Scales of Environmental Justice Essays

Scales of Environmental Justice Essays Scales of Environmental Justice Paper Scales of Environmental Justice Paper Abstract This paper examines the spatial point pattern of industrial toxic substances and the associated environmental justice implications in the San Francisco Bay Area, California, USA. Using a spatial analysis method called Ripley’s K we assess environmental justice across multiple spatial scales, and we verify and quantify the West Oakland neighborhood as an environmental justice site as designated by the US Environmental Protection Agency. Further, we integrate the ISCST3 air dispersion model with Geographic Information Systems (GIS) to identify the number of people potentially affected by a particular facility, and engage the problem of non-point sources of diesel emissions with an analysis of the street network. Introduction Environmental Justice Theory Environmental injustice has multiple meanings to different people, but can be thought of simply as occurring when a particular social group is disproportionately burdened with environmental hazards (Pellow 2000). Pellow defines environmental racism, an environmental justice issue, as the institutional rules, regulations, and policies of government or corporate decisions that deliberately target certain communities for least desirable land uses, resulting in the disproportionate exposure of toxic and hazardous waste on communities. Environmental inequality addresses structural questions that focus on social inequality of power, resources and environmental burdens. Environmental justice cases in California have focused recently on air pollution exposure from urban traffic (Houston et al. 2004), especially with regards to public schools (Morello-Frosch et al. 2002b; Pastor et al. 2002), and public policy for health risk measurements (Dunsby 2004). Additionally, environmental justice has been addressed in California’s water management (Haughton 1998), Toxic Releases Inventory (TRI; volume and location of emissions from facilities), and treatment, storage, and disposal facilities (TSDF), though the geographic focus has been primarily on Southern California and Los Angeles rather than the San Francisco Bay Area (Boer et al. 1997; Morello-Frosch et al. 2002a; Morello-Frosch et al. 2001; Sadd et al. 1999). The US Environmental Protection Agency’s (EPA) Air Toxics and Environmental Justice teams at the Region 9 Office recently focused their assessment of justice and equality on the health impacts of air toxics on a dense minority and low-income area in West Oakland, California in the San Francisco Bay Area through a sequence of events. Citizens suspected that their health was at risk from odorous releases from the many facilities in the area. Additionally, heavy diesel truck (vehicular) traffic was increasingly becoming a problem due to truck routes through their residential neighborhood (Pacific Institute and Coalition for West Oakland Revitalization 2003). The community of West Oakland mobilized (for mobilization on a transit issue, see Rodriguez 1999) and approached the US EPA for help. A working relationship does not in itself push a community to the top of the US EPA list. Two additional factors made West Oakland a top priority: an unusually high number of pollutant sources and a high density of minority and lowincome people, all of which were matters of high public awareness in the region. On this base of perception, there was then need for valid empirical demonstration of congruent clustering of emissions and of minority residential populations. The agency needed to know the numbers of pollutant sources and the population structure in West Oakland before designating this area for priority attention. We analyzed environmental justice in West Oakland across multiple scales with a point pattern analysis of spatial statistics new to environmental justice. Our objectives were to pinpoint statistically significant clusters of point source polluters and examine the surrounding demographics. Through Geographic Information Systems (GIS) we investigated non-point source pollution in addition to estimating the demographics affected by the most dangerous point-source polluter via a Gaussian plume model. The purpose of this research was to answer and quantify questions of scale in environmental justice. The underlying processes that lead to environmental injustices can be political, economic, historical, and social. Politically, this could be lack of representation or participation, lack of lobbying power, greed among politicians, NIMBYism (not in my backyard), unequal power in the legal system and inadequate laws (Cole 1992), zoning (Maantay 2001), and inadequate regulations/enforcement/permitting (Levenstein and Wooding 1998; Mank 1999; Weinberg 1999). Economic processes include suburban sprawl (Ellis et al. 2002), widening income gaps (Krugman 2002), capitalism externalities (Levenstein and Wooding 1998), and market dynamics (Been 1994). Historical processes vary for different peoples, but these might include slavery, Jim Crow laws1 , land ownership (Romm 2000), disenfranchisement, persecution, anti-immigrant laws, genocide, access to health care (US Commission on Civil Rights 2003), and immigrant work programs (Marentes 2004). Social processes include stereotypes (Bobo 2001), ra cism (Pulido 1996), language barriers, segregation, hegemony, social construction, affirmative inaction and mismatched attitudes (Blackwell et al. 2002). Critics of environmental justice have cited methodological problems (Friedman 2003; Oakes et al. 1996; Yandle and Burton 1996), alternative causal interpretations such as market dynamics of capitalism (Been 1994), and misplaced priorities such as poverty over pollution (Foreman 1998). Further, it has been argued that potentially hazardous industries would provide compensation such as jobs for minorities and that increased wealth leads to increased health benefits (Adler 1999; Simon 2000; Tiebout 1956), but these arguments neglect the scale at which wealth and health benefits relate to the detrimentally affected local people. Facilities may not employ local residents nor pay an equitable wage (Pellow 2000). Further, even if facilities are not directly polluting, accidents can occur (Bolin et al. 2000). Legislatively, environmental justice in the U.S. is addressed in a number of media. The 14th Amendment mandates equal protection under law, but intent must be shown to prove discrimination (such as clearly selective enforcement, unequal municipal services, or statements by government officials). Title VI of the Civil Rights Act prohibits discrimination by Federally funded programs, and the Supreme Court requires intent to be shown if a lawsuit is brought under Section 601, yet only disparate impact for administrative complaints under Section 602 (Mank 1999; Weinberg 1999). Disparate impact may be evaluated through five steps (Mank 1999): Identify the affected population, especially