Atmospheric dispersion

Quick links: TAPRI Non-point sources Source term sensitivity Thompson buildings study GMOs Historical dispersion modelling Meteorological data for dispersion modelling

CERC was set up to create practical models using the latest developments in the understanding of the dispersion of pollutants in the atmospheric boundary layer. The resulting ADMS suite of models now comprise:

• ADMS 6 for modelling emissions from small numbers of industrial or agricultural sources in detail;
• ADMS-Urban and ADMS-Roads for modelling emissions from urban areas, in particular those from road sources and domestic combustion;
• ADMS-Airport for modelling emissions from urban areas including aircraft take-off and landing emissions; and
• GASTAR, ADMS-STAR and ADMS-Puff for modelling short-term hazardous releases including those of dense gases and radioactive substances.

The principles behind these models, with references to underlying fundamental research publications, are given in the ADMS Technical Specification documents. CERC's understanding of atmospheric dispersion has also been applied to modelling wind turbine wakes and anthropogenic heat emissions.

Thermal Transport of Air Pollution from Regulated Industries (TAPRI) project

The UK Environment Agency‘s Thermal Transport of Air Pollution from Regulated Industries (TAPRI) research project assessed the impacts of thermal air flows on atmospheric pollutant dispersion. The project explored how certain local weather patterns - thermal flows such as sea breezes, warm city air, or cold air from hillsides - can affect how air pollution disperses from industrial and waste sites. Real-world examples, data analysis, and computer modelling were used to explore how often these weather patterns occur, how much they impact air quality, and how they might change in the future.

A number of reports resulting from the project, comprising summary reports, stakeholder consultation and an exploration of the project outcomes, have been published. CERC produced one of these reports, entitled Case studies on thermal flow conditions. In the report, CERC’s analyses of meteorological, land use and topographical data for selected ‘pilot areas’ are presented, leading to a better understanding of thermal flows. CERC explored approaches that could be used to model air quality impacts of thermal flows, using the models ADMS and KLAM-21 and using Numerical Weather Prediction (NWP) data.

The work was carried out by CERC in collaboration with Air Quality Consultants (AQC) and John Moncrieff (University of Edinburgh School of Geosciences), as part of Defra’s Research Development and Evidence Framework Agreement.

Dispersion from non-point sources

Many regulated sources have complex geometries near or at ground level, for instance agricultural emissions of ammonia and particulates from pig and poultry farms, litter and manure storage and land spreading. Other 'non-point' sources include composting sites where bio-waste emits fungi and bacteria, for instance from windrows. These sources are usually in rural areas which may be close to protected nature sites, with particular concerns about exposure to pollutants. In addition, agricultural and composting facilities are sources of odour, where the impacts on nearby residential areas must be assessed.

CERC, A S Modelling & Data Ltd. and Professor Akula Venkatram (University of California, CA, US) carried out 'A Review of the Limitations and Uncertainties of Modelling Pollutant Dispersion from Non-point Sources'. This work was funded by the UK Atmospheric Dispersion Modelling Liaison Committee (ADMLC). The report presents a detailed literature review and results from four validation studies (focussing on ADMS and AERMOD); it also includes a good practice guidance section. The report describing the outcomes of this review was published in April 2016 and is available for download from the ADMLC publications webpage. Some of this work was presented at the 17^th conference on Harmonisation within Atmospheric Dispersion Modelling for Regulatory Purposes; the extended abstract describing this work is also available.

Source term sensitivity

Defining source terms is an essential part of dispersion modelling. A substance can be released into the atmosphere in many different ways, particularly in accidental release situations. It is important to have an understanding of the sensitivity of the model results to each of the source term input parameters. CERC and GT Science & Software Ltd carried out a 'High Level Review of the Sensitivity of Dispersion Model Predictions to Individual Source Term Parameters'. This work was funded by the UK Atmospheric Dispersion Modelling Liaison Committee (ADMLC).

The review discusses the main issues of source term sensitivity, describes a range of commonly-used dispersion models, and presents the results of detailed sensitivity tests. The source terms examined in the review include evaporating pools, catastrophic failures of pressurised vessels, jet releases, spray releases, warehouse fires and pool fires. The report describing the outcomes of this review was published in January 2017 and is available for download from the ADMLC publications webpage.

Evaluating ADMS buildings dispersion module using Thompson study

Detailed data from the Thompson study of dispersion from sources near buildings have been used to evaluate and improve the ADMS buildings dispersion model. This wind tunnel study used four different building geometries and multiple source locations relative to each building geometry to examine the variation of buildings effects on dispersion. ADMS model performance has been evaluated against the whole dataset and also in detail for each buildings geometry in order to identify potential improvements to the ADMS model code.

Dispersion study of GMOs

CERC was a partner in the EU Framework 6 project SIGMEA with the goal of Genetically Modified Organism (GMO) mapping across Europe. ADMS with its advanced treatment of area sources was used for the dispersion modelling aspects of the problem. CERC also carried out investigations into the dispersion of pollen for Ministry for Agriculture, Fisheries and Farming (MAFF), now part of Defra. The work is reported in the publication below.

Learn more

• Walklate PJ, Hunt JCR, Higson HL and Sweet JB,2004:A model of pollen-mediated gene flow for oilseed rape.Proc. R. Soc. Lond. B March 7, 2004 271:441-449, DOI 10.1098/rspb.2003.2578 (Royal Society website)

Historical dispersion modelling

Economics researchers at the University of Bristol have investigated the impact of industrial pollution during the Industrial Revolution on the social structures of English cities. To inform the research, CERC ran ADMS 5, using statistical meteorological data, to generate pollution footprints for over 50 English towns and cities at the peak of the Industrial Revolution. Dr Stephan Heblich, Dr Alex Trew and Dr Yanos Zylberberg then combined the pollution distributions with census data to investigate the persistence of social divisions relative to pollutant concentrations. The results of the research were published in 2016.

Meteorological data for dispersion modelling

Historically, meteorological data for dispersion modelling has been obtained from measurements taken at nearby met observational stations. More recently, however, the use of data from Numerical Weather Prediction (NWP) models has become increasingly common, due to improvements in resolution, accuracy and availability of this data, along with reduced availability of local observed data due to the closure of some observational stations.

CERC were commissioned by the Atmospheric Dispersion Modelling Liaison Committee (ADMLC) to examine the effects of NWP model grid resolution on dispersion modelling, in both regulatory and emergency planning contexts, and to make recommendations for the use of NWP data in dispersion modelling.

The report, published in January 2024, includes an evaluation of NWP outputs in comparison with meteorological observations at eight UK sites, a comparison of ADMS and AERMOD dispersion outcomes for idealised sources at four locations with observed and modelled meteorological data, and an investigation of the interaction between fine-scale NWP data and local flowfield modelling using the FLOWSTAR model for flow over complex terrain in ADMS.