Abstract:
This proposal aims to use a community-supported weather forecast model to study precipitation,
snowpack dynamics, and streamflow in Wyoming, a key headwaters region for the nation.
We plan to use a recently incorporated hydrology component to this model, the Weather
Research and Forecasting (WRF) model. This coupled land-atmosphere model will be run
over a 30 year period (1980-2009) driven by actual weather (using a “reanalysis” product)
at a sufficiently fine resolution (2 km) to capture orographic precipitation and runoff,
which are very terrain-sensitive. Our preliminary simulations indicate that WRF accurately
captures observed seasonal precipitation and snowpack build-up in Wyoming. Evapotranspiration,
infiltration, and groundwater release into streams are more uncertain, mainly because
of lack of data. Therefore streamflow data will be used to calibrate certain land
surface parameters. This model “training” will be based on the 30 year retrospective
run. The rather long simulation time is needed to validate statistical probabilities
of precipitation amounts at timescales ranging from hourly to annual, 1 April snowpack
water loading, and streamflow at various times of the year for all streams in Wyoming
at locations upstream of the first reservoir.
The proposal aims to answer two questions: firstly, how well does WRF Hydro simulate
the observed year-to-year variations in precipitation, snowpack dynamics, and streamflow
in the headwaters region of Wyoming? And secondly, how is the distribution of these
parameters expected to change in a 2050s climate? As to the latter, a pseudo-global
warming technique will be used to perturb the retrospective reanalysis with the anticipated
change according to the consensus global model guidance under IPCC’s most likely scenario.
This technique preserves low-frequency general circulation patterns and the characteristics
of storms entering the domain. The model then will be rerun over 30 years with perturbed
conditions, and any changes in the probability density functions of the above-mentioned
parameters will be examined. Thus we aim to quantify changes in water supply parameters
in Wyoming not just in an average sense, but also in terms of probabilities of water
excesses and shortages.
Broader Impact: Water is essential to the economy of Wyoming and the arid western
USA. This proposal combines our current understanding of the Earth system to quantify
the variability of seasonal water supply in Wyoming, and offers guidance about how
this variability may evolve in the next few decades. This work will benefit the state’s
water allocation decision process within the framework of interstate water treaties,
as well as agricultural and forestry interests. This work would not be possible without
our access to the NCAR Wyoming Supercomputer and to the NCAR-developed WRF model,
which we apply to Wyoming.