MODELING THE EFFECTS OF CLIMATE CHANGE ON PEAK FLOWS IN THE STILLAGUAMISH RIVER

The Stillaguamish watershed in northwest Washington State encompasses an area of about 1700-sq-km and has a drainage the ranges from sea level to over 2000 m of elevation. Seasonal snowpack within the watershed historically exists above 1,000 m, but modeling projections indicate that snow lines will recede to higher elevations as temperatures rise through the 21^st century. As additional landscape becomes exposed to winter rainfall rather than snow, more rapid runoff will increase peak flows and flood severity, particularly in the lowlands of the watershed which are developed and populated. Peak flows can also be detrimental to salmonid populations, which are of economic and cultural importance to the Stillaguamish Tribe of Indians. To quantify peak flows in the Stillaguamish River through the 21st century we use the physically based Distributed Hydrology Soil Vegetation Model (DHSVM) and new historical and projected meteorological data processed with the Weather Research and Forecasting (WRF) numerical model. Projected peak-flow estimates will aid in the development of updated flood maps and inform future salmonid recovery efforts.

We establish the digital spatial characteristics of the Stillaguamish watershed for the DHSVM at a 150 m grid resolution and calibrate the model using 1-hr historical WRF meteorological gridded surface data with a 6-km resolution. We calibrate streamflow to multiple established stream gauges in the basin and snowpack to regional SNOTEL stations. The calibrated model is then used to simulate the effects of projected climate warming into the 21st century using global climate models of the CMIP5 with one RCP4.5 and 12 RCP8.5 forcing scenarios that are dynamically downscaled using the WRF model at 1-hr time steps. Modeling results indicate that the Stillaguamish watershed will transition from a mixed rain-and-snow watershed to a purely rain dominant watershed into the 21st century. Peak flows increase in frequency and magnitude in the winter months due to lower snowpack and an increase in projected high intensity storm evens.