CAT (Canyon Air Temperature) is a parametric model that predicts site-specific air temperature in an urban street canyon for extended periods on the basis of data from a reference station in the region. In addition to a rudimentary description of the two sites, it requires as input only time-series of meteorological parameters measured at standard stations, which serve as descriptors of the constantly evolving meso-scale weather.

First proposed with fixed values for moisture availability based on surface cover at the two sites (Erell and Williamson, 2006), the revised formulation of CAT now estimates the effect of moisture availability on sensible heat flux using an empirical parameterization scheme based on the Local-scale Urban Meteorological Parameterization Scheme (LUMPS) described by Grimmond and Oke (2002). The value of the Priestly-Taylor coefficient a is evaluated at hourly intervals, taking into account direction-dependent upwind surface cover and the effect of atmospheric stability on the spatial extent of the source area. Stability is estimated by a simplified indicator based on the difference between air temperature at screen height and the surface temperature of the ground (approximated by its sol-air temperature).

Data from the Goteborg street canyon experiment (Eliasson et al, 2006) and reference data from the SMHI weather station about 1.8km away were used to assign empirical weightings for source areas for moisture at distances of up to 1000 metres from each of the sites. The distance of source areas contributing to atmospheric moisture at the two sites was found to increase with increasing atmospheric stability - with the significant exception of 'super stable' conditions associated with the development of the largest canyon heat islands. In such conditions – nights with clear sky conditions and zero wind for at least two hours – the value of a was found to depend only upon surface cover at the immediate vicinity of the sites.

In addition to being able to 'predict' the Goteborg street canyon air temperature, the revised CAT model was also tested against the original Adelaide data set. Although the moisture availability is now calculated by the model, rather than being provided as a user input, predictions of air temperature show improvement with respect to the previous formulation. The ability of CAT to predict canyon air temperature in all weather conditions in locations as diverse as Adelaide (see figure) and Goteborg suggests that the methodology may be used with some confidence to generate site-specific air temperature data for use in building thermal simulation modelling.

References

Eliasson, I., Offerle, B., Grimmond, C.S.B. and Lindqvist, S., 2006. Wind fields and turbulence statistics in an urban street canyon. Atmospheric Environment, 40, 1-16.

Erell, E. and Williamson, T., 2006. Simulating air temperature in an urban street canyon in all weather conditions using measured data at a reference meteorological station, International Journal of Climatology, 26, 1671-1694.

Grimmond, C.S.B, and Oke, T.R., 2002. Turbulent heat fluxes in urban areas: Observations and a Local-scale Urban Meteorological Parameterization Scheme (LUMPS). Journal of Applied Meteorology 41, 792-810.

Figure: Comparison of air temperature at reference site in Adelaide with measured and predicted temperatures at street canyon site, 8 days in June 2000.