Stage‐Wise Stochastic Deep Learning Inversion Framework for Subsurface Sedimentary Structure Identification

Zhan, Chuanjun; Dai, Zhenxue; Soltanian, Mohamad Reza; Zhang, Xiaoying

doi:10.1029/2021gl095823

Cited by 46 publications

(25 citation statements)

References 61 publications

(85 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While the modeling framework can reproduce principal preferential flow paths and associated paleo‐channels, reproducing multi‐layered and intersecting subsurface channel networks encountered in wide braided river systems likely necessitates more complex geostatistical approaches (Brunetti et al., 2019; Pirot et al., 2015; Renard & Allard, 2013; Siirila‐Woodburn & Maxwell, 2015). A promising way forward for the detection of paleo‐channels are machine learning‐based approaches, which may allow including additional information into the modeling framework while simultaneously reducing the computational burden of ISSHM inversion (Sun, 2018; Wang et al., 2021; Zhan et al., 2022; Zhu & Zabaras, 2018).…”

Section: Discussionmentioning

confidence: 99%

Buried Paleo‐Channel Detection With a Groundwater Model, Tracer‐Based Observations, and Spatially Varying, Preferred Anisotropy Pilot Point Calibration

Schilling

Partington

Doherty

et al. 2022

Geophysical Research Letters

View full text Add to dashboard Cite

Alluvial sand and gravel (ASG) aquifers are highly heterogeneous and exhibit strong, spatially variable anisotropy, often interspersed by buried paleo‐channels of increased hydraulic conductivity. Groundwater flow and solute transport is often characterized by preferential flow caused by anisotropic properties in ASG aquifers. Connected ASG subsurface structures such as buried paleo‐channels, however, are difficult to reproduce with commonly used techniques, and anisotropy is rarely considered in applied groundwater models. To ease the notoriously difficult problem of how to consider anisotropy, we propose a novel modeling framework based on calibration of an integrated surface‐subsurface hydrological model via spatially varying, preferred anisotropy pilot point inversion. The inversion leverages hydraulic and tracer‐based observations representing multiple spatial and temporal scales. We demonstrate the applicability of the framework on a real‐world ASG site used for drinking water production, and we quantify the information content of observations to identify connected paleo‐channels and provide guidance for optimal field‐data acquisition.

show abstract

Section: Discussionmentioning

confidence: 99%

Buried Paleo‐Channel Detection With a Groundwater Model, Tracer‐Based Observations, and Spatially Varying, Preferred Anisotropy Pilot Point Calibration

Schilling

Partington

Doherty

et al. 2022

Geophysical Research Letters

View full text Add to dashboard Cite

show abstract

“…The results suggest that V RZ largely controls the tailing behavior, where large RZs would have more capacity to capture solutes and thus lead to a heavier tailing. Our new finding of relationship between n and β can be possibly used for upscaling transport process via deterministic method [29] for complex porous media in nature [30].…”

Section: Non-fickian Transport With Bimodal Tailing Process Depending...mentioning

confidence: 91%

The Effects of Pore Geometry on Late Time Solute Transport with the Presence of Recirculation Zone

Yang

Gu³

et al. 2022

Energies

View full text Add to dashboard Cite

The solute transport process in porous media is central to understanding many geophysical processes and determines the success of engineered applications. However, fundamental understanding of solute transport in heterogeneous porous media remains challenging especially when inertial effects are significant. To address this challenge, we employed direct numerical simulations in a variety of intrapore geometries at a high Reynolds number (Re = 10) flow regime, where recirculation zones (RZs) are present with significant inertial effects. We find that the volume of RZs depends on pore geometries. Moreover, RZs serve as an immobile domain that can trap and release solutes that lead to non-Fickian transport, characterized by the early arrival and heavy tailing of breakthrough curves and bimodal residence time distributions (RTDs). Lastly, the late time portion of RTDs is fitted to the power law function with determined exponent n, where n depends on the pore geometries and consequently the volume of RZs. Our study sheds light on the mechanisms of an immobile zone on the solute transport, especially improving our understanding of late time transport tailing in pressurized heterogeneous porous media.

show abstract

“…10.1029/2022WR032429 5 of 26 whether x ∼ p data (x) or x ∼ p G (x) (Zhong et al, 2019;Zhong et al, 2020;T. Wang, Trugman, et al, 2021;Zhan et al, 2021). The goal of the generator is to approximate the distribution of p G (x) to p data (x).…”

Section: Dcgan For Permeability Field Reconstructionmentioning

confidence: 99%

Integration of Deep Learning and Information Theory for Designing Monitoring Networks in Heterogeneous Aquifer Systems

Chen

Dai

Dong

et al. 2022

Water Resources Research

Self Cite

View full text Add to dashboard Cite

Groundwater monitoring networks are direct sources of information for revealing subsurface system dynamic processes. However, designing such networks is difficult due to uncertainties in the spatial heterogeneity of aquifer parameters such as permeability (k). This study combines deep learning and information theory with an optimization framework to address network design problems in heterogeneous aquifer systems. The framework first employs a generative adversarial network to parameterize heterogeneous k distribution using a low-dimensional latent representation. Then, surrogate models are developed based on the deep neural networks to perform uncertainty quantification of pressure heads and solute concentrations at locations of pre-designed candidate monitoring stations. The monitoring stations are then ranked using the greedy search algorithm based on the maximum information minimum redundancy (MIMR) criterion. In order to depict the importance of each candidate monitoring location, the hotspot maps of the selection probability (P s ) are derived from MIMR repetition results. Comprehensive monitoring networks derived from the hotspot maps are then conducted as the final monitoring stations to improve monitoring information compared to MIMR results. Additionally, nine entropy quantization strategies are compared to evaluate their effects on monitoring network optimization results. Results indicate that caution should be taken when selecting entropy quantization strategies to achieve the accuracy required for model calibration and to improve the efficiency of monitoring optimization. Considering high-dimensional uncertainties associated with aquifer parameters, the developed framework can provide important insights for monitoring network designs in various earth observational projects. Plain Language SummaryGroundwater is widely distributed throughout the crust and lithosphere playing a critical role in geological evolution processes and environmental problems. This study presents an integrated framework for designing groundwater monitoring networks in heterogeneous aquifer systems. The proposed model takes advantage of the strength of deep-learning algorithms and deals with high-dimensional and highly non-linear problems of entropy-based methods for designing the monitoring networks. The framework can efficiently provide monitor locations for complex aquifers. The monitoring data collected by the designed network can significantly reduce uncertainties associated with high-dimensional parameters and contain the least amount of redundant information. Further, the proposed method is significant in practice as credible groundwater forecasting requires model calibrations that use monitoring data.

show abstract

Stage‐Wise Stochastic Deep Learning Inversion Framework for Subsurface Sedimentary Structure Identification

Abstract: Delineating subsurface sedimentary structure is one of the most significant challenges in earth sciences and energy related applications (

Cited by 46 publications

References 61 publications

Buried Paleo‐Channel Detection With a Groundwater Model, Tracer‐Based Observations, and Spatially Varying, Preferred Anisotropy Pilot Point Calibration

Buried Paleo‐Channel Detection With a Groundwater Model, Tracer‐Based Observations, and Spatially Varying, Preferred Anisotropy Pilot Point Calibration

The Effects of Pore Geometry on Late Time Solute Transport with the Presence of Recirculation Zone

Integration of Deep Learning and Information Theory for Designing Monitoring Networks in Heterogeneous Aquifer Systems

Contact Info

Product

Resources

About