Simulation of deforestation in eastern Amazonia using a high-resolution model

Abstract

This work evaluates the impact of deforestation on the climate of the eastern portion of the Amazon basin. This region is primarily an area of native tropical rainforest, but also contains several other natural ecosystems such as mangroves and savanna. It is the most densely populated area in Amazonia, and has been significantly affected by deforestation. In this study, numerical simulations were performed with a high spatial resolution, regional model that allows for consideration of mesoscale aspects such as topography, coastlines and large rivers. To evaluate the present situation and to predict potential future effects of deforestation on the climatic conditions of this region, two, one-year model simulations were made. In the first, ‘‘control simulation’’, an attempt was made to match the existing surface vegetation. The biophysical parameters used were derived from recent studies of similar Amazonregion ecosystems. In the second run, ‘‘deforested simulation’’, the forested-area biophysical parameters were replaced by those corresponding to the pasture areas of the region. The higher-resolution regional modelling revealed important climatic features of the deforestation process, displaying someassociatedmesoscaleeffects that are not typically represented in similar Global Circulation Model simulations. Near coastal zones and along large rivers, deforestation resulted in reduced cloud cover and precipitation. However, increased cloudcoverandprecipitation waspredictedoveruplandareas, especially onslopesfacingrivervalleys.Themodelledsurface sensible and latent heatfluxesalso presented bothpositive and negative anomalies. The magnitudes of these anomalies were greater during the dry season.Windspeednearthe surfacewas the meteorological variable that presented the most significant change due todeforestation. The reduction in roughness coeff icient resulting from the shift from forest to pasture produced increased windspeeds near the Atlantic coast. The greater windspeeds diminished local humidity convergence and consequently reduced rainfall totals in nearby regions. The results obtained from these higher-resolution simulations show that, in general, orography, coastline profile and the distribution of large rivers play important roles in determining anomaly patterns of precipitation, wind, and energy exchange associated with deforestation in eastern Amazonia.