Spatial Investigation of Legacy Pollutants within a Confirmed Cancer Cluster in Three Houston, TX Neighborhoods

Abstract

Two cancer clusters near potential contamination sites have recently been identified in Northeast Houston, Texas. A Creosote plume from the Englewood Rail Yards wood treatment processes during the early 1900s to the 1980s is one suspected cause identified by community groups. We present a case study that demonstrates how complex spatial analysis and Geographic Information Systems (GIS) can be used to identify spatial relationships between soil contaminant risk variables (total incremental lifetime cancer risk (LCR), the total concentration of polycyclic aromatic hydrocarbons (PAH), benzo(a)pyrene, naphthalene, and pyrene) and the CDC's Environmental Justice Index (EJI). The results identify spatial patterns between the distribution of PAH contaminants and areas with the highest cumulative environmental injustice impacts. The bivariate LISA High-High (HH) clusters delineate areas near the creosote plant where high PAH concentrations and increased cancer risk converge with heightened environmental burdens. Low-Low (LL) clusters, signifying areas with lesser PAH exposure and environmental strains, predominantly spanned the southeastern sectors of the study site. Furthermore, the spatial outliers yielded from bivariate LISA highlight spatial discrepancies: areas with low soil contaminant hazards surrounded by high EJI values were evident in the southwest, while regions of amplified soil contaminant risk juxtaposed against reduced EJI scores were primarily spotted in the southeastern quadrant. By discerning the nuanced distribution and potential exposure points of soil contaminants, spatial analytics enable public health professionals to craft targeted interventions, ensuring that mitigation efforts are targeted where they are most needed.

Fig. 1

Study site map of a) the Fifth Ward, Kashmere Gardens, and Denver Harbor/Port Houston, TX super neighborhoods and soil sampling locations, and the PAH risk distributions for b) total PAH, c) IELCR, d) benzo(a)pyrene, e) naphthalene, and f) pyrene

Fig. 2

Univariate LISA analysis results for the five risk variables

Fig. 3

Bivariate LISA analysis results for the EJI summation scores and the five risk variables