Estimation of Dynamic Storage across Northern Catchments

Friday, 19 December 2014
Marjolein Van Huijgevoort1, Doerthe Tetzlaff1, James M Buttle2, Sean Kevin Carey3, Hjalmar Laudon4, James P McNamara5 and Chris Soulsby1, (1)University of Aberdeen, Aberdeen, United Kingdom, (2)Trent University, Peterborough, ON, Canada, (3)McMaster University, Hamilton, ON, Canada, (4)SLU Swedish University of Agricultural Sciences Umeå, Umeå, Sweden, (5)Boise State Univ, Boise, ID, United States
The volume of water stored within a catchment and its partitioning between groundwater, soil moisture, snowpack, vegetation and surface water ultimately characterise the state of the hydrologic system. While storage is relatively straightforward to conceptualize and define, it is difficult to measure. The response time of discharge to precipitation events and the partitioning of water are both largely determined by the amount of storage. Although this important role of storage is widely recognized, the processes within catchments and the influence of storage on the sensitivity of a catchment are still not completely understood. Here we present results from a study in which we estimated dynamic storage in five different northern headwater catchments within the VeWa (Vegetation effects on water flow and mixing in high-latitude ecosystems) project using the HBV-light model. All five catchments are located in high-latitude regions (Scotland, US, Canada and Sweden), but have contrasting catchment characteristics. These catchments allowed for inter-site testing of how to conceptualize catchment water storage and its relation to catchment runoff. Based on the output data from HBV, the relation between storage and discharge for each catchment was investigated. Changes in dynamic storage across the catchments and differences in calibrated parameters were studied and linked to catchment characteristics for better process understanding. Different scenarios were used to determine the influence of storage on the sensitivity of the catchment. Results showed that catchments with a large storage potential were less sensitive to changes in the input data, as expected. Even though a simple model was used here for the simulation, the results can still contribute to a better understanding of dynamic storage by the comparison across catchments.