A Further Examination of the Urban Maximum of Thunder Rainfall in London, 1951-60

The maximum of thunder rain over London's urban area is examined in an attempt to isolate those storms which contributed most thunder rainfall. First, an analysis is made of the distribution of thunder rain of different intensities to see whether the urban effect varied with storm intensity. Secondly, within previously established synoptic types and associated vertical distributions of temperature, humidity, wind speed and direction over south-east England, calculations are made of the mean depth and movement of clouds in the thunder atmosphere. It is shown that urban induced thunder clouds are capable of precipitating over London in the summer season only. Thirdly, summer thunder rainfall of different intensities and in different synoptic types is investigated. The whole analysis suggests that the urban effect is only strongly operative in six synoptic types and that, within these types, fairly heavy ( > I2 mm/day) warm frontal falls are the most important. Two examples of the latter type are presented. IN a previous paper (B. W. Atkinson, 1968) it was shown that a maximum of thunder rainfall existed over London. It was suggested that this urban maximum was a summer occurrence and probably resulted from increases in temperature, humidity, turbulence, and the number of condensationand ice-nuclei over the city. Recent observations of these elements in cities lend support to these preliminary suggestions. For instance, W. A. Hass et al. (1967) have shown that even one incinerator causes substantial lift to tetroons over New York city; G. W. Paltridge (1968) has observed horizontal temperature anomalies of 6°C in the lowest 90o m caused by individual buildings; T. J. Chandler (1967) has conclusively demonstrated for the first time that absolute humidities are sometimes higher in cities than in the nearby country; I. R. Graham (1968) has shown that turbulence increases markedly over Fort Wayne, Indiana; and V. J. Schaefer (1966) has suggested that the lead oxide content of automobile exhausts may provide ice nuclei in amounts which, if found at appropriate altitudes, would be adequate for precipitation initiation. The present paper builds on the results of the previous one (Atkinson, 1968) and reports on a three-pronged analysis of the observed urban thunder rain maximum. In effect, the total thunder rainfall distribution (Fig. I) is dissected in an attempt to discover which storms actually gave rise to the urban maximum. The first part of the 'dissection' is to see whether the urban effect varied with the intensity of the thunder outbreaks. The second part of the analysis is undertaken to answer the following questions: if storm clouds are triggered by the urban area, would the precipitation fall on the urban area? Alternatively, if the urban maximum exists, is it a result of precipitation from clouds initiated by the conurbation or by some other nonurban feature? Such questions lead us to enquire about the lifetime of clouds and the distance they travel from inception to precipitation. The third part of the analysis is a logical extension of the second. If, in fact, clouds could be initiated by and precipitate on the urban area, it does not necessarily mean that they actually do so. Consequently, the distribution of rainfall in different synoptic types is analysed to see in which types the urban maximum occurs. This 97 This content downloaded from 157.55.39.207 on Thu, 20 Oct 2016 04:13:18 UTC All use subject to http://about.jstor.org/terms