COMBINING MEASUREMENTS OF TRANSPIRATION AND STABLE ISOTOPES OF WATER TO DETERMINE GROUNDWATER DISCHARGE FROM FORESTS

Discharge of saline groundwater from Eucalyptus forests on a semi-arid floodplain was directly determined by first measuring transpiration r ates from the forests, and then partitioning the transpiration flux in to groundwater discharge and soil water depletion. This partitioning w as achieved by identifying the source of the transpired water with nat urally occurring stable isotopes of water. Transpiration rates were lo w, being about 0.3 mm day-1 from three E. largiflorens sites and up to 2 mm day-1 from an E. camaldulensis site. There was no significant va riation in transpiration across seasons, indicating that transpiration was limited by environmental factors other than evaporative demand. D espite its salinity (electrical conductivities of 11-33 dS m-1), the g roundwater was used by the forests at all sites and all times, and mad e up 100% of transpiration in more than half of the measurements, and 40-80% in the remainder. There was some consistency in water uptake pa tterns. E. camaldulensis tended to take up shallow soil water and grou ndwater simultaneously, as did trees at one of the E. largiflorens sit es. At the driest sampling time, however, groundwater was the only sou rce of water for trees at both of these sites. Trees at the remaining two E. largiflorens sites generally relied solely on the groundwater. The tree water source results indicate that groundwater discharge flux es were between 40 and 100% of the transpiration fluxes at these sites . These groundwater discharge fluxes were small in terms of regional g roundwater balances, but would be important in the salinisation of the soils. Additionally, uptake of water from the soil profile by the tre es substantially increased groundwater discharge compared with dischar ge from the soils had they been bare of vegetation.