Figure 5: How the percentage of residents falling into the “Extremely Low Income” and “Above Middle Income” household income groups correlates with a neighborhood’s average PM2.5 level.
Figure 5: How the percentage of residents falling into the “Extremely Low Income” and “Above Middle Income” household income groups correlates with a neighborhood’s average PM2.5 level.
First, we visualized how the percentage of residents falling into certain income groups correlated with average PM2.5 levels. These income groups are determined by the U.S. Department of Housing and Urban Development and are based off of the median household income in Los Angeles for a family of four ($73,100 in 2019). For simplicity, we analyzed the lowest and the highest income groups: “extremely low income”, which constitutes families making less than 30% of the median income; and “above middle income”, which include families making above the median income. If a community is composed of many residents in the “extremely low income” grouping, it is likely that the neighborhood is socioeconomically disadvantaged. Like Figures 1 and 2, each neighborhood is represented as a dot, and the acceptable PM2.5 limit is indicated as a purple dashed line.
Figure 5 demonstrates that communities composed of 30%+ “extremely low income” residents consistently experience PM2.5 levels above the acceptable EPA limit. The low variance in the extremely low income data points above 30% indicates that the number of residents in this income group is a decisive factor in determining a community’s exposure to pollution. On the other hand, the relationship between the number of “above middle income” residents and pollution is opposite. Neighborhoods with few residents in this income group experience consistently high PM2.5 levels, but above 20%+, the data points begin to trend downwards. It is important to note that neighborhoods with most residents in the “above middle income” grouping have a lot of variance in PM2.5 levels, with some neighborhoods experiencing relatively low levels and others still having levels above the EPA limit; therefore, other factors other than socioeconomic status are likely playing a role in determining pollutant exposure. Nevertheless, the number of low income residents in a neighborhood is extremely important and correlates well with PM2.5 levels, with more low income residents increasing that community’s exposure to air pollutants.
Figure 6: The average PM2.5 levels measured at neighborhoods according to poverty level. Error bars represent standard deviations of the respective quartiles’ PM2.5 levels.
Another method to parameterize socioeconomic status is the percentage of residents who fall under the poverty threshold set by the U.S. Census Bureau, as illustrated in Figure 6. Unlike the previous visualizations, Figure 6 is a bar graph; this choice was because we grouped the neighborhood into four quartiles, depending on how many of their residents fell under the poverty line. The leftmost bar (lowest 25%) represents neighborhoods with the lowest levels of residents in poverty and thus include the most affluent neighborhoods. Likewise, the rightmost bar includes neighborhoods with high levels of poverty and thus less affluent communities. The standard deviation of each quartile subgroup’s PM2.5 concentrations was calculated and used as the error bars.
The most affluent neighborhoods with the lowest levels of poverty experience an average PM2.5 concentration of 11.434 𝜇g/cubic meter, which is lower than the other three subgroups. As one goes up in quartiles, the PM2.5 levels increase, with the most severe air pollution experienced by the top 25% neighborhoods in poverty. GIS analysis conducted by Jerrett et al. found similar results, as they found income and unemployment correlated well with air pollution. However, they found that property values were more significant than both income and unemployment when predicting pollutant levels, which was a variable not included in our analysis (Jarrett et al. 2001).
The error bars for each of these subgroups is relatively large, causing them to overlap with each other. Therefore, there is quite a bit of variance in PM2.5 levels even when accounting for poverty levels, indicating that other confounding variables, like weather or demographic factors, may be contributing to air pollution. Nevertheless, despite this limitation, these averages generally corroborate with Figure 5 and support the argument that less socioeconomically advantaged communities experience greater levels of PM2.5 pollution.
Figure 7: Spatial distribution of drinking water contaminants in Los Angeles.
We see that along the coastal areas and near the beaches, drinking water contaminants appear to be less prevalent, as indicated by the relatively low index. In contrast, neighborhoods in the east side of Los Angeles generally have higher indices and thus more water pollution. Since neighborhoods on the westside of Los Angeles (Beverly Hills, Westwood, Malibu) tend to be more affluent than ones downtown or close to the Inland Empire, we argue that the spatial distribution of water contamination demonstrates a potential relationship between socioeconomic factors and water quality. A report by the Safe Water Alliance, the Environmental Justice Coalition for Water, and the International Human Rights Law Clinic also observed similar results, with rural areas, tribal lands, and urban communities of color experiencing high levels of contaminants like arsenic and nitrates. For instance, Maywood, a low-income area in Los angeles, was found to have high levels of lead, mercury, and other toxic pollutants, and clean-up efforts there have been limited (Environmental Justice Coalition for Water 2014, 20).
However, it is also important to note that many affluent neighborhoods, such as Beverly Hills, are not in the lowest bracket of drinking water contaminants. Instead, they have contamination indices around 300-500, so this relationship between socioeconomic status and water pollution may have other factors confounding our analysis. For instance, drinking water infrastructure is not necessarily restricted to a specific census-tract. As a result, the act of analyzing drinking water on a census-tract level may not account for infrastructure spanning across multiple tracts, potentially complicating our findings.
Figure 8: A map of Los Angeles neighborhoods, with color representing a neighborhood’s income and the size of the dot representing the percentage of tree canopy cover.
Noting that maps convey geospatial arguments, we utilized latitude and longitude coordinates of Los Angeles neighborhoods as a platform to investigate how spatial distribution of greenspace is related to socioeconomic status. This interactive map connects one’s location, their neighborhood’s median household income, and the neighborhood’s green space urban tree canopy covering. Furthermore, it also includes the neighborhood’s water contamination index and its average PM2.5 concentration to summarize our previous analyses.
The colors, which represent a neighborhood’s income, demonstrate that many Los Angeles neighborhoods may be considered lower-income, as indicated by the large number of red circles. Since the size represents the percentage of urban tree canopy, the map argues that more affluent areas (colored in green) appear to have more green space and less air and water pollutants. Although we acknowledge that there are lots of data points (i.e. neighborhoods) missing due to filtering null values, the map still shows a correlation between socioeconomic status and greenspace and summarizes many of the trends from Figures 1-4 regarding water and air pollution.
Created with Mobirise website templates