Utilising CoDA methods for the spatio-temporal geochemical characterisation of groundwater; a case study from Lisheen Mine, south central Ireland

Lisheen Mine in County Tipperary, Ireland exploited an underground Pb/Zn massive sulphide deposit hosted in Carboniferous (Mississippian) carbonates. During the extraction phase, the mine workings (located at an average depth of 170 m below ground level), were continuously pumped to lower the groundwater level. Following mine closure in 2015, pumping ceased and eight groundwater wells in the surrounding area were sampled monthly over an 11-month period to monitor the effects of groundwater rebound. These wells draw water from the upper 30 m of a limestone/dolostone aquifer and the monthly samples were analysed for the concentration of 31 elements and compounds (SO4, Cl, NO3, F, NH4, NO2, P, Ca, Na, K, Mg, Fe, Mn, Cu, Zn, Pb, Al, Ni, Ba, As, Hg, B, Cr, Cd, Mo, Ag, Co, Sr, Be, Sb and U). All of the water can be described as Ca-HCO3 type as expected. Standard methods for analysing groundwater geochemistry data (e.g. piper diagrams etc.) are useful, differentiating groundwaters with first-order contrasting chemical signatures, for example, distinguishing Ca-HCO3-type from Na-HCO3-type water. Samples from the 8 monitoring wells appear to be broadly similar, using this approach. However, these major ion methods fail to further distinguish between different groundwaters. The use of multivariate statistical analytical techniques has become more common in groundwater studies in recent years, allowing the interaction of all the elements and compounds to be considered simultaneously. Compositional Data Analysis (CoDA) was used on the Lisheen dataset to gain a better understanding of the spatial and temporal variation in groundwater geochemistry. Ilr-ion plotting, Hierarchical Cluster Analysis (HCA) and Principal Component Analysis (PCA) through CoDA highlights the elements and compounds that account for the majority of the variance and at Lisheen these are nitrate, manganese, ammonium, sulphate and potassium. By displaying these data visually in a CoDA bi-plot, each location can be reliably 'geochemically fingerprinted' despite similar concentrations of major ions within a relatively small geographical area (<30 km(2)). Relabeling the bi-plot observations by date of recovery reveals how one particular groundwater well (PH) subtly varies over time, most likely as a result of seasonal land-use changes (input of compounds associated with fertiliser). This type of statistical analysis has broad applications in hydrology and hydrogeology including contaminant tracing and interaction, environmental studies, land-use planning and mineral exploration.

Automated summary

The Lisheen Mine in County Tipperary, Ireland experienced groundwater rebound following closure in 2015. By utilizing multivariate statistical analytical techniques, the spatial and temporal variation in groundwater geochemistry can be better understood, allowing for accurate geochemical fingerprinting of groundwater.