Python tools for cloud-native coastal analytics.
You can install coastpy with pip in a Python environment where GDAL and pyproj are already installed.
pip install coastpy
However, if you start from scratch, it's probably easier to install with conda:
conda env create -f https://raw.githubusercontent.com/TUDelft-CITG/coastpy/refs/heads/main/environment.yaml
The data that is produced with this software can be directly accessed via the cloud using tools like DuckDB; see the tutorials and analytics other for other access methods (Python) and latest usage instructions.
Cross-shore coastal transects are essential to coastal monitoring, offering a consistent reference line to measure coastal change, while providing a robust foundation to map coastal characteristics and derive coastal statistics thereof.
The Global Coastal Transect System consists of more than 11 million cross-shore coastal transects uniformly spaced at 100-m intervals alongshore, for all OpenStreetMap coastlines that are longer than 5 kilometers.
# Download all transects located in the United States.
duckdb -c "COPY (SELECT * FROM 'az://coclico.blob.core.windows.net/gcts/release/2024-08-02/*.parquet' AS gcts WHERE gcts.country = 'US') TO 'United_States.parquet' (FORMAT 'PARQUET')"
# Download transects by bounding box.
duckdb -c "COPY (SELECT * FROM 'az://coclico.blob.core.windows.net/gcts/release/2024-08-02/*.parquet' AS gcts WHERE bbox.xmin <= 14.58 AND bbox.ymin <= -22.77 AND bbox.xmax >= 14.27 AND bbox.ymax >= -23.57) TO area_of_interest.parquet (FORMAT 'PARQUET')"
# Or, download the data in bulk using AZ CLI
az storage blob download-batch \
--destination "./" \
--source "gcts" \
--pattern "release/2024-08-02/*.parquet" \
--account-name coclico
The Coastal Grid dataset provides a global tiling system for coastal analytics. It supports scalable data processing workflows by offering coastal tiles at varying zoom levels (5, 6, 7, 8, 9, 10) and buffer sizes (500 m, 1000 m, 2000 m, 5000 m, 10000 m, 15000 m).
Better installation and usage instructions will come when we build the documentation. For now, to run the tutorials, analytics or scripts proceed as follows.
- Install Git or GitHub Desktop
- Clone CoastPy
- Install Miniforge
- Open a Miniforge prompt (finder/spotlight)
- Run the following commands:
# Make sure to update if you already had miniforge installed
conda update --all --yes
# Create the software environment
conda env create -f https://raw.githubusercontent.com/TUDelft-CITG/coastpy/refs/heads/main/environment.yaml
- Open Miniforge
- Change to the directory where CoastPy was cloned by using
cd /path/to/coastpy
- Activate the software environment by
mamba activate coastal
- Launch Jupyter lab by
jupyter lab
- Navigate to the tutorials folder in Jupyter lab
@article{CALKOEN2025106257,
title = {Enabling coastal analytics at planetary scale},
journal = {Environmental Modelling & Software},
volume = {183},
pages = {106257},
year = {2025},
issn = {1364-8152},
doi = {https://doi.org/10.1016/j.envsoft.2024.106257},
url = {https://www.sciencedirect.com/science/article/pii/S1364815224003189},
}
Interested in contributing? Check out the contributing guidelines. Please note that this project is released with a Code of Conduct. By contributing to this project, you agree to abide by its terms.
coastpy
was created by Floris Calkoen. The software is licensed under the terms of the
MIT license. Data licenses are typically CC-BY-4.0, and can be found in the respective
STAC collection.