A Novel Approach to Uncertainty Quantification in Groundwater Table Modeling by Automated Predictive Deep Learning

Shahri, Abbas Abbaszadeh; Shan, Chunling; Larsson, Stefan

doi:10.1007/s11053-022-10051-w

Cited by 105 publications

(15 citation statements)

References 78 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Uncertainty can be described as a situation involving incomplete or unknown information [225]. In general, three sources should be considered to quantify uncertainty [226], including the physical variability of the equipment, data and model errors. Traditional methods normally used statistical methods, polynomial chaos expansion method, Perturbation method and Monte Carlo simulation are common methods for uncertainty quantification.…”

Section: Challenges and Future Workmentioning

confidence: 99%

Recent Advances of Deep Learning in Geological Hazard Forecasting

Wang¹,

Sun²,

Chen³

et al. 2023

Computer Modeling in Engineering &Amp; Sciences

View full text Add to dashboard Cite

Geological hazard is an adverse geological condition that can cause loss of life and property. Accurate prediction and analysis of geological hazards is an important and challenging task. In the past decade, there has been a great expansion of geohazard detection data and advancement in data-driven simulation techniques. In particular, great efforts have been made in applying deep learning to predict geohazards. To understand the recent progress in this field, this paper provides an overview of the commonly used data sources and deep neural networks in the prediction of a variety of geological hazards. KEYWORDSGeological hazard; deep learning; neural networks; geohazard data sources; earthquake; volcanic 1 Introduction Geological hazards are caused by endogenous and exogenous geological processes or anthropogenic factors on earth [1,2], which encompass a broad range of abnormal stratigraphic activity or extreme changes in geological environments, such as landslides, avalanches, debris flows, earthquakes, etc. Geological hazards pose a serious threat to human life and property [3]. Only by accurately forecasting the occurrence of geohazards can emergency measures be devised to mitigate the damage caused by the impending disaster. Geological hazard forecasting refers to the use of logical reasoning, numerical simulation, and comprehensive analysis based on historical geological hazard activity patterns, formation conditions, occurrence mechanisms, and other factors to speculate and assess the development and changes of geological hazards and the possible degree of danger and damage in a certain period in the future. Geohazard forecasting has attracted tremendous attention nowadays, particularly considering natural resource scarcity, environmental degradation, population expansion, sustainable development, and world economic integration. Many efforts have been made in forecasting geological hazards in the past decades. The "3S technology", including Global Position System (GPS), Geographic Information System (GIS) [4], and Remote Sensing (RS), has been widely applied in the field of geological hazard forecasting [5]. The

show abstract

Section: Challenges and Future Workmentioning

confidence: 99%

Recent Advances of Deep Learning in Geological Hazard Forecasting

Wang¹,

Sun²,

Chen³

et al. 2023

Computer Modeling in Engineering &Amp; Sciences

View full text Add to dashboard Cite

show abstract

“…We focused on thermal heating of the crust because it is usually neglected or not considered, but it has a significant influence. Nevertheless, a somewhat simple probabilistic model has been developed which respects the real variability of input and output quantities [9,[15][16][17][18][19][20][21][22][23][24][25].…”

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

“…Other possible approaches connected with groundwater investigations and uncertainty quantification are presented in [12,23]. In reference [23], there is a state-of-the-art in approach for uncertainty quantification in geomechanics. However, this approach cannot yet be applied to the complicated problem of crust-mantle interaction because of the lack of global information about the crust.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Analysis of the Influence of Thermal Loading on the Behaviour of the Earth’s Crust

2023

View full text Add to dashboard Cite

The article focuses on the deformation and strain-stress analysis of the Earth’s crust under external thermal loading. More specifically, the influence of cyclic changes in the surface temperature field on the stress and displacement inside the crust over a two-year time span is investigated. The finite element program MSC.Marc Mentat was used to calculate the stresses and displacements. For practical analysis reasons, the Earth’s crust is simplified as a planar, piecewise homogeneous, isotropic model (plane strain), and time-varying temperature functions of illumination (thermal radiation) from the Sun are considered in the local isotropy sections of the model. Interaction between the Earth’s crust and mantle is defined by the Winkler elastic foundation. By applying a probabilistic approach (Monte Carlo Method), a new stochastic model of displacements and stresses and new information on crustal displacements relative to the Earth’s mantle were obtained. The results proved the heating influence of the Sun on the Earth’s crust and plate tectonics.

show abstract

“…Despite its importance, UQ for ML model predictions is challenging and the development of a high-quality and computationally efficient UQ method, which produces precise InD uncertainty and identifies OOD samples, is even more challenging. Some UQ-for-ML methods have been applied in hydrological modeling, including Bayesian neural networks (Lu et al, 2019), Gaussian processes (Zhu et al, 2020;Klotz et al, 2022), Monte Carlo dropout (Fang et al, 2020;Lu et al, 2021) and other dropout or ensemble-based approaches such as ensembles with variance analysis (Song et al, 2020), ensemble at inference (Althoff et al, 2021), and ensembles with random weights drop-off (Abbaszadeh Shahri et al, 2022). The Bayesian neural networks are computationally expensive and impractical for large-scale, deep-learning models (Gal and Ghahramani, 2016a).…”

mentioning

confidence: 99%

Uncertainty quantification of machine learning models to improve streamflow prediction under changing climate and environmental conditions

Liu

Painter

et al. 2023

Front. Water

View full text Add to dashboard Cite

Machine learning (ML) models, and Long Short-Term Memory (LSTM) networks in particular, have demonstrated remarkable performance in streamflow prediction and are increasingly being used by the hydrological research community. However, most of these applications do not include uncertainty quantification (UQ). ML models are data driven and can suffer from large extrapolation errors when applied to changing climate/environmental conditions. UQ is required to quantify the influence of data noises on model predictions and avoid overconfident projections in extrapolation. In this work, we integrate a novel UQ method, called PI3NN, with LSTM networks for streamflow prediction. PI3NN calculates Prediction Intervals by training 3 Neural Networks. It can precisely quantify the predictive uncertainty caused by the data noise and identify out-of-distribution (OOD) data in a non-stationary condition to avoid overconfident predictions. We apply the PI3NN-LSTM method in the snow-dominant East River Watershed in the western US and in the rain-driven Walker Branch Watershed in the southeastern US. Results indicate that for the prediction data which have similar features as the training data, PI3NN precisely quantifies the predictive uncertainty with the desired confidence level; and for the OOD data where the LSTM network fails to make accurate predictions, PI3NN produces a reasonably large uncertainty indicating that the results are not trustworthy and should avoid overconfidence. PI3NN is computationally efficient, robust in performance, and generalizable to various network structures and data with no distributional assumptions. It can be broadly applied in ML-based hydrological simulations for credible prediction.

show abstract

A Novel Approach to Uncertainty Quantification in Groundwater Table Modeling by Automated Predictive Deep Learning

Cited by 105 publications

References 78 publications

Recent Advances of Deep Learning in Geological Hazard Forecasting

Recent Advances of Deep Learning in Geological Hazard Forecasting

Analysis of the Influence of Thermal Loading on the Behaviour of the Earth’s Crust

Uncertainty quantification of machine learning models to improve streamflow prediction under changing climate and environmental conditions

Contact Info

Product

Resources

About