Mapping the spatial transferability of knowledge-guided machine learning: Application to the prediction of drain flow fraction.

Machine learning (ML) methods continue to gain traction in hydrological sciences for predicting variables at large scales. Yet, the spatial transferability of these ML methods remains a critical yet underexamined aspect. We present a metamodel approach to obtain large-scale estimates of drain fraction at 10 m spatial resolution, using a ML algorithm (Gradient Boost Decision Tree).

Using Expert Participation to Evaluate the Accuracy of Hand-Drawn Water-Table Maps.

Water-table maps are fundamental to hydrogeological studies and a manual, hand-drawn method is still commonly used to produce them. Despite this, the accuracy and variability of such maps have received little attention in international literature. In a unique experiment, 63 groundwater professionals drew water-table equipotential contours based on the same dataset of point measurements and were asked to infer flow directions and predict groundwater elevations at predefined locations.

Including Vertical Fault Structures in Layered Groundwater Flow Models.

Accurate representation of groundwater flow and solute transport requires a sound representation of the underlying geometry of aquifers. Faults can have a significant influence on the structure and connectivity of aquifers, which may allow permeable units to connect, and aquifers to seal when juxtaposed against lower permeability units. Robust representation of groundwater flow around faults remains challenging despite the significance of faults for flow and transport.

Using Sealed Wells to Measure Water Levels Beneath Streams and Floodplains.

The design of wells beneath streams and floodplains has often employed with tall standpipes to prevent incursion of surface water into the well during flood events. Here, an approach has been presented to minimise the infrastructure demands in these environments by sealing the well top (e.g.