Measuring Environmental Equity with Geographical Information Systems

Environmental justice, or the equitable distribution of environmental hazards, is currently attracting more attention than perhaps any other environmental issue. Last year, the White House issued an executive order that requires federal agencies to consider the impacts of their decisions on environmental equity, and the U.S. Environmental Protection Agency has created a special office to facilitate such analyses. Both actions were motivated by concern that racial minorities and the poor may be shouldering a disproportionate share of the impacts of environmental hazards.

Indeed, racial minorities and the poor, who in many cases are one and the same people, typically do have greater exposure to environmental hazards than those who are more economically advantaged. The poor often live in areas that are likely to have more environmentally undesirable facilities—for example, factories, power plants, waste incinerators, and so on—than the areas where other groups live. And, unfortunately, these subpopulations may include a disproportionate number of young children or elderly people, two groups that are generally believed to be especially susceptible to the health effects of pollution.

Before cases of existing environmental inequity can be remedied, they must first be identified. Fortunately, a new information technology has emerged that can be used to provide real data about environmental equity impacts in any selected location. This technology—known as geographical information systems, or GIS—is a type of software that was originally developed for combining different types of spatial data, such as information about a region's topographical features and its distribution of natural resources. Using GIS requires appropriate data and expertise, but most of the data are already available, and some systems are friendly enough to be learned easily.

GIS is becoming an increasingly important technology for analyzing environmental equity. The information provided injects an essential degree of objectivity into environmental justice deliberations. In turn, this objectivity helps decision makers to establish priorities based on information about which hazards create the greatest disparities in impacts and which groups of people are most affected.

In the Center for Risk Management at Resources for the Future, we are conducting a study that demonstrates the potential of GIS to shed light on the distribution of environmental burdens. An important part of the study is a comparison of proximity-based measurements and risk-based measurements of environmental equity. Below we describe how we have used GIS to generate both kinds of measurements in an analysis of environmental equity with respect to industrial hazards in Pittsburgh and surround ing Allegheny County, Pennsylvania, circa 1990.

This analysis differs in three important respects from related analyses of environmental equity that have been done elsewhere. First, it considers not only chronic hazards in the form of air pollution from industrial facilities but also acute hazards in the form of potential exposure to accidents involving the airborne release of toxic chemicals from facilities where the chemicals are stored. Second—and more important—our analysis of equity is based not only on proximity to hazards but also on the actual health and safety risks associated with each kind of hazard, separated and combined. These dimensions of equity are absent in most related studies, which measure equity based only on people's proximity to hazardous facilities. Third, for selected facilities, our analysis will trace changes in the distribution of environmental hazards using historical data on hazards, land use, property values, demographics, and other agents or indicators of change, whether legal, political, or economic.

Our study using GIS reveals the need to look beyond aggregate results when analyzing equity and to be cautious when using worst-case assumptions. Also, somewhat surprisingly, it shows that, in the face of hazards that have the potential to affect large areas (such as major accidental chemical releases), most of those who would be exposed do not belong to the most disadvantaged groups of people—that is, racial minorities and the poor.

When our study is completed this fall, we expect it to benefit a wide audience, ranging from community groups to professional peers, by demonstrating how to assemble data germane to environmental equity, how to analyze the data using moderately priced GIS software, and how to interpret the results. We believe our approach will be particularly useful in showing how to measure "outcome inequity," that is, in determining whether one socioeconomic group bears more of the burden of a particular environmental hazard than another. In addition, the part of our study that deals with the use of historical data to examine the evolution of environmental inequities should prove useful in showing how the distribution of burdens changes with time and—perhaps—in helping understand why.

Proximity-based measurements

One way to measure environmental equity is based on people's proximity to facilities that pose environmental hazards. In our analysis of environmental equity with respect to industrial hazards in Allegheny County in 1990, we used a GIS to avoid some of the pitfalls that attend more simplistic approaches to proximity-based equity measurements.

Previous approaches hinged on a comparison of the percentage of minorities or poor people in the census areas that contain environmentally hazardous facilities with the percentage of minorities or poor people in nearby census areas that do not contain such facilities. This kind of approach is problematic for several reasons. First, it draws no distinction between areas that are home to only one facility and those that host two or more facilities. Second, it does not account for the possibility that the hazardous facility or facilities may be so close to the edge of the host area that a neighboring area is affected as much, if not more. Third, and perhaps most important, this approach does not consider that census tracts and counties do not generally represent either the affected neighborhood or the range of the hazard associated with a facility. A more sensible way to represent both is to construct an imaginary circle centered at each facility, although the question of how large the radius of the circle should be is open to question. In the case of facilities in urban areas, a radius of one or two miles seems reasonable, since neighborhoods do not usually extend any further than that.

