Feedbacks between the atmospheric boundary layer and terrestrial ecosystems: implications for global-scale modelling

摘要

The importance of representing realistic vegetation in General Circulation Models (GCMs) is now widely recognised (Cox et al. 1999). Feedbacks with the vegetation may become particularly important when extrapolating into future climates with elevated CO_2 and temperature. However, very few attempts have been made to include dynamic interactions between vegetation and climate in GCM simulations (Betts et al. 1997). Dynamic vegetation models are generally too large and complex to be included in on-line simulations. There is therefore a need for models which are computationally simpler, but represent the key feedbacks accurately. The short-term feedbacks between vegetation and the atmosphere have been considered in several studies. McNaughton & Jarvis (1991) showed that the sensitivity of transpiration to a change in stomatal conductance decreased with increasing spatial scale, as the number of negative feedbacks increased. Jacobs and deBruin (1997) used a coupled planetary boundary layer (PBL)-vegetation model to study the influence of the PBL-vegetation feedbacks on surface conductance g_s and regional transpiration E. The effect of elevated atmospheric CO_2 concentration was also examined. The results confirmed the negative feedback between the PBL and transpiration. A positive atmospheric feedback on g_s was also observed, where an initial decrease in g_s caused by elevated CO_2 concentration reduces transpiration and thereby reduces humidity of the air within the canopy, causing the stomata to close further. Huntingford & Monteith (1998) and Huntingford & White (in prep.) reduced a PBL-vegetation model to a simple algebraic relationship between the Priestley-Taylor coefficient and g_s. The two coefficients of this relationship provide an analytical means of quantifying the strength of PBL-vegetation feedbacks.