JutulDarcy.jl

Appearance

JutulDarcyRe-thinking reservoir simulation in Julia

High-performance porous media and reservoir simulator based on automatic differentiation

Getting started
View on Github
Run .DATA file
JutulDarcy
💥

Physical systems

Immiscible, compositional, geothermal and black-oil flow

⚙️

Differentiability

Compute sensitivities of parameters with high-performance adjoint method

🏃

High performance on CPU & GPU

Fast execution with support for MPI, CUDA and thread parallelism

What is this? ​

JutulDarcy.jl is a general high-performance purpose porous media simulator toolbox (CO2 sequestration, gas/H2 storage, oil/gas fields) written in Julia based on Jutul.jl, developed by the Applied Computational Science group at SINTEF Digital.

A few highlights:

  • Immiscible, black-oil, compositional, CO2-brine and geothermal systems

  • Fully differentiable through adjoint method (history matching of parameters, optimization of well controls)

  • High performance, with optional support for compiling MPI parallel binaries

  • Consistent discretizations

  • Industry standard input formats - or make your own model as a script

  • 3D visualization and tools for post-processing of simulation results

Quick start guide ​

Getting started is the main setup guide that includes the basics of installing Julia and creating a Julia environment for JutulDarcy.jl, written for users who may not already be familiar with Julia package management.

Installation ​

If you want to get started right away: Install Julia and add the following packages together with a Makie backend library to your environment of choice using Julia's package manager Pkg:

julia
using Pkg
Pkg.add("GLMakie")    # Plotting
Pkg.add("Jutul")      # Base package
Pkg.add("JutulDarcy") # Reservoir simulator

Perform a simulation ​

To verify that everything is working, we have a minimal example that runs an industry standard input file and produces interactive plots. Note that interactive plotting requires GLMakie, which may not work if you are running Julia over SSH:

julia
using JutulDarcy, GLMakie
# Load the SPE9 dataset
spe9_dir = JutulDarcy.GeoEnergyIO.test_input_file_path("SPE9")
file_path = joinpath(spe9_dir, "SPE9.DATA")
case = setup_case_from_data_file(file_path)
# Simulate and plot
result = simulate_reservoir(case)
plot_reservoir(case, result, fancy = true)

We can then continue to produce interactive plots for the wells:

julia
plot_well_results(result)

There is also support for plotting summary data:

julia
plot_summary(result)

This can then be used to create dashboard that summarize the field behavior for specified values:

julia
# Plot field rates and rates for two specific wells in field units
plot_summary(result, plots = ["FWPR,FOPR,FGPR", "PRODU2:WOPR,WWPR,WGPR", "PRODU4:WOPR,WWPR,WGPR"], unit_system = "Field")

Python bindings ​

Alternatively, the code has a Python package that can be installed using pip:

sh
pip install jutuldarcy

Examples ​

The examples are published in the documentation. For a list of examples categorized by tags, see the Example overview page.

To get access to all the examples as code, you can generate a folder that contains the examples locally, you can run the following code to create a folder jutuldarcy_examples in your current working directory:

julia
using JutulDarcy
generate_jutuldarcy_examples()

These examples can then be run using include("jutuldarcy_examples/example_name.jl") or opened in an editor to be run line by line.

Citing JutulDarcy ​

If you use JutulDarcy for a scientific publication, please cite the main paper the following way:

O. Møyner, (2024). JutulDarcy.jl - a Fully Differentiable High-Performance Reservoir Simulator Based on Automatic Differentiation. Computational Geosciences (2025), Open Access: https://doi.org/10.1007/s10596-025-10366-6

Show BibTeX
bibtex
@article{jutuldarcy,
  title={JutulDarcy.jl - a fully differentiable high-performance reservoir simulator based on automatic differentiation},
  author={M{\o}yner, Olav},
  journal={Computational Geosciences},
  volume={29},
  number={30},
  year={2025},
  publisher={Springer}
}

  • Fimbul.jl is built on top of JutulDarcy.jl and contains extensions and examples that cover advanced geothermal systems, including closed-loop and enhanced geothermal systems.

  • jutul-agent is an agentic interface that has specializations for JutulDarcy that allows setting up cases, visualization and interaction with other agents.

A few of the packages used by JutulDarcy ​

JutulDarcy.jl builds upon many of the excellent packages in the Julia ecosystem. Here are a few of them, and what they are used for:

...and many more directly, and indirectly - see the Project.toml and Manifest files for a full list!