In places that experience snow and ice, road clearing and deicing operations are a necessity to ensure that road networks remain open and safe for travel. Such operations, however, are costly to both taxpayers and the environment making it all the more important that they are used in an efficient manner. Efficient use of road treatment resources takes experience on the part of the road network manager as well as access to reliable road surface temperature (RST) data which are used to determine when roads are conducive to snow and ice accumulation. On major roads and highways, road surface temperature is primarily obtained via road weather information systems (RWIS), thermal mapping, or a combination of the two methods. RWIS data are collected remotely from roadside weather stations which transmit meteorological readings and RST to a central computer running a predictive model such as HS4Cast (Hertl and Schaffar, 1998) or METRo (Crevier and Delage, 2001). RWIS are, however, limited in their usefulness because they only provide forecasts at their specific point locations. In reality, road surface temperatures can vary as much as 10°C at any given time depending on spatial location due to a number of interacting meteorological and geographical parameters (Shao et al., 1996). Thermal mapping was first described in the 1980s as a method to obtain RST in areas between roadside weather stations, thereby incorporating the spatial component of RST prediction (Gustavsson and Bogren, 1988). This method uses an infrared camera attached to a vehicle which travels along a subject route collecting data serving as a thermal “fingerprint” of the road surface that displays spatial variations of RST. When combined with RWIS data for verification, thermal mapping has proven to be an effective and economical method to visualize RST for large road networks (Shao et al., 1996).
RWIS and thermal mapping, however, are not universally used and may be impractical for certain areas such as southeastern Massachusetts that are in close proximity to the ocean and have very limited access to in situ road temperature data. This region of New England frequently experiences dramatic horizontal gradients of air temperature within short distances especially along the coast due to the influence of relatively warm ocean winds. This, combined with the unpredictable nature of ocean storms, introduces complexity to models and creates a challenge for road network managers to identify where and when conditions are right for the accumulation of ice and snow on roadways. Roadside weather stations for RWIS exist in this area, but are usually restricted to major state roads and are too few to verify thermal maps. As a result, local jurisdictions are required to decide when to dispatch road crews primarily based on visual interpretations of road conditions, which can be inefficient for large areas. There is much research describing methods to create point specific forecasts of RST on major roads, but little addressing the needs of local road networks without RWIS. Considering this fact, this ongoing project attempts to develop an alternative to thermal mapping and RWIS by indirectly estimating road surface temperature using Geographic Information Systems (GIS) and numerical modeling with metrological and geographical parameters.
Covert, J. & Hellstrom, R. (2014, June). An Indirect Method for Predicting Road Surface Temperature in Coastal Areas with Snowy Winters. Paper presented at the 71st Eastern Snow Conference, Appalachian State University, Boone, NC.
Virtual Commons Citation
Covert, Jason and Hellström, Robert (2014). An Indirect Method for Predicting Road Surface Temperature in Coastal Areas with Snowy Winters. In Geography Faculty Publications. Paper 11.
Available at: https://vc.bridgew.edu/geography_fac/11