| Helga Leitner | ||
|
Helga Leitner |
||
| POSITION STATEMENT While I have worked on socio-spatial inequalities in urban areas and inequities in urban and regional development in the past, currently my research regarding issues of inequality and equity centers on questions of environmental justice and equity in US cities. Specifically, I am interested in 1) determining the extent of environmental inequity and injustice (both in terms of distributive as well as procedural justice), 2) identifying factors and conditions creating patterns of inequality and injustice, and 3) identifying ways to address environmental inequities. In order to determine whether principles of environmental justice have been violated, a large number of empirical studies have been carried out to ascertain whether minority and low-income populations are disproportionately exposed to industrial pollution. Typically, such studies of environmental justice, focusing on the inequalities in exposure to toxic hazards among sub-populations, are cast within the tradition of distributive justice, addressing equity of outcomes.As a result, such studies are often referred to as environmental equity analysis. During the past few years I have been working with Eric Sheppard and Bob McMaster at the University of Minnesota investigating theoretical and methodological issues in environmental equity research, using the Twin Cities metropolitan area as a case study. As part of this research we have identified a number of problems and limitations of environmental equity analysis, some of which are detailed below. Use of spatial data, models, techniques, software etc. Studies of environmental equity are inherently spatial in nature: Debates have to do with who lives how far from toxic hazards, and why those hazards and communities are located where they are. Thus any analysis of environmental equity or inequity requires selection of a spatial methodology that measures as precisely as possible degrees of inequity in exposure among different sub-populations. During the past decade Geographic Information Systems (GIS) software combined with various correlation and multiple regression analyses has increasingly been utilized in environmental equity analysis. The findings of these studies often conflict with one another. They range from identifying strong associations between the location of minority populations and toxic facilities to those finding no association, and results suggesting that low income communities of color are less at risk than white and higher income communities. A close examination of these studies reveals that these different results are at least in part a consequence of differences in data used and in measures of potential exposure, applied to different kinds of places, at different geographic scales (city, county, metropolitan, state, national), using data with different levels of spatial resolution (blocks, block groups, census tracts). In terms of the data used there is clearly a lack of consensus as to which variables are most meaningful for assessing environmental equity. To date much of the analysis has focused analyzing the relationship of a single variable (e.g. race, income, age) with risk of exposure. There exist, however, complex relationships among different variables, specifically between race and income, which should be part of the analysis. In terms of measures of potential exposure the most common GIS-based approaches are �spatial coincidence� and analytical buffering. The former estimates and compares the characteristics of the population in enumeration units (e.g. census tracts, block-groups) that contain environmental hazards (e.g. industrial toxic emissions, toxic waste sites) with characteristics of the population in enumeration units which do not contain such environmental hazards. In analytical buffering, GIS is used to compute circular buffers of varying radiuses around hazardous sites, or to make zonal calculations along major highways.Differences in population characteristics within and outside buffers and zones are measured in order to ascertain whether vulnerable populations are disproportionately located within buffers and zones close to hazardous sites and routes.Thus far little attention has been paid to the impact of different sizes of buffers, or to the significance of the differences identified using such GIS-based measures of environmental equity.In our research we used a geographic randomization methodology for assessing whether observed inequities are unusually high by comparison to those that might have resulted by chance. Our findings suggest that simulations of sampling distributions are necessary to make reasonable judgments about the �significance� of observed environmental equity results. Most studies also treat the simple existence of a hazardous site as a surrogate for potential exposure, ignoring important differences in the toxicity and quantity of chemicals and the spatial diffusion of toxic releases. The application of plume dispersion models within GIS is an attempt to account for these differences, by integrating the toxicological characteristics of the chemicals emitted or stored, physical characteristics of the sites, and atmospheric conditions to identify the geographic area and population likely to be affected by a plume. Such models frequently also entail simplifying assumptions, for example about average wind direction in an area and about topography, which often is assumed to be flat, which can lead to erroneous results. Plume models are also much more time consuming to apply, and it may be that under certain circumstances reasonable approximations can be gained from the use of simpler buffers. This suggests a need for sensitivity analysis to determine, for example, whether geometric approximations of analytical buffers are close enough to accurately specified physical diffusion models, at desired levels of data resolution, to act as less labor and time intensive surrogates. In spatial analysis of environmental equity, it has now become clear that the choice of the geographic scale of the study area (e.g. states, metropolitan areas, counties, municipalities) and of the spatial resolution of data within that study area (e.g. zip codes, census tracts, block-groups) influence the results of the analysis. For example, in our work at the county scale we identified a strong relationship between persons of color and TRI sites, whereas at the scale of the city of Minneapolis, we observed a stronger income-based rather than race-based pattern of inequity. This suggests the need for a discussion of what constitutes the most appropriate scale and resolution in environmental equity analysis. Another limitation of most environmental equity analyses is their restriction to measuring risk of exposure to toxic chemicals, rather than also incorporating health measures. Most environmental equity analyses are static in nature, documenting the degree of geographic association between risk of exposure and population characteristics at a particular point in time. These do not allow the analyst to draw any conclusions about factors causing the association, because patterns may reveal little about the underlying processes. For example, in the case of studies documenting a close geographical association between risk of exposure and communities of color, the community of color may have occupied the area before the noxious facility moved in, but it is equally possible that the prior presence of the facility depressed property prices, encouraging subsequent in-migration of a community of color to take advantage of affordable housing. In the former instance, a legally actionable case of environmental racism exists if it can be shown that race was a factor in the facility location decision. In the latter instance there can be no legally actionable case, although race may still be a factor as a result of discrimination in labor and housing markets. This suggests the need for historical geographical analysis to ascertain reasons behind environmental inequity. In terms of addressing environmental inequities, we have worked at the micro-scale, helping inner city neighborhood organizations explore the potential relevance of GIS-based neighborhood environmental inventories for mitigating environmental hazards at the neighborhood level. We see this as one way to promote procedural justice in the dynamics of toxic facilities. �Best practices,� �Learning Resources,� and Workshops As I have suggested above, there are numerous methodological issues that need to be addressed in order to improve environmental equity analysis. A workshop on environmental equity analysis bringing together social scientists and experts on spatial data analysis would be highly advantageous for this purpose. |
||