In its different versions, the Atmospheric Dispersion Modelling System (ADMS) for calculation of pollutant concentrations is now used in the UK for the majority of studies of air pollution both for industrial areas and in urban centres. Sharon Watts, CERC, outlines ten years of ADMS, from the development of the system as a scientific concept, through early versions of the model for a single point source, to a multi-source system and general acceptance by the user community.The well known Atmospheric Dispersion Modelling System (ADMS) for calculation of pollutant concentrations in air is now available in three different packages: SCREEN for preliminary assessments, Version 3 (Industrial) for industrial releases and Urban for urban air quality. In its different versions the model is now used in the UK for the majority of studies of air pollution both for industrial areas and in urban centres. In addition, ADMS is used in many cities throughout the world and is being translated into a number of different languages. Overall there are some 500 licenses worldwide.
In the 1980s the two atmospheric dispersion models used in the UK were R-91 and ISC. R-91 is a series of algorithms, which can easily be coded, based on recommendation of the Atmospheric Dispersion Working Group of the National Radiological Protection Board (NRPB), whilst ISC became the preferred model of the US EPA and is available as public domain software. Both models were based on similar physics, the essential features being a simple parameterisation of the atmospheric boundary layer, and hence dispersion, by means of the Pasquill stability category. Such an approach implies, incorrectly, that the atmospheric boundary layer has similar dispersion properties throughout its depth.
During the 1970s and 1980s field and laboratory measurements and theoretical advances increased our understanding of the boundary layer, so that it became widely accepted that the vertical profile of both mean wind and turbulence showed significant change with height and could be described in terms of the boundary layer parameters H (height) and L (Monin Obukhov length). This allowed new advances and practical parameterisation of atmospheric dispersion and was the fundamental basis for the development of ADMS.
After an initial study, CERC, with technical assistance from National Power and the UK Meteorological Office, received sponsorship from a number of organisations, including the UK Health and Safety Executive and the forerunner of the Environment Agency. This led to ADMS version 1, a single source model based on the new parameterisations including advanced algorithms for dispersion around complex terrain and in the neighbourhood of major site buildings, and a concentration fluctuation model for calculation of probabilities that short time scale concentration peaks are exceeded - an algorithm now widely used for the assessment of odour impacts. Subsequently, major investment from CERC and sponsorship from the 11 major UK-based organisations led to the developments of ADMS 2 (multiple sources, point, line, area volume sources) and ADMS 3 (improved performance). After initial resistance to change, widespread acceptance of the system has benefited from extensive validation, showing it to significantly outperform R-91 and ISC. The development of ADMS-Urban is being widely used by local authorities as a modelling tool for the National Air Quality Strategy.