-
Multimodal machine learning for 3-dimensional characterization of hidden groundwater and geothermal resources
Authors:
Michael J. Friedel,
Nicole Lautze,
Erin Wallin,
Roland Gritto,
Alain Bonneville,
Massimo Buscema,
Steve Martel
Abstract:
We present a multimodal machine learning (MML) workflow to assimilate and simultaneously predict the 3d distribution of numeric and categorical features along a groundwater-geothermal continuum. Success of the MML workflow relies on a transductive learning algorithm that projects field modalities onto a single embedding space (hypersurface). Multimodalities can include any combination of measured…
▽ More
We present a multimodal machine learning (MML) workflow to assimilate and simultaneously predict the 3d distribution of numeric and categorical features along a groundwater-geothermal continuum. Success of the MML workflow relies on a transductive learning algorithm that projects field modalities onto a single embedding space (hypersurface). Multimodalities can include any combination of measured (point field) and derived (multiphysics based numerical model inversions, data driven machine learning, and multiphysics informed machine learning) features. The proposed MML workflow is applied to assimilate randomly shuffled subsets of Hawaii Play Fairway modalities and predict subsurface geophysical, geologic, and hydrogeologic features at the Islands of Lanai and Hawaii. Despite challenging field data characteristics (disparate, scale dependent, spatially limited, sparse, and uncertain), the MML workflow yields a single 3d transdisciplinary model that generalizes well to independent data presented to the trained model. The predicted features are used to identify hidden groundwater and geothermal resources at Lanai, and geothermal resources at Hawaii. Other interpreted subsurface features at Lanai include basalt, batholith, dike swarm, pluton, sill, mantle, Moho, and 3d geothermal stratigraphic units; whereas interpreted subsurface features at Hawaii include 3d velocity layering, 3d earthquake-fault associations, 3d fault systems; basalt, oceanic crust, magmatic underplating, lithospheric flexure, mantle, and Moho. This study provides new capabilities for characterizing continuous subsurface groundwater and geothermal features for sustainable living in the Hawaiian Islands and other geothermal sites worldwide.
Keywords: Machine learning, 3d hidden groundwater resources, hidden geothermal resources fault system, oceanic crust, magmatic underplating, lithospheric flexure, Moho, Hawaii
△ Less
Submitted 22 December, 2023;
originally announced December 2023.
-
Multiphysics-decision tree learning for improved variably saturated subsurface parameter estimation and reduced-order simulation
Authors:
Michael J Friedel,
Massimo Buscema
Abstract:
A novel multiphysics-decision tree learning algorithm is presented for (1) estimating transport properties in the variably saturated subsurface governed by explicitly coupled equations for water, heat, and solute transport; and (2) providing reduced order simulation of time-dependent pressure head, temperature, and concentration with subsurface properties and/or changing surface boundary condition…
▽ More
A novel multiphysics-decision tree learning algorithm is presented for (1) estimating transport properties in the variably saturated subsurface governed by explicitly coupled equations for water, heat, and solute transport; and (2) providing reduced order simulation of time-dependent pressure head, temperature, and concentration with subsurface properties and/or changing surface boundary conditions. We demonstrate that the proposed algorithm results in about one order of magnitude less error in estimated parameters than the traditional multiphysics numerical inversion. We further show that the multiphysics-decision tree learning algorithm reduces the computational burden associated of traditional parameter estimation with reductions in the number Jacobian sensitivity calculations by as much as 90% and the number of iterations required for convergence by up to an order of magnitude. A natural outcome following convergence of the proposed learning algorithm is the reduced order set of supervised decision tree learning models for predicting the pressure head (Random Forest), and the temperature and concentration (Ensemble Gradient Boosting) given knowledge of time, depth, and remaining pair of state variables. The supervised reduced-order modeling is extended to unsupervised machine learning for the simultaneous prediction of state variables by training a Self-Organized Map using the joint multiphysics-decision tree learning property estimates, stochastic boundary conditions, and subsurface state field measurements. The reduced-order machine learning models provide a computationally efficient alternative for studying the effects of changing subsurface water, heat, and solute transport properties and/or surface boundary conditions on coupled subsurface pressure head, temperature, and concentration.
△ Less
Submitted 15 December, 2023;
originally announced December 2023.
-
A simple stacked ensemble machine learning model to predict naturalized catchment hydrology and allocation status
Authors:
Michael J. Friedel,
Dave Stewart,
Xiao Feng Lu,
Pete Stevenson,
Helen Manly,
Tom Dyer
Abstract:
New Zealand legislation requires that Regional Councils set limits for water resource usage to manage the effects of abstractions in over-allocated catchments. We propose a simple stacked ensemble machine learning model to predict the probable naturalized hydrology and allocation status across 317 anthropogenically stressed gauged catchments and across 18,612 ungauged river reaches in Otago. The t…
▽ More
New Zealand legislation requires that Regional Councils set limits for water resource usage to manage the effects of abstractions in over-allocated catchments. We propose a simple stacked ensemble machine learning model to predict the probable naturalized hydrology and allocation status across 317 anthropogenically stressed gauged catchments and across 18,612 ungauged river reaches in Otago. The training and testing of ensemble machine learning models provides unbiased results characterized as very good (R2 > 0.8) to extremely good (R2 > 0.9) when predicting naturalized mean annual low flow and Mean flow. Statistical 5-fold stacking identifies varying levels of risk for managing water-resource sustainability in over-allocated catchments; for example, at the respective 5th, 25th, 50th, 75th, and 95th percentiles the number of overallocated catchments are 73, 57, 44, 23, and 22. The proposed model can be applied to inform sustainable stream management in other regional catchments across New Zealand and worldwide.
△ Less
Submitted 4 December, 2023;
originally announced December 2023.
