledge about the probabilities of water excess occurrence enables the planning sunflower crop management, mainly about the sowing period. The objective of this study was quantify and analyze the occurrence of water excess through numerical study for sunflower crop sown on different dates, considering the storage capacity of water in different soils of the Central Region of Rio Grande do Sul and determine the probability distribution functions that better represent the water excess occurrence. Mathematical modeling was used to simulate the sunflower development, considering a medium maturity genotype. The simulation was done for 14 sowing dates in each year of the time series of 43 years of diary weather data from 1968 to 2011 of Main Climatological Station of Santa Maria (ECPSM). The number of days with water excess in different development stages of culture was determined for main soils of the region covered by the ECPSM. Capacity values of available water storage (CAD) were considered according soil and culture characteristics, being the water excess calculated using the Dialy Water Balance, represented by accumulated water in the soil exceeding the CAD. The distribution probability functions tested for the number of days with water excess were Normal, Exponential, Gamma, Lognormal and Weibull, and the Kolmogorov-Smirnov and Chi-square tests were used to check the fit of functions. The weather conditions were different for the different sowing dates simulated in 43 years, so that differentiate the simulations of the development cycle of the sunflower. Thus, difference was observed in the number of days with water excess between sowing dates for all sub-periods analyzed. In terms of functions, there was at least one function that is fitted in most of the cases, and when there was no possible fit, the empirical frequency was used to analyze the risk. The Weibull had the greatest number of adjustments for developmental cycle as well as for subperiods. Early August until mid-September sowing dates are those with the greatest number of days with water excess, considering the whole development cycle of sunflower, regardless of the soil CAD.