Simulation of a cold-air pool in Utah's Salt Lake Valley

The Persistent Cold-Air Pool Study (PCAPS), which took place in Utah's Salt Lake Valley during winter 2010-2011, provides a rich dataset of targeted observations at scales appropriate for better understanding the dynamical evolution of persistent cold-air pools. We examine the influence of the land use and land cover datasets available within the Weather Research and Forecasting (WRF) numerical model on the model's ability to accurately simulate a persistent cold-air pool (CAP). A modified version of the most recently released land use dataset, 2011's National Land Cover Database (NLCD 2011), was used to model a CAP that occurred from 1 January 2011 to 8 January 2011. Modifications to the NLCD 2011 dataset included reducing the areal extent of the Great Salt Lake to reflect the lake state at that time as well as changing the characteristics of a number of land use classifications (e.g., urban and barren land) to more closely match albedo observations obtained during PCAPS. Snow cover obtained from North American Mesoscale (NAM) reanalysis was also modified to better match observations. The resulting model simulation for the 1-8 January 2011 period was notably improved compared to an ‘out-of-the-box' run for the same period relying on the default U.S. Geological Survey (USGS) and unmodified NAM reanalysis snow cover data. The most substantive improvements were observed within the Salt Lake and Cache Valleys, where modifications to the areal extent of the Great Salt Lake and improved snow cover allowed for a more realistic simulation. The time of model initialization relative to the onset of the CAP was found to be a less critical factor than the improvements to land use and snow cover.