A Land–Atmosphere Interaction Theory for the Tropical Deforestation Problem

Abstract

A theoretical framework is developed in understanding the mechanisms and processes determining the response of the land?atmosphere system to tropical deforestation. The analytical approach is made possible by simplifications in the vertical from the quasi-equilibrium moist convective closure, and in the horizontal from the dynamical temperature homogenization process. The theory emphasizes the energy and water balance. It highlights the interaction among processes of moist convection, cloud, radiation, and surface hydrology while each individual process is simplified. The zero surface energy flux condition, due to the small heat capacity of land, makes land?atmosphere interaction distinctly different from ocean?atmosphere interaction. This imposes a constraint on the sensitivity to the details of surface energy partitioning. Consequently, land surface temperature is largely a response to the energy and water balance, rather than a forcing as in the case of sea surface temperature. Results from a wet-season surface albedo change case compare well with a recent RCCM2/BATS simulation, with the theory depicting the mechanisms and the roles of the intertwining processes. The precipitation has a significant decrease, initiated by ground radiative forcing as increased surface albedo reflects more solar radiation into space. A positive feedback by moisture convergence is essential for this tendency, with another positive feedback from reduced evaporation providing further enhancement. These are opposed by a negative feedback due to the reduced magnitude of negative cloud radiative forcing as cloud cover decreases. This sheds light on the higher sensitivity in some GCM studies with prescribed clouds. The cloud radiative forcing also has a negative feedback on the initial cooling tendency in ground temperature. Together with reduced evaporation, this leads to little change in the ground temperature. Sensitivities of precipitation and ground temperature changes to individual processes are found to depend on the reference state parameter values, implying a sensitivity of anomaly response to simulated climatology for GCMs. The analysis here also serves as an example of the tight coupling between convection, large-scale atmospheric dynamics, and land processes in the tropical land?atmosphere system.