those in close proximity to the facility; Determine the demographics of the affected population through mapping technology such as GIS; Determine the universe(s) of facilities and total affected population(s), especially the cumulative pollution burden of neighboring facilities; Conduct a disparate impact analysis both by examining the racial or ethnic composition within the affected population and by comparing that composition to unaffected populations in other relevant areas; Determine the significance of the disparity through the use of standard statistical methods. We follow the steps outlined above in this research. The U.S. Environmental Protection Agency (US EPA) historically has failed to enforce Title VI because of conflicts with the agency’s primary goal to reduce pollution (Mank 1999). In 1994, President Clinton issued Executive Order 12898, which mandated all Federal agencies to address environmental justice in minority and low-income populations. Although the US EPA created an Office of Environmental Justice for guidance (U.S. Environmental Protection Agency 1992; U.S. Environmental Protection Agency 1998), the Office of Inspector General2 released a review report stating that the US EPA had not been consistently implementing the intent of the Executive Order (Office of Inspector General 2004). Among the Office of Inspector General’s findings were that the US EPA had recently deemphasized minority and low-income populations in environmental justice, and that the methods of analysis, including the use of GIS, had been inconsistent. The US EPA must follow a methodology by whi ch disparate impact can be assessed, and populations can be analyzed appropriately and consistently. Spatial Analyses of Environmental Justice In this paper we apply a point pattern statistical approach to environmental justice research that avoids pre-determined units of analysis to identify appropriate scales of analysis. A number of studies have integrated point pattern analysis into a GIS framework and have explored the value of this approach to epidemiology (Gatrell and Bailey 1996; Kingham et al. 1995) in the context of detecting clusters (Bhopal et al. 1992; Fotheringham and Zhan 1996; Gatrell et al. 1996). Our spatial point pattern analysis is based on Ripley’s K-function (Ripley 1976), which has been broadly applied in ecological spatial patterns; examples include landscape dynamics of forest disease (Kelly and Meentemeyer 2002), distribution patterns of herbs (Kenkel 1993), desert shrubs (Prentice and Werger 1985; Skarpe 1991), and tropical forest trees (Sterner et al. 1986). OBrien et al. (1999) used Ripley’s K to assess the spatial and temporal distribution of canine cancers in Michigan. Barff (198 7) analyzed the second-order point pattern of manufacturing plants in Ohio for economic and social justice. Ripley’s K examines the test statistic across various spatial scales and reveals the scale at which the pattern of events is operating most strongly. Furthermore, our analysis avoids the use of census tracts3 , which are politically defined and can change with time. Ripley’s K addresses the distributive theory of equality in questioning whether or not certain communities are burdened with a disproportionate number of facilities. Not only does the statistic provide agencies such as the US EPA with sound backing of statistical significance and a link to equality and justice theory, but it also helps guide policies at the appropriate political-spatial scale- from international, national, and regional/state to county, city, and neighborhood. Initial studies in environmental justice showed that emissions are concentrated in minority relative to predominantly white residential areas, with consequently differential health impacts among racial groups (Bullard 1994; United Church of Christ Commission on Racial Justice 1987). However, critics were quick to point out flaws in the analysis with emphasis on scale- results and conclusions can change depending on the range of space and time analyzed (Anderton et al. 1994; Friedman 2003). Maantay (2002) reported that there is a need to develop more accurate methods for determining the geographic extent of exposure and the characteristics of the affected populations, to use dispersion modeling and advanced proximity analysis, and neighborhood-scale analysis. Rhodes (2003) suggested the use of several different spatial measurement units with geographic problems and to be sensitive if or when the indications of environmental justice problems change. We implement the conclusions of thes e researchers here, with particular attention to varying spatial scale, neighborhood-level analysis and dispersion modeling. The standard spatial scale of analysis for environmental justice with GIS in the US has been the census tract level (Bowen et al. 1995; Buzzelli et al. 2003; Cutter et al. 2002; Szasz and Meuser 2000; Yandle and Burton 1996). The focus on the census tract assumes that point sources and the population are distributed uniformly throughout the census tracts, which are inconsistent in size and shape. Researchers have attempted to avoid the census tract level with proximity-based assessments of demographics within a certain radius of a given facility. The exact radius is often fitted, subjective, or arbitrary (Anderton et al. 1994), however, and should depend on air movement. Radius sizes have ranged from 0.5 km (Dunn and Kingham 1995; Dunn et al. 2001; Dunn et al. 1995), 0.8 km (Baden and Coursey 2002), and 1.6 km (Bolin et al. 2000; Glickman and Hersh 1995), to 4.0 km (Anderton et al. 1994); Pastor et al. (2004) examined a range of 0.8 km, 1.6 km, and 4 km buffers. Additionally, these a uthors employed a range of statistical tests, such as Chisquare, Cramer’s V (Cramà ©r 1999), raised incidence modeling (Diggle and Rowlingson 1994), and the Townsend index of deprivation (Townsend et al. 1988). Some authors were content with producing maps with no statistics. Scholars have debated what the unit of analysis should be, from the â€Å"community† that is based on census block groups and travel time (Taquino et al. 2002) to raster-based (Mennis 2002) or multi-scalar analysis (Williams 1999). There has also been criticism that the choice of target predetermines the evaluation of social impact. We take a different path: scale becomes a variable rather than a pre-determined measure. We seek to identify the scales at which clusters, and congruence among them, are more and less likely to exist for different analytical purposes. Methods Data The main demographic information (e.g. population density, race breakdown) was obtained from the US Census Bureau’s 2000 survey at census tract, block group, and block spatial scales. We used TRI for 1999 (to compare to the 2000 Census data) for the point pattern analysis4 . The annual TRI records the volume and location of self-reported releases from private sector and federal facilities. All industries that meet the following criteria