Probabilistic delineation of subsurface connected pathways in alluvial aquifers under geological uncertainty

Abstract

The present work aims to address the issue of the delineation of subsurface pathways carved in sedimentary aquifers from a stochastic perspective. A probabilistic approach to recognize preferential alluvial pathways is proposed relying only upon borehole data. These data can be employed to identify locations where the surficial aquifer is characterized by highly permeable (alluvial) hydrofacies. A Monte Carlo (MC) approach is set up to simulate stochastic realizations of local alluvial soil ratios in a bi-dimensional regularly gridded domain. Alluvial pathways are defined as the collection of cells whose simulated alluvial ratios are greater or equal to a first threshold, imposed on the simulated sediments ratio, and sharing a minimum occurrence probability, equal to a second probabilistic threshold. Then, probable highly permeable subsurface pathways throughout the domain for a real case-study application, within the province of Lecco (Lombardy, Northern Italy), are appraised. Subsurface alluvial pathways delineated upon geological data show channel-like patterns, like their surface counterparts. Subsurface discharge peaks in the central portion of these pathways, where permeable sediment probability is higher than elsewhere. Moreover, subsurface alluvial pathways features are consistent with available hydrogeological and piezometric information. The present approach has been also compared with a Multiple-Points Statistics approach, proving better at capturing the spatial patterns of connected subsurface pathways. Therefore, the present approach seems promising to be employed in the preliminarily understanding of large-scale groundwater circulation patterns, in particular for poorly monitored areas or serving as further analysis background.