Global climate change due to the buildup of greenhouse gases in the atmosph
ere has serious potential impacts on water resources in the Pacific Northwe
st. Climate scenarios produced by general circulation models (GCMs) do not
provide enough spatial specificity for studying water resources in mountain
watersheds. This study uses dynamical downscaling with a regional climate
model (RCM) driven by a GCM to simulate climate change scenarios. The RCM u
ses a subgrid parameterization of orographic precipitation and land surface
cover to simulate surface climate at the spatial scale suitable for the re
presentation of topographic effects over mountainous regions. Numerical exp
eriments have been performed to simulate the present-day climatology and th
e climate conditions corresponding to a doubling of atmospheric CO2 concent
ration. The RCM results indicate an average warming of about 2.5 degrees C,
and precipitation generally increases over the Pacific Northwest and decre
ases over California. These simulations were used to drive a distributed hy
drology model of two snow dominated watersheds, the American River and Midd
le Fork Flathead, in the Pacific Northwest to obtain more detailed estimate
s of the sensitivity of water resources to climate change. Results show tha
t as more precipitation falls as rain rather than snow in the warmer climat
e, there is a 60 percent reduction in snowpack and a significant shift in t
he seasonal pattern of streamflow in the American River. Much less drastic
changes are found in the Middle Fork Flathead where snowpack is only reduce
d by 18 percent and the seasonal pattern of streamflow remains intact. This
study shows that the impacts of climate change on water resources are high
ly region specific. Furthermore, under the specific climate change scenario
, the impacts are largely driven by the warming trend rather than the preci
pitation trend, which is small.