The Impact of Multi-Depth Subsurface Sensor Data on Road Surface Model Forecasts

Precise road weather forecasts are crucial for road maintenance crews as good predictions of pavement temperature and state can help them to apply correct treatment on road surface and therefore reduce costs and save lives.

Accurate prediction of road surface temperature and state requires precise observations of both road surface and subsurface temperatures. Road Weather Information System (RWIS) stations typically include a surface and a subsurface temperature sensor that can be, for example at 30 cm depth. Road surface models utilize these observations to create an initial subsurface temperature profile that is used as a starting point for the forecasts. The vertical levels in road models go down to 1-2 m depth, where typically no observations are available and the temperature must be estimated. That can result in incorrect temperature profile and also translate into inaccurate pavement temperature and state predictions.

Ingesting data from multi-depth subsurface sensors into a road surface model will initialize the model with correct temperature profile and holds promise to improve the surface temperature predictions. This study will show some results of surface temperature forecasts from a road surface model with and without subsurface data.