Water on road surfaces significantly degrades the available skid resistance. Indeed, 25% of injury accidents occur on wet roads in France due to lack of skid resistance. Some research results advocate an exponential decrease in skid resistance with increase in water depth. The above conclusions show the importance of knowing the water depth present on road surfaces during and after rainfall events in order to evaluate the available skid resistance for driver safety. In this paper, a prediction model of water depth on road surfaces during and after rainfall events is presented. This model takes into account the road surface characteristics such as macrotexture, slope and porosity, and weather conditions such as temperature, wind speed, relative humidity and rainfall rate. The model is based on the kinematic wave simplification of the Saint Venant equations, which represents the balance of flows from rainfall, run-off, infiltration and evaporation. The model is validated from comparisons to experimental measurements of water depth performed on four different surfaces of the IFSTTAR test track. The experimental device for the validation is composed of sensors that measure water depth, rainfall intensity, wind velocity, air humidity and temperatures. The model predictions on these four surfaces are very satisfactory.
