The climatic significance of tropical forest edges and their representation in global climate models

Abstract

An emerging theme in global climate modelling is whether land covers created in the clearance of tropical humid forests influence water exchange between remnant forest patches and the atmosphere, and, if so, how this affects regional and global water exchange. Fieldwork presented in this thesis ascertains whether the amount of water transferred to the atmosphere from a humid tropical forest situated in Sabah, Northern Borneo, Malaysia, differs between its edge and interior due to the influence of surrounding clearings through horizontal heat transfer. Using satellite imagery to measure the shape and size of tropical forests, field measurements of water transfer were extrapolated to continental and global levels to infer how differences in water exchange with the atmosphere between forest edges and interiors may influence regional and global forest-atmosphere water exchange.
Mean sap flow in trees within 50 meters of a forest-clearing boundary was found to be 73% greater than that in trees further into the forest; an observation supported by the decreased canopy temperature also recorded there. Evaporation from the forest canopy constituted a high fraction of annual rainfall (33%), but showed no edge effect similar to that of sap flow. Edge plots, however, expressed evapotranspiration rates 22% lower than forest interiors (657-890 mm yr-1), owing to the lower number and size of trees there. One edge plot, however, exhibited evapotranspiration 49.5% greater than that of forest interiors. Gradients of air temperature, vapour pressure deficit and wind speed from the adjacent clearing to the forest interior indicated that warm, dry air moving from the clearing to the forest was the most credible cause of increased sap flow of trees near the forest edge. This hypothesis was supported by a strong correlation between the amount of vapour in the air moving from the clearing and tree water use. It was estimated that the influence of differences in water transfer to the atmosphere between the edges and interiors of tropical forest would not alter global water transfer to the atmosphere by more than 0.25-4%, or by 4-7% in the most fragmented tropical continent, Africa.
However, it remains unclear whether the inclusion of tropical forest edge effects within climate models is necessary, as the pioneering nature of this thesis, and of existing studies reviewed within it, means that solid conclusions will be dependent upon future work. This thesis concludes with suggestions for future research that will most effectively consolidate the provisional conclusions and recommendations herein.