Chapter 4: Satellite and Remote Sensing

Satellite vs. Ground Measurements: Complementary Strengths

Satellites excel at: Spatial coverage, source detection, regional budgets

Ground stations excel at: Precision, long-term trends, validation

Optimal approach: Combine both for comprehensive monitoring

How Passive Sensors Work

1. Sunlight source: Sun illuminates Earth's surface
2. Atmospheric path: Light passes through atmosphere (down and back up)
3. Absorption: GHG molecules absorb specific wavelengths
4. Reflection: Surface reflects remaining light
5. Satellite detection: Spectrometer measures received spectrum
6. Retrieval algorithm: Inverts spectrum to derive GHG concentration

Key concept: More absorption → less light reaching satellite → higher GHG concentration

Understanding XCO₂ and XCH₄

Definition: Total column amount of gas divided by total column of dry air

XCO₂ = (Total CO₂ column) / (Total dry air column)

Units: ppm (parts per million)

Example:
  Total CO₂ column: 8.8 × 10²¹ molecules/cm²
  Dry air column: 2.12 × 10²⁵ molecules/cm²
  XCO₂ = (8.8 × 10²¹) / (2.12 × 10²⁵) × 10⁶ ≈ 415 ppm

Why use XCO₂ vs. surface concentration?

• Insensitive to surface pressure variations
• Comparable to ground-based FTIR measurements
• Integrates entire atmospheric column

Retrieval Process Overview

1. Forward model: Simulate expected spectrum given atmosphere state
2. Comparison: Compare simulation with actual measurement
3. Iteration: Adjust atmospheric state to minimize difference
4. Convergence: Repeat until best fit achieved
5. Output: Retrieved GHG profile or column amount

Key challenges:

• Aerosol and cloud interference
• Surface reflectance variability
• Temperature and pressure profiles
• Spectroscopic uncertainties
• Instrument calibration drifts

Methane Plume Detection

Detection threshold (approximate): TROPOMI can detect CH₄ sources on the order of several tonnes per hour; high-resolution satellites such as GHGSat can detect sources on the order of several hundred kg per hour

Detection algorithm:

1. Calculate anomaly: ΔCH₄ = Observed - Background
2. Identify contiguous enhanced pixels
3. Apply statistical threshold (e.g., > 3σ)
4. Compute integrated mass enhancement (IME)
5. Estimate emission rate using wind speed

Emission rate (kg/hr) ≈ IME × U_eff / L
(U_eff = effective wind speed, L = plume length scale (square root of plume area))

Major sources detected:

• Oil & gas infrastructure leaks
• Coal mine ventilation shafts
• Large landfills
• Natural gas processing plants
• Pipeline ruptures

How Satellites Are Validated

1. Pre-launch: Laboratory calibration, algorithm testing on synthetic data
2. Commissioning: Initial on-orbit performance verification
3. Ground-based FTIR: TCCON provides primary validation (XGas)
4. Aircraft campaigns: Vertical profiles for retrieval validation
5. In-situ networks: Surface and tower data for flux inversions
6. Inter-satellite comparison: Cross-validation between missions

Inverse Modeling Concept

Forward problem: Fluxes → Concentrations (atmospheric transport)

Inverse problem: Concentrations → Fluxes (infer sources/sinks)

Basic equation:

y = H × x + ε

where:
  y = observations (satellite XCO₂)
  x = state vector (fluxes at surface)
  H = observation operator (transport model)
  ε = observation error

Goal: Find x that best fits y

Solution approach (Bayesian):

x̂ = xb + BH^T(HBH^T + R)^-1(y - Hxb)

where:
  xb = prior fluxes (best guess)
  B = prior error covariance
  R = observation error covariance
  x̂ = optimized (posterior) fluxes

Standard Components for Satellite Data

Data ingestion API: The SDK call below is an example that wraps the specification's Phase 3 §8 REST endpoint GET /api/v1/ghg/concentration.

// Fetch satellite XCO2 data for region
const satelliteData = await wia.satellite.query({
  mission: 'OCO-2',
  gas: 'CO2',
  product: 'XCO2',
  level: 'L2',
  bbox: [lon_min, lat_min, lon_max, lat_max],
  date_range: ['2025-01-01', '2025-12-31'],
  quality_filter: 'good'
});

// Returns standardized GeoJSON with metadata
{
  type: 'FeatureCollection',
  mission: 'OCO-2',
  processing_version: 'v10r',
  features: [...]
}

Multi-mission fusion:

• Harmonize different satellite products
• Bias correction to common reference
• Optimal averaging based on uncertainties
• Gap-filling with data fusion techniques

Quality assurance:

• Automated comparison with TCCON
• Inter-satellite consistency checks
• Anomaly detection and flagging
• Uncertainty propagation through processing chain