Flood Risk Prediction Dashboard

Full-stack web application for real-time flood risk prediction using a high-performance XGBoost model trained on All-India datasets.

Model V3 Upgrade (Latest)

The system has been upgraded to Model V3, featuring:

• Satellite Integration: Real-time soil moisture from NASA POWER API (GEOS-5/SMAP).
• Advanced ML: High-accuracy XGBoost model replacing the legacy Random Forest.
• 52 Features: Comprehensive feature engineering including rainfall acceleration, SMAP history, and geographic risk indices (TWI, CN).

Features

• Real-time Satellite & Weather Data: Combines Open-Meteo precipitation with NASA's satellite soil wetness data.
• 4-Tier Risk Classification: Low, Moderate, High, and Very High risk levels with professional iconography and alerts.
• Model Metrics Dashboard: Live performance indicators (AUC: 0.962, F1: 0.50) displayed on the user dashboard.
• Modern UI: Professional dark-themed dashboard with dynamic risk scales and interactive map integration.

---

Project Structure

FloodRiskPredictor/
├── backend/
│   ├── app.py                 # Flask API server
│   ├── model_loader.py        # ML model management (V3 support)
│   ├── weather_fetcher.py     # Weather & NASA API integration 
│   ├── feature_engineering_v3.py # 52-feature engineering pipeline
│   └── requirements.txt       # Python dependencies
├── models/
│   └── new_model/             # Model V3 weights and scalers
├── frontend/
│   ├── src/
│   │   ├── components/        # React components (Dashboard, Alerts, etc.)
│   │   ├── utils/             # API client
│   │   └── App.css            # Styles with 4-tier risk themes
└── README.md

---

Quick Start

Prerequisites

• Python 3.8+
• Node.js 16+
• npm or yarn

Backend Setup

# Navigate to backend directory
cd backend

# Install Python dependencies
pip install -r requirements.txt

# Run Flask server
python app.py

Backend will run on http://localhost:5000

Frontend Setup

# Navigate to frontend directory
cd frontend

# Install dependencies
npm install

# Start development server
npm run dev

Frontend will run on http://localhost:3000

---

API Endpoints

POST /predict

Predict flood risk for a city using Model V3 (52 features).

Request:

{
  "city": "Patna"
}

Response:

{
  "city": "Patna",
  "temperature": 28.5,
  "confidence": 87.5,
  "risk_level": "very_high",
  "risk_tier": 4,
  "model_version": "v3",
  "model_metrics": { "auc": 0.9623, "f1": 0.5032 }
}

GET /cities

Get list of available cities.

GET /health

Check server and model status.

---

Supported Cities

City	Latitude	Longitude
Patna	25.5941°N	85.1376°E
Guwahati	26.1445°N	91.7362°E
Gorakhpur	26.7606°N	83.3732°E
Malda	25.0104°N	88.1328°E
...And 20+ more major Indian cities

---

Technology Stack

Backend:

• Flask - API Framework
• XGBoost - Primary Machine Learning Model
• NASA POWER API - Satellite Soil Moisture (GWETROOT)
• Open-Meteo API - Current and Historical Rainfall

Frontend:

• React 18 - UI Library (Vite based)
• Axios - Data fetching
• Modern CSS - 4-tier risk styling and animations

---

ML Model (V3)

• Algorithm: XGBoost (Extreme Gradient Boosting)
• Feature Count: 52
• Metrics: 0.9623 AUC, 0.496 MCC
• Critical Features: SMAP 7-day Avg, Rainfall Acceleration, Soil Moisture Proxy, Topographic Wetness Index (TWI).

---

Usage

1. Select City: Choose from the dropdown to start live analysis.
2. View Metrics: Click the "Model Performance" panel to see live model accuracy stats.
3. Monitor Alerts: High and Very High risk alerts provide actionable emergency text.

---

Production Deployment

Backend (Python)

# Install Gunicorn
pip install gunicorn

# Run with Gunicorn
gunicorn -w 4 -b 0.0.0.0:5000 app:app

Frontend (React)

# Build for production
npm run build

# Serve static files with any web server
# dist/ folder contains built files

Environment Variables: Update API endpoint in frontend/src/utils/api.js to production URL.

---

Security Notes

• Add rate limiting for API endpoints
• Implement API key authentication for production
• Use HTTPS in production
• Add input validation and sanitization

---

Future Enhancements

• Multi-city monitoring dashboard
• Historical flood data visualization
• Email/SMS alerts for high risk
• Map-based city selection
• User authentication
• Prediction history tracking

---

License

This project is for educational and research purposes.

---

Contributing

Contributions welcome! Please open an issue or submit a pull request.

---

Support

For issues or questions, please open a GitHub issue.

---

Built for flood risk awareness and prevention