Influência de diferentes fontes e escalas de informação do relevo sobre a estimativa de cheias a partir do Hidrograma Unitário Instantâneo de Nash

Abstract

Water is a natural resource indispensible to life and essential to regional economic development. Due to climate change, anthropic interferences, and rapid population growth, the occurrence of flood-related natural hazards in watersheds has increased. Watershed flood-related studies allow the estimation of peak streamflow and direct surface runoff hydrograph resulting from single or multiple rainfall events. However, such estimations are directly dependent on existing streamflow historical series, which might be troublesome in developing countries due to the lack of streamflow gauging stations. In this context, indirect flood estimation methods stand out. Among the different flood estimation methods presented in the literature, the Unit Hydrograph (UH), Instantaneous Unit Hydrograph (IUH), and Geomorphological Instantaneous Unit Hydrograph (GIUH) have caught researchers’ attention. The Nash’s IUH (NIUH) is one of the most widely used IUH models. Several geomorphological approaches have been developed for NIUH, thus relating its parameters to watershed and drainage network physical characteristics. During the characterization of watersheds in geoprocessing softwares, the main information is that of relief obtained from a Digital Elevation Model (MDE) which can be obtained from topographic maps or from radar images (e.g. SRTM) and sensor images (e.g. ASTER). The main objective of this study was to evaluate the applicability and reliability of different relief data sources and scales for determination of geomorphological parameters required to estimate floods from NIUH. This study took into account data sets from five experimental watersheds with different physiographical characteristics, which have hydrological monitoring. The DEMs analysed in this study were obtained from topographical maps in the 1:50,000 scale, SRTM images with 30 and 90-meter spatial resolution, TOPODATA database, ASTER images. For the smallest watershed, an in situ topographic survey was also carried out for DEM derivation. Four geomorphological approaches for the NIUH were selected; two of them were based on the GIUH theory, whereas, the others were adjusted from empirical studies conducted in different watersheds. Based on the results obtained for the analysed watersheds, the main conclusions were: a) the main watercourse’s slope and Horton’s and Schumm’s ratios are the most sensitive parameters to relief data sources and scales; b) flat watersheds are the most susceptible to altimetry errors, which increase as the watershed area decreases; c) it is not possible to identify any combinations of geomorphological approach-DEM that better or worse describe all the analyzed watersheds when assessing watershed slope or drainage area independently; d) the geomorphological approaches which do not depend on streamflow speed information presented satisfactory results when compared to those based on GIUH theory; and e) the combination of different approaches enables to satisfactorily estimate the behavior of direct surface runoff hydrographs and their peak streamflow and time in all the considered watersheds.