12 ANALYSIS OF ROAD TEMPERATURE PREDICTION
12.1 Comparing Predicted and Measured Road Temperatures
The heat balance at the Earth's surface is complicated by heat and moisture flows
occurring in many modes. Heat can be transferred to and from the surface from above by
all three principal methods of heat transfer: conduction, convection, and radiation.
Radiation components include both shortwave and longwave. Evaporation and
condensation are additional modes of heat transfer at the surface. Phase change heat
transfer is also a factor with freezing and melting of moisture occurring frequently in cold
regions. Heat and mass transfer with the soil underlying the surface also impacts the
surface energy balance. If the surface itself is assumed to have no finite mass, then its
temperature will be determined by the balance of these energy transfers at that point.
Accurately modeling all of these heat flow components to achieve an accurate surface
temperature prediction is a challenging task.
Equally challenging is the task of making an accurate measurement of surface
temperature, particularly on the surface of a high traffic roadway. Essentially any effort
to measure a surface temperature will in some manner disturb the surface temperature
distribution by introduction of a material with differing thermal or optical properties. A
temperature sensor placed right at the surface may not mimic the surface itself with
respect to thermal, optical, and textural properties, each an important factor in heat
transfer with the surrounding.
A third complicating factor in comparing measured and modeled results for surface
temperature is the input data used to drive the surface temperature model. Most often
complete measured weather data is not available at the site where the comparison is to be
done and local dissimilarities due to terrain as well as variation in weather over even
moderate distances can generate significant errors. Our task here is compounded by the
fact that for the operational use of our surface temperature model we need to rely on
forecasted weather data with its inherent inaccuracies. In the post analysis we can at least
eliminate this impact by using measured weather data, understanding fully that it is still
encumbered with the other limitations discussed above.
Hence when making comparisons between modeled and measured surface temperatures
it's imperative that we keep in mind the limitations of each result. While we are keenly
aware of the limitations of the surface temperature model presented here, and our ability
to generate precise weather to drive the model, we did not have a good grasp of what
expected measurement errors might be. For this reason we decided to make some surface
temperature measurements at several of the locations where the roadway temperatures
were being measured. We used two devices, a FLIR ThermaCAM S60 infrared camera
and a portable infrared surface thermometer. Measurements were taken under both day
and night clear sky conditions when we felt thermal and optical properties of the roadway
139
Previous Page
