SEALS

SEALS (Spatial Economic Allocation Landscape Simulator) downscales coarse or regional land-use change projections to fine-resolution (300 meter) LULC maps globally in less than 30 minutes on a laptop.

All documentation are generated from QMDs in docs/ and are hosted at the Seals Homepage, and the User Guide, which includes pages for installation , quickstart, a guided first run, the scenarios CSV reference, and several model linkage exercises.

Install Summary

See Installation for detailed instructions.

1. First install Hazelbean via condaforge

conda create -n <your_env>
conda activate <your_env>
conda install hazelbean

2. Clone this repository:

git clone https://github.com/jandrewjohnson/seals

3. Install cloned SEALS repository. In the root of the repository, with <your_env> activated, run:

pip install -e . --no-deps

--no-deps assumes the dependencies are already in the environment from conda. Note that running SEALS on a Windows PC requires having a C/C++ compiler installed, which is described in more details on the installation page.

Run it

conda activate <your_env>
cd seals/seals
python run_seals.py

Outputs land in a Project file just outside of the repo, e.g. at ~/Files/seals/projects/seals/, containing input/, intermediate/, output/, and one folder per task. Missing base data is downloaded on demand.

Run it a second time and it finishes almost immediately — every task is behind an existence check, so completed work is skipped.

For a faster first run, run_seals_test.py uses the same task tree with a pared scenarios CSV (baseline plus one BAU, one projection year, Rwanda).

Repository layout

Hazelbean, ProjectFlow and the Earth-Economy Devstack

SEALS relies on Hazelbean 2.0.0 or later, which is a Python package that provides high-performance geospatial functions and ProjectFlow, which enables parallel computation of a task tree. Hazelbean, and the rest of the Earth-Economy Devstack is documented in the devstack docs.