Evaluation of statistical models in atmospheric dispersion modelling

When a gas is released into the atmosphere, there are inevitably random fluctuations in its concentration due to turbulence. Great recent improvements in instrumentation have shown how large these are. This has been acknowledged by CEC (the Commission of the European Communities), for example in that part of its Environment and Climate Programme (Framework IV) dealing with accidents. Likewise, many governmental and commercial organisations now recognise the need for regulatory models to include the effects of fluctuations. Models have been, and still are being, developed. Such models must be stochastic if they are to be realistic; a typical output is the probability that a given concentration is exceeded at a given point and time. Current CEC policy on Model Evaluation does not deal with such models, and this paper focuses on how it may be amended to do so satisfactorily. It is, incidentally, suggested that some re-evaluation of current policy might be in order.

[1]  T. G. McRae,et al.  Analysis of Burro series 40-m3 lng spill experiments , 1982 .

[2]  P. Sullivan,et al.  Spectral analysis of heavy contaminant clouds , 1997 .

[3]  Christopher D. Jones,et al.  Effect of atmospheric stability on concentration fluctuations and wake retention times for dispersion in the vicinity of an isolated building , 1995 .

[4]  Eugene Yee,et al.  Statistical characteristics of concentration fluctuations in dispersing plumes in the atmospheric surface layer , 1993 .

[5]  Hans Ejsing Jørgensen,et al.  Field experiments with dispersion of pressure liquefied ammonia , 1997 .

[6]  Rex E. Britter.  The evaluation of technical models used for major-accident hazard installations. Report to Commission of the European Communities Directorate General 12 , 1991 .

[7]  Steven R. Hanna,et al.  Hazardous gas model evaluation with field observations , 1993 .

[8]  David J. Ride,et al.  A practical method of estimating toxic loads in the presence of concentration fluctuations , 1995 .

[9]  S R Hanna,et al.  Air quality model evaluation and uncertainty. , 1988, JAPCA.

[10]  R. F. Griffiths.  The use of probit expressions in the assessment of acute population impact of toxic releases , 1991 .

[11]  D. M. Lewis.  Monte Carlo estimates of dosages of gases dispersing in the atmosphere , 1997 .

[12]  Søren Ott,et al.  A collection of data from dense gas experiments , 1995 .

[13]  M. G. Dodson,et al.  Ignition probabilities in turbulent mixing flows , 1981 .

[14]  Steven R. Hanna,et al.  Confidence limits for air quality model evaluations, as estimated by bootstrap and jackknife resampling methods , 1989 .