RoadSurf – Open-source library for predicting road conditions

Ice and snow cause hazardous road conditions every winter, especially in northern countries. Accurate road surface temperature and road condition forecasts are essential for road maintenance personnel to keep the roads safe for driving. With accurate forecasts, the roads can be salted before the freezing occurs and thus the roads can be kept free of ice. Finnish Meteorological Institute has used its road weather model called RoadSurf for over two decades to predict road conditions. RoadSurf is one dimensional model and calculates the heat transfer in the road and the heat balance at the road surface. The model requires atmospheric values, like air temperature, wind speed and radiation as input values and calculates the road surface temperature and amounts of ice, snow, frost, and water on the road. Shadowing caused by the surrounding objects can have a large effect on the road surface temperature. If sky view factor and local horizon angles are known at the to forecast location, the model will use those to modify the radiation input. Soon other institutes and companies can also use RoadSurf, as the main parts of the model will be published as an open-source Fortran library. The library will not include all the model features like friction calculation, but it is programmed so that implementing one’s own features is easy. When the model is implemented to new locations, some model parameters might need calibration to better fit the local road structure. The model is light to run and can be run in parallel mode, so the forecasts can be made easily to thousands of road points. The quality of the forecasts made by RoadSurf has been assessed previously in many verification studies, and new studies will be made to ensure that the new open-source library performs well.