Uncertainty-based saltwater intrusion prediction using integrated Bayesian machine learning modeling (IBMLM) in a
Jina Yin1, Yulu Huang1, Chunhui Lu1
1The National Key Laboratory of Water Disaster Prevention, Hohai University, Nanjing, China; Yangtze Institute for Conservation and Development, Hohai University, Nanjing, China.
Journal of Environmental Management
|February 23, 2024
Summary
This study introduces an integrated Bayesian machine learning modeling (IBMLM) method to improve groundwater predictions by accounting for model structure and parameter uncertainties. The IBMLM offers more accurate and reliable saltwater intrusion predictions than standalone models.
Area of Science:
- Hydrogeology
- Machine Learning
- Environmental Modeling
Background:
- Physically based groundwater models are computationally intensive.
- Existing machine learning surrogates often neglect structural uncertainties, impacting reliability.
- Saltwater intrusion (SWI) is a critical issue in coastal aquifers.
Purpose of the Study:
- To develop a flexible integrated Bayesian machine learning modeling (IBMLM) method.
- To explicitly quantify uncertainties from both structure and parameters of machine learning surrogates.
- To improve the accuracy and reliability of groundwater predictions, specifically for saltwater intrusion.
Main Methods:
- Integrated Bayesian Machine Learning Modeling (IBMLM) framework.
- Incorporation of Expectation-Maximization (EM) algorithm and Bayesian Model Averaging (BMA).
- Application of Artificial Neural Network (ANN), Support Vector Machine (SVM), and Random Forest (RF) within the IBMLM.
Main Results:
- IBMLM demonstrated high predictive accuracy with R values > 0.98 and NSE values > 0.93.
- IBMLM provided superior predictions compared to individual machine learning models.
- The method effectively predicted saltwater intrusion without complex physical simulations.
Conclusions:
- Explicitly considering machine learning model structure uncertainty enhances prediction accuracy and reliability.
- IBMLM offers a computationally efficient alternative to traditional groundwater models.
- The IBMLM framework is applicable in data-scarce regions for hydrogeologic modeling.
Related Concept Videos
01:29Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving
54
Mechanistic models play a crucial role in algorithms for numerical problem-solving, particularly in nonlinear mixed effects modeling (NMEM). These models aim to minimize specific objective functions by evaluating various parameter estimates, leading to the development of systematic algorithms. In some cases, linearization techniques approximate the model using linear equations.
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
54
01:30Typical Model Studies
359
Fluid mechanics model studies often utilize scaled-down systems to predict fluid behavior in full-scale environments, such as river flows, dam spillways, and structures interacting with open surfaces. Maintaining Froude number similarity in river models is crucial, as it replicates surface flow features like wave patterns and velocities.
359
01:27Uniform Depth Channel Flow
74
Uniform depth channel flow keeps fluid depth consistent along channels such as irrigation canals. In natural channels, such as rivers, approximate uniform flow is often assumed. This condition occurs when the channel’s bottom slope matches the energy slope, balancing potential energy lost from gravity with head loss due to shear stress. This balance prevents depth changes along the channel length, resulting in a steady, uniform flow.Uniform flow in open channels with a constant cross-section...
74
01:12Modeling and Similitude
267
Scaled modeling is a fundamental technique in engineering, enabling the study of large and complex systems by creating smaller, manageable replicas that recreate critical characteristics of the original. In hydrology and civil infrastructure, for example, scaled models of dams help analyze water flow, turbulence, and pressure. This method allows for accurate predictions of real-world behavior within a controlled environment, significantly reducing the cost and time involved in full-scale...
267