have mandatory reporting: (a) the production facility’s primary Standard Industrial Classification is manufacturing; (b) the facility has 10 or more full-time employees; and (c) the facility manufactures or processes over 25,000 pounds of at least one of the over 600 TRI chemicals, or uses more than 10,000 pounds of at least one TRI chemical (U.S. Environmental Protection Agency 1997; U.S. General Accounting Office 1991a; U.S. General Accounting Office 1991b). Specifically, TRI chemicals in West Oakland included: chlorine, chromium, nitrate compounds, and zinc compounds, though the carcinogen acetaldehyde is perhaps the most harmful. Data are provided for each release medium, including air, water, underground injection, land, and off-site transfers. A critical methodological issue is the accuracy of the locations of point-source hazards found in US EPA data sets (e.g., Glickman and Hersh 1995; Scott et al. 1997; Stockwell et al. 1993). But, the US EPA TRI data offers the most comprehensive measures of industrially released hazardous emissions in the United States, particularly after we checked for locational validity by geocoding each TRI location to its correct street address (Bolin et al. 2000; Daniels and Friedman 1999; Krieg 1998b; Mitchell et al. 1999). Addresses used in geocoding were verified as the facility address rather than the office address. The BAAQMD health risk screening was based on the US EPA’s Industrial Source Complex Short Term (ISCST3) air dispersion model (U.S. Environmental Protection Agency 1995; U.S. Environmental Protection Agency 1999), and those risks were estimated in accordance with procedures adopted by Cal/EPA’s Office of Environmental Health Hazard Assessment (OEHHA) for the Air Toxics Hot Spots Program. The ISCST3 model has been validated successfully for a number of pollutants in a variety of locations (e.g., Elbir 2002; Kumar et al. 1999; Lorber et al. 2000; Mazzeo and Venegas 2004). Dispersion is a term that describes advection (horizontal movement) and diffusion (mixing) of gases. Air dispersion models can range from simple models that require minor computation to complex threedimensional models that require extensive data and computation, the type of which depends on the scale of the problem and the input data available. Chakraborty and Armstrong (1995) detail the methodology by which to determine demographic composition of a population affected by the release of toxic substances. They describe the Geographic Plume Analysis approach that takes a dispersion model, which outputs a dispersal â€Å"footprint†, and superimposes it on a demographic database. We follow this approach here. The data were processed in a GIS database containing air pollutant source data (U.S. Environmental Protection Agency 1997) for US EPA Region 9 (Arizona, California, Nevada and the Pacific Islands) and detailed demographic information for West Oakland from the 2000 US Census (U.S. Bureau of the Census 2000). We integrated the database with ESRI’s ArcGIS 8.1 and S-Plus 6.0. The main statistical processing of point pattern analyses relied on SPlus 6.0, and analyses and map production on the ArcMap component of ArcGIS 8.1. The data were projected into Albers Equal Area with the North American Datum 1983 to preserve area measurements. Analysis Clusters of point source polluters were identified through point pattern analysis that combined intensity distributions and Ripley’s K. Intensity is defined as the mean number of points per unit area; intensity distributions reveal first-order properties of a spatial point process and variation through space to assess the spatial dependence between points. First-order properties of a spatial point process describe how the mean number of points per unit area (the intensity) varies through space. For a stationary process, the intensity is assumed to be constant over the bounded region of interest. We initially followed a weighted edge correction (Ripley 1977), though Lancaster and Downes (2004) specify that edge correction is not necessary for length-scale assessment of clusters. Intensity distributions show where clusters are occurring, Ripley’s K reveals statistical significance of those apparent clusters. Ripley’s K is a second-order (variance of distances) function for spatial point pattern and is used to detect spatial randomness (Ripley 1976). A spatial point pattern is a collection of points irregularly located within a bounded region of space (e.g. pollutant sources within a county). The data set may consist of locations only, or it may be a marked point process, with data values associated with each location (e.g. longitude/latitude with associated emissions). The analysis is termed â€Å"second-order† because of its focus on the variance of the test statistic across a series of progressively larger areas- the size of the step is set to reveal the inter-event distances at which clustering, if present, is strongest. By examining the test statistic at various spatial scales (e.g. region, county, city), the scale at 6 which the pattern of events (points) is operating most strongly (highest statistical significance and confidence) can be determined. Ripley’s K e xamines the null hypothesis of Complete Spatial Randomness (CSR) for a mapped spatial point pattern. CSR is defined by the following criteria: a) the intensity of the point pattern does not vary over the bounded sampling region, and the pattern follows a homogeneous Poisson distribution; b) there are no interactions among the points. Ripley’s K can reject the null hypothesis that the spatial pattern of points is random. After the data were analyzed by the Ripley’s K method, a plot of count K(h) versus distance h revealed deviations as expected under CSR. The deviation was tested for statistical significance. One test employed the calculation of constant approximate confidence intervals around CSR (Getis and Franklin 1987; Szwagrzyk and Czerwczak 1993). Another test used Monte Carlo methods to determine statistical significance of the results by determining the amount of variation to be expected in sample statistics from computergenerated data (e.g., Manly 1991). In the context of spatial pattern analysis, Monte Carlo methods simulate randomly generated plots of the same dimensions of the observed plot thus creating confidence intervals from the highest and lowest values of K(h) (Haase 1995). We plot ((K(h)/Ï€) 0.5 – h) or simply (L(h) – h) against h to show the deviation of K(h) from CSR. If the deviation of the sample statistic from zero expectation is positive and above the upper limit of the confidence interval, then a clumped distribution can be assumed, while negative deviation indicates a regular pattern, otherwise the null hypothesis of CSR cannot be rejected (Haase 1995). Our GIS analysis followed the point pattern analysis with an examination of the communities and pollutant sources in the proximity of identified clusters of point source polluters. We determined race and