Explicit K-Means Clustering Assisted Direct Model for Identification of Groundwater Pollution Sources

Abstract

A computationally inexpensive and data parsimonious direct model for identification of groundwater pollution sources that captures their location, spatial extent and flux intensities at a chosen reference time is presented in this study. The model is based upon instantaneous micro-level mass balance that is depicted by a pseudo steady state solute transport equation. Clusters of polluting locations were identified by finite differencing the solute transport equation and applying K-means clustering algorithm. The model is validated by the problem taken from Singh, Datta, and Jain (2004) and is also illustrated by applying it to an area lying between river Krishni and Hindon. Validation results showed that the model captured closely the pollution source locations; their corresponding spatial extents and flux intensities. Model captured the sources even when the concentration plumes are far away from the source locations due to advection dominated contaminant transport. Illustration results showed that the model identified the leachate flux intensity from irrigation in the form of a regional source. Further, it established a few local sources arising possibly out of localized industrial and/or municipal waste water disposal.

Keywords:

• Groundwater pollution
• source identification
• K-means clustering
• direct model

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data Availability Statement

Data available on request due to privacy/ethical restrictions: The data that support the findings of this study are available on request from the corresponding author, [initials]. The data are not publicly available due to [restrictions e.g., their containing information that could compromise the privacy of research participants]. Data utilized during the study appear in the submitted manuscript. The code utilized is proprietary and cannot be shared as the authors are in the process of acquiring the code’s copyright.