Date of Award


Document Type


Degree Name

Master of Science (MS)



First Advisor

Nicolas Perdrial


Urban soils around the world have been found to possess elevated concentrations of toxic trace metals such as As, Cd, Cu, Pb, Mn, Hg, Zn known to pose human health risks. Tightening environmental legislation and further elucidation of the detrimental health impacts from trace metals has necessitated more efficient means of contamination assessment, as well as greater public awareness. Within this thesis, I sought to develop an array of tools to holistically approach the socially relevant environmental challenges derived from heavy metal soil contamination. These tools consist in providing means to simplify Pb, Zn and Cu analysis in-situ, develop strategies to increase participatory sampling and outreach, and characterize Pb contamination in NE US cities through GIS.To improve pXRF accuracy and precision for metals in soils, it is necessary to produce measurement corrections as a function of affecting variables (moisture, organic matter content and grain size heterogeneity). Urban forest soil samples were subjected to pXRF measurement of Pb, Cu and Zn under artificially increasing soil moisture, organic matter, and particle size heterogeneity for correction development. A correction equation was successfully obtained for moisture effects but was not feasible for organic matter and particle size heterogeneity trials, highlighting the difficulty to accurately determine contamination in-situ for all metals. Application of the soil moisture correction equation on 120 surface soils proved successful at minimizing the effects of moisture on measured Pb, Cu, and Zn concentrations. However, similar performance to a simple dilution-based correction equation suggested that empirical correction may not be necessary. To generate a comprehensive dataset on lead distribution within the Burlington (VT) area while simultaneously empowering at-risk communities on lead contamination, I carried out a community science project based on a novel educational/outreach partnership project model. This program was designed to recruit high school students as community scientists to sample soil and water from their homes for analysis at UVM. The community science project successfully incorporated a diverse group of young community scientists into a project important to their community’s health. This also enabled mass sampling in areas of concern, and we have identified 19 properties (out of 228) with soil Pb concentrations above the EPA safe level. Remote implementation, necessitated by Covid-19, resulted in easily transferable project content organized into a project website for easy dissemination and reproduction. To properly identify soil contaminated areas, as well as understanding key distribution factors, spatial prediction of trace metals is an important tool. Utilizing surface soil samples collected in a gridded fashion from three New England cities, I performed areal kriging to predict the distribution of soil Pb as well as identify effective cofactors. Despite the highly variable concentrations typical of soil Pb, areal kriging provided a means to minimize the effects of small-scale Pb distribution heterogeneity. Incorporation of structure age summarized to the census block level provided slight improvements in model accuracy and minimized underestimation of Pb concentrations. The results of these studies have demonstrated that our ability to address trace metal contamination may be improved upon through further development of identification and education methodologies. Soil contamination is a strong environmental justice challenge that deserves greater attention and my thesis developed promising tools to provide affordable and accurate soil analysis, empower affected communities and incorporate social variables into contamination assessment.



Number of Pages

109 p.