Air Monitoring Stations Near Los Angeles Highways
This GIS mapping project explores the policy problem of particulate matter exposure around major highways in Los Angeles County. Children are at an increased risk of harm from particulate matter exposure. Recognizing this concern, in 2003 the California State Legislature enacted a law that no new schools be built less than 500 feet from very busy roadways. However, there are currently 41 schools in Los Angeles County that are less than 500 feet (or approximately 275 meters) away from a major highway, and these schools are predominately in lower income census tracts. The elementary school data came from the UCLA Mapshare Los Angeles Schools shapefile, where I selected only elementary schools to include in my study.
I mapped these 41 schools as a part of my midterm project, demonstrated in the image above.
Air Monitoring Stations and Distance from Elementary Schools
An important impetus for change regarding particulate matter exposure for these children is to ensure accurate measurements of air quality take place in order to set and enforce air quality standards. Currently, there are 19 air monitoring stations in Los Angeles County, which are mapped above, alongside a map of the 41 elementary schools that are 150 meters or closer to a major highway. The air monitoring stations for the county come from original data I collected from the South Coast Air Quality Management District website. I input the address data for each of the air monitoring stations, and geocoded the addresses in order to map the locations.
The above map demonstrates the distance of the closest air monitoring station to each of the 41 elementary schools that are within 150 meters of a major highway. To map these distances I used the network analysis tool of closest facility, and was able to determine which air monitoring station was closest to each of the 41 elementary schools.
As this chart demonstrates, 16 schools are 4,000 meters or less from an air monitoring station, 8 schools are between 4,000 meters and 6,000 meters from an air monitoring station, and 17 schools are more than 6,000 meters away from an air monitoring station. These schools are all within 150 meters of a major highway, so the schools also serve as indicators of how close air monitoring stations are to major highways.
This map lays out the various air monitoring stations within Los Angeles County along with the major LA County highways. The ‘major highways’ were derived from the LA County Highway shapefile from the UCLA Mapshare through selection by attributes (where FCC=A15). The highways are mapped with 150 meter buffers, and the air monitoring stations are mapped with 275 meter buffers. These buffers were created through the ArcGIS buffer tool. The Natural Resources Defense Council has sued the South Coast Air Quality Management District for not accurately measuring air quality because the air monitoring stations are too far away from highways to measure the dense particulate matter exposure near highways. NRDC believes air monitoring stations should be around 900 feet (approximately 275 meters) from a major highway, thus my decision to map a 275 meter buffer around each of the air monitoring stations.
By extracting information from my buffer, I discovered there is only one air monitoring station in LA County that is within 275 meters of a major highway. This station is mapped in the layout above. This reveals there is a need for air monitoring stations in LA County closer to highways, which have higher levels of particulate matter.
Variables to Determine Ideal Locations for New Air Monitoring Stations
There are three variables I used to determine what areas new air monitoring stations should be located. The first variable is the distance from elementary schools that are within the 150 meter buffer of a major highway. It is more desirable for new monitoring stations to be close to these schools. The second variable is the distance of new air monitoring stations from a major highway. In this instance, it is more desirable for new monitoring stations to be close to highways. The third variable is the distance from existing air monitoring stations. In this instance, is it more desirable for new monitoring stations to be further away from other air monitoring stations to maximize efficiency.
The next step in determining where new air monitoring stations would be ideally located is a hot spot analysis. A hot spot analysis provides information on where there are clusters of my three values spatially. The first step was to rasterize my three variables: distance from elementary schools that are within 150 meters of a major highway, distance from a major highway, and distance from existing air monitoring stations. After rasterization, I reclassified my variables to be broken down into numeric categories 1 through 5, where 1 is the most desirable scenario and 5 is the least desirable scenario. I used the modeling tool to reclassify:
Reclassification of Variables
The above map demonstrates the reclassification of the distance from elementary schools that are within 150 meters of a major highway. The areas with the lightest color green are the most desirable locations for new air monitoring stations depending on their distance from the aforementioned elementary schools. The darker the gradation of green, the less desirable the location is for new air monitoring stations in terms of this particular variable.
This map is the reclassification of the distance from major highways variable. The lightest color, which represents 1, is the most desirable location for new air monitoring stations if this is the only variable to consider, and the darker the gradation, the less desirable the location. As is evident, the most desirable locations surround the major highways.
