Exploring the Model Space of Airborne Electromagnetic Data to Delineate Large‐Scale Structure and Heterogeneity Within an Aquifer System

Kang, Seogi; Knight, Rosemary; Greene, Todd J.; Buck, Caitlin E.; Fogg, Graham E.

doi:10.1029/2021wr029699

Cited by 12 publications

(4 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Resistivity models from airborne electromagnetic (AEM) surveys can be used to estimate aquifer parameters and aquifer frameworks (N. K. Christensen, Ferre, et al., 2017; N. K. Christensen, Minsley, & Christensen, 2017; N. K. Christensen et al., 2016; Ikard et al., 2023; Kang et al., 2021; Knight et al., 2018; Korus et al., 2017; Marker et al., 2015; Minsley, Foks, & Bedrosian, 2021; Minsley, Rigby, et al., 2021). The relationship between resistivity and hydraulic conductivity can be influenced by factors such as porewater chemistry and the presence of clay minerals.…”

Section: Introductionmentioning

confidence: 99%

Three‐Dimensional Probabilistic Hydrofacies Modeling Using Machine Learning

Kawo,

Korus,

Kishawi

et al. 2024

Water Resources Research

View full text Add to dashboard Cite

Characterizing the 3D distribution of hydraulic properties in glacial sediments is challenging due to fine‐scale heterogeneity and complexity. Borehole lithological data provide high vertical resolution but low horizontal resolution. Geophysical methods can fill gaps between boreholes, providing improved horizontal resolution but low vertical resolution. Machine learning can combine borehole and geophysical data to overcome these challenges. However, few studies have compared multiple machine learning methods for predicting hydrofacies in glacial aquifer systems. This study uses colocated airborne electromagnetic resistivity and borehole lithology data to train multiple machine learning models and predict the 3D distribution of hydrofacies in glacial deposits of eastern Nebraska, USA. Random Forest, Gradient Boosting Classifier, Extreme Gradient Boosting, Multilayer Perceptron, and Stacking Classifier were used to model 3D probabilistic distributions of hydrofacies (sand and clay) at a grid size of 200 m × 200 m × 3 m. Comparison of the predicted 3D hydrofacies models shows that the probability distributions and the contrasts between hydrofacies vary. The classification metrics show that the Stacking Classifier model performed better than other machine learning models in predicting hydrofacies. Multi‐Layer Perceptron and Stacking Classifier models show sharp vertical transitions between the low and high sand probability while other machine learning models show gradual transitions. K‐means clustering was used to translate the Stacking Classifier model into a 4‐class hydraulic conductivity model. This study shows that machine learning methods advance our understanding of glacial hydrogeology by improving the vertical and horizontal resolution of hydrofacies distribution and resolving aquifer‐aquifer and stream‐aquifer connections.

show abstract

Section: Introductionmentioning

confidence: 99%

Three‐Dimensional Probabilistic Hydrofacies Modeling Using Machine Learning

Kawo,

Korus,

Kishawi

et al. 2024

Water Resources Research

View full text Add to dashboard Cite

show abstract

“…While this is developed on a single infiltration pond, Knight et al (2018) made use of similar relationships to map aquifer systems for optimal recharge conditions using Airborne Electromagnetic measurements. This approach has recently been extended by Kang et al (2021) to ease interpretation of such data over large domains. However, these small and large scale investigations are often disconnected, which is mainly due to missing instrumentation that can acquire data rapidly over different scales.…”

Section: Introductionmentioning

confidence: 99%

3D hydrogeophysical characterization of managed aquifer recharge basins

et al. 2022

View full text Add to dashboard Cite

Aquifers are increasingly stressed. Managed aquifer recharge provides a potential solution to mitigate this stress and provide sustainable groundwater resources. Subsurface properties are known to have a strong control on the infiltration rates that can be achieved. However, these properties are often highly heterogeneous and difficult to assess with conventional probing techniques. Here, we show the application of 3D geophysical imaging to assess the recharge potential and its variation across several basins used for managed aquifer recharge. We link in-situ measurements of saturated hydraulic conductivity with the electrical resistivity of the subsurface to establish petrophysical relationships and use those relationships to estimate the distribution of hydraulic conductivity throughout the five recharge basins. Our results show a considerable variability in the hydraulic properties, i.e., soil texture and saturated hydraulic conductivity, that have a direct impact on potential infiltration rates. We use the 3D hydraulic property distributions to model groundwater recharge and provide estimates for infiltration rates and volumes, and use this approach to assess the impact of management activities on groundwater recharge performance. Having such data not only enables us to predict infiltration rates, but also provides means for optimizing such water infrastructure.

show abstract

“…The airborne electromagnetic (AEM) method is potentially a highly effective means of imaging large‐scale structure for the development of a groundwater model (Barfod et al., 2018; Foged et al., 2013; He et al., 2014; Kang et al., 2021; Knight et al., 2018; Sattel & Kgotlhang, 2004; Wynn, 2002). Inversion of the acquired AEM data recovers a resistivity model of the subsurface.…”

Section: Introductionmentioning

confidence: 99%

Improved Imaging of the Large‐Scale Structure of a Groundwater System With Airborne Electromagnetic Data

Kang

Knight

Goebel

2022

Water Resources Research

Self Cite

View full text Add to dashboard Cite

With climate change and population growth, there is increasing concern about the depletion of groundwater resources. Evidence of this is a petition signed by over 1,300 scientists from 100 countries calling for action to ensure that groundwater benefits society now and into the future (Gleeson et al., 2019). A groundwater model is the foundation on which to build sustainable groundwater management. Required to inform the development of the groundwater model is information about the subsurface that captures spatial heterogeneity at the level needed as the input for flow modeling.All groundwater models include some representation of the large-scale structure of the groundwater system-the hydrogeologic units and other major features relevant for modeling flow. We define "large" to be the scale at which we represent the architecture of the subsurface in terms of the boundaries between areas with significant differences in hydrogeologic properties, for example, the boundary between an aquitard and aquifer, the top of the bedrock. Information derived from driller's logs, recorded when wells are drilled, is typically used to build the model. The logs potentially provide valuable information at point locations, but the quality of the driller's logs

show abstract

Exploring the Model Space of Airborne Electromagnetic Data to Delineate Large‐Scale Structure and Heterogeneity Within an Aquifer System

Cited by 12 publications

References 47 publications

Three‐Dimensional Probabilistic Hydrofacies Modeling Using Machine Learning

Three‐Dimensional Probabilistic Hydrofacies Modeling Using Machine Learning

3D hydrogeophysical characterization of managed aquifer recharge basins

Improved Imaging of the Large‐Scale Structure of a Groundwater System With Airborne Electromagnetic Data

Contact Info

Product

Resources

About