Análise da missão GPM (Global Precipitation Measurement) na estimativa da precipitação sobre território brasileiro

• Ano de defesa: 2018
• Autor(a) principal: Gadêlha, André Nóbrega
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraíba; Brasil; Engenharia Cívil e Ambiental; Programa de Pós-Graduação em Engenharia Civil e Ambiental; UFPB
• Palavras-chave em Português: GPM; IMERG; Precipitação; Brasil; Precipitation; Brazil; CNPQ::ENGENHARIAS::ENGENHARIA CIVIL
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/123456789/13132

Precipitation is a very important variable for the Earth's water, climate and energy cycles, playing a prominent role in maintaining human and animal life. Accurate measurements of this variable are of great need for studies that seek to better understand these cycles. In some regions or seasons, these measurements are difficult to obtain, hampering various research in this field. The use of sensors installed on satellites to estimate rainy events has shown to be useful in such circumstances over the last four decades and their results have become better and better. Global Precipitation Measurement (GPM) is a space mission for this purpose, launched in 2014 by NASA and JAXA space agencies. With a spatial resolution of 0.1º and a time resolution of 30 minutes, the GPM has to show good results when estimating the spread around the globe through IMERG, a mission product that obtains data from several merged satellites. This study aims to compare IMERG results across Brazil, comparing with precipitation data from interpolated rainfall stations. This comparison is shared by evaluating specific statistical parameters to determine the assertiveness, similarity and precision of the space mission, with presentation of spatial distribution in maps, tables, graphs and dispersion and boxplots. For all applied metrics, in a pixel-pixel comparison, they exhibit concentrated and symmetric values. Some error in the east region of the northeast region are revealed, with underestimates and detection failures. There is a substantial improvement of all metrics when the time scale of evaluation is reduced to month or year. On the average of the pixels, the IMERG overestimated the annual precipitation for 2016, even without catching rainfall greater than 3,600 mm / year. The estimates are improved with the increasing the evaluated area. IMERG still has a better performance in the Midwestern areas to assess similarity, and a Northeast region for precision assessment. A Northern region presented a worse prediction. In the annual scale, comparing the regions of the country, IMERG was better in the Southeast and Midwest and worse in the North and Northeast.