Using Remote Sensing and Machine Learning to Locate Groundwater Discharge to Salmon-Bearing Streams

Abstract

We hypothesized topographic features alone could be used to locate groundwater discharge, but only where diagnostic topographic signatures could first be identified through the use of limited field observations and geologic data. We built a geodatabase from geologic and topographic data, with the geologic data only covering ~40% of the study area and topographic data derived from airborne LiDAR covering the entire study area. We identified two types of groundwater discharge: shallow hillslope groundwater discharge, commonly manifested as diffuse seeps, and aquifer-outcrop groundwater discharge, commonly manifested as springs. We developed multistep manual procedures that allowed us to accurately predict the locations of both types of groundwater discharge in 93% of cases, though only where geologic data were available. However, field verification suggested that both types of groundwater discharge could be identified by specific combinations of topographic variables alone. We then applied maximum entropy modeling, a machine learning technique, to predict the prevalence of both types of groundwater discharge using six topographic variables: profile curvature range, with a permutation importance of 43.2%, followed by distance to flowlines, elevation, topographic roughness index, flow-weighted slope, and planform curvature, with permutation importance of 20.8%, 18.5%, 15.2%, 1.8%, and 0.5%, respectively. The AUC values for the model were 0.95 for training data and 0.91 for testing data, indicating outstanding model performance.

Real‐Time Monitoring and Postprocessing of Thermal Infrared Video Images for Sampling and Mapping Groundwater Discharge

“…At the study sites, the locations of the cold springs did not conform to the surface topography (Figure 2). Although there are catchments where the topography can be used to predict the location of groundwater discharge zones (Gerlach et al, 2022;Leach et al, 2017), no correspondence between the surface topography and aquifer boundaries has been reported for Quaternary volcanic catchments (Fujimoto et al, 2016;Jefferson et al, 2006) and volcanic ash plateaus (Ikeda et al, 1999) similar to our sites, which can be attributed to the high subsurface permeability. Thus, our results demonstrate the importance of field surveys for mapping springs at sites underlain by a highly permeable geology.…”

Identification of Thermal Refuges and Water Temperature Patterns in Salmonid-Bearing Subarctic Rivers of Northern Quebec

“…Machine learning and remote sensing combinations have been used to identify the location of thermal refuges and GW-SW interaction zones in numerous rivers (e.g., [17,[42][43][44]). In these studies, key landscape parameters (e.g., channel confinement, location of dry valleys) have been noted as potential predictors for the location of thermal refuges.…”

Exploring landscape and geologic controls on spatial patterning of streambank groundwater discharge in a mixed land use watershed

“…Surface landforms have long been invoked to conceptually explain spatial patterns of groundwater connectivity to surface waters (Winter et al, 1998). Locally, steep streambanks or streambed slopes can cut across hydraulic gradients and create groundwater discharge across the land surface, while high stream curvature can cause the convergence of groundwater flow paths to streams (Eschbach et al, 2017; Gerlach et al, 2022; Wawrzyniak et al, 2016). At the valley scale, the degree of channel confinement can control the prevalence of groundwater discharge, with partially confined valleys having the highest prevalence of streambank groundwater discharge relative to confined or unconfined settings (Dugdale et al, 2015).…”

“…While surface topography can exert strong controls on the spatial patterning of groundwater discharge, characteristics of underlying soils and bedrock geology are important as well. Gerlach et al (2022) found that the depths and locations of water bearing geologic formations derived from borehole data predicted the occurrence of groundwater discharges across several salmon bearing streams in Alaska, USA. Gerlach et al (2022) also found that the occurrence of bedrock outcrops was important in creating hillslope and streambank discharge features.…”

“…At the valley scale, the degree of channel confinement can control the prevalence of groundwater discharge, with partially confined valleys having the highest prevalence of streambank groundwater discharge relative to confined or unconfined settings (Dugdale et al, 2015). Landform‐derived metrics, such as topographic wetness (a function of slope and the upstream contributing area; Buttle et al, 2004) and terrain ruggedness (indicative of soil texture properties; Gerlach et al, 2022), have shown promise in identifying features that focus flow toward streams and have become more accurate with the increasing availability of high resolution topographic data.…”