Skip to main content

AI Demos by industry

Environmental risk & remediation AI Demos

Examine contaminant sources, groundwater transport, and remediation choices. Compare uncertainty and subsurface environmental risks using transparent modeling evidence. Browse interactive demonstrations relevant to this area, with methods, screenshots, and resource links.

Browse by industry

Choose an industry to explore relevant demonstrations. Research & scientific AI brings together methods used across industries.

5 of 5 entries

Browse entries by name

Geysers Seismicity

The Geysers workflow presents seismic activity within a major operating geothermal field, connecting earthquake locations with their spatial organization over a decade of observations. We have performed various clustering and spatiotemporal ML analyses. We present 3D views of our closest server to make it easier to examine where events concentrate and how different groups occupy the field. The demo provides an interactive starting point for discussing induced seismicity and selecting patterns for further scientific investigation.

  • Explores earthquake observations from the Geysers geothermal field over 2010–2019.
  • Groups events into prepared spatial clusters to reveal differences in event density and location.
  • Supports pan, zoom, hover, and comparison of two clustering views.
  • Uses prepared observational visualizations; browsing does not fit a new seismic model or establish a causal explanation.
  • A recent research paper titled “Deep-Embedded-Clustering of Microseismicity Identifies Multiple Failure Mechanisms at The Geysers Geothermal Field” summarizes preliminary ML results.

Model Diagnostics

A close match to observed data leaves important questions about a model's reliability and the information contained in its parameters. This demo combines calibration with held-back checking, sensitivity analysis, and an examination of whether individual inputs can be constrained. The resulting views help users distinguish influential parameters from weakly supported ones and identify where additional evidence could improve an analysis.

  • Combines model calibration with a held-back data check.
  • Examines which inputs influence predictions and which parameters are poorly constrained.
  • Includes an optional unused input to illustrate an identifiability warning.
  • Encourages joint review of fit, validation, sensitivity, and identifiability rather than judging a model by fit alone.
  • Applies our open-source code Mads.jl

Bayesian Information Gap Decision

Environmental decisions often depend on both monitoring evidence and uncertainty that extends beyond a single assumed scenario. This demo combines Bayesian evidence with information-gap stress testing to compare contaminant source-termination choices. By showing how far each option can tolerate departures from its assumptions, it helps users discuss robustness alongside a selected limit on modeled failure probability.

  • Combines Bayesian monitoring evidence with information-gap stress testing for contaminant source termination.
  • Compares four choices against a selected maximum acceptable modeled failure probability.
  • Measures how much uncertainty each choice tolerates before exceeding that limit.
  • Supports comparison under deep uncertainty; the model informs the decision rather than making it for the operator.

Groundwater Source Unmixing

Groundwater sampled at a well can contain a mixture of chemical influences from several sources. This demo applies NMFk to separate those mixtures into candidate source fingerprints and estimate their contributions at individual wells. Examining alternative source counts, stability, and reconstruction behavior helps users develop and compare explanations for observed groundwater chemistry.

  • Separates mixed groundwater chemistry into hidden source fingerprints using NMFk.
  • Compares candidate source counts and repeat-run stability.
  • Relates each recovered source to contributions at individual wells.
  • Can hide one measurement to examine its reconstruction, while preserving the distinction between measured and estimated chemistry.

Contaminant Source Identification

Monitoring wells record the downstream consequences of a release, while its original timing and movement through groundwater may remain uncertain. This synthetic inverse problem uses concentration histories to investigate a release window and an associated transport speed. Bringing the monitoring layout and estimated source behavior together helps users understand how observations constrain a possible contamination history.

  • Uses monitoring-well concentration histories to infer a contaminant release window.
  • Estimates the groundwater transport speed connecting a release to downstream monitoring wells.
  • Provides quick and stability-check depths for the synthetic inverse problem.
  • Presents the monitoring layout, concentration histories, source timing, and transport estimate together for interpretation.