A statistical approach for identification of potential pollution incidents due to lignite mining activity in a surface water stream

Abstract

The present study aims to the development of a methodology for the identification of potential pollution incidents in surface water bodies, by the utilization of statistical techniques, using monitoring data collected from a multi-parametric sensors system. Demonstration of the proposed methodology carried out in a surface water stream located within a lignite mine area in Northern Greece; an on-line water quality monitoring system using multi-parametric sensors was installed in three representative sites of the study area, and the corresponding measured parameters included pH, temperature, conductivity, nitrate ions (), sulfate ions (), and chloride ions (Cl⁻) for a period of about 10 months. This system comprised of a probe containing the measuring elements (electrodes and sensors) and additional hardware/software for data collection, processing and transfer through the internet (wireless telemetry). The data collected from the monitoring station were treated using a statistical approach, aiming to reveal potential correlations between the physicochemical parameters and the volumetric flows of water drained from the lignite mines into the surface stream, in order to identify and justify pollution incidents from lignite mining activity. The results indicated that 14 potential pollution incidents were recorded during the entire monitoring period, between May 2010 and October 2010. Most active months within the study period, for the appearance of potential pollution incidents were May, June, and July for pH, Cl⁻, and ; however conductivity was not considered as an index of potential pollution incident, since it was not correlated to significant changes in water quality. The detailed analysis of the periods where potential pollution incidents were identified revealed that similar trends in the variation of the physicochemical parameters were observed in 12 out of 14 cases, justifying the proposed statistical approach as a method for the identification of potential pollution incidents.

Keywords:

• Identification of pollution incidents
• Lignite mining activity
• Statistical technique
• Surface water stream quality
• Multi-parametric sensors system
• Ptolemaida––Greece

Acknowledgments

This research has been co-financed by the European Union (European Social Fund––ESF) and Greek national funds through the Operational Program “Education and Lifelong Learning” of the National Strategic Reference Framework (NSRF)—Research Funding Program: Heracleitus II. Investing in knowledge society through the European Social Fund.