C-Star
C-Star is an open-source system for building, running and sharing regional ocean simulations. It is developed by ocean and biogeochemical modelers and scientific software engineers at [C]Worthy to support Monitoring, Reporting and Verification (MRV) of ocean-based carbon dioxide removal. Today it runs UCLA-ROMS coupled to the MARBL biogeochemistry model, on a laptop or on a supported HPC system, from the same description of the run.
How it works
C-Star is organized around two ideas. An application is something you can run: the ROMS-MARBL model, Forge (the tool that builds a new ROMS-MARBL domain), or a smaller data-transformation or analysis step. A blueprint is a YAML file holding every input an application needs to produce its result: which model code to build, which input files to use, which settings to apply. Given the same blueprint, an application produces the same result on any supported machine, which is what makes a C-Star simulation shareable and reproducible.
A simple ocean modeling project moves through three steps:
- Describe the domain. The Forge wizard, a form that runs in your browser or as a Jupyter notebook, walks you through choosing a model configuration, a region and grid, forcing datasets, and a run window. It writes a forge blueprint.
- Generate the inputs. Running that forge blueprint downloads the source datasets, builds the grid, initial conditions, boundary and surface forcing, and rivers and tides where requested, and writes the ROMS namelist. Its output is a ROMS-MARBL blueprint pointing at everything the model needs.
- Run the simulation. Running the ROMS-MARBL blueprint fetches and compiles the model code and executes the simulation.
To run several simulations as one unit, a workplan lists blueprints as steps with dependencies between them. C-Star schedules the steps, submits them to the cluster's job scheduler, and tracks their status, so a chain such as a spin-up followed by an experiment and its control can be launched and monitored with a single run ID. Reusable pieces of a domain description (model configurations, domains, forcing selections, output settings and saved blueprints) live in a catalog; C-Star ships a small bundled catalog to start from.
Installation
C-Star is published on conda-forge
as cstar-ocean. One install gives you the cstar command line, the
ROMS-MARBL application and Forge, including its wizard.
On a laptop or workstation, install cstar-ocean-standalone, which bundles
the compilers, MPI and netCDF toolchain that ROMS needs:
conda create -n cstar-env -c conda-forge cstar-ocean-standalone
conda activate cstar-env
cstar --version
On a supported HPC system, install cstar-ocean alone and let the site's
environment modules provide the toolchain. The
installation guide
covers the HPC variant, installing from source, and what to set up after
installing.
Quick start
The three steps above, on a laptop, with a deliberately tiny Western Indian Ocean domain that processes in minutes:
# 1. Build a forge blueprint in your browser; save or download it when done.
cstar forge wizard
# 2. Generate the domain. The last lines of output name the ROMS-MARBL
# blueprint Forge wrote and the command to run it.
cstar blueprint run forge_blueprint.yaml
# 3. Compile ROMS-MARBL and run the simulation.
cstar blueprint run ~/cstar/blueprint_runs/forge/<name>/blueprints/B_<name>.yaml
The same command runs every blueprint; the blueprint's application field
decides which application handles it. GLORYS ocean reanalysis supplies the
initial and boundary conditions, so
register for Copernicus Marine
before the first run. The
end-to-end tutorial
walks through each step in detail.
Documentation
The full documentation lives at https://c-star.readthedocs.io.
- Getting started: installation, dataset registration and configuration
- Terminology and concepts
- User guide: blueprints, workplans, catalog, wizard
- Domain generation with Forge
- Supported systems and running on HPC
- API reference
- Contributor guide
Related projects
- UCLA-ROMS: the Regional Ocean Modeling System with MARBL biogeochemistry, the model C-Star builds and runs
- ROMS-Tools: generates ROMS input files from source datasets and analyzes model output; Forge drives it
- MARBL: the Marine Biogeochemistry Library coupled to ROMS
Feedback and contributions
Bug reports and feature requests are welcome as GitHub Issues. We also accept contributions as pull requests; the contributor guide covers the development environment, the test suite and building the documentation.
License
C-Star is openly available and permissively licensed under the Apache License 2.0. Copyright 2026 [C]Worthy LLC.
Metadata
Release files for cstar-ocean 0.16.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| cstar_ocean-0.16.0.tar.gz | 3.7 MB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| cstar_ocean-0.16.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 5.3 MB
Release files / cstar_ocean-0.16.0.tar.gz
| Download URL | cstar_ocean-0.16.0.tar.gz |
|---|---|
| Size | 3.7 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
9a0ec0d7b3756c5507504ced1f833ad60c6d49d3a7a7dbc57ce5f904c943d44f
|
|
BLAKE2b-256 checksum How to use checksums |
b9c8ce485fc890cc4047150c2632844a55589d543ca4d6625b04ec231bec1d3c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 8, 2026.
Transparency logRelease files / cstar_ocean-0.16.0-py3-none-any.whl
| Download URL | cstar_ocean-0.16.0-py3-none-any.whl |
|---|---|
| Size | 1.6 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
a301ad529d7091b8927abce96efa4e89a3393fbd6da9b040262ad444ab83eb17
|
|
BLAKE2b-256 checksum How to use checksums |
0dacfc28ccdeefd5e813b9ec966d3cca0a230a6cb646d0dcace93b83984fd6a9
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 8, 2026.
Transparency log