Análise de desempenho de campos de chuva pelo satélite TRMM na Paraíba, para fins de modelagem hidrológica distribuída

• Autor(a) principal: Santos, Aderson Stanrley Peixoto
• Data de Publicação: 2014
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UFPB
• Texto Completo: https://repositorio.ufpb.br/jspui/handle/tede/5541
• topic: Estimativa; Verificação; Índices de desempenho; Sensoriamento remoto; Tabela de contingência; Estimates; Verification; Skill Score; Remote sensing; Table of confusion; ENGENHARIAS::ENGENHARIA CIVIL

Precipitation is a weather element that serves as the key to most types of analysis input parameter. However, traditional systems of collecting budget record historical, geographical and temporal in obtaining, processing and transfer problems. This fact has mobilized the rise of techniques such as orbital remote sensing (ORS), increasingly applied in studies aimed at monitoring and forecasting through Hydrological Models such as the Distributed Hydrological Models (DHM). However, because they are considered the ground truth science requires intercomparison of results from orbital sensors with them in order to have the analytical scrutiny of verification of responses from performance. However, the methodologies adopted in the comparative analysis of their results - with terrestrial information - usually do not report the nature of the error information, if basing only on larger or smaller link the information. Therefore, this paper aims to contribute to science through the methodological increase with use of statistics related to performance indicators, to assess the particularities involved in the errors associated with each sensor element for measuring rain. Methodologically compared the information of rain sensor 3B42 algorithm s 7 version of Tropical Rainfall Measurement Mission (TRMM) with historical information from rain gauges belonging to Agência Executiva de Gestão de Águas do Estado da Paraíba (AESA), with analysis from 1998 to the year 2011, were methodologically compared followed two analytical biases. At first, there was the spatio-temporal similarity to pixel level (pixel-to-pixel) in what is called approach 1 (A1); and second, there was the similarity of the results in quantitative precipitation bands. Results in analytical approach 1 (A1), the TRMM showed good performance on measures of agreement for negative events correct. According to the second approach (A2) the performances presented to corroborate the depreciation of the estimated sensor.