Groundwater constraints on simulated transpiration variability over South Eastern Australian forests.

A land surface scheme with and without groundwater-vegetation interactions is used to explore the impact of rainfall variability on transpiration over drought-vulnerable regions of Southeastern Australia. We demonstrate that if groundwater is included in the simulations there is a low correlation between rainfall variability and the response of transpiration to this variability over forested regions. Groundwater reduces near-surface water variability enabling forests to maintain transpiration through several years of low rainfall, in agreement with independent observations of vegetation greenness. If groundwater is not included, the transpiration variability matches the rainfall variability independent of land cover type. Our results suggest that omitting groundwater in regions where groundwater sustains forests will (a) likely overestimate the likelihood of forest die- back during drought; (b) overestimate a positive feedback linked with declining transpiration and a drying boundary layer and (c) underestimate the impact of land cover change due to inadequately simulating the different responses to drought for different land cover types.

Figure 8: The difference in the standard deviation of the deseasonalized (by removing the mean annual cycle) total column soil moisture (in kg m-2) between the NoGW and GW simulations. The stippling (dots) indicate regions covered by at least 60% grasses and the areas with at least 85% forest cover are marked with an x.