Estimativas de curvas IDF e curvas de permanência na Amazônia sob a influência de mudanças climáticas

Abstract

The impacts on global water resources may be more intense due to climate change, making access to water more difficult and, consequently, maintaining life. In the Amazon, the effect may be even worse, as it is one of the regions most vulnerable to these changes. Representative Concentration Pathways (RCPs) scenarios are essential tools for General Circulation Models (GCMs) and Global Hydrological Models (GHMs) to simulate future climate change. Intensity, Duration and Frequency (IDF) curves and flow duration curves are fundamental for the elaboration of hydraulic projects and risk management. Thus, the objective of this study was to elaborate projections of IDF curves for the Tapajós watershed in RCP 4.5 and 8.5, using data from GCMs HadGEM2-ES, CanESM2 and MIROC5. Another objective was to analyze variations in the permanence curves and available volumes of the Amazon River using data from the GHM WaterGAP2 forced by MIROC5 and HadGEM2-ES (in RCPs 6.0 and 8.5). The projected IDF curves were compared with the existing IDF, elaborated using a stationary method. The base permanence curves were created from the last 20 years of observed flows and compared with the curves of future scenarios (from 2020). They were calculated from decadal volumes. The biggest differences for the projected IDF curves were in MIROC5 (143.15% in RCP 8.5) and the smallest differences were in HadGEM2-ES (4% in RCP 4.5) both for the 100-year return period. The spatial resolutions of each GCM influenced their IDF curves, since CanESM2 did not present satisfactory results and MIROC5 was the one that best represented the possible future differences. WaterGAP2 presented the classification “Very Good” for most stations according to statistical validation indicators. Most of the extreme flows were for 2080 to 2099. For WaterGAP2 (MIROC5), most volumes were below the century's decadal average, increasing from 2060. For WaterGAP2 (HadGEM2-ES) projections the volumes are usually close or below the decadal average, falling from 2060 onwards. MIROC5 is the most suitable for studies of climate projections in the Amazon.