disslucc

Overview

disslucc brings land use and cover change (LUCC) modeling to Python, on top of DisSModel. It implements the components of INPE’s LuccME — demand, potential and allocation — in a raster-only, script-first package:

• Continuous allocation (CLUE-like) and discrete allocation (CLUE-S-like).
• Spatial-lag potential, saturation, and demand computed from land-use layers.
• Validation shared across both paradigms.

It is an alternative that runs next to TerraME/LuccME, not a replacement for them. Agreement with TerraME/LuccME — iteration counts, per-year error and timings — is checked in LambdaGeo/disslucc-benchmark, which pins the disslucc release it ran against.

Quick example

from dissmodel.core import Environment
from disslucc import DemandInline, PotentialLinearRegression, AllocationClueLike

demand = DemandInline(values=[...], land_use_types=["forest", "urban"])
potential = PotentialLinearRegression(backend=backend, demand=demand, ...)
allocation = AllocationClueLike(backend=backend, demand=demand, potential=potential, ...)

Environment(end_time=7).run()

The script is the complete experiment — reproducible with git clone, pip install and python script.py. When automatic provenance matters more than simplicity (production, CI), disslucc.executors provides ModelExecutor and ExperimentRecord as a second entry point, with the same math and identical results.

One repository for both paradigms

disslucc consolidates the former disslucc-continuous and disslucc-discrete repositories into a single package, mirroring how the original LuccME keeps continuous and discrete components side by side. Once the migration completes, the previous repositories will be archived and remain citable. See the DisSLUCC research project for the broader context.

Installation

pip install disslucc                 # from PyPI
pip install "disslucc[examples]"     # + matplotlib, for the examples