Assessment of Drought Constraints on Transpiration and Canopy Conductance in Mature Aspen and Jack Pine Stands

Soil water content can exert an important control on tree growth processes, an important consideration for the boreal forest in Central Canada where climate change may bring about increased drought frequency. Empirical evidence has shown the strong coupling between conductance and photosynthesis, and the effect that drought can have on both variables. In models, canopy conductance and carbon uptake are often explicitly related through feedback loops that may or may not include the consideration of soil water content. Models that are aimed towards landscape-level applications usually require the scaling up of response functions to coarse spatial and temporal scales. Although at such scales, crude models of stomatal response to environmental variables are sufficient, empirical data for parameter estimates or for validation of response functions on such scales are difficult to obtain. From 2001 to 2003, the Canadian Prairies and adjacent boreal forest were subjected to a severe drought that affected two sites over which continuous measurements of ecosystem exchanges of CO2 and water vapour had been made for a number of years: an aspen stand on a sandy clay loam soil, and a jack pine stand on coarse sand. The continuous measurements of CO2 exchanges provided an ecosystem-level experiment of drought impacts on canopy conductance. The objectives of this work were therefore to 1) determine the rooting depth at these sites based on measurements of transpiration and soil moisture 2) quantify the relationship between soil water content and canopy conductance for contrasting forested sites 3) determine whether these relationships were modified by the temporal scale at which they were analysed and 4) verify the applicability of published relationships for quantifying the impact of drought on gas exchanges. The results show that depth of water uptake (rooting depth) varies with soil texture, and tends to be shallower in coarser-textured soils. In this large drought event, soil water content was the most important variable in explaining the variability of canopy conductance of the aspen stand, but was far less important for the jack pine site where extreme drainage maintained the stand under near permanent drought conditions. Aggregation of temporal scale from half-hourly to monthly time steps tends to linearise physiological responses to environmental variables, and obliterates some that are critical at finer scales. At the monthly scale, solar radiation and soil moisture are the only significant variables for modelling canopy conductance. Finally, this empirical analysis offers much needed validation of proposed soil water modifiers that are necessary to properly represent the impact of present and future droughts on forest productivity. Our results for a sandy clay loam (aspen) are in agreement with those found in the literature, but a different relationship was found for a sandy soil with rapid drainage (jack pine).