GIS and Societal Risk to Hazardous Materials


Helga Leitner, Robert McMaster, Roger Miller, and Eric Sheppard
Department of Geography
University of Minnesota

E-mail: eqj6139@maroon.tc.umn.edu / mcmaster@atlas.socsci.umn.edu
rpmiller@maroon.tc.umn.edu / shepp001@maroon.tc.umn.edu




Introduction

Over the past twenty years, advances in digital cartography and geographic information systems (GIS) have made mapping and spatial analysis accessible not only to geographers and other spatial scientists, but also to society in general. GISs, which are computer-based methods for encoding, storing, analyzing, and representing (mapping) spatial data, now can be found in all levels of government and the private sector. Locally for instance, Hennepin County government maintains a state-of-the-art GIS for property records, criminal analysis, and parks/recreation, while Dakota Electric has developed a detailed spatial database for maintaining public utility records. Concomitantly, the application of geographic information systems to assess environmental and technological risk is increasing. One can find examples of GIS used in assessing the risk from air toxics, monitoring the quality of groundwater, analyzing the human response to earthquakes, and wildfire management planning. In most instances, while the development of a database focusing on the actual risk is adequate, or even quite detailed, the information on human activity--what may be called the basic geodemographics--is poorly developed. Accounting for societal needs must become an integral component of GIS in the domain of risk analysis.



Public Policy, Geodemographics, and Risk Assessment

Emergency Planning and Community Right-To-Know

The Emergency Planning and Community Right to Know Act of 1986 establishes requirements for Federal, State, and local governments and industry regarding emergency planning and Community Right-to-Know" reporting on hazardous and toxic chemicals. The purpose of the Community Right-to-Know provision is to help increase the public's knowledge of and access to information on the presence of hazardous chemicals in their communities and releases of these chemicals into the environment. States and communities will, with these data available, be better able to improve chemical safety and protect public health. The Emergency Planning and Community Right-to-Know Act (SARA Title III) has four components: emergency planning, emergency release notification, community Right-to-Know reporting requirements, and toxic chemical release inventory.

Emergency Planning

The emergency planning legislation is designed to promote State and local governments' emergency response and preparedness capabilities through both better coordination and planning. At the state level, The Emergency Planning and Community Right-to-Know Act requires the Governor of each state to create a State Emergency Response Commission (SERC). According to the Title III Fact Sheet, the SERCs include individuals from a variety of public agencies, including those that deal with the environment, natural resources, emergency services, public health, occupational safety, and transportation. It is also the responsibility of the SERC in each state to appoint local emergency planning districts and Local Emergency Planning Committees (LEPC) for each district. Most often, the planning districts are established at the county level although some substate planning districts have been created. Among other tasks, the SERC is responsible for establishing procedures for receiving and processing public requests for SARA Title III data. The LEPC for each district normally includes a mixture of individuals from the following organizations: elected state and local government, police, firefighters, civil defense, public health, environmental, hospitals, transportation, representatives of facilities, community groups, the media. As of October 17th, 1988, the LEPC was responsible for creating an emergency response plan.

According to the Title III Fact Sheet, this plan must include the following:

-identify facilities and transportation routes of extremely hazardous substances.
-describe emergency response procedures, on-site and off-site
-designate a community coordinator and facility coordinator to implement the plan
-outline emergency notification procedures
-describe methods for determining the occurrence of a release and the probable affected area and population
-describe community and industry emergency equipment and facilities and identify the persons responsible for them
-outline evacuation plans
-describe a training program for emergency response personnel
-present methods and schedules for exercising emergency response plans.


This emergency response plan, initially reviewed by the SERC, is also reviewed annually by the LEPC. It should be noted that the LEPCs first area of concern is with the threshold planning quantities of the 360 extremely hazardous substances published in the Federal Register. The list and threshold planning quantities are based on toxicity, reactivity, volatility, dispersability, combustibility, or flammability of the substance.

