Numerical assessment of changes in land–atmosphere interactions during the rainy season in South America using an updated vegetation map

Abstract

AbstractLand Use and Land Cover Change (LULCC) is a key driver of changes in land–atmosphere interactions, playing an essential role in climate and weather patterns in South America (SA). This study used a modelling approach to assess the land–atmosphere changes introduced by LULCC over the Brazilian territory. Numerical experiments of early (2006), neutral (2004) and late (2008) rainy season onset years were performed in the Integrated BIosphere Simulator (IBIS) to assess the land‐surface parameters, and in Brazilian Atmospheric Model (BAM) to assess the atmospheric variables. The models were run with a natural (NAT) vegetation map and an updated (UP) vegetation map, which incorporated realistic Brazil deforestation up to the early 2000s. The differences between the two maps were particularly larger over 25°–15°S and 50°–40°W, where the Atlantic Forest and Cerrado biomes were replaced by pasture. IBIS experiments showed that over the area of larger LULCC there was an increase in albedo of up to 8% while reductions in net radiation, surface heat fluxes, surface temperature and surface roughness were noticed. Land–atmosphere feedbacks assessed with BAM experiments showed that LULCC contributed to a drier and colder (stable) atmosphere over central‐east SA that opposes the necessary conditions for the rainy season onset. The changes included increases in surface pressure and low‐level wind associated with reductions in 2 m temperature and surface roughness. These changes contributed to decrease the cloud formation and less precipitation in all three rainy season onset years, but particularly in the late rainy season onset year during September–October–November (SON). The atmospheric changes induced by LULCC show a pattern of dry atmosphere (reduced precipitation), similar to those of late onset years, and indicate that LULCC enhances the late rainy season onset condition over central‐east SA. Additionally, this study highlights the importance of considering up‐to‐date vegetation maps because these changes significantly affect both surface and atmospheric variables.