Generating Dynamic Prescription Maps for Winter Road Treatment via Sun-Shadow Simulation

Snow and ice can severely decrease road transportation efficiency and cause deadly threats to drivers. To cope with this issue, a significant amount of workforce and resources are arranged for winter roadway treatments each year. These anti-icing treatments typically apply the same amount of chemicals along the entire section of roadway, regardless of how much roadway has direct exposure to the sun and how much roadway may be in shadows of vegetation or buildings. There is a need to develop systematic methods to identify sections of the road with significant shadows that reduce solar heating of the road and to generate custom maps that vary application rates according to presence of significant shadows. In this work, we present an open-source sun-shadow simulator capable of large-scale road analyses for dynamic prescription maps. The National Renewable Energy Laboratory's Solar Position Algorithm was used to estimate the time- and location-dependent incident angles of the Sun's rays, while state-wide LiDAR data from the United States Geological Services were utilized to locate obstacles. By comparing with on-site photos and satellite imagery, the simulation results were shown to be highly accurate. Furthermore, site-specific direct sun radiation energy values were evaluated for a 38-km-long highway segment as a case study. The resulting prescription maps make it possible to dynamically adjust pre-treatment amount at three-meter resolution, increasing efficiency and cost-effectiveness while reducing the environmental impact.