Machine learning for sustainable geoenergy: uncertainty, physics and decision-ready inference

Abstract

Geoenergy projects (CO2 storage, geothermal, subsurface H2 generation/storage, critical minerals from subsurface fluids, or nuclear waste disposal) increasingly follow a petroleum-style funnel from screening and appraisal to operations, monitoring, and stewardship. Across this funnel, limited and heterogeneous observations must be turned into risk-bounded operational choices under strong physical and geological constraints - choices that control deployment rate, cost of capital, and the credibility of climate-mitigation claims. These choices are inherently multi-objective, balancing performance against containment, pressure footprint, induced seismicity, energy/water intensity, and long-term stewardship. We argue that progress is limited by four recurring bottlenecks: (i) scarce, biased labels and few field performance outcomes; (ii) uncertainty treated as an afterthought rather than the deliverable; (iii) weak scale-bridging from pore to basin (including coupled chemical-flow-geomechanics); and (iv) insufficient quality assurance (QA), auditability, and governance for regulator-facing deployment. We outline machine learning (ML) approaches that match these realities (hybrid physics-ML, probabilistic uncertainty quantification (UQ), structure-aware representations, and multi-fidelity/continual learning) and connect them to four anchor applications: imaging-to-process digital twins, multiphase flow and near-well conformance, monitoring and inverse problems (monitoring, measurement, and verification (MMV), including deformation and microseismicity), and basin-scale portfolio management. We close with a pragmatic agenda for benchmarks, validation, reporting standards, and policy support needed for reproducible and defensible ML in sustainable geoenergy.

Figures (3)

• Figure 1: Funnel $\times$ Stack framing: the subsurface project funnel (screening $\rightarrow$ appraisal $\rightarrow$ operations $\rightarrow$ closure) repeated through the ML stack (sensing $\rightarrow$ inference $\rightarrow$ simulation $\rightarrow$ decision). The uncertainty ribbon highlights that uncertainty is generated at early funnel stages, shaped by new data and models as the project matures, and must ultimately be expressed as decision risk (operating envelopes and trigger criteria).
• Figure 2: Conceptual map linking the four bottlenecks (data/labels, uncertainty, scale bridging, and trust/governance) to method families that fit geoenergy constraints (hybrid physics--ML, probabilistic UQ, structure-aware representations, and multi-fidelity/continual learning), and to four anchor applications (imaging-to-process digital twins, multiphase conformance, MMV/inversion, and basin-scale portfolios). The validation strip emphasizes that out-of-distribution checks, physics residual/stress tests, calibration diagnostics, and blind challenges should be routine across methods and applications. Application labels can be read across systems: imaging-to-process is most relevant to pore-driven reactivity (CCS/geothermal/minerals), multiphase near-well effects to storage/deliverability (CCS/H$_2$), MMV to plume/thermal front tracking and seismicity constraints (CCS/EGS/H$_2$), and portfolios to basin-scale planning across all.
• Figure 3: Illustrative decision output for sustainable geoenergy: an operating envelope / Pareto front showing trade-offs under uncertainty (e.g., performance versus pressure footprint and induced seismicity risk). The deliverable is a calibrated, risk-bounded frontier.