Passive Sampling with Active Carbon Fibres in the Determination of Organic Pollutants in Groundwater

Control of organic contaminants in groundwater by passive sampling and multivariate statistical analysis

Organic contaminants in groundwater are among the most challenging chemical compound contaminants today, particularly when it comes to understanding their occurrence, origin, and relations in groundwater, as well as the transport processes, fate, and environmental impacts involved. This paper presents the use of active carbon fibre (AFC) passive sampling and multivariate statistical processing of the results to predict the possible occurrence of organic compounds (OCs) in groundwater and to determine the origin of various anthropogenic activity. This study aims to deepen our knowledge on the control of OCs in groundwater by introducing a multi-analytical and multi-elemental holistic approach, using the Dravsko polje aquifer, the largest intergranular aquifer in Slovenia, as an example. The occurrence of OCs in groundwater was determined by means of ACFs and compared against the characteristics of the recharge area and the type of compounds detected. We combined hierarchical cluster analysis (HCA) and principal component analysis (PCA) to identify the relationship between different OCs in groundwater. The relationships between their occurrence, environmental setting and type of compound were determined using multiple linear regression (MLR). From the total of 343 organic compounds detected using passive sampling, 47 were included in further statistical analysis. MLR shows that the environmental setting is one of the most important factors affecting the different types of pollutants in groundwater. MLR models were calculated for different sources of pollution (agricultural, urban, and industrial) based on the environmental setting, land use, agglomeration, infrastructure networks, and hydrogeological characteristics of the aquifer. By means of HCA and PCA, we identified the relationships between different OCs in groundwater. As expected, the strongest correlations were found between primary compounds and their degradation products (e.g. atrazine and desethylatrazine) and compounds of similar use (e.g. atrazine and propazine, also desethylatrazine and propazine, atrazine and simazine). Some of them were also found to have a similar molecular structure (e.g. palmitic and stearic acid, 5-methoxygramine and 5-methoxytryptamine). The use of the same substances in different environments (agricultural/urban) makes them markers of both (different) origins. Therefore, it is particularly important to determine the combination of markers of different origin using multivariate statistical methods, especially in the case of mixed land use. This study identifies the main factors influencing the distribution of groundwater OCs and thus contributes to a more comprehensive understanding of the vulnerability of shallow groundwater to surface-derived contamination in similar environments.