Dispersion models atmospheric stability

TABLE 26-28 Atmospheric Stability Classes for Use with the Pasquill-Gifford Dispersion Model... 

The Britter and McQiiaid model was developed by performing a dimensional analysis and correlating existing data on dense cloud dispersion. The model is best suited for instantaneous or continuous ground-level area or volume source releases of dense gases. Atmospheric stability was found to have little effect on the results and is not a part of the model. Most of the data came from dispersion tests in remote, rural areas, on mostly flat terrain. Thus, the results would not be apphcable to urban areas or highly mountainous areas. 

Gaussian dispersion models based on Turner s Workbook of Atmospheric Dispersion Estimates, PHS Pub. No. 999-AP-26. Different air stabilities and wind speeds are used. 

In another review, Hoffert discussed the social motivations for modeling air quality for predictive purposes and elucidated the components of a model. Meteorologic factors were summarized in terms of windfields and atmospheric stability as they are traditionally represented mathematically. The species-balance equation was discussed, and several solutions of the equation for constant-diffusion coefficient and concentrated sources were suggested. Gaussian plume and puff results were related to the problems of developing multiple-source urban-dispersion models. Numerical solutions and box models were then considered. The review concluded with a brief outline of the atmospheric chemical effects that influence the concentration of pollutants by transformation. 

Source-dispersion and receptor-oriented models have a common physical basis. Both assume that mass arriving at a receptor (sampling site) from source j was transported with conservation of mass by atmospheric dispersion of source emitted material. From the source-dispersion model point of view, the mass collected at the receptor from source j, Mj, Is the dependent variable which Is equal to the product of a dispersion factor, Dj (which depends on wind speed, wind direction, stability, etc.) and an emission rate factor, Ej, 1. e. , ... 

Some AQ models rely on standard meteorological products which usually do not include turbulence, atmospheric stability, mixing height, and dispersion coefficients... 

One commonly used suite of models that is based on Gaussian plume modeling is the Industrial Source Complex (ISC) Dispersion Models (US EPA, 1995). This suite includes both a short-term model (ISCST), which calculates the hourly air pollutant concentrations in an area surrounding a source, as well as a long-term model (ISCLT), which calculates the average air pollutant concentrations over a year or longer. ISCLT uses meteorological data summarized by frequency for 16 radial sectors (22.5° each) this data format is referred to as a stability array (STAR). Within each sector of STAR, joint frequencies of wind direction, wind speed, and atmospheric stability class are provided. 

Flux is a somewhat abstract entity and does not relate this information directly. It quantifies the chemical movement rate across an interface plane into a receiving media such as the air boundary layer (BL) in the above example. Only when it is coupled with an air dispersion model does it produce concentrations in air. In the case of a large soil surface area source, a simple relationship exists between flux and concentration. For neutral air stability conditions in the atmospheric BL with steady-state wind speed v (m/sec), the concentration in air, c (mg/m ), can be approximated by... 

Atmospheric dispersion modelling requires an accurate and reliable description of the weather, including descriptions of the prevailing airflow, the atmospheric stability and the effects of flow structures at the spatial and temporal scales of interest. 

Leuning, R. (2000). Estimation of scalar source/sink di.stributions in plant canopies using Lagrangian dispersion analysis corrections for atmospheric stability and comparison with a multilayer canopy model Boundary-Layer Meteorol. 96, 293-314. 

A case study is performed assuming an instantaneous release of the toxic liquid acrylonitrile from a rail tankwagon. After the release of the toxic liquid a pool of 600m is formed from which evaporation occurs, leading to a vapour cloud. This vapour cloud travels with the wind and disperses. The degree of dispersion is determined by the wind speed, the stability of the atmosphere and the surface roughness. The stability of the atmosphere is indicated by the pasquill-stabUity class. By day, the most common atmospheric stability class is class D and a wind speed of 5 m/s is assumed, this weather condition is abbreviated with D5. At night class F is the most common atmospheric stability, associated with a wind speed of 1.5 m/s this weather condition is abbreviated as FI.5. The evaporation and dispersion calculations are performed with EFFECTS 7.6. A neutral gas model is used for the dispersion calculations. 

Turbulent diffusivity is by nonzero wind speed the function of time, location, wind speed and meteorological conditions. The turbulent diffusivity is implemented to the model on the basis of Sutton s definition of coefficients of dispersion. Coefficients of dispersion Oj = (Ty, oi are dependent on atmospheric stability and distance from the source. 

The ability to model the spread of gases considerably denser than air is particularly interesting to chlorine processors. Relative degrees of atmospheric stability have less effect on the dispersion of heavy gases. Some of the specific characteristics of chlorine also modify the phenomenon of dispersion [73]. Ice tends to form near the release point, for example. There is also the possibility of entrainment of liquid chlorine by the formation of mist or by a choking flow that produces a two-phase mixture. The latter leads to rainout and revaporization of liquid at some distance from the source. 

The passive gas dispersion models are usually based on the Gaussian plume model. In Gaussian models, atmospheric dispersion is taken into account through empirical dispersion coefficients that vary by atmospheric turbulence class (stability class) and distance from source. Dilution by the wind is taken into account through division by wind speed. No consideration, however, is given to the difference of the density between the ambient air and the gas, other than to calculate an initial plume rise if the release is hot (buoyant plumes rise according to relatively well-established approximations and then behave as a plume characterized by Gaussian concentration profiles). Because of this, these models must only be used for gas mixtures with a density approximately the same as that of air. 

Recommended Atmospheric Conditions for Use in Risk Analysis. The main input parameters used in dispersion models for estimating the downwind extents of hazard zones are atmospheric stability, wind speed, and wind direction. For any location in Canada, joint frequency distribution for these variables can be obtained from the Atmospheric Environment Service of Environment Canada, in terms of ... 

Analyzing the performance of a dispersion model by comparing the model output with observed values should result in the establishment of some performance measure. Often such a performance measure consists of one or more statistical parameters, but it can also be a written statement or a graphical presentation. Here we restrict ourselves to statistical measures, keeping in mind that they should be applicable to whole data sets as well as to specific subsets of data in order to examine trends of the performance with some external parameters (e.g., atmospheric stability or downwind distance). 

---