Derivation of Climatology - Based CONUS PoP Guidance Grids

The creation of probability of precipitation (PoP) guidance grids in mountainous terrain can be challenging especially given the current resolution of most available models and ensembles. Nearly all of the models and ensembles have grid spacing too coarse to accurately depict the spatial variability of precipitation in complex terrain. In this study, climatological 24-hour PoPs were generated from the RFCQPE (River Forecast Center Quantitative Precipitation Estimation) data on an approximately 4km grid. Various point-based guidance from models (such as the MEX, MAV, MET MOS guidance), as well as the Weather Prediction Center (WPC) forecast point values, can be mapped to these climatological 24-hour PoP grids to create PoP guidance grids, which account for observed variations in PoP in complex terrain. A climatological-based PoP grid creation scheme can easily be used throughout the CONUS, even at locations with minimal spatial variation due to local effects. One of the outcomes of the Weather-Ready Nation initiative is to have accurate and spatially consistent forecast grids for the entire CONUS. Climatology-based PoP grids could be one of the many spatially-consistent guidance grids available for the production of national PoP grids through the “Blender Project.” One challenge in creating these grids has been the discontinuities between boundaries of the River Forecast Centers (RFC). These differences are most notable in locations which lack a high density of radar and surface observations, mainly in the intermountain western US. A scheme which utilizes Parameter-elevation Relationships on Independent Slopes Model (PRISM) data near the RFC boundaries was used to smooth these differences. Another challenge is that only four years (2011-2014) of RFCQPE grids for all RFCs were easily available. A limitation of this short time period is that it likely does not capture many of the climatological regimes. However, this brief period allowed for the robust use of COOP and Community Collaborative Rain, Hail and Snow Network (CoCoRaHS) data to adjust grids. Over 5000 data points of observed 24-hour PoPs were used to adjust the PoPs from the RFCQPE grids. The current terrain adjustment procedure for WPC PoP grids uses PRISM data to locally adjust the PoP to complex terrain, assuming that the PoP is locally proportional to average monthly precipitation. The analysis of RFCQPE data indicates that this assumption greatly exaggerates the change in PoP in a local area. The use of climatological grids could provide a more realistic spatial variation in PoPs.