Comparison of contaminant-specific risk maps for an urban aquifer: Patiño aquifer case

Abstract

Most studies focused on producing contamination risk maps for aquifers usually calibrate a risk model utilizing some proxy or measure of contamination. The most common way of doing these risk maps is by calibrating some index-based model, such as DRASTIC. However, few studies focus and outline the differences between contamination risk maps that have been created for the same area but calibrated with different contaminants. Here, we present a modification of the well-known DRASTIC vulnerability model, in which anthropogenic parameters were added, and then calibrated with different contaminants (total nitrogen and total coliforms). The model incorporated anthropogenic parameters such as “land use”, “density of cesspools” and “major transport routes” as indicators of possible contamination sources. The pre- and post-calibration correlation coefficients for the total nitrogen risk maps were 0.073 and 0.522, respectively, while the pre- and post-calibration correlation coefficients for the total coliform maps were 0.357 and 0.7, respectively. After calibration and validation, the geostatistics of the maps were compared to understand how similar or different they were from each other. The comparison of the two maps showed that on average only 15% of the area are similar which implies that one risk map cannot suffice when describing the risk of contamination of a particular aquifer. This study is the first attempt of an in-depth analysis of the risk of contamination of the Patiño aquifer, where we show that 42% of the aquifer is at medium to high risk of contamination by either total nitrogen or total coliforms.