How the New Mexico Department of Health Uses Private Well Nitrate Data to Make Decisions

BACKGROUND: About 20% of New Mexico’s population, or 400,000 people, rely on a private well as their sole source of drinking water. Nitrate levels can occur above health advisory concentrations. At high concentrations, nitrate can pose human-health risk. Due to limited resources, data are needed to prioritize activities. Communities need to be prioritized for water quality testing and the development of wastewater infrastructure. A classification method was developed to prioritize vulnerable sub-populations that may have inadequate testing or concentrations of nitrate that are hazardous to human health.

METHODS: The New Mexico Department of Health’s Private Well Program (PWP) collects and maintains private well water quality data from a variety of partners. The PWP utilized this dataset to map 20,000 water quality records in New Mexico and aggregate data by census tracts for nitrate. Census tracts were scored based on geographic, geologic, socio-demographic, land use, and water quality data. These rankings were then classified into four contamination risk groups for nitrate: low, moderate, high and very high. A cluster analysis was then used to identify vulnerable sub-populations. Variables included in the cluster analysis were: median household income and the percent of childhood poverty, low educational attainment, renter occupied, non-white, and of homes older than 1950. A novel Bayesian profile regression approach was employed that fit the joint distribution of indicators of socio-demographic vulnerabilities to identify the final at-risk population clusters. A map of the four contaminant risk groups was overlaid onto a map of the vulnerable population clusters.

RESULTS: Of the 499 census tracts in New Mexico, 16 census tracts resulted with nitrate contamination index scores that were very high, 76 high, 117 moderate and 290 low. Six of the census tracts that ranked as very high were located within a single county. When overlaid with the vulnerable population clusters for this particular county, we identified which census tract was the highest priority for intervention. 

CONCLUSIONS:  A novel technique of scoring census tracts by factors known to affect water quality combined with a cluster analysis of indicators of population-level vulnerabilities enabled the PWP to determine where and how to focus interventions. State-wide, ten census tracts have been chosen for well water testing events for the next fiscal year based on our results. Concerning nitrate, discussions at the county-level on how best to help the people of the highest priority census tracts have begun.