Averaging for Wind Speed in a Given Direction in the Concentration Equation for Pollutants

J. Neumann Department of Atinospheric Sciences, The Hebrew University, Jerusalem, Israel

Search for other papers by J. Neumann in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

In the equation for the concentration of pollutants from a steady continuous point source, in a stationaryturbulent flow, the factor 1/u enters (u is the mean wind for a given stationary situation). If we are interestedin the concentration along a given wind direction and u denotes the wind speed in that direction and if weseek the average concentration for a class of flow situations (e.g., for the class of statically stable flows),each member of the class representing an individual stationary situation, then the averaging to be appliedis to 1/u and not to u. On the assumption (verified by some examples) that the distribution of u isa "humped" gamma distribution (standard deviation σ less than the average u of u for the class as a whole),we show that the average of 1/u equals 1/(u[1-(σ/u)2]}. Thus the average of 1/u is greater than 1/uand the resulting concentration estimate is larger than the one that would be obtained by the incorrect useof 1/u.

Abstract

In the equation for the concentration of pollutants from a steady continuous point source, in a stationaryturbulent flow, the factor 1/u enters (u is the mean wind for a given stationary situation). If we are interestedin the concentration along a given wind direction and u denotes the wind speed in that direction and if weseek the average concentration for a class of flow situations (e.g., for the class of statically stable flows),each member of the class representing an individual stationary situation, then the averaging to be appliedis to 1/u and not to u. On the assumption (verified by some examples) that the distribution of u isa "humped" gamma distribution (standard deviation σ less than the average u of u for the class as a whole),we show that the average of 1/u equals 1/(u[1-(σ/u)2]}. Thus the average of 1/u is greater than 1/uand the resulting concentration estimate is larger than the one that would be obtained by the incorrect useof 1/u.

Save