Skip to content

TEMPO_MULTI_GAS_SUMMARY.md

Latest commit

 

History

History
359 lines (291 loc) · 11.9 KB

TEMPO_MULTI_GAS_SUMMARY.md

File metadata and controls

359 lines (291 loc) · 11.9 KB
Raw

TEMPO Multi-Gas Enhancement Summary

Overview

Enhanced the TEMPO satellite data integration to fetch multiple gases (NO2 and O3), estimate AQI from satellite measurements, and display the data in an OpenWeather-compatible format with full health recommendations support.

Changes Made

Backend Changes

1. New Function: estimate_aqi_from_gas_concentrations()

Location: backend/app/services.py (lines 669-731)

Purpose: Converts TEMPO tropospheric column density measurements to ground-level concentrations and estimates AQI.

Conversion Logic:

  • NO2: Converts molecules/m² to μg/m³ using scaling factor (value / 1e15) * 60
  • O3: Converts molecules/m² to μg/m³ using scaling factor (value / 1e15) * 75
  • AQI Estimation: Uses EPA breakpoints to calculate sub-indices for each pollutant
  • Final AQI: Returns the maximum of all sub-indices (worst pollutant drives overall AQI)

Returns:

(aqi: int, pollutants: Dict[str, float])
# Example: (3, {'no2': 85.3, 'o3': 75.2, 'pm2_5': 0, 'pm10': 0, 'co': 0, 'so2': 0, 'nh3': 0})

2. New Function: fetch_tempo_multi_gas()

Location: backend/app/services.py (lines 734-873)

Purpose: Fetches NO2 and O3 data from TEMPO satellite and formats the response like OpenWeather API.

Key Features:

  • Fetches multiple gases (NO2, O3TOT) in parallel
  • Handles partial data (continues if one gas fails)
  • Estimates AQI and pollutant concentrations
  • Returns OpenWeather-compatible format with additional tempo_details section

Response Format:

{
  "success": true,
  "source": "NASA TEMPO Satellite",
  "coordinates": {"lat": 40.7127, "lon": -74.0060},
  "data": {
    "list": [{
      "dt": 1696598400,
      "main": {"aqi": 3},
      "components": {
        "no2": 85.3,
        "o3": 75.2,
        "pm2_5": 0,
        "pm10": 0,
        "co": 0,
        "so2": 0,
        "nh3": 0
      }
    }]
  },
  "tempo_details": {
    "data_date": "2025-09-16",
    "elevation_m": 10.0,
    "measurements": {
      "NO2": {
        "value": 893667669428336.8,
        "unit": "molecules/cm²",
        "scientific_notation": "8.94e+14"
      },
      "O3TOT": {
        "value": 1.2e+15,
        "unit": "molecules/cm²",
        "scientific_notation": "1.20e+15"
      }
    },
    "note": "AQI estimated from satellite gas measurements...",
    "metadata": {
      "satellite": "TEMPO (Tropospheric Emissions: Monitoring of Pollution)",
      "coverage_area": "Continental United States",
      "temporal_resolution": "Hourly during daylight",
      "spatial_resolution": "~2.1 km x 4.4 km"
    }
  }
}

3. New API Endpoint

Location: backend/app/routes.py (lines 114-133)

Endpoint: GET /air-pollution/tempo-current

Parameters:

  • lat (float, required): Latitude coordinate
  • lon (float, required): Longitude coordinate
  • start_date (string, required): Start date in format YYYY-MM-DD
  • end_date (string, required): End date in format YYYY-MM-DD

Example Request:

GET /air-pollution/tempo-current?lat=40.7127&lon=-74.0060&start_date=2025-09-06&end_date=2025-10-06

Frontend Changes

1. Updated TEMPO Data Fetching

Location: frontend/src/pages/Index.tsx (lines 80-125)

Changes:

  • Switched from /air-pollution/tempo to /air-pollution/tempo-current
  • Removed gas parameter (now fetches multiple gases automatically)
  • Updated toast messages to mention "NO2 and O3 measurements"
  • Check for tempo_details.data_date instead of data_date

