|Title||Sensitivity of Oregon Watersheds to Streamflow Changes Due to Climate Warming: A Geohydrological Approach|
|Other Date||24-May-2011 (iso8601)|
|Note||Presented at The Oregon Water Conference, May 24-25, 2011, Corvallis, OR.|
|Abstract||A key challenge for resource and landscape managers is to predict the consequences of climate warming on streamflows and water resources. Different approaches are being developed to forecast the direction, magnitude, and timing of future streamflow changes in specific landscapes. One approach that is being utilized in the Pacific Northwest involves coupling downscaled climate predictions to macroscale hydrologic models, such as the Variable Infiltration Capacity (VIC) model. VIC is typically parameterized and calibrated in selected watersheds, and then applied to a regional scale that includes larger population of uncalibrated watersheds.
Summer streamflows are sensitive to both changes in the timing of snowpack accumulation and melt, and intrinsic, geologically-mediated differences in the efficiency of landscapes in transforming recharge (either as rain or snow) into discharge. Here we explore the importance of this effect by using geologically focused “bottom-up” approach to empirically characterize the sensitivity of late-summer streamflows to climate warming for a range of basins across Oregon. We define sensitivity as the slope of the relation between annual precipitation and summer streamflow, characterized as 7-day low flow and total summer flow. Drainage efficiency was defined in terms of the: 1) rate of recession (K) of the streamflow hydrograph; and 2) ratio of base flow to total flow (Base Flow Index or BFI). We compare our sensitivity results with those derived from VIC simulated streamflow.
Using the bottom-up approach, we found that the both K and BFI are good predictors for streamflow sensitivity to climate change. Fast-draining basins (high K / low BFI) are much less sensitive to changes in annual precipitation, whereas slow-draining basins (low K / high BFI) are much more sensitive. For basins where VIC was calibrated, downscaled VIC simulations are similar to empirical data. Uncalibrated basins, however, do not show a clear relationship with drainage efficiency, meaning that VIC may under predict sensitivity of summer stramflows to climate change in uncalibrated groundwater-dominated watersheds. This implies that spatial heterogeneity in aquifer properties must be explicitly incorporated into parameterization and calibration schemes if the full range of hydrologic response to warming is to be captured across the landscape.