open-DARTSopen-DARTS

Delft Advanced Research Terra Simulator

Open-source simulation for next-generation subsurface energy

Build, extend and run multiphysics simulations in Python — from your laptop to HPC.

Geothermal · CO2 storage · Hydrogen · Reactive transport · THMCS

pip install open-darts
Read the documentation Browse example models
supporting points: 0 of 117p = – bar · z = –
Operator-based linearization evaluates expensive physics only where needed and reuses the results across cells, iterations and model runs. Define physics in Python – run at HPC scale. Voskov, 2017 · Khait & Voskov, 2017 · Novikov & Voskov, 2026

Why open-DARTS?

Conventional simulatoropen-DARTS
Fixed physicsUser-defined physics
ProprietaryOpen source
CPU-centricCPU + GPU
Separate module for each formulationFully coupled THMCS
Specialized simulatorPython-extensible framework
THMCFully coupled, fully implicit thermal, flow, mechanics and chemistry
C++ core with Python APIOpenMP on CPU, CUDA kernels on GPU, with the nonlinear loop in Python
Flexible thermodynamicsUser-defined state specification and physical constitutive models
Advanced inverse capabilityAdjoint gradients for data assimilation and optimization
/in-action

See open-DARTS in action

Problems simulated with open-DARTS. Every animation comes from a model in darts-models that you can run yourself following the 3 steps below. Open-DARTS installs from PyPI as a regular Python package, no compiler needed; for the full tutorials, see the documentation.

Install

python -m pip install open-darts

Use a fresh virtual environment with a recent Python.

Get the examples

git clone -b main https://gitlab.com/open-darts/darts-models.git

Ready-made models, including the teaching material.

Run a model

Pick a problem: each tab shows the animation, what it found and the command that runs it. Run times are measured on an AMD Ryzen Threadripper PRO 7945WX with 12 threads.

DARTS ecosystem

Five tools that work together. Choose one to see three papers behind it.

/research

Capabilities and applications

Methods and applications from the first DARTS papers on: geothermal energy, CO2 and hydrogen storage, and subsurface monitoring, earlier and recent work side by side.

/theses

PhD theses

Doctoral theses at TU Delft that developed or applied DARTS, newest first.

  1. A thermodynamics-based simulation framework for modelling of geological CO2 sequestrationMichiel Wapperom · Defended 2026 · TU Delft repository ↗
  2. Improving uncertainty quantification of geological carbon storage with data assimilationGabriel Serrão Seabra · Defended 2026 · TU Delft repository ↗
  3. A finite volume framework for accurate modeling of fault reactivation in poroelastic rocksAleksei Novikov · Defended 2024 · TU Delft repository ↗
  4. General-purpose inverse modeling framework for energy transition applications based on adjoint method and operator-based linearizationXiaoming Tian · Defended 2023 · TU Delft repository ↗
  5. Modelling of coupled reservoir and wellbore for energy transitionKiarash Mansour Pour · Defended 2023 · TU Delft repository ↗
  6. Towards the uncertainty quantification of fractured karst systems: reactive transport and fracture networksStephan de Hoop · Defended 2022 · TU Delft repository ↗
  7. The impact of heterogeneity on geothermal production: simulation benchmarks and applicationsYang Wang · Defended 2021 · TU Delft repository ↗
  8. Simulation of foam in enhanced oil recovery (EOR) and carbon capture and storage (CCS) applicationsXiaocong Lyu · Defended 2021 · TU Delft repository ↗
  9. Delft Advanced Research Terra Simulator: general purpose reservoir simulator with operator-based linearizationMark Khait · Defended 2019 · TU Delft repository ↗
/news

News

Releases, training workshops and conference talks, newest first.

  1. Open-DARTS 2.0 releasedA major release with the open-source solver stack and tons of new options for physics. Run your multi-million-cell model right after pip install, with hydrates, salts and hysteresis. See all details in the changelog.
  2. ECMOR 2026, PortoFive talks: geochemistry coupling, induced seismicity and DAS, coupled wells, CO2 hydrates, and LLM agents driving open-DARTS through an MCP server.
  3. SPE Reservoir Simulation Conference, GalvestonSeven talks: interpolation strategies for OBL, thermal-compositional CO2 storage in depleted fields, DARTS-well, the SPE11 convergence study with impurities, machine-learning-assisted upscaling, representative heterogeneity for direct-use geothermal, and physics-informed neural networks for CO2 flooding.
  4. DARTS-flash workshopTwo days at TU Delft celebrating the open-source DARTS-flash release, with talks from DTU, UiB, EBN and UPPA/TotalEnergies.
  5. Open-DARTS paper in JOSSThe software paper is published in the Journal of Open Source Software.
  6. First open-DARTS workshopMore than 50 participants: a day on open-source reservoir simulation and a day of hands-on training.
A talk at the DARTS-flash workshop at TU Delft
DARTS-flash workshop, TU Delft, November 2024.
Participants of the first open-DARTS workshop in Delft
Participants of the first open-DARTS workshop, Delft, March 2023.
/team

The team

Researchers at the Department of Geoscience and Engineering, TU Delft, working with partners at Stanford University and in industry.

Former members

Contact us

Ran into a problem, or have a question? Send us an email and the team will get back to you.

darts@tudelft.nl

To help us help you

Include the version (pip show open-darts), your operating system and Python version, the full error message and, if you can, a small script that shows the problem.