Search

summary Streambed temperature and heat fluxes are important for aquatic habitats as well as in the development and improvement of water temperature models. In the present study, measured streambed temperatures at different depths were used as a tracer to predict the magnitude and direction of groundwater flow using an advection–conduction heat transport model. This analysis was carried out under different conditions, namely under natural surface water temperature conditions (i.e., as measured in the field), under steady-state conditions (e.g. under stream ice cover) and for conditions where the surface water temperatures followed a sinusoidal function. In Catamaran Brook, results from the advection–conduction numerical model showed good agreement between predicted and observed streambed temperatures with root-mean-square errors (RMSEs) ranging between 0.07 C to 0.6 C. A comparison of streambed fluxes showed that the heat flux by conduction was more important during the summer period for upwelling conditions (mean value 96 W m � 2

Summary Global climate models (GCMs) project significant changes to regional and globally-averaged precipitation and air temperature, and these changes will likely have an associated impact on groundwater recharge. A common approach in recent climate change-impact studies is to employ multiple downscaled climate change scenarios to drive a hydrological model and project an envelope of recharge possibilities. However, each step in this process introduces variability into the hydrological results, which translates to uncertainty in the future state of groundwater resources. In this contribution, seven downscaled future climate scenarios for a northern humid-continental climate in eastern Canada were generated from selected combinations of GCMs, emission scenarios, and downscaling approaches. Meteorological data from the climate scenarios and field data from a small unconfined aquifer were used to estimate groundwater recharge with the soil water balance model HELP3. HELP3 simulations for the period 2046–2065 indicated that projected recharge was most sensitive to the selected downscaling/debiasing algorithm and GCM. Projected changes in average annual recharge varied from an increase of 58% to a decrease of 6% relative to the 1961–2000 reference period. Such a large range in projected recharge provides very little useful information regarding the future state of groundwater resources. Additional results from recent comparable studies are compiled and discussed. Based on the results obtained from the present case study and the other studies reviewed, the limitations of current approaches for projecting future recharge are identified, and several suggestions for research opportunities to advance this field are offered.