Our environmental equity analysis accounts for all three of the above shortcomings. In this analysis, we divided Allegheny County's industrial facilities into two types: those that may pose chronic hazards and those that may pose acute hazards. We refer to the former facilities as TRI facilities after the Toxic Release Inventory (a national database of reports of industrial air pollution), from which we obtained the location of these facilities and information about emissions from each. We refer to the latter facilities as EHS facilities because they store "extremely hazardous substances"; the risks associated with these facilities arise in the event of an accidental chemical release rather than from continual, routine chemical releases. We were able to identify these facilities using the federally required reports that indicate where EHSs are stored in quantities above a certain threshold.

Next, we constructed circles with radii of one-half mile, one mile, and two miles around each TRI and EHS facility. Then, for each radius, we divided Allegheny County into two parts, one being the area formed by the circles and their overlapping portions, and the other being the rest of the county (the areas outside of all the circles). We made this division for the EHS facilities and then, separately, for the TRI facilities. In each case, the combined area within the circles, which may not all be contiguous, is what we call the "close-proximity region"—the region where people live in close proximity to the facilities. We assumed that, for a given choice of radius, the close-proximity region is homogeneous with regard to proximity effects—that is, the hazard burden is the same no matter which facility you are close to or how close you are to it, as long as you live within the region.

Using a GIS, we then calculated the proportion of nonwhite residents and poor residents inside and outside the close-proximity region of the sixty-two facilities in the county that stored large quantities of EHSs in 1990. We found that nonwhite residents made up 16 percent of the population inside the close-proximity region but only 11 percent of the population outside this region in 1990 (see figure, p. 4, top). Similarly, poor residents made up 16 percent of the population inside the close-proximity region but only 10 percent of the population outside that region in 1990 (see figure, p. 4, center). Thus the percentage of nonwhites and the percentage of the poor among people who live close to the EHS facilities are slightly higher than those elsewhere in the county.

When we calculated the percentage of nonwhite residents and poor residents inside and outside the close-proximity regions of the county's TRI facilities, we obtained similar results: the percentages of nonwhites and the poor living inside the close-proximity regions were greater than the percentages of these people living outside those regions.

Risk-based measurements

Risk-based equity measurements are superior to proximity-based equity measurements because they take into account other major factors on which risk depends—factors that can actually change the picture of environmental equity given by proximity-based measurements. These other factors include the probability of an accidental release of chemicals; the size of the area affected by such a release (which depends, in turn, on the substance released, the quantity released, the nature of the release, the release rate, and the weather at the time of the release); and the wind direction at the time of release. Risk also depends on the toxicity of the chemical released and on the level of exposure of the population of concern.

While the use of GIS to measure environmental equity is still in its infancy, we feel safe in making certain observations about this practice. Given the widespread availability of census and TRI data, as well as the increasing availability of user-friendly GIS packages, the capability to produce proximity-based estimates of industrial air pollution hazards is within the reach of many interested parties. Naturally, such estimates should not be considered the "last word" on environmental equity, since TRI data and EHS storage data are self-reported, TRI facilities are but one source of air pollution, and proximity is not a surrogate for risk. Other pollution sources and any environmental hazards that are of a nonpolluting nature or that are unrelated to health effects also can be readily subjected to a proximity-based analysis, provided that the data are available, complete, and "clean." GIS may be a new technology, but the oldest maxim in computing—"garbage in, garbage out"—still applies.

Risk-based analysis of environmental equity is another matter entirely. Such analysis is still the province of specialists, requires much more data than proximity-based equity analysis, and yields results that are more difficult to interpret. However, these obstacles will become less formidable as more research of the kind we are conducting is done, as better risk assessment software becomes available, and as risk education and communication improve in general. In the meantime, much more research is needed on how to combine risks, especially those that are difficult to measure in common units, such as carcinogenic and noncarcinogenic risks (or even health and ecological risks), and those that do not merely sum when accumulated, namely health risks that are exacerbated in the presence of certain other health risks.

In the near future, two principal benefits are likely to emerge from the use of GIS to measure environmental equity. One is the capability of concerned parties, such as public interest groups or government agencies, to use GIS as a screening tool to evaluate a region and determine which facility or facilities are contributing to the inequitable distribution of risks in the region. The other benefit is the contribution that GIS can make to the process of facility siting. Ideally, all the stakeholders in this process—whether they be industries, government agencies, or community groups—would participate in the process of identifying and evaluating the candidate locations for an undesirable facility, with the assistance of a GIS.

If, at some point in the future, an inventory of risk estimates could be developed for each region—whether it be a city or county—the facilities considered in the screening or siting process could be evaluated not only in terms of the absolute risk they pose to each population group of concern but also according to their relative contributions to the overall risk burden of each group.

One overarching policy issue should be confronted in the not-too-distant future: Is it ultimately better for all parties concerned to spread out a region's environmental hazards in order to achieve short-term equity? This would be the outcome of making immediate, piecemeal improvements in the status quo. Or is it better to concentrate these hazards in one or more "hazard zones" and effect long-term equity by reducing the associated risks and putting programs in place to enable affected residents to relocate over time? This issue goes well beyond use of GIS to measure environmental equity, although GIS could help in such a policy analysis.