Identifying key parameters to differentiate groundwater flow systems using multifactorial analysis

SummaryMultivariate techniques are useful in hydrogeological studies to reduce the complexity of large-scale data sets, and provide more understandable insight into the system hydrology.In this study, principal component analysis (PCA) has been used as an exploratory method to identify the key parameters that define distinct flow systems in the Selva basin (NE Spain). In this statistical analysis, all the information obtained in hydrogeological studies (that is, hydrochemical and isotopic data, but also potentiometric data) is used. Additionally, cluster analysis, based on PCA results, allows the associations between samples to be identified, and thus, corroborates the occurrence of different groundwater fluxes.PCA and cluster analysis reveal that two main groundwater flow systems exist in the Selva basin, each with distinct hydrochemical, isotopic, and potentiometric features. Regional groundwater fluxes are associated with high F− contents, and confined aquifer layers; while local fluxes are linked to nitrate polluted unconfined aquifers with a different recharge rates.In agreement with previous hydrogeological studies, these statistical methods stand as valid screening tools to highlight the fingerprint variables that can be used as indicators to facilitate further, more arduous, analytical approaches and a feasible interpretation of the whole data set.

► Multivariate analysis detects key parameters in a range-and-basin hydrological system.
► PCA and cluster analysis link hydrogeological dynamics and pollution sources.
► F−, NO3- and isotopes stand as fingerprint variables in the study area.
► Statistical methods are demonstrated as screening tools to interpret complex data sets.