Navegando por Assunto "Daily precipitations"

Agora exibindo 1 - 1 de 1

Acesso aberto (Open Access)
Modelagem estocástica de função cumulativa de probabilidades de precipitação diária na região hidrográfica tocantins-araguaia (RHTA)
(Universidade Federal do Pará, 2019-03-28) PROGÊNIO, Mayke Feitosa; BLANCO, Claudio José Cavalcante; http://lattes.cnpq.br/8319326553139808
Knowing the temporal and spatial behavior of the probability of occurrence of rainfall is indispensable for the planning and management of agricultural and agroindustrial activities. However, in some river basins the available historical precipitation series are generally short and with a large number of faults, which makes statistical analyzes difficult. Thus, the objective of the work was to develop a stochastic model of cumulative function of daily precipitation probabilities in the Tocantins Araguaia hydrographic region (TAHR). The model is of the parametric type, in which precipitation occurrences were determined through the first-order Markov chain (MC) and the precipitation quantities were estimated by 4 cumulative probability functions (CPFs): exponential simple, exponential a two parameters, mixed exponential and gamma. The parameters of the CPFs were estimated by the Maximum Likelihood Method. The simulation process was performed separately for each rainfall station, without considering the spatial correlation between them. The developed model was applied in 196 rainfall stations distributed in 3 homogeneous regions (HR) of precipitation in TAHR. The results showed that the MC of the 1st order was able to reproduce satisfactorily the amount of dry and rainy days. However, in areas heavily influenced by long series of drought, the results were not satisfactory. In relation to the estimated precipitated quantities, the Kolmogorov-Smirnov (KS) test and the probability-probability (P-P) graph showed that the mixed exponential was the one that presented better adherence to the observed data for most months of the year, with the exception of the less rainy months of June, July and August in RH II and RH III, and in the months of September, October and November for RH I, for which the gamma function was more efficient, these results were also confirmed by the low Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) values. Thus, the model developed showed to be efficient in the estimation of average daily rainfall in TAHR, in addition, the use of more than one CPF gave the model greater capacity to estimate rainfall in different locations and seasons.