Estudo hidrometeorológico da bacia amazônica

Abstract

The change in patterns of hydrometeorological variables has been a major concern in the scientific community and society as a whole. All concerns they are in the sense of understanding how the impact on the environment is. Thus, the objective of this project was to study hydrometeorological aspects (precipitation and evapotranspiration) in different regions of the Amazon with different levels of deforestation. For this, rainfall data from the Climate Prediction Center (CPC) with monthly frequency and spatial coverage of 0.5º latitude and longitude (300 x 120 grid points) that comprise the climatological period from 1981 to 2010 for the North, East and Southern regions of the Amazon. Also dry severity index data were used (Standardised Precipitation Evapotranspiration Index – SPEI) on monthly scales (SPEI-1) and annual (SPEI-12) with the purpose of analyzing the variability and identifying the severity of the drought, for the same climatological period of precipitation for each region of the study. The methodology consisted in using stochastic model results of the Auto Regressive Integrated to Moving Averages (ARIMA), from which it was possible to analyze the dependent variable (precipitation) as a function of time in each region. In addition, a climatological simulation (1981-2010) was carried out for Tropical South America from RegCM4 using the Grell parameterization scheme. From the precipitation and temperature of the model, climatological series of SPEI-1 and SPEI-12 were generated for the regions, being possible the comparative analysis between observation and model. Additionally, the Southern Oscillation Index (SOI) was analyzed in order to establish a relationship between the large oceanic forcings (notably the El Niño Southern Oscillation) and the hydrometeorological response on the studied regions. For the analysis of the temporal scale of the dominant variability for these variables, the Wavelet Transform (TW) methodology was applied. The results indicated a greater variability in the rainfall in the eastern region where there is a marked increase in deforestation, which is part of the deforestation arc and in the South, where deforestation is already well established. While for the North region, the pattern remained without great variability. The ARIMA model was adequate, with good adjustment for monthly precipitation for the study regions. With regard to the RegCM4, this one ,it underestimated precipitation for the entire climatological series of all regions, but was able to adequately represent the variability of climatology for both precipitation and SPEI, mainly on the annual temporal scale (SPEI-12). As well as, it was better adjusted to identify conditions of moderate drought severity, mainly for the North and East regions.