Multiple Data Imputation\nMethods Advance Risk Analysis\nand Treatability of Co-occurring Inorganic Chemicals in Groundwater

Accurately assessing and managing risks associated with\ninorganic\npollutants in groundwater is imperative. Historic water quality databases\nare often sparse due to rationale or financial budgets for sample\ncollection and analysis, posing challenges in evaluating exposure\nor water treatment effectiveness. We utilized and compared two advanced\nmultiple data imputation techniques, AMELIA and MICE algorithms, to\nfill gaps in sparse groundwater quality data sets. AMELIA outperformed\nMICE in handling missing values, as MICE tended to overestimate certain\nvalues, resulting in more outliers. Field data sets revealed that\n75% to 80% of samples exhibited no co-occurring regulated pollutants\nsurpassing MCL values, whereas imputed values showed only 15% to 55%\nof the samples posed no health risks. Imputed data unveiled a significant\nincrease, ranging from 2 to 5 times, in the number of sampling locations\npredicted to potentially exceed health-based limits and identified\nsamples where 2 to 6 co-occurring chemicals may occur and surpass\nhealth-based levels. Linking imputed data to sampling locations can\npinpoint potential hotspots of elevated chemical levels and guide\noptimal resource allocation for additional field sampling and chemical\nanalysis. With this approach, further analysis of complete data sets\nallows state agencies authorized to conduct groundwater monitoring,\noften with limited financial resources, to prioritize sampling locations\nand chemicals to be tested. Given existing data and time constraints,\nit is crucial to identify the most strategic use of the available\nresources to address data gaps effectively. This work establishes\na framework to enhance the beneficial impact of funding groundwater\ndata collection by reducing uncertainty in prioritizing future sampling\nlocations and chemical analyses.