Land-Cover Characterization and Aridity Changes of South America (1982–2006): An Attribution by Ecohydrological Diagnostics

Abstract

Abstract To quantify impacts of climate change and anthropogenic activities on land surface dynamics a novel diagnostic tool is introduced, an application to the South American continent is presented, and the results are compared with observational studies. The diagnostics are performed in an ecohydrological state space spanned by surface flux ratios of excess energy U (loss by sensible heat H over supply by net radiation N) versus excess water W [loss by runoff (Ro) over gain by precipitation P]. The attribution of a changing state is deduced by rotating the (U, W) coordinates of its trajectory onto the external (or climate) and internal (or anthropogenic) forcings dependent of the regional state of aridity at the origin of the trajectory of change. Vegetation greenness (NDVI) is included in the attribution analysis as an active tracer. The first and second periods (1982–93 and 1994–2006; ERA-Interim) are chosen for change attribution analysis. 1) State space climates are characterized by a bimodal distribution with two distinct geobotanic regimes (forest and steppe) of high and moderate vegetation greenness. 2) Area changes between the first and second period are attributed to external (or climate)-induced (about 83%) and internal (or human)-induced (17%) processes. 3) In regions of significant (U, W) changes the significant vegetation greenness decreases in 36.2% (increases in 63.8%) of the area independent of vegetation type and aridity. 4) In these regions the water-limited areas tend to become drier, while energy-limited parts get wetter.