wetland-suitability-ml-ahp-2025

Code supplement: Identifying suitable locations for in-stream wetlands using machine learning and AHP approaches

This repository contains the code and data references for the paper evaluating in-stream wetland suitability using Random Forest (RF) and Analytic Hierarchy Process (AHP) models, comparing spatial resolution and data source effects.

Folders and files

• data/: references to local and global input datasets (10 m and 50 m), and environment.yml for conda setup
• model_rf/: training, validation, and final prediction scripts for the RF model
• model_ahp/: scripts and matrices used for AHP-based suitability modelling
• validation/: ROC-AUC analysis and comparison against historical wetland maps