Modelling environmental equity: access to air quality in Birmingham, England

Many studies in the USA have noted inequities with regard to the socioeconomic status or racial character of communities and their relative exposure to environmental disamenities. In this paper the authors focus particularly on the environmental equity of air pollution in the English city of Birmingham. Using statistical methodologies they examine the pattern of exposure to two key air pollutants: carbon monoxide (CO) and nitrogen dioxide (NO2) across certain population groups in the city. Estimated emission levels of CO and NO2 were mapped by using modelled associations between vehicle densities and measured emissions at existing monitoring stations. These data were input to a geographical information system (GIS) for subsequent comparisons with population maps. Three types of variables were considered to distinguish possibly disadvantaged populations: age profile, ethnic make-up, and poverty indicators. From the 1991 Census, relevant statistics were derived for each enumeration district in the city. No relationship could be established on the age variable (that is, neither children nor pensioners appear to differ from the general population in their likely exposure patterns). However, there was a striking relationship between modelled emissions and poverty indicators and ethnicity. The effects are difficult to separate out but there is strong evidence to suggest that the two factors (poverty and ethnicity) operate in an independent manner. The implications of these findings, with regard to the causes of the disparities and the likely impacts of possible efforts to improve air quality in Birmingham, are discussed. DOI:10.1068/a34184 ôAuthor for correspondence. The hazardous environmental exposure often associated with environmental inequities include multiple sources of air pollution, waste treatment and disposal facilities, waterquality concerns, and the cumulative impacts associated with living in urban or rural areas. Studies to document the relationship between the geographic distribution of environmental pollution and minority populations were published in the USA during the 1970s (for example, CEQ, 1971). Environmental equity became a national issue in 1982 when attention was focused on the proposed siting of a landfill for polychlorinated biphenyls (PCBs) in a predominantly black county in North Carolina (USGA, 1983). More recently, a plethora of US-based empirical studies (including Brajer and Hall, 1992; Breen, 1993; Brooks and Sethi, 1997; Burke, 1993; Morello-Frosch et al, 2001; Perlin et al, 1995; Zimmerman, 1993) have confirmed that minority racial groups are more likely than are white people to live in areas close to toxic waste facilities or with higher than average pollutant emissions. To understand the processes by which environmental inequalities may be operating, there is a need to separate ethnic from poverty effects. Although the assumption may be made that more impoverished populations live in areas with higher levels of contamination, empirical studies have found the relationship to be inconsistent. Perlin et al (1995) in the USA and Jerrett et al (1996) in Canada reported positive relationships between income and pollutant emissions. Similarly, McLeod et al (2000) observed that densely populated urban areas in the United Kingdom with the highest air pollutant levels also tended to contain predominantly wealthy populations. Morello-Frosch et al (2001) undertook a multivariate analysis that verified the joint importance of both income and racial factors in calculated cancer risks. Zimmerman (1993) noted that a disproportionate number of inactive hazardous waste disposal sites were located in predominantly black and Hispanic neighbourhoods in the USA, but they also observed that the percentage of persons below the poverty line in these communities largely matched that of the nation as a whole. This finding suggests that racial character rather than poverty may have biased the location of the disposal sites in those communities. However, Graham et al (1999) note that there may also be historical factors to consider. They examined the population surrounding 82 coke plants or oil refineries in the USA and found a disproportionate number of poor and nonwhite residents in the most recent census. Nevertheless, Graham et al noted that the plants and refineries were built near the year 1900, when the populations in the areas were largely white, suggesting that changes in the racial character of the communities had taken place since the construction of the refineries. Relatively little research has been undertaken to examine the relationship between air quality, ethnicity, and poverty outside the USA. The US lead is, in some part, a result of long-established and lengthy legislation relevant to air pollution in that country. In the United Kingdom, since December 1997, each local authority has been charged with the task of reviewing and assessing air quality in its area. This involves measuring air pollution and trying to predict how it will change in the future. The aim of the review is to ensure that the national air quality objectives, laid out in the Air Quality Strategy published by the government in January 2000 (DETR, 2000), will be achieved. If a local authority finds any places where the objectives are not likely to be achieved, it must declare an Air Quality Management Area (AQM) there. An AQM may range in size from a few streets to an entire urban zone. The requirements for the development of AQMs set out in the Air Quality Strategy has meant that a considerable amount of work is ongoing within UK local authorities to model spatial variations in air quality across small geographical areas. By combining the output of this work with information on population demographic 696 J S Brainard, A P Jones, I J Bateman, A A Lovett, P J Fallon