income distributions in addition to the presence or absence of environmental goods (e.g, parks). With the BAAQMD health risk screening based on the US EPA’s ISCST3 air dispersion model, we evaluated in the West Oakland cluster the potential exposure by the facility Red Star Yeast (LaSaffre Yeast Corp.), which the US EPA determined posed the greatest health risk due to carcinogenic emissions of acetaldehyde (personal communication Grow 2001). LaSaffre Yeast, which operates in over 180 countries, ranked 8th of all SF Bay Area facilities for cancer health risks, and 2nd in Oakland for air pollution health risks (Greenaction.org 2003). Finally, we assessed mobile source pollution with an examination of the road network and travel routes within and around the neighborhood. Results First- and Second-Order Spatial Analysis An initial examination of point sources in the US EPA’s Region 9 in California shows clusters unsurprisingly in the major population centers of the San Francisco Bay Area and Los Angeles. With the question of West Oakland in mind, we first analyzed all TRI sources from 1999 in the San Francisco Bay Area cluster, which includes the counties of Alameda, Contra Costa, Marin, San Mateo and San Francisco (Figure 1a). An intensity distribution revealed multiple peaks, with the two largest clusters located in the East Bay (Figure 1b). Our Ripley’s K test found that the two major clusters were statistically significant whereby the deviation of the sample statistic from zero expectation was positive and above the confidence interval (Figure 1c). The distribution of Ripley’s K above the upper confidence interval indicates clustering, between the confidence intervals indicates random spatial pattern, and below the lower confidence interval indicates a regularly distributed pattern. The y-axis represents the deviation of the sample statistic from CSR; the units are in count numbers, but have been transformed as per Haase (1995). The x-axis represents distance (units are in degrees of longitude and latitude here), and the distance shows the extent of the clustering. Second, we examined the East Bay (Alameda County) TRI clusters specifically, again with the intensity distribution and Ripley’s K (Figures 2a, b, and c). The intensity distribution shows the evident positioning of facilities along the west side, but cannot distinguish clearly individual clusters within that swell of facilities. Fig. 1. (a) TRI (1999) facilities in the San Francisco (SF) Bay Area. (b) Intensity distribution for TRI (1999) sources in the SF Bay Area of points per grid cell in longitude by latitude. (c) Ripley’s K test for TRI (1999) facilities in the SF Bay Area. The distribution above the upper confidence interval indicates clustering at the scales of degrees. The distribution between the confidence intervals indicates a random spatial pattern. Below the lower confidence interval would indicate a regular spatial pattern. The y-axis represents the deviation of the sample statistic from CSR; the units are in transformed count numbers. The x-axis represents distance (units are in degrees). We were able to distinguish the individual clusters within that swell with the Ripley’s K plot as evidenced by the three peaks in the clustered area of the plot (Figure 2a). Fig. 2. (a) TRI (1999) facilities in Alameda County. (b) Intensity distribution for TRI (1999) facilities in Alameda County. (c) Ripley’s K test for TRI (1999) facilities in Alameda County. Because the US EPA was interested in validating the identification of West Oakland as an  environmental justice site, we examined in depth the City of Oakland cluster (Figure 3a). The intensity distribution illustrates the two clusters of facilities, but cannot determine if those peaks are statistically significant clusters (Figure 3b). The Ripley’s K plot shows that there is still statistically significant clustering occurring, though the data have become limited at this small of a spatial scale so this is the smallest scale at which we can examine with Ripley’s K (Figure 3c). Fig. 3. (a) TRI (1999) facilities in the City of Oakland. (b) Intensity distribution for TRI (1999) facilities in the City of Oakland. (c) Ripley’s K test for TRI (1999) facilities in the City of Oakland. GIS Analysis for West Oakland After determining the presence, scale, and location of clusters, the next step in an environmental justice framework is to examine the communities within the extent of those clusters. Using the US Census Bureau 2000 survey, we analyzed the block level data (a block is roughly equal to a city block) for the cluster at West Oakland. The greatest density of West Oakland residents is situated in the center and along the eastern freeway border of the area; the Fig. 5. Median household income by block group (frequency) in US Dollars. western industrial port is largely uninhabited. We created a broad race distribution map based on the racial majority for each block (Figure 4). This map shows that the majority of the West Oakland community is African American. In West Oakland, African Americans (Black) comprise 65% of the population, the rest of the population is made up of Caucasians (White; 9%), Latinos (Hispanic; 7%), Asians (9%), and â€Å"Other† (racial mixes, Native Americans/American Indians, Hawaiian; 10%). The median household income per year is roughly $20-25,000US, which is lower than that of the surrounding areas (Figure 5). There are 14 schools in West Oakland, half of which are within 600 m of a TRI facility. There are some parks (an environmental good), but the newest one is rarely used as it is adjacent to  the freeway and a Superfund site . There are no hospitals in West Oakland. Figures 6. (a) and (b) Extent of Red Star Yeast emissions based on BAAQMD’s Health Risk Screening Analysis on population density per square mile. (c) Total number of people potentially affected by Red Star Yeast emissions. We split air pollutant sources into stationary (e.g. factories) and mobile (e.g. vehicles) sources. Stationary sources are scattered throughout West Oakland, but they are not equal to each other in their relative health threat to the community. Now, from a TRI database with thousands of facilities the US EPA can focus on the facilities that pose the greatest health threats- in the case of West Oakland, the carcinogenic emissions of acetaldehyde by Red Star Yeast. We integrated the BAAQMD air modeling analysis of Red Star Yeast’s emissions into the GIS (Figures 6a and b). We overlaid the ISCST3 air dispersion model on top of the block level population layer in order to determine the number of people potentially affected by Red Star Yeast (Figure 6c). The resulting estimate of 5,628 people is derived from the maximum extent of the BAAQMD model, including the whole of those blocks cut by the model, but the number of people who live within the high-concentration areas is under 268. Furthermore, the air dispersion model is drawn in much more detail closer to the facility, whereas