Improved Trend Evaluation and Workflow Optimisation of Groundwater Monitoring Programmes - Alternate Approaches for Small Chlorinated-Solvent Datasets

Abstract

Groundwater monitoring programmes (GMP) for sites contaminated with chlorinated solvents often rely on qualitative assessments that rely on the judgement of the evalua tor. However, the datasets encountered in Sweden are often small, typically fewer than 20 wells, and present challenges such as skewed distributions, irregular sampling inter vals, and limited spatial coverage, which could reduce the reliability of traditional trend evaluation methods. This thesis investigates alternative and complementary analytical approaches that can strengthen the understanding of chlorinated solvent behaviour and movements over time in groundwater. The results demonstrate that small datasets require analytical approaches that go beyond qualitative interpretation. Statistical characterisation confirmed pronounced skewness, heteroskedasticity, and non-normality, supporting the use of non-parametric methods such as the Mann-Kendall test and Sen’s slope. Importantly, visual trend assessments showed substantial uncertainty, with only a minority of wells classified with high con fidence, underscoring the limitations of qualitative judgement alone. Site-wide plume evaluation using the Ricker Method revealed strong sensitivity to interpolation choice and well placement, while scenario testing highlighted how removing key wells can impair plume mass estimation. Although no single software tested fully meets GMP needs, GWSDAT offers strong visual and spatial functions but limited statistical depth, ProUCL provides the most robust statistics, and AquaChem is broad but time-intensive. GWSDAT ranked highest overall, but with results being sensitive to user preferences. Overall, the findings show that reliable interpretation of small chlorinated solvent datasets requires combining multiple analytical perspectives. A workflow integrating non-parametric trend testing, individual constituent evaluation, and plume mass investigation can sub stantially improve decision making regarding sampling frequency, plume stability, and long-term monitoring optimisation.