This map is the reclassification of the distance from existing air monitoring stations. Again, 1 is the most desirable location for a new monitoring station if distance from existing air monitoring stations is the only variable to consider. However, in this instance the most desirable regions are on the outskirts of the county, because the majority of existing air monitoring stations are clustered more toward the center of the county.
This map compares all three previous maps side by side. However, none of the variables alone really creates the analysis necessary to determine where new air monitoring stations should be located. For example, if distance from existing air monitoring stations was the only factor, a new air monitoring station would be built in the northeast corner of the county, which is far away from the targeted population of children in schools near major highways. Therefore, an index is necessary of all three variables.
Index of the Variables
After rasterizing and reclassifying each variable as was demonstrated in the previous maps, I created an index of the three variables in order to conduct a hot spot analysis of the areas where new air monitoring stations would be most beneficial. Because I do not believe the three variables are equal, I gave weights to each. The distance from a major highway is the most predominant variable, I believe, so I gave it a weight of 5. While schools are the driving variable, I am most worried about schools by highways, and making distance from highway the most heavily weighted variable will also encompass other important populations beyond schools, including low income populations that live near the major highways. I gave the distance from elementary schools that are within 150 meters of a major freeway a weight of 3, because I find it to be the second most important factor being that these schools are the target population. I gave distance from existing air monitoring stations a weight of 2, because I believe that it is important in terms of efficiency, but my greatest concern is equity, which is why it is given the lowest weight, but a weight greater than 1. I then used the Map Algebra Raster Calculator to implement my weights using the equation: [(“h_reclass”*5)+ (“s_reclass”*3) + (“a_reclass”*2)]/3.
This map shows the result of the index. This map does not include highways, schools, or air monitoring stations to get a clearer view of the best places for new air monitoring stations. Again, the areas with the lightest color are the most ideal locations for new air monitoring stations, and the desirability reduces as the gradation becomes darker. Because I gave weights to my variables, the scale is no longer from 1-5, but it represents the same five classifications of data.
Adding back in the schools, air monitoring stations, and highways to the layout, my suggestion would be to place the air monitoring station(s) in one of the more congested areas where there are large groupings of schools, and a conversion of the lightest color from the hot spot analysis. The particular area I chose to highlight also has a conversion of major highways.
Conclusions and Policy Implications
GIS provided an effective platform to conduct an analysis determining the locations in which new air monitoring stations would ideally be located. GIS also allowed for the visual representation of air monitoring stations and revealed the large gaps in air quality monitoring. These gaps likely originate from budgetary issues. However, due to whatever reason from which the lack of air monitoring stations stems, the extent of the air quality problems that face those who live, work, and go to school near major highways is largely unrecorded. Mandating air quality monitoring stations to be near highways protects the children who are at the greatest risk of exposure to particulate matter, because these children not only go to school near major highways, but likely live near a major highway as well.
 Beate Ritz, MD, PhD, and Michelle Wilhelm, PhD. “Air Pollution Impacts on Infants and Children.” Accessed February 1, 2012. http://www.environment.ucla.edu/reportcard/article.asp?parentid=1700
 Southern California Particle Center and Supersite. Accessed, February 1, 2012. http://www.scpcs.ucla.edu/news/Freeway.pdf
 “Annual Air Quality Monitoring Network Plan.” South Coast Air Quality Management District. http://www.aqmd.gov/tao/AQ-Reports/AQMonitoringNetworkPlan/AQnetworkplan.htm
 “Lawsuit Seeks Justice for 1.2 Million Residents Living Near SoCal Freeways: Lack of Air Monitors Fails to Capture Full Extent of Preventable Pollution.” Natural Resources Defense Council. Accessed March 10, 2012. http://www.nrdc.org/media/2012/120103.asp.
“Lawsuit Seeks Justice for 1.2 Million Residents Living Near SoCal Freeways: Lack of Air Monitors Fails to Capture Full Extent of Preventable Pollution.” Natural Resources Defense Council. Accessed March 10, 2012. http://www.nrdc.org/media/2012/120103.asp.
 “How Hot Spot Analysis: Getis-Ord Gi (Spatial Statistics) Works,” ArcGis Desktop 9.3 Help. January 2009. Accessed March 14, 2012. http://webhelp.esri.com/ArcGISdesktop/9.3/index.cfm?TopicName=How%20Hot%20Spot%20Analysis:%20Getis-Ord%20Gi*%20(Spatial%20Statistics)%20works