uncertainty leads to a generalized rough rectangle for the maximum bounds. The highest concentrations of acetaldehyde emitted from the facility effect those populations living closest to Red Star Yeast, and the associated health effects are subject to the assumptions of the model. Here, we found the estimated number of people potentially exposed to Red Star Yeast’s emissions based on the block level demographic information within the boundaries of the ISCST3 model. The mobile source in West Oakland is primarily from heavy diesel truck traffic through the community to the Port of Oakland. Chemicals in diesel pollution may cause cancer, harm the reproductive system and aggravate asthma (Kagawa 2002; Morgan et al. 1997). The Pacific Institute and Coalition for West Oakland Revitalization (2003) offered a number of ideas to alleviate these problems. With GIS we assess two of their recommendations: install traffic barriers on prohibited streets, and create a designated truck route not through the neighborhood. As communicated by the US EPA, the truck drivers indicated that there are only a few gasoline (petrol) stations near the port and those stations are unavoidably in the middle of the densely populated neighborhoods (personal communication Grow 2001). We mapped the population density, road network, port terminals, and all gasoline stations (15) within a 2.4-km Fig. 7. Road network, port terminals, gasoline stations, and an alternative driving route around the high population density. (1.5-mile) radius of the center of West Oakland along with the major terminals (Figure 7). Based on the two recommendations and Figure 7, we found an alternative driving route around the high population density. The roads currently used- which cut through the community- are crossed off with X’s, and the alternative route is highlighted. In sum, the key results here are: 1) detection of the cause of the problem- the three roads that run into the community; and 2) recommendation of a solution to the problem- an alternative route around the community. Without site familiarity, the US EPA would have little sense of road spacing and gasoline station locations. The GIS can provide a clear picture to the arguments posed by the residents and truckers on transit routes and gasoline stations. Discussion We used Ripley’s K combined with GIS to identify not only statistically significant areas of clusters, but also the scales at which those clusters exist. This research focused on narrowing down the extensive region- and statewide datasets into local neighborhoods that can be applied with local remedies. It is not unusual that a hierarchy of politics and economics exists across spatial scales for environmental justice (Simmons 2004). At the local level, new issues emerge that may not otherwise be evident at larger scales, such as the road network and transportation problem. Corrective justice, which is the notion that polluters should be punished and held responsible for cleanups and should compensate or repair communities damaged by historic pollution, can be implemented at the local level (Lazarus 1993). But, intermediate scales of clusters were also seen in, for example, the clustering of facilities in the East Bay relative to the San Francisco Bay Area as a whole. We focuse d on West Oakland not only as a directly applicable problem, but also as a means to raise and answer broader questions and purposes to be applied generally. As the spatial scale becomes smaller we necessarily lose the amount of data points to work with and the analysis and interpretation of the intensity distributions and Ripley’s K plots likewise changes. At the scale of the San Francisco Bay Area, clustering is most dominant in the East Bay, though geography influences clustering and edge effects as well (water and topography constrain potential sites at this scale that play less of a role at other sites and scales). The intensity distribution can show where clumping occurs, but the largest peaks are likely the most important clusters because the peaks are based on a scale relative to one another. At the scale of Alameda County, the intensity distribution reveals little about individual clustering, as it seems that the point sources are evenly spread along the water (large peak on the west side). The Ripley’s K plot, however, is more informative at the county scale where geography and edge effects are minimized. Three clear peaks in the plot point to three areas of clusters all within the broad cluster along the water. Analysis of one of those three peaks- the City of Oakland- shows two areas where facilities are located. The intensity distribution shows a cluster where West Oakland is defined, and another more intense peak (the facilities are clustered closer together) further south. The Ripley’s K plot is not as clear at this scale because the data have become so sparse relative to the larger scales. As with other EPA hazard data, there are recognized limitations to the TRI data. In addition to not providing human exposure information, the data are restricted to large manufacturing facilities and exclude releases from smaller firms, landfills and abandoned industrial sites, hazardous waste facilities, and power plants (Bolin et al. 2000). Pastor et al. (2004) report that it is difficult to make time series or longitudinal comparisons in TRI emission reporting due to periodic changes in the reporting requirements, but this paper focuses primarily on issues of  spatial scale. Further epidemiological work is still needed. The literature on GIS and spatial analysis for environmental justice has focused on census tracts or proximity-based assessments within variable radii from facilities. Critics have argued the need to assess environmental justice across different spatial scales; additionally, dispersion modeling and neighborhood-scale analysis has been called for (e.g., Maantay 2002). Further, no clear consensus on appropriate spatial statistics has emerged, because each statistic addresses different types of questions. We add to the environmental justice literature a method that: avoids census tracts and radii-based proximity assessments; assesses environmental justice across large and small spatial scales; integrates a welldeveloped air dispersion model with demographic data; and includes a measure of statistical significance for cluster evaluation. The Office of Inspector General criticized the US EPA’s de-emphasis of minority and low-income populations in addressing Executive Order 12898. While the criticism is certainly justified, the US EPA must grapple with changing political power and administrative changes that lead to these shifts in emphasis. The method we present here approaches the issue of environmental justice without starting with race or income, since such a starting point might lead to bias in data interpretation. In other words, our results showed that clusters of polluters are statistically pinpointed first, and then the surrounding