2. New Display Format (OpenWeather-like)

Location: frontend/src/pages/Index.tsx (lines 721-833)

Features:

  • AQI Display: Large AQI number with description and icon (same as OpenWeather)
  • Pollutant Cards: Shows NO2 and O3 with "From TEMPO" badge
  • Raw Measurements: Collapsible section showing tropospheric column density in scientific notation
  • Metadata: Displays satellite info, data date, and elevation
  • Satellite Icon: Blue satellite icon to distinguish from OpenWeather data

Visual Example:

╔══════════════════════════════════════╗
║   🛰️  Estimated AQI from Satellite  ║
║                                      ║
║              3                       ║
║             AQI                      ║
║   Moderate - Sensitive groups...    ║
║              [icon]                  ║
╠══════════════════════════════════════╣
║  Pollutant Concentrations (μg/m³)   ║
║  (Satellite-derived estimates)       ║
║                                      ║
║  ┌────────┐  ┌────────┐            ║
║  │ NO₂    │  │ O₃     │            ║
║  │ 85.3   │  │ 75.2   │            ║
║  │ TEMPO  │  │ TEMPO  │            ║
║  └────────┘  └────────┘            ║
╠══════════════════════════════════════╣
║  ▶ Raw Satellite Measurements       ║
║  (expandable)                        ║
╠══════════════════════════════════════╣
║  Satellite: NASA TEMPO               ║
║  Data Date: 2025-09-16              ║
║  Elevation: 10 m                     ║
║  Note: AQI estimated from...        ║
╚══════════════════════════════════════╝

3. Health Info Tab Integration

Location: frontend/src/pages/Index.tsx (lines 863-891)

Changes:

  • Unified Data Handling: Both OpenWeather and TEMPO now use data.list[0] structure
  • Health Recommendations: TEMPO data now provides health advice (with disclaimer)
  • Satellite Badge: Blue info box appears below health recommendations when using TEMPO
  • Disclaimer: "AQI estimated from tropospheric column density. Actual ground-level may vary."

Benefits:

  • ✅ TEMPO users now get personalized health recommendations
  • ✅ LLM-generated advice works with TEMPO data
  • ✅ Clear indication that AQI is satellite-estimated

How It Works

Data Flow

  1. User selects TEMPO data source → Frontend changes data source to "tempo"
  2. User selects location → Frontend calls /air-pollution/tempo-current
  3. Backend searches for data → Searches backwards up to 30 days for available satellite data
  4. Backend fetches NO2 & O3 → Downloads data from NASA Earthdata
  5. Backend converts measurements → Converts tropospheric column density to ground-level μg/m³
  6. Backend estimates AQI → Calculates AQI from converted concentrations
  7. Backend returns formatted data → Returns OpenWeather-compatible structure
  8. Frontend displays data → Shows AQI, pollutants, and satellite details
  9. Health Info works → HealthInfoTab receives AQI and pollutant data like OpenWeather

AQI Estimation Algorithm

# 1. Convert NO2 from molecules/m² to μg/m³
no2_ugm3 = (no2_molecules_m2 / 1e15) * 60

# 2. Convert O3 from molecules/m² to μg/m³  
o3_ugm3 = (o3_molecules_m2 / 1e15) * 75

# 3. Calculate NO2 AQI sub-index (EPA breakpoints)
if no2_ugm3 <= 53:    no2_aqi = 1  # Good
elif no2_ugm3 <= 100: no2_aqi = 2  # Fair
elif no2_ugm3 <= 360: no2_aqi = 3  # Moderate
elif no2_ugm3 <= 649: no2_aqi = 4  # Poor
else:                 no2_aqi = 5  # Very Poor

# 4. Calculate O3 AQI sub-index
if o3_ugm3 <= 54:     o3_aqi = 1  # Good
elif o3_ugm3 <= 70:   o3_aqi = 2  # Fair
elif o3_ugm3 <= 85:   o3_aqi = 3  # Moderate
elif o3_ugm3 <= 105:  o3_aqi = 4  # Poor
else:                 o3_aqi = 5  # Very Poor

