An initial evaluation of the performance of the Pavement Precipitation Accumulation Estimation System (PPAES)

Road maintenance personnel are responsible for keeping roads safe for travel and thus plow and chemically treat roads upon which wintertime precipitation has fallen. In order to manage this activity, road maintenance personnel require real-time information regarding where and how much precipitation has fallen. Our ability to determine these, however, is limited. Surface observation data are limited owing to lacking spatial and temporal densities and because few surface observation sites provide real-time information regarding wintertime precipitation accumulation. While Road Weather Information Systems (RWISs) have great potential to help in this regard since they are located next to roadways where estimates of precipitation occurrence and accumulation are needed, they generally do not provide information regarding wintertime precipitation accumulation. Radar data, on the other hand, provide much improved spatial and temporal coverage. These data, though, are limited because low-elevation radar beams commonly overshoot winter precipitation systems at moderate to distant ranges and because radar-based precipitation estimates typically have greater errors than do surface-observation-based estimates. Satellite data are limited by their temporal resolution and by the fact that they have not commonly been applied to relatively weak wintertime systems.

In response to the road maintenance personnel need for improved information regarding where and how much precipitation has fallen, the Pavement Precipitation Accumulation Estimation System (PPAES) has been developed. PPAES utilizes surface, radar, satellite, and model/analysis data, and is designed to take advantage of the strengths of these data sets so as to provide as accurate of information regarding wintertime precipitation occurrence and accumulation as possible. A key aspect of PPAES is its design to provide these estimates along roadways, where road maintenance personnel need them the most. Novel aspects of this system include the use of GOES-derived cloud top pressure data in conjunction with radar and surface observation data to enhance areal coverage.

PPAES precipitation occurrence and accumulation products are being validated during the 2005-06 and 2006-07 winters. Precipitation occurrence is being tested using ASOS/AWOS and RWIS (Road Weather Information Station) precipitation occurrence data from across North and South Dakota. Performance measures [e.g., False Alarm Ratio (FAR), Probability of Detection (POD), and the Critical Success Index (CSI)] are being computed for multiple instantiations of PPAES, including with and without satellite data and model/analysis data. PPAES performance regarding precipitation accumulation is being tested using data from the field site that has been established by the Surface Transportation Weather Research Center (STWRC), within which this research is embedded. Any events that produce wintertime precipitation at the field site are being used, with the events being stratified according to precipitation amount and wind speed. These strata are being used since PPAES precipitation accumulation accuracy is expected to depend upon precipitation amount and since the performance of surface precipitation gauges (e.g., the Geonor at the field site) depends upon wind speed.

PPAES design, development, and validation will be discussed. Emphasis will be placed on initial validation results for events from the first half of the 2005-06 winter. In addition, continuing development and future challenges will be considered.