Dependence of rainfall on vegetation: theoretical considerations, simulation experiments, observations, and inferences from simulated atmospheric soundings

Several simulation experiments, aimed at some selected tropical regions, have been carried out with a number of different state-of-the-art global climate models (GCMs). Most of the earlier experiments were conducted with GCMs using highly simplified parametrizations of the biosphere while others, particularly the more recent ones, used GCMs with much more realistic parametrization of the biosphere. The influences of three different aspects of the vegetation, namely the surface albedo, the evapotranspiration, and the surface roughness, have been examined in the former studies whereas the influences of realistic vegetation anomalies, either observed or anticipated, have been examined in the latter studies. Almost all these studies have shown that vegetation increases rainfall. This paper demonstrates how the dependence of convective rainfall (which is the major component of the total rainfall in tropics) on the local vegetation can be inferred by probing into interactions among vegetation, boundary layer, and moist convection. Our sensitivity tests on a large variety of simulated atmospheric soundings strongly in support the rationale behind these inferences.