Risk analysis

We will detail a project that uses geographic information systems (GIS) to analyze technological risk in the Twin Cities area. First, we describe a risk assessment for the Twin Cities area that involves an analysis of the TRI (toxic release inventory) data, gathered under the Community Right To Know legislation, and basic geodemographic characteristics of the Twin Cities, including race, income, and housing data using GIS. This analytical stage will attempt to identify those regions of the Twin Cities most susceptible to an air-borne toxic release. It will be necessary, of course, to also identify regions of the metro area where significant percentages of lower income and minority groups reside. Such analysis will attempt to identify regions within the Twin Cities that are, according to Massey and Denton, hypersegregated , using measures such as unevenness in the distribution of the minority within the total population, residential isolation of minorities from non-minorities, the degree of clustering of minorities in a single area, the degree to which minorities live near the central business district, and the concentration of minorities into densely-populated neighborhoods. In the few existing studies, including those by McMaster (1991), and Burke (1993), strong spatial associations were discovered between the location of both minority and lower income populations and the location of hazardous materials. This condition has been labeled by some environmental racism , although the term implies a planned positioning of environmental hazards, which may not always be valid. The McMaster study, using grid- based GIS analysis and modeling, found strong correlations amongst minority (including African-American, Asian, and Hispanic) and hazardous materials sites in Santa Monica, California. Applying tract-level analysis of 1990 Los Angeles census data Burke, likewise, found associations amongst TRI (Toxic Release Inventory) sites, lower income groups, and minorities. Both studies were preliminary, and much work remains in identifying logical methods, for the identification of, and subsequent mitigation of this problem.

Included in the analysis will be day-care centers, schools, and other institutions likely to house sensitive populations. In order to account for varying meteorological conditions, our plans are also to include a Gaussian plume-dispersion model in this analysis that will provide the spatial distribution of given TRI substances, in parts per million, given specific wind speed, direction and chemical type/amount. The end result of this project will be the development of a risk model for the Twin Cities, a spatial analysis of sensitive and minority populations related to this risk, and an attempt to articulate the degree of environmental in justice that results from the storage and manufacture of hazardous materials. An ancillary component of this study will complete a geodemographic analysis at a variety of scales--block, census tract, and neighborhood--to determine the effect of resolution on risk assessment. The measures for determining the aforementioned hypersegregated regions of a city, for instance, are not consistent as one moves from the more regional tract-level resolution to the finer block-level. What effect does this have on risk correlation? Is there an optimal resolution for such urban analysis (e.g., Are block-level data simply too noisy for such studies), and what is the effect of mixing both different geometries--point, line, and areal data--and different resolutions. Such research falls within what is called the modifiable areal unit problem , or MAUP. The careful identification of such segregated populations is, of course, crucial for including societal concerns in risk studies.

Building on the results from this GIS-based risk analysis, the issue "risk perception" will be addressed in a subsequent study. All individuals carry preconceived ideas of risk to hazardous materials. Some are not concerned with purchasing a home several blocks from a TRI site; others, because of concerns about the potential toxicity, position themselves great distances from such sites. It should be noted that the same spatial decisions are made regarding other societal risks/problems, such as crime, flooding, and airport noise. We intend to build, starting at the neighborhood level, a perceived risk surface to technological hazards for the Twin Cities. How does this compare with the actual--derived in stage one-- surface? How does such a perceived surface affect spatial decision making with regard to hazardous materials? Are there cultural/ethnic differences in the perception of risk? Most importantly, how can perceived, in contrast to actual, risk be represented within a GIS (visualized)? The GIS community can not assume that an actual distribution is always the best for estimating societal behavior; most individuals make spatial decisions based on their perception of risk--the mental map as articulated in the geographical literature.

Contribution to discipline

A growing area of concern in GIS deals with its relationship with society, including issues of privacy, legal ownership of data, ethical considerations, and how alternative views of social space can be represented in what are, admittedly, almost exclusively Euclidean-based systems. Through addressing both a series of conceptual and empirical questions, we feel the research design proposed here could provide the framework for other geographical studies that address GIS and society issues (mixing regional and community levels of analysis). Thus far, most of the studies in this area have been written either from a critical social theory perspective (skeptical of the potential use of GIS) or from the GIS specialist s standpoint (overly optimistic about the potential for GIS). This study attempts to bridge these two views by looking at both the spatial data/analysis side of GIS, as well as several case studies that address the actual impact of GIS on a societal concern: the actual and potential risk to hazardous materials in the Twin Cities.



Return to

List of Position Papers     Initiative 19 Page