demographics are examined next rather than the other way around. Unfortunately, minorities and low-income populations are often coupled with these clusters nonetheless, hence the rationale behind the environmental justice movement, though certainly many minorities and poor people live in clean environments. This analysis also ties into â€Å"disparate impact† as was discussed on Section 602 in Title VI of the Civil Rights Act. Disparate impact is now connected to communities living within statistically significant clusters, or a disparate number, of TRI facilities relative to other communities. The US EPA can apply these results to implement policies, but based on what notions of justice or equality? Theories of justice (Miller 1999; Rawls 1999; Wenz 1988) range from utilitarian (greatest good for greatest number), libertarian (greatest individual benefits without harm), communitarian (community over individual), and egalitarian (greatest benefit to least advantaged- maximize the minimum). Theories of equality (Sharder-Frechette 2002) include distributive justice (equal apportionment of social benefits and burdens), participative justice (equal rights to self-determination in societal decision-making), and procedural justice (equal distribution of enforcement, monitoring and othe r processes). Some theories of justice and equality simply are insufficient to base policies on. A utilitarian approach, for instance, might justify disproportionate environmental burdens if society as a whole is better off economically due to production, but Lejano et al. (2002) has already found that there is no justice in this approach, at least for air quality policies in Southern California. Procedural justice has been argued as insufficient in the environmental justice in California and the Southwestern US to advance environmental equality (Pulido 1994). Although the State of California’s EPA has emphasized participative justice in environmental justice policies (California Environmental Protection Agency 2003), this may be more difficult to apply at a federal level for the US EPA. Other theories of justice and equality need to be reinforced through policies. The egalitarian approach is lacking because West Oakland, considered as the least advantaged in terms of income, exposure to po llution and access to environmental goods, is not realizing the greatest benefits. The US EPA must implement distributive justice because these social burdens are not equally apportioned. Areas like West Oakland may be obvious areas for targeting environmental justice research due to the high amounts of pollution coupled with high minority concentrations, and the communities mobilized to solve their local injustice found there. For future inquiry, however, other communities may be less mobilized and empowered or the clustering of pollutant sources may be less obvious than those studied here. Although the spatial data analysis techniques were used to verify existing environmental justice areas, these methods can be used to identify new areas for study relatively 12 quickly and efficiently. The creation of potential environmental injustice areas may be proactively avoided if the US EPA has an efficient data monitoring strategy whereby clusters of possible polluters would trigger a statistical alarm. Certainly, not every cluster would necessarily be an environmental justice site, but this would at least provide a mechanism for the US EPA to focus their efforts for further research. Additionally, if the air dispersion model can be applied to many facilities simultaneously, then a possible aggregate impact could be assessed. It is important that the US EPA continue to use GIS and spatial data analysis to approach these issues and expedite the process to the enforcement stage. Following the completion of this analysis (but before publication), the US EPA applied our results in combination with community members, activist groups6 , and Federal, State and local agencies to apply pressure on Red Star Yeast to significantly restrict emissions or face sanctions and be shut down. Further, BAAQMD did not renew Red Star Yeast’s air emissions permit. Subsequently, Red Star Yeast announced that they would close their facility in West Oakland due to â€Å"market conditions† and â€Å"challenging California environmental conditions.† From an environmental justice standpoint for the community, the closure was a victory that came from activism, science, and involvement with government agencies. References Adler, J.H. 1999. Theres no justice in EPAs environmental justice policy. Corporate Environmental Strategy. 6:183-189. Anderton, D.L., A.B. Anderson, J.M. Oakes and M.R. Fraser 1994. Environmental Equity: The Demographics of Dumping. Demography. 31:229-248. Baden, B.M. and D.L. Coursey 2002. The locality of waste sites within the city of Chicago: a demographic, social, and economic analysis. Resource and Energy Economics. 24:53-93. Barff, R.A. 1987. Industrial clustering and the organization of production: A point pattern analysis of manufacturing in Cincinnati, Ohio. Annals of the Association of American Geographers. 77:89-103. Been, V. 1994. Locally Undesirable Land Uses in minority neighborhoods: Disproportionate siting or market dynamics? Yale Law Journal. 103:1383-1421. Bhopal, R.S., P. Diggle and B. Rowlingson 1992. Pinpointing clusters of apparently sporadic cases of Legionnaires disease. British Medical Journal. 304:1022-1027. Blackwell, A., S. Kwoh and M. Pastor 2002. A Changing America. In Searching for the Uncommon Common Ground New Dimensions of Race in America Eds. A. Blackwell, S. Kwoh and M. Pastor. Norton, New York, pp. 21-26. Bobo, L. 2001. Racial Atttitudes and Relations at the Close of the Twentieth Century. In America Becoming Racial Trends and their Consequences Eds. N. Smelser, W.J. Wilson and F. Mitchell. National Research Council, Washington, D.C., pp. 264-301. Boer, J.T., M. Pastor, J.L. Sadd and L.D. Snyder 1997. Is there environmental racism? The demographics of hazardous waste in Los Angeles County. Social Science Quarterly. 78:793-810. Bolin, B., E. Matranga, E.J. Hackett, E.K. Sadalla, K.D. Pijawka, D. Brewer and D. Sicotte 2000. Environmental equity in a sunbelt city: the spatial distribution of toxic hazards in Phoenix, Arizona. Environmental Hazards. 2:11-24. Bowen, W.M., M.J. Salling, K.E. Haynes and E.J. Cyran 1995. Toward Environmental Justice: Spatial equity in Ohio and Cleveland. Annals of the Association of American Geographers. 85:641- 663. Bullard, R. 1994. Environmental Justice for all: Its the right thing to do. Journal of Environmental Law and Litigation. 281 Buzzelli, M., M. Jerrett, R. Burnett and N. Finklestein 2003. Spatiotemporal perspectives on air pollution and environmental justice in Hamilton, Canada, 1985- 1996. Annals of the Association of American Geographers. 