Eo-1 Hyperion Measures Canopy Drought Stress In Amazonia
The central, south and southeast portions of the Amazon Basin experience a period of decreased cloud cover and precipitation from June through November. There are likely important effects of seasonal and interannual rainfall variation on forest leaf area index, canopy water stress, productivity and regional carbon cycling in the Amazon. While both ground and spaceborne studies of precipitation continue to improve, there has been almost no progress made in observing forest canopy responses to rainfall variability in the humid tropics. This shortfall stems from the large stature of the vegetation and great spatial extent of tropical forests, both of which strongly impede field studies of forest responses to water availability. Those few studies employing satellite measures of canopy responses to seasonal and interannual drought (e.g., Bohlman et al. 1998, Asner et al. 2000) have been limited by the spectral resolution and sampling available from Landsat and AVHRR sensors. We report on a study combining the first landscape-level, managed drought experiment in Amazon tropical forest with the first spaceborne imaging spectrometer observations of this experimental area. Using extensive field data on rainfall inputs, soil water content, and both leaf and canopy responses, we test the hypothesis that spectroscopic signatures unique to hyperspectral observations can be used to quantify relative differences in canopy stress resulting from water availability.