Avaliação das estimativas de chuva do satélite TRMM no estado da Paraíba

• Ano de defesa: 2014
• Autor(a) principal: Soares, Alexleide Santana Diniz
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraí­ba; BR; Engenharia Cívil e Ambiental; Programa de Pós-Graduação em Engenharia Urbana e Ambiental; UFPB
• Palavras-chave em Português: Sensoriamento remoto; Estimativas de chuva -Paraíba; Remote sensing; Tropical Rainfall Measuring Mission - TRMM; ENGENHARIAS::ENGENHARIA CIVIL
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/tede/5530

The spatial and temporal variability is a precipitation feature and constitutes a factor of complexity for developing rainfall studies. Moreover, the low density of rain gauge stations and errors in data collection in the field increase the difficulties in implementing studies in this research area. However, such researches are essential considering that it is from them that we can carry out flood and drought forecasts, understand the hydrological regime of rivers, soil moisture, temperature changes, among others. Thus, the spatial rainfall estimates obtained through satellites data are important because, although present uncertainties, when compared with punctual data measured in the field can provide good indicators of the spatial distribution of rainfall for a given area. In this research, we evaluate the potential of rainfall estimates from TRMM (Tropical Rainfall Measuring Mission) sensor to represent the spatio-temporal variability of precipitation in the State of Paraíba, in the Northeast of Brazil. In this study we considered daily time series of 14 years length of rainfall data collected by AESA (Agência Executiva de Gestão das Águas do Estado da Paraíba) in 269 rainfall gauges and rainfall data estimated from TRMM satellite for a spatial mesh of 198 grid points covering the Paraíba State and which have been interpolated to the rain gauge locations using the inverse squared distance method. Comparisons were made considering the accumulated rainfall in different periods of time: daily, three days, seven days and monthly. With respect to spatial factors, the comparisons were developed based on punctual values in rain gauges stations, areal averages over sub-basins and mesoregions, and topographic profile. The statistical analyzes of comparison between the observed and estimated rainfall were developed based on the average rainfall, the linear correlations, the mean absolute error and root mean square error considering each accumulated period. Regarding the daily precipitation, the majority of the rain gauges (91%) showed correlation coefficients ranging from 0.5 to 0.7. This correlation increases for considering 3 days-rainfall, with values ranging from 0.5 to 0.7 in 56% of rain gauges, and of 0.7-0.8 for 42% of rain gauges. For the 7 days-rainfall, 58% of the rain gauges presented correlations ranging from 0.7 to 0.8, while for the monthly rainfall 95% of the rain gauges obtained correlations higher than 0.8. Therefore, the results indicate that the TRMM satellite provides better estimates when data are accumulated in larger time intervals. The monthly analysis showed that March and April are the months with higher correlation between observed and estimated precipitation, and that in the first months of the year the estimated and observed values have better approximations for all types of analyzes. It was also verified a good estimation potential in the analysis of seasonal variability of precipitation. Moreover, it was observed that the satellite presents the largest errors in the areas with the largest amount of rainfall. In the sub-basins and in the mesoregions of the state the rainfall regime was estimated quite closely. We concluded that the TRMM satellite presents very good skill in reproducing the observed rainfall measured in the gauge stations over the Paraíba state, becoming an important data source for helping the water resources planning and decision making