# 5. Overall AQI = worst pollutant
overall_aqi = max(no2_aqi, o3_aqi)

Comparison: Before vs After

Before

Feature Status
Gases Fetched NO2 only
AQI ❌ Not available
Pollutant Concentrations ❌ Only raw column density
Health Recommendations ❌ Not available
Display Format Custom satellite display
Compatibility TEMPO-specific

After

Feature Status
Gases Fetched NO2 + O3
AQI ✅ Estimated (1-5 scale)
Pollutant Concentrations ✅ Converted to μg/m³
Health Recommendations ✅ Full support with LLM advice
Display Format OpenWeather-compatible
Compatibility Works with existing components

Usage

Backend API

# Fetch TEMPO multi-gas data with AQI estimation
curl -X GET "http://localhost:8000/air-pollution/tempo-current?lat=40.7127&lon=-74.0060&start_date=2025-09-06&end_date=2025-10-06"

Frontend

  1. Select NASA TEMPO as data source
  2. Search for a US location
  3. Click "Use My Location" or select from search
  4. View:
    • Estimated AQI with icon
    • NO2 and O3 concentrations
    • Raw satellite measurements
    • Personalized health recommendations

Important Notes

Accuracy Considerations

⚠️ AQI Estimation Disclaimer:

  • AQI is estimated from tropospheric column density
  • Conversion factors are approximations based on typical atmospheric conditions
  • Actual ground-level concentrations may differ
  • TEMPO measures vertical column, not surface-level
  • More accurate for relative trends than absolute values

Data Limitations

📊 TEMPO Data Characteristics:

  • Coverage: Continental US only
  • Temporal: Hourly during daylight hours
  • Processing Delay: 2-3 days typical
  • Resolution: ~2.1 km x 4.4 km
  • Gases Available: NO2, O3, HCHO (currently using NO2 + O3)

When to Use TEMPO vs OpenWeather

Use TEMPO when:

  • ✅ Location is in continental US
  • ✅ Need satellite-based measurements
  • ✅ Want to see tropospheric column density
  • ✅ Interested in spatial patterns across regions
  • ✅ OK with 2-3 day data delay

Use OpenWeather when:

  • ✅ Need real-time data
  • ✅ Location is outside US
  • ✅ Want ground-level sensor measurements
  • ✅ Need all pollutants (PM2.5, PM10, CO, SO2, etc.)
  • ✅ Prefer validated AQI calculations

Future Enhancements

Potential Improvements

  1. Add HCHO (Formaldehyde)

    • Include third gas in multi-gas fetch
    • Update AQI estimation with HCHO sub-index

  2. Improve Conversion Factors

    • Use atmospheric profile data
    • Account for seasonal variations
    • Consider boundary layer height

  3. Validation

    • Compare with ground stations when available
    • Adjust conversion factors based on validation

  4. Historical Trends

    • Fetch multiple dates
    • Show temporal trends
    • Compare with previous weeks/months

  5. PM2.5 Estimation

    • Research potential PM2.5 proxies from TEMPO
    • Explore satellite-based PM estimation

Testing

Test the New Endpoint

# Start backend
cd backend
python -m uvicorn main:app --reload --host 0.0.0.0 --port 8000

# Test TEMPO multi-gas endpoint
curl -X GET "http://localhost:8000/air-pollution/tempo-current?lat=40.7127&lon=-74.0060&start_date=2025-09-06&end_date=2025-10-06" | jq

Expected Response

{
  "success": true,
  "source": "NASA TEMPO Satellite",
  "data": {
    "list": [{
      "main": {"aqi": 3},
      "components": {
        "no2": 85.3,
        "o3": 75.2,
        ...
      }
    }]
  },
  "tempo_details": {
    "measurements": {
      "NO2": {...},
      "O3TOT": {...}
    },
    ...
  }
}

Conclusion

✅ TEMPO data is now fully integrated with the rest of the application ✅ Health recommendations work with satellite-estimated AQI ✅ Display format matches OpenWeather for consistency ✅ Users get useful air quality info even with satellite data

The TEMPO satellite data source is now a first-class citizen in the application, providing estimated AQI and health recommendations just like OpenWeather, while maintaining the unique satellite measurement details for users who want to explore the raw data.