Making federal hazard intelligence accessible to every community — regardless of technical capacity or budget.

→ Live Dashboard

The United States faces an accelerating disaster crisis. FEMA has declared over 3,400 major disasters since 2000, obligating more than $372 billion in federal aid. Yet the communities most at risk — small municipalities, rural counties, and underserved populations — often lack the technical capacity to interpret and act on the federal datasets that exist precisely to help them.

ResilienceMap is a free, open-source platform that closes this gap. It ingests authoritative federal datasets (FEMA, USGS, NOAA, US Census), computes composite risk scores per census tract, and delivers them through a public REST API and interactive map dashboard — with zero GIS expertise required on the consumer side.

These figures are not projections — they are live statistics computed from authoritative federal sources on every page load.

The US has world-class federal hazard datasets. The problem is access:

• FEMA NFHL flood zone data requires specialized GIS software to query
• USGS seismic hazard data is published as PDFs and shapefiles, not APIs
• NOAA NWS alerts are machine-readable but require parsing complex CAP/XML feeds
• CDC Social Vulnerability Index adds a critical equity dimension but lives in a separate silo

No single tool unified these datasets at the census-tract level with a public API. Emergency managers in small counties, researchers studying climate displacement, and journalists covering disaster policy all had to build their own integrations from scratch.

ResilienceMap provides the missing infrastructure layer.

• 📡 Live data — NOAA storm alerts, USGS earthquakes, NIFC wildfires refresh every 5 minutes
• 🗺️ Census-tract risk scoring — composite score weighted by flood, seismic, storm, wildfire, and social vulnerability
• 🔌 Public REST API — query risk by county FIPS code, retrieve active alerts, export GeoJSON
• 📊 Analytics dashboard — disaster declarations by year/type/state, federal spending trends
• 👥 Population exposure estimator — live estimate of people in active hazard zones by state
• 🔔 Configurable alerting — webhook/email triggers on NOAA watches and warnings
• 🐳 One-command deployment via Docker Compose
• 🔄 Automated data refresh — GitHub Actions pipeline keeps federal spending totals current monthly

# 1. Clone the repo
git clone https://github.com/henok256/resiliencemap.git
cd resiliencemap

# 2. Copy environment config
cp .env.example .env

# 3. Start all services (API + PostGIS + dashboard)
docker compose up --build

# 4. Run initial data ingestion
docker compose exec api python -m ingestion.fema.ingest_flood_zones
docker compose exec api python -m ingestion.usgs.ingest_earthquakes

# 5. Open the dashboard
open http://localhost:8000/dashboard

Full API docs available at http://localhost:8000/docs (Swagger UI).

┌─────────────────────────────────────────────────────────────────┐
│                         Data Sources                             │
│   FEMA NFHL  ·  USGS Earthquake  ·  NOAA NWS  ·  US Census     │
│   NIFC Wildfires  ·  HIFLD Infrastructure  ·  CDC/ATSDR SVI    │
└────────┬───────────────┬──────────────┬──────────┬──────────────┘
         │               │              │          │
         ▼               ▼              ▼          ▼
┌─────────────────────────────────────────────────────────────────┐
│                   Ingestion Layer (Python)                        │
│       GeoPandas · Requests · Scheduled via GitHub Actions        │
└──────────────────────────┬──────────────────────────────────────┘
                           │
                           ▼
┌─────────────────────────────────────────────────────────────────┐
│                PostGIS (PostgreSQL + spatial)                     │
│   census_tracts · flood_zones · seismic_hazard ·                 │
│   storm_alerts · risk_scores · disaster_costs                    │
└──────────────────────────┬──────────────────────────────────────┘
                           │
                           ▼
┌─────────────────────────────────────────────────────────────────┐
│                     FastAPI Backend                               │
│              Risk API · Hazard API · Alert API                   │
└──────────────────────────┬──────────────────────────────────────┘
                           │
                           ▼
┌─────────────────────────────────────────────────────────────────┐
│                  Leaflet.js Map Dashboard                         │
│   Live counters · Analytics charts · Export tools                │
└─────────────────────────────────────────────────────────────────┘

Documented in full at docs/methodology.md. The composite risk score per census tract is a weighted function of:

The social vulnerability component ensures the platform highlights not just physical exposure but community capacity to withstand and recover — a dimension consistently identified by FEMA as underrepresented in existing tools.

resiliencemap/
├── app/                    # FastAPI application
│   ├── api/routes/         # Endpoint handlers
│   ├── core/               # Config, logging, settings
│   ├── db/                 # Database session, migrations
│   ├── models/             # SQLAlchemy ORM models
│   └── schemas/            # Pydantic request/response schemas
├── ingestion/              # Data ingestion modules
│   ├── fema/               # FEMA NFHL flood zone + disaster costs ingestion
│   ├── usgs/               # USGS earthquake data ingestion
│   └── noaa/               # NOAA NWS storm alerts ingestion
├── processing/             # Risk scoring algorithms
├── dashboard/              # Leaflet.js frontend (local dev)
├── docs/                   # Live GitHub Pages dashboard + methodology
├── tests/                  # Unit and integration tests
├── scripts/                # DB seed, migration helpers
├── .github/workflows/      # CI + automated FEMA data refresh
├── docker-compose.yml
├── Dockerfile
└── pyproject.toml

Contributions are welcome. Please read CONTRIBUTING.md before opening a PR.

Areas where help is particularly valuable:

• GIS / geospatial engineers
• Emergency management professionals who can validate risk models
• Frontend contributors (Leaflet.js / data visualization)
• Researchers working on climate risk, disaster policy, or environmental justice

MIT — see LICENSE. Free to use, adapt, and deploy.

If you use ResilienceMap in research, policy work, or emergency planning, please cite:

@software{resiliencemap2025,
  title  = {ResilienceMap: Open-Source US Disaster Risk Intelligence Platform},
  author = {Mengesha, Henok},
  year   = {2025},
  url    = {https://github.com/henok256/resiliencemap},
  note   = {Free public dashboard: https://henok256.github.io/resiliencemap/}
}