Análise da Precipitação na Bacia do Rio Tietê com Dados de Postos Pluviométricos e Sensoriamento Remoto

Abstract

Considering the importance of water for human existence, the Water Resources Plans wereinstituted as an instrument to manage water resources, provided for in Federal Law No.9,433/1997. When the Plans are developed at the State scale, they are called PERH (StateWater Resources Plan). To the management to be carried out, it is necessary that the datasource for water modeling does not have great uncertainties. In Brazil there is a lack of datawith the ideal quality, either due to poor distribution and lack of maintenance of the posts, oreven due to the lack of data in certain determined periods, thus, it becomes necessary to studyand evaluate other forms of data collection, that is when the remote sensing arises. In thepresent work, an analysis of precipitation will be carried out from the year 1981 to 2020 in theTietê river basin, the most populous in the state of São Paulo, identifying dry and wet years, inaddition to its spatial distribution. The analysis was made comparing two data sources, thestations operated by DAAE-SP (Department of Water and Electricity of the State of SãoPaulo) and the product CHIRPS (Climate Hazards Group InfraRed Precipitation with StationData) which provides daily rainfall data on a global scale from remote sensing, obtainedthrough the Google Earth Engine platform. For the analysis, the annual average ofprecipitation and the Rainfall Anomaly Index were observed considering the two sources ofdata at the scale of the watershed, then an analysis of these data was carried out with apunctual investigation of 3 stations. Finally, the years in which the ENSO events (EL NiñoSouth Oscillation) occurred were observed to assess the impact on the annual average in thestudied watershed. The results indicated that the sources presented very close results,confirming the same dry and wet years, changing only their intensities, from the IAC scatterplot an R² greater than 0,8 was found, which confirms the idea of a good data representation.Furthermore, when comparing ENSO events with the rainfall history, it was not possible tofind a sequence of events coinciding with the behavior of precipitation in the watershed