Ethel-Ogallo/Wetland_Inundation_DE-Africa

A comprehensive workflow analyzing wetland dynamics in the Zambezi Delta using Digital Earth Africa ## Wetland Dynamics Analysis in the Zambezi Delta This repository contains a comprehensive workflow for analyzing wetland dynamics in the Zambezi Delta region of Mozambique using satellite data from the Digital Earth Africa (DE Africa) STAC Catalog. The analysis leverages Landsat data to derive key environmental indicators that reveal spatial and temporal changes in wetland characteristics over a 7-year period (2017–2024). ### Overview Wetlands are critical ecosystems providing essential services including flood control, water filtration, carbon sequestration, and biodiversity habitat. This project monitors and analyzes wetland dynamics by: - **Temporal Coverage**: 2017–2024 (7 years of continuous monitoring) - **Study Area**: Zambezi Delta, Mozambique - **Satellite Data**: Landsat imagery via DE Africa STAC Catalog - **Key Indicators**: - **Tasseled Cap Wetness (TCW)**: Measures soil and vegetation moisture - **Water Observations from Space (WOfS)**: Detects surface water presence - **Fractional Cover (FC)**: Assesses vegetation density and bare soil proportion ### Repository Structure ``` Wetland_Inundation_DE-Africa/ ├── STAC- wetland_inundation.ipynb # Main analysis notebook ├── get_product_config.py # DE Africa product config utility ├── env.yml # Conda environment specification ├── Paper_Ogallo.pdf # Full seminar paper with methodology & results ├── output/ # Generated visualizations │ └── wetland_dnamics_monthly.gif # Animated wetland dynamics └── README.md ``` ### Main Workflow (`STAC- wetland_inundation.ipynb`) The notebook implements the following analysis pipeline: 1. **Data Acquisition**: Queries and loads Landsat data via STAC API 2. **Preprocessing**: Applies cloud masking and data cleaning 3. **Index Calculation**: Computes TCW, WOfS, and FC indicators 4. **Temporal Analysis**: Aggregates data to monthly/seasonal patterns 5. **Visualization**: Generates maps, time ser …