Snowpack in the NARCCAP Regional Climate Models

This study assesses the ability of the regional climate models used in the North American Regional Climate Change Assessment Program (NARCCAP) to simulate snow water equivalent (SWE) over North America. As part of NARCCAP, six regional climate models (RCMs) were forced with NCEP-DOE Reanalysis 2 data to estimate biases associated with model configuration and physics. In this study, SWE from the NCEP-driven RCMs is compared against an ensemble of off-the-shelf, gridded 'observation based' SWE products compiled from a number of sources. This ensemble observed SWE product is intended to capture some of the known uncertainty associated with large-scale SWE measurements and includes data from microwave satellite products, snow pack models, SWE from reanalysis products, and SWE from land-surface models forced with observed meteorological conditions. Where available, the observed SWE product has been compared against raw snow data such as SNOTEL and other cooperative snow survey sites. To estimate biases in the RCMs, we compare the simulated SWE to the spread of SWE found from the different observational sources. Biases in SWE found in each model are then analyzed in relationship to simulated temperature, precipitation and albedo. Given that SWE is directly related to temperature, precipitation and the choice of land-surface parameterization scheme, the evaluation of SWE in the NARCCAP models allows us explore the processes level issues that result in model biases.