How indigowizard/wildfire-burn-severity works
Project type and purpose
• This is a single-page web application built with Streamlit.
File structure
• Top-level:
• .github (CI/docs)
• .streamlit/config.toml (Streamlit settings)
• app.py (entry point for Streamlit UI)
• requirements.txt (Python dependencies)
• src/ (helper modules and assets, including the GIF preview)
• No separate “frontend” vs. “backend” folders—both UI and business logic live in app.py and src/.
• External configuration (.streamlit) and standard Python packaging files (.gitignore, LICENSE).
README for additional insights
• Confirms use of Streamlit, GEE Python API, geemap, Folium.
• Describes user workflows: upload GeoJSON or click map, select dates/cloud filter, view map layers via Layer Control.
• Mentions two versions (Streamlit is primary).
How to represent this architecture
a. Main components
1. Streamlit Frontend (app.py)
2. Business Logic / Data Processing (modules under src/)
3. External Service: Google Earth Engine API
4. Visualization Libraries: geemap, Folium
5. Environment / Runtime: Python 3.x container or VM (with dependencies from requirements.txt)
6. (Optional) Configuration: .streamlit/config.toml
b. Relationships & data flows
• User interacts with the Streamlit UI →
• UI passes parameters to processing modules →
• Processing modules call GEE API →
• GEE returns satellite imagery (ImageCollections) →
• Modules compute NBR/NDWI indices →
• Modules feed results to geemap/Folium to build map layers →
• Streamlit renders the interactive map back to the user
c. Architectural patterns & design principles
• Monolithic single‐process application (no microservices).
• Clear separation of concerns:
• Presentation layer (Streamlit)
• Domain layer (NBR/NDWI calculations in src/)
• External data layer (GEE API)
• Dependency management via requirements.txt
d. Key technologies & frameworks
• Frontend/UI: Streamlit
• Mapping/Visualization: geemap, Folium
• Data service: Google Earth Engine Python API
• Supporting libs: geopandas (for GeoJSON), numpy, pandas (if used)