93:557-573. California Environmental Protection Agency 2003. Final Report of the Advisory Committee on Environmental Justice to the Cal/EPA Interagency Working Group on Environmental Justice. California Environmental Protection Agency, Sacramento, CA, pp. 13-35. Chakraborty, J. and M.P. Armstrong 1995. Using geographic plume analysis to assess community vulnerability to hazardous accidents. Computers, Environment and Urban Systems. 19:341-356. Cole, L. 1992. Empowerment as the key to environmental protection: The need for environmental poverty law. Ecology Law Quarterly. 19:634-683. Cramà ©r, H. 1999. Mathematical Methods of Statistics. Princeton University Press. Cutter, S.L., M.S. Scott and A.A. Hill 2002. Spatial variability in toxicity indicators used to rank chemical risks. American Journal of Public Health. 92:420-421. Daniels, G. and S. Friedman 1999. Spatial inequity and the distribution of industrial toxic releases: evidence for the 1990 TRI. Social Science Quarterly. 80:244-262. Diggle, P. and B. Rowlingson 1994. A conditional approach to point process modelling of elevated risk. Journal of the Royal Statistical Society. Series A 157:433-440. Dunn, C.E. and S.P. Kingham 1995. Modelling air quality and the effects on health in a GIS framework. In Innovations in GIS 3 Ed. D. Parker. Taylor Francis, London, pp. 205-213. Dunn, C.E., S.P. Kingham, B. Rowlingson, R.S. Bhopal, S. Cockings, C.J.W. Foy, S.D. Acquilla, J. Halpin, P. Diggle and D. Walker 2001. Analysing spatially referenced public health data: a comparison of three methodological approaches. Health Place. 7:1-12. Dunn, C.E., J. Woodhouse, R.S. Bhopal and S.D. Acquilla 1995. Asthma and factory emissions in northern England: addressing public concern by combining geographical and epidemiological methods. Journal of Epidemiology and Community Health. 49:395-400. Dunsby, J. 2004. Measuring environmental health risks: The negotiation of a public right-to-know law. Science, Technology Human Values. 29:269- 290. Elbir, T. 2002. Application of an ISCST3 model for predicting urban air pollution in the Izmir metropolitan area. International Journal of Environment and Pollution. 18:498-507. Ellis, J., C. Mason, B. Shamasunder and C. Garzà ³n 2002. Brownfields Neighborhood Revitalization. In Second National People of Color Environmental Leadership Summit Summit II. Urban Habitat, Oakland, CA, pp. 1-12. Foreman, C. 1998. The Promise and Peril of Environmental Justice. Brookings Institution Press, Washington, D.C. 1-33 pp. Fotheringham, A.S. and F. Zhan 1996. A comparison of three exploratory methods for cluster detection in point patterns. Geographical Analysis. 28:200-218. Friedman, D. 2003. The Environmental Racism Hoax. American Enterprise Institute. Gatrell, A.C. and T.C. Bailey 1996. Interactive spatial data analysis in medical geography. Social Science and Medicine. 42:843-855. Gatrell, A.C., T.C. Bailey, P.J. Diggle and B.S. Rowlingson 1996. Spatial point pattern analysis and its application in geographical epidemiology. Transactions of the Institute of British Geographers. 21:256-274. Getis, A. and J. Franklin 1987. Second-order neighborhood analysis of mapped point patterns. Ecology. 68:473-477. Glickman, T. and R. Hersh 1995. Evaluating environmental equity: the impacts of industrial hazards on selected social groups in Allegheny County Pennsylvania. Resources for the Future, Washington, DC. Greenaction.org 2003.  Action Alert. Greenaction.org. Grow, R. 2001. Air Grants and Program Integration Office, San Francisco, CA. Haase, P. 1995. Spatial pattern analysis in ecology based on Ripleys K-function: Introduction and methods of edge correction. Journal of Vegetation Science. 6:575-582. Haughton, G. 1998. Geographical equity and regional resource management: water management in southern California. Environment and Planning B: Planning Design. 25:279-298. Houston, D., J. Wu, P. Ong and A. Winer 2004. Structural disparities of urban traffic in Southern California: Implications for vehicle-related air pollution exposure in minority and high-poverty neighborhoods. Journal of Urban Affairs. 26:565- 592. Kagawa, J. 2002. Health effects of diesel exhaust emissions a mixture of air pollutants of worldwide concern. Toxicology. 181:349-353. Kelly, M. and R.K. Meentemeyer 2002. Landscape dynamics of the spread of sudden oak death. Photogrammetric Engineering and Remote Sensing. 68:1001-1009. Kenkel, N.C. 1993. Modeling Markovian dependence in populations of Aralia nudicaulis. Ecology. 74:1700-1706. Kingham, S.P., A.C. Gatrell and B. Rowlingson 1995. Testing for clustering of health events within a Geographical Information System framework. Environment and Planning A. 27:809-821. Krieg, E. 1998b. Methodological considerations in the study of toxic waste hazards. The Social Science Journal. 35:191-201. Krugman, P. 2002. For Richer. In The New York Times, Sunday Magazine, p. 62. Kumar, A., N.K. Bellam and A. Sud 1999. Performance of an industrial source complex model: Predicting long-term concentrations in an urban area. Environmental Progress. 18:93-100. Lancaster, J. and B.J. Downes 2004. Spatial point pattern analysis of available and exploited resources. Ecography. 27:94-102. Rodriguez, J.A. 1999. Rapid transit and community power: West Oakland residents confront BART. Antipode. 31:212-228. Romm, J. 2000. The Coincidental Order of Environmental Justice. In Justice and Natural Resources Eds. K.M. Mutz, G.C. Bryner and D.S. Kenney. Island Press, Washington, DC. Sadd, J.L., M. Pastor, J.T. Boer and L.D. Snyder 1999. Every breath you take : The demograhics of toxic releases in southern California. Economic Development Quarterly. 13:107-123. Scott, M., S. Cutter, C. Menzel, M. Ji and D. Wagner 1997. Spatial accuracy of the EPAs environmental hazards databases and their use in environmental equity analysis. Applied Geographic Studies. 1:45-61. Sharder-Frechette, K. 2002. Distributive Justice, Participative Justice, and the Principle of Prima Facie Political Equality. In Environmental Justice: Creating Equality, Reclaiming Democracy. Oxford University Press, Oxford, pp. 23-34. Simmons, C.S. 2004. The political economy of land conflict in the Eastern Brazilian Amazon. Annals of the Association of American Geographers. 94:183- 206. Simon, T.W. 2000. In defense of risk assessment: A reply to the environmental justice movements critique. Human and Ecological Risk Assessment. 6:555- 560. Skarpe, C. 1991. Spatial patterns and dynamics of woody vegetation in an arid savanna. Journal of Vegetation Science. 2:565-572. Sterner, R.W., C.A. Ribic and G.E. Schatz 1986. Testing for life historical changes in spatial patterns of four tropical tree species. Journal of Ecology. 74:621- 633. Stockwell, J., J. Sorensen, J. Eckert and E. Carreras 1993. The US EPA geographic information system for mapping environmental releases of toxic release inventory (TRI) chemicals. Risk Analysis. 13:155-164. Szasz, A. and M. Meuser 2000. Unintended, inexorable: The production of environmental inequalities in Santa Clara County, California. American Behavioral Scientist. 