An empirical method for estimating the spatial range and wind speed of country breezes using horizontal temperature gradients

This paper presents an empirical method for estimating the spatial range and wind speed of country breezes on the basis of horizontal temperature gradients which avoids the need for the (difficult) measurement of low thermally induced pressure gradients. This approach, which is still in the verification stage, allows country breeze speeds to be estimated using stationary temperature measurements or measurement trips and facilitates the analysis of interacting near-surface winds (country breezes, down-slope winds and mountain winds).

The method is based on stationary air temperature measurements along ventilation aisles from the surrounding area to the built-up area. The conclusions concerning the constancy of the country breeze become more precise as the number of stations is increased. An equation for estimating maximum country breeze speed using only a temperature gradient standardized with reference to distance is derived from a tried and tested pressure gradient approach on the basis of the ideal gas law. Using this equation, it is possible to determine whether flow between the stations is thermally induced. In order to determine the actual country breeze speed, a roughness term can be added to the equation to take into account the surface roughness, which normally increases at thermal cliffs.

The method was tested and verified in Cologne, Germany, on the basis of air temperature measurements (2 m above ground level) and wind measurements (4 m above ground level) made at seven near-surface stations along an open area extending from the surrounding countryside on the right bank of the River Rhine into the city center on 44 clear nights, mainly in the summer, between July 1995 and June 1996. The roughness z₀ was determined within an accuracy of 0.2 m.

The locations of thermally induced country breezes determined on the basis of temperature gradients between stations were in good agreement with the analysis of results obtained at stations. Between the surrounding countryside and the inner-city area, it was possible to identify several separate micro-scale country breezes between individual thermal cliffs. The maximum calculated country breeze speeds were approx. 4.3 m/s, whereas actual country breeze speeds were < 1.3 m/s. In the case of calculations made using the roughness term, the deviation between actual and estimated speeds was only - 0.1 m/s. The standard deviation was 0.3 m/s. Currently, the correlation coefficient between actual measured and calculated country breeze speeds is only r=0.53. The reasons for this relatively poor agreement are the high variation of relatively low wind speeds and the determination of roughness between stations, which is still rather imprecise.

Possibilities are presented to optimize the equation parameters for improving the statistical correlation between measured and calculated speeds of country breezes.