Subsurface storage in small coastal BC watersheds and uncertainty in salt dilution gauging

Many small municipalities are reliant on surface water supply for their municipal water needs. Studies of water resources are seen as cost prohibitive and rarely implemented, even under development pressures and climate insecurities. Natural tracers can provide an economic alternative to extensive monitoring networks. This study used electrical conductivity (EC) and stable isotopes of water to characterize the subsurface water storage with regards to streamflow contribution, transit times and dominant flow processes in two small, mountainous watersheds of coastal southwestern BC. Long-term EC and periodic discharge measurements were used to establish an EC-discharge relationship. Transit times were evaluated by comparing the diel cycles in EC and temperature during snowmelt, and responses of EC to fall rain events. The relative contribution of snowmelt, shallow subsurface water and deeper subsurface water was calculated by hydrograph separation using EC and stable isotopes as tracers. Discharge measurements were made using the salt dilution gauging method and the second part of this study identified the appropriate measurement model, significant uncertainty sources, and how to combine these to a discharge uncertainty using standardized methods. Snowmelt was identified as a significant contributor to streamflow during the spring and summer months. The summer was characterized by more higher elevation groundwater storage, fed by snowmelt, than lower elevation or deeper groundwater storage. Using both tracers, the Harvey Creek stream consisted of 12% of reacted pre-event water attributed to deeper subsurface flows, 32% of unreacted pre-event water representative of high elevation headwater streams, and 55% of snowmelt. Discharge and EC were related by a power law relationship that appears distinct between spring to summer and fall to winter. Transit times differed with season and were 5 to 24 hours, although unquantified processes likely influenced this. The economic tracer usage had other limitations, most notably the requirement for sufficient spatial and temporal coverage for a good understanding of hydrological processes. The uncertainty evaluation noted a preferred discharge and calibration model and found that calibration prediction represented the largest uncertainty source for a well performed test. The major effects influencing discharge measurements were coarse integration resolution and non-linearly varying background EC.Applied Science, Faculty ofCivil Engineering, Department ofGraduat