43:602-632. Szwagrzyk, J. and M. Czerwczak 1993. Spatial patterns of trees in natural forests of East-Central Europe. Journal of Vegetation Science. 4:469-476. Taquino, M., D. Parisi and D.A. Gill 2002. Units of analysis and the Environmental Justice hypothesis: The case of industrial hog farms. Social Science Quarterly. 83:298-316. Tiebout, C.M. 1956. A pure theory of local expenditures. Journal of Political Economy. 64:416-424. Townsend, P., P. Phillimore and A. Beattie 1988. Health and deprivation: Inequality and the North, Routledge, London. U.S. Bureau of the Census 2000. Statistical Abstract of the United States: 2000. U.S. Department of Commerce, Washington, DC. U.S. Environmental Protection Agency 1992. Environmental equity reducing risk for all communities. U.S. Environmental Protection Agency, Washington, DC. U.S. Environmental Protection Agency 1995. Draft users guide for the industrial source complex (ISC3) dispersion models (Revised). User Instructions, Vol. I. Office of Air Quality Planning and Standards, Research Triangle Park, NC. U.S. Environmental Protection Agency 1997. Toxic release inventory relative risk-based environmental indicators methodology. U.S. Environmental Protection Agency, Washington, DC. U.S. Environmental Protection Agency 1998. Interim Guidance for Investigating Title VI Administrative Complaints Challenging Permits. U.S. Environmental Protection Agency, Washington, DC. U.S. Environmental Protection Agency 1999. Appendum: users guide for the industrial source complex (ISC3) dispersion models, User Instructions. Office of Air Quality Planning and Standards, Research Triangle Park, NC. U.S. General Accounting Office 1991a. Toxic chemicals: EPAs toxic release inventory is useful but can be improved. Government Printing Office, Washington, DC, pp. 19-121. U.S. General Accounting Office 1991b. Toxic chemicals: EPAs toxic release inventory program: more comprehensive data and better public outreach needed. Government Printing Office, Washington, DC, pp. 91-75. United Church of Christ Commission on Racial Justice 1987. Toxic Wastes and Race in the United States. United Church of Christ, New York, NY. US Commission on Civil Rights 2003. A Quiet Crisis: Federal Funding and Unmet Needs in Indian Country. US Commission on Civil Rights, Washington, D.C., pp. 1-139. Weinberg, P. 1999. Equal Protection. In The Law of Environmental Justice Ed. M. Gerrard. American Bar Association, pp. 3-22. Wenz, P. 1988. Principles of Justice. In Environmental Justice Ed. P. Wenz. State University Press of New York, New York, pp. 22-36. Williams, R.W. 1999. Environmental injustice in America and its politics of scale. Political Geography. 18:49-73. Yandle, T. and D. Burton 1996. Reexamining environmental justice: a statistical analysis of historical hazardous waste landfill siting in metropolitan Texas. Social Science Quarterly. 77:477-492.

How To Choose Building Plans - 10 Steps to Your Dream Home

How To Choose Building Plans - 10 Steps to Your Dream Home Whether you are building a new house or remodeling an older home, youll need plans to guide you through the project. Here are some tips to help you choose the best building plans for your needs. How to Choose the Right Building Plan Create a Spreadsheet of Needs. Talk with your family. Discuss what each of you wants. What are your needs now and what will your familys needs be in the future? Should you plan for future aging in place? Write it down.Observe. Look at how you live and where you spend most of your time in your house or apartment. Why spend the time and money to build or remodel? If its just because you like change, maybe no building plan will satisfy.Reflect on homes you have visited. What features did you especially enjoy? Look at the way other people live. Is that lifestyle really what you want?Consider the features of your land. Where is the sunlight best? Which direction offers the greatest views and the cooling breezes? Could remodeling capture a piece of nature overlooked by builders of another time?Select exterior finishing details with care. Know if youll be building in an historic district, which may restrict exterior modifications.Browse through building plan catalogs for ideas. You dont hav e to buy stock plans, but these books can help you visualize possibilities. Public libraries may have these popular books on their shelves. Use the webs search function offered by online directories of building plans. Houses from sites like Houseplans.com have often been designed as custom homes before being offered as stock plans. Some plans are specs (speculative) and many are often more interesting than plain vanilla catalog plans.Select a floor plan that most closely matches your ideal. Do you need adaptability? Perhaps you should consider a house without walls. Pritzker Prize-winning architect Shigeru Ban designed Naked House (2000) with movable interior modules - a unique solution that you wont find in a house plan catalog.Estimate your building costs. Your budget will determine many choices you make in the design of your home.Consider hiring an architect to personalize your building plan, or to create a custom design. What Comes First, the House or the Site? Architect William J. Hirsch, Jr. writes, Its a good idea to have a basic concept of what kind of house you want before selecting a site because the type of house will dictate to some extent the nature of the site that makes the most sense for you. Likewise, if you have your heart set on the land first, the house design should fit the site. It might take four months to build a house, but the planning could take years. Additional Tips Choose your floor plan first and your exterior faà §ade second. Most plans can be finished in nearly any architectural style.Its usually best to purchase your land before you select your building plan. The land establishes the amount of area and the type of terrain you have to build on. To build an energy-efficient structure, try to follow the sun as it crosses your lot. Pre-purchasing the land also helps you budget the rest of your project.Be sure to budget for landscaping and finishing touches.Listen actively. Reflect back what you hear when you talk with family members. You might be surprised to find out that your children or in-laws plan to live with you. Do You Have the Confidence? Jack Nicklaus (b. 1940) has been called the greatest professional golfer of all time. So, what does he know about design? Plenty. Nicklaus is said to have had an interesting strategy when he played profession sports - he competed against the golf course instead of other players. Nicklaus knew the ins and outs of all the courses he played - he figured out what he liked and what he didnt like about golf course design. And then, he formed a company. Nicklaus Design promotes itself as the worlds leading design firm. You have lived in the spaces chosen by your parents. Now its your turn to decide. Source Hirsch, William J. Designing Your Perfect House: Lessons from an Architect. Dalsimer Press, 2008, p. 121