This paper reviews road condition prediction models. These models, based on energy balance and statistical approaches, are widely used in north-west Europe to forecast pavement temperature, wetness and snow/ice cover. They are used as tools by highway maintenance organizations for the management and control of winter maintenance fleets, an activity costing over $2 billion per year in North America.
This paper has been prepared within the FORETELL program, a joint state and federal initiative to integrate Intelligent Transportation Systems (ITS) with advanced weather systems prediction using mesoscale models. FORETELL aims to support operational highway maintenance management and traveler information systems, eventually throughout North America. The purpose of this review is to help provide an informed basis for choice between the various modeling approaches currently in use. It examines the operation of the models as explained in the available literature.
An issue of particular interest within this paper is how best to integrate road condition models with mesoscale forecasting models. The paper is written both for meteorologists and engineers, assuming little prior knowledge, while not avoiding in-depth analysis.
The overall contents of the paper are as follows. First, the main models developed in the UK, Sweden, Denmark and Germany are outlined. Then, the models' bases are described in detail, tracing parameterizations back to the original literature wherever possible. Successive sections describe the basis of shortwave solar radiation modeling, incoming and outgoing longwave radiation, convection and latent heat transfers, and finally conduction between the road surface and its lower layers. The issue of traffic effects is briefly discussed. In addition, an overview of the road condition modeling approaches used in the FORETELL program will be given. Finally, the paper outlines some unsolved issues in road condition modeling, concerned with snow drifting and packed snow condition evolution through time.