A comparison of turbulence classification schemes

Various schemes for classifying the atmospheric turbulence in terms of the Pasquill A-F classes are compared and assessed. This classification is necessary in some aspects of air pollution control strategies that involve the use of dispersion models based on the Gaussian formulation. In particular the ΔT and σ_θ methods of the Nuclear Regulatory Commission, the STAR program, Richardson and bulk Richardson numbers, and a parameter defined as the ratio of wind speeds at two levels are compared. The results indicate that in terms of overall distributions, some of the methods compare favorably to the σ_θ method which is used as the basis of comparison. However, simultaneous turbulence category determinations by the various methods, on an hourly basis, do not correlate well except for the neutral class. Possible reasons for the discrepancies are discussed. The implications for modeling of concentration estimates are briefly evaluated.