Coupled ground- and satellite-based assessment of regional evaporation and ecosystem vulnerability in tropical wetlands

Abstract:

Satellite information plays a key role in tropical wetland monitoring and management. It is used to shed light on inundation dynamics of floodplains, to improve the understanding about eco-hydrological characteristics of floodplain ecosystems, and to quantify a wetland’s water balance. Seasonal wetlands in the tropics are very sensitive to changes in hydrological processes. Upstream land use alterations such as the installation of hydroelectric infrastructure or agricultural water withdrawal directly influence the ecosystem by changing the inundation dynamics. Large uncertainties due to missing in-situ measurements caused by remoteness complicate the quantification of a wetland’s water balance, where evaporation is considered to be its major water flux. We developed a spatially explicit approach to quantify daily evaporation considering the impact of inundation dynamics as the dominant controls and assessed the vulnerability of the Brazilian Pantanal wetland against the background of human induced impacts on the inundation process. In a first step a widely used water index (mNDWI) was calculated from MODIS surface reflectance products (MOD09A1) to differentiate between land and water for dry and wet years comparing and validating it with two years of continual in-situ water level measurements at different locations in our study area in the Northern Pantanal. Later on, we used the mNDWI to determine the water available for evaporation based on a recently developed approach (Schwerdtfeger et al., 2014, HESSD) to simulate evaporation fluxes on a large spatial scale. To set our evaporation results in the context of ecosystem vulnerability we defined the range of wet and dry years in the Pantanal for the last twelve years by means of local precipitation data and calculated yearly evaporation with our new approach. Considering now alterations of the inundation extent determined by the mNDWI in our model input parameters, our approach allowed us to make propositions about future ecosystem state in terms of alternated inundation dynamics and their impact on evaporation flux putting our results in context with the twelve observed years.