evoland visualizes, optimizes, localizes and and determines predictable land use shifts
The evoland-plus R package (or evoland for short) provides tools for analyzing and projecting land use evolution. The package implements a statistically calibrated, constrained model for predicting locations of future land use / land cover change (LULCC).
The fundamental purpose of evoland-plus is to:
- Gather and process land use/land cover data from multiple sources
- Calibrate statistical models to understand historical land use transitions
- Project future land use patterns using predictive models and allocation algorithms
- Support scenario analysis through interventions and parameter modifications
The package follows a three-phase workflow:
- Data Preparation: Marshalling heterogeneous data sources into a unified database structure
- Calibration: Feature selection, training Markovian transition models, and parameterizing allocation strategies
- Prediction + Allocation: Recursive model evaluation to generate future land use projections
This represents a re-implementation and enhancement of the land use model published at https://github.com/ethzplus/evoland-plus-legacy, building on substantial prior experience while modernizing the technical implementation.
You can use renv to set up a development environment with all dependencies.
When you run renv::init() from inside the project directory, you'll be asked to either install dependencies from the DESCRIPTION file, or to discover them from all project files.
The former should be faster and sufficient, given that dependencies should always be declared in DESCRIPTION.
During development, devtools::load_all() acts as though you were calling library(evoland) on an installed package.
This makes it easy to rapidly reload code when you've changed something.
This package uses pkgdown, see http://ethzplus.github.io/evoland-plus.
- Tutorials & How-to Guides: Package vignettes and examples in R man pages
- Reference Documentation: R man pages using the standard
?function_name - Explanation & Design Rationale: See the project wiki for
- Detailed explanation of the modeling approach
- Database schema and architecture decisions
- Development guidelines and coding standards
This project is licensed under the AGPL-3 License - see the LICENSE.md file for details.
Don't hesitate to get in contact with @mmyrte and/or @blenback if you'd like to contribute!
This work builds upon the foundational research and implementation available at https://github.com/blenback/LULCC-CH.