CLOSE OVERLAY

Supplementary Frost Forecasts for Iowa Roads and Bridges

Project Details
STATUS

Completed

START DATE

12/10/13

END DATE

05/31/15

RESEARCH CENTERS InTrans
SPONSORS

Federal Highway Administration State Planning and Research Funding
Iowa Department of Transportation

Researchers
Principal Investigator
Gene Takle

About the research

Frost on roads and bridges in Iowa represents a potential hazard to the motoring public (Takle, 1990). Iowa Department of Transportation (DOT) contracts with a private agency for daily forecasts of bridge and roadway frost for guidance on mitigation and amelioration of frost conditions of traveled surfaces but seeks a supplemental forecast to assist in determining uncertainty at times when forecast skill is low. The Agronomy Department runs a legacy weather forecast model (MMS) in real time that can be used for creating forecast variables needed to drive bridge and roadway forecast models. Takle (1990) provided an overview of frost occurrence on roads and bridges in Iowa, documenting the spatial and temporal occurrences. Greenfield and Takle (2006) developed a bridge frost prediction model (BridgeT) that can be applied to Iowa bridges, and Knollhoff et al. (2003) provided estimates of the amount of frost accumulation required to be observable by standard IDOT methods of detection. Environment Canada (2013) has created a model, METRo (Model ofthe Environment and Temperature of Roads), for forecasting frost on both roads and bridges. METRo uses observations provided by a road weather station (road weather information system, RWIS) and the atmospheric forecast model, to predict the roads conditions such as freezing rain, accumulation of snow, frost and thaw for a specific location. METRo is written in python and fortran 90 and performs a 48-h forecast on a standard desktop computer in less than 2 seconds. The project proposes to run the MMS forecast model on a daily basis to produce 72-h forecasts of temperature, dew-point temperature, precipitation, wind speed and solar radiation at the standard observing level as well as skin temperature for all model grid points in Iowa at 12-km resolution. These output data will be used to drive BridgeT and METRo on a daily basis for all MMS grid points in Iowa to provide frost and temperature forecasts for roads and bridges in Iowa. The forecast model will be initialized with the 00 UTC (6 PM local standard time) and 12 UTC (6 AM local standard time) data and will provide output by 09 UTC and 21 UTC, respectively, to run the frost models. Surface temperature and output of the frost models will be available online at 12 UTC (6 AM) and 00 UTC (6 PM), respectively, for indicating surface temperature and frost occurrence for the following 30 hours and surface temperature and frost outlook for the subsequent 36 hours. Forecasts will be issued with twice daily valid times for the period of 26 November 2013 through 15 April2014. Frost forecast probability will be indicated by a color scheme indicating HIGH, MEDIUM, and LOW probability of frost for each county. These probabilities will be determined by use of frost accumulation results of Knollhoff (2003). Twice daily maps of surface temperature and frost occurrence on roads and bridges in all Iowa counties. A comparison of surface temperatures at validation time will be compared with observed values as measured by the road weather information system (RWIS) network. All of this information will be archived and provided as necessary to support analyses of this modeling approach.

 

 

TOP