Recent developments in identification of unknown contamination sources and monitoring network design for contaminated groundwater systems

In order to design an effective aquifer contamination remediation strategy two important steps are: identification of unknown groundwater pollution sources once contamination is detected in an aquifer, and efficient and effective monitoring of contaminant plume movement. James Cook University, Australia and CRC-CARE are collaborating in developing comprehensive and easy to use computer software and state of art methodologies that at can be utilized for identification of unknown pollution source magnitudes, location, and time of activity; optimal design of a contamination monitoring network that can be implemented in any contaminated groundwater site incorporating site specific information. One part of the collaborative research between James Cook University team and CRC-CARE has resulted in the development of software that enables water resources managers and engineers to solve the difficult problems of identifying sources of pollution in a contaminated groundwater systems, and to design optimal monitoring networks that can detect the extent and movement of contaminants in contaminated groundwater systems. The developed computer software make it possible for practitioners with limited knowledge of hydrogeology and pollutant transport processes to address the source identification issue. These software are expected to be immensely useful for proper management of contaminated sites with unknown sources of contamination. The capabilities of the two developed software, the contamination source identification and the monitoring network design, are briefly introduced. Figure 1 shows the architecture of the two developed software packages. There are various challenging issues that make the source characterization problem very difficult to solve. Some of the complexities arise due to uncertainties in modelling the aquifer system and the associated physical processes, due to sparsity of measurement data in typical contaminated aquifers, due to errors in field measurements, and due to difficulty in predicting the complex geochemical processes involving reactive chemical species such as those occurring in mine sites, and also due to the non-uniqueness of the aquifer response to various stresses i.e., contaminant injection. Few recent advances towards improving the available methodologies are focussed towards making the source identification methodology more versatile and valid under different real-life scenarios e.g., urban spills and abandoned mine sites where very complex reactive geochemical processes are occurring. The extension of source identification methodology to incorporate complex geochemical reactive environment, incorporation uncertainties through fuzzy quantification, incorporation of uncertainties through Fuzzy quantification, use of Fractal Singularity Index to delineate potential pollution plumes, development of new surrogate or meta models for ensuring the computational feasibility of the developed methodologies are some of the recent advances and possible future trends. The current state of development of these methodologies for more accurate and reliable contaminant source identification and the improvements achieved by implementing a designed monitoring network will be presented. (author)