5.1.4 — Carbon Intelligence — maturity: live
Industrial Emissions Attribution
Pinpointing which industrial facilities — steel mills, refineries, cement plants, petrochemical complexes — are responsible for specific greenhouse gas plumes, using satellite-derived spectroscopy and thermal imaging.
Satellite spectroscopy now pinpoints which facility, flare stack, or furnace is responsible for excess industrial CO₂ — turning self-reported inventory guesswork into verifiable, plant-level accountability.
Governments signing up to net-zero commitments face an immediate credibility problem: self-reported facility emissions from heavy industry are systematically under-declared, sometimes by 40–70% relative to independent atmospheric measurements. Without an independent space-based check, regulators are negotiating carbon budgets and handing out allowances on the basis of figures that emitters themselves supply. The gap between reported and actual emissions is not an accounting rounding error — it is a structural policy failure that invalidates any downstream carbon market or border carbon adjustment.
A sovereign constellation couples shortwave-infrared (SWIR) spectrometers tuned to CO₂ and CH₄ absorption bands with thermal infrared (TIR) sensors that identify furnace and flare heat signatures, and optional synthetic aperture radar to confirm plant operational status. Data fused at ground level allows analysts to isolate individual stacks and attribute measured column concentrations to specific assets using Gaussian plume inversion and Lagrangian transport modelling. At ~30m spatial resolution and daily revisit, this stack moves emissions attribution from a statistical estimate to a facility-level fact.
The operational outcome is leverage: a regulator armed with satellite-derived attribution can challenge a steel mill's annual emissions declaration before the compliance window closes, not three years later in a court dispute. That leverage feeds directly into carbon pricing, trade-exposed sector policy and bilateral enforcement of carbon border adjustment mechanisms like the EU CBAM. Nations that depend on a commercial provider or a foreign government programme for these numbers are, in effect, outsourcing the integrity of their own industrial policy.
Frequently asked
Can a satellite actually tell which specific factory or power plant is responsible for a CO₂ plume?
Yes, with caveats. Instruments like NASA's OCO-3 and ESA's forthcoming CO2M can detect XCO₂ enhancements of 1–3 ppm above background at spatial resolutions down to 2–3 km², which is sufficient to distinguish a large steel mill from a neighbouring cement plant in most industrial zones. Attribution still requires cross-referencing the plume's downwind geometry with wind-field data and a facility registry — it is a probabilistic inference, not a direct read. For tightly clustered facilities, confidence intervals widen and ground-based supplementation becomes necessary.
How does satellite attribution compare to self-reported emissions under existing national registries?
Studies using Copernicus CAMS and GHGSat data consistently find discrepancies of 10–40% between satellite-derived estimates and operator-reported figures for individual industrial sites, with self-reporting skewing low. The EDGAR v8 global inventory (JRC, 2023) acknowledged structural underreporting in the metals and cement sectors. Satellite attribution does not replace self-reporting but acts as an independent cross-check that regulators — and increasingly carbon border adjustment authorities — can use to trigger audits.
Why should a government own this capability rather than just buying data from GHGSat or Planet?
Three reasons: custodial control, continuity, and negotiating leverage. A sovereign constellation means the government controls the tasking schedule (you can re-observe a suspicious facility tonight, not wait for a commercial revisit window), owns the raw calibration chain admissible in its own courts, and is never subject to a commercial vendor withdrawing service under foreign-government pressure. When a major emitter is also a major trading partner, the independence of your evidence source becomes a geopolitical asset.
What orbit is best for industrial emissions attribution?
Low Earth orbit (LEO), specifically sun-synchronous orbits at 500–600 km altitude, are the operational standard. They provide consistent illumination geometry for passive spectrometers, allow small-satellite constellations to achieve daily to near-daily revisit for major industrial corridors, and keep downlink latency low enough for near-real-time analysis. GEO would provide continuous stare but the angular resolution required for 2–5 km attribution footprints is not achievable from 35,786 km with affordable apertures.
How many satellites does a nation actually need to monitor its own industrial sector?
A minimum viable constellation for national-scale attribution — covering major industrial sites with 3–5 day cloud-free revisit — is typically 4–8 microsatellites in complementary sun-synchronous planes, based on the architecture demonstrated by GHGSat's 16-satellite fleet and ESA's two-satellite CO2M design. Larger nations with dispersed heavy industry (refineries, smelters, power plants across a continental landmass) may require 12–20 satellites or augmentation agreements with allied constellations to maintain consistent coverage.
Is satellite-derived CO₂ attribution data legally admissible as evidence in emissions trading enforcement?
This varies by jurisdiction. Within the EU ETS, the legal evidentiary basis remains operator-reported data verified under Regulation 2018/2067, but the European Commission's MRV reform discussions (2024–2025) are actively considering satellite data as a tier-2 cross-check trigger for inspections. In the absence of a national legal framework explicitly recognising satellite attribution, governments typically need to use the data to initiate regulatory audits rather than levy penalties directly. Establishing this legal pathway is a policy task, not a technical one, and is best done when the government owns the data rather than licences it.
Can satellite CO₂ attribution detect fraud in voluntary carbon markets (VCMs)?
Increasingly, yes. Attribution satellites can verify whether a facility claiming carbon credits from an efficiency upgrade actually shows a statistically significant reduction in its measured CO₂ column enhancement — independent of the methodology documents. Organisations such as the Integrity Council for the Voluntary Carbon Market (ICVCM) and Berkeley Carbon Trading Project have begun referencing satellite cross-checks as a best-practice MRV supplement. Sovereign ownership of this capacity gives a government independent audit rights over VCM projects operating within its borders.
What is the difference between CO₂ attribution and methane attribution, and do they require different satellites?
CO₂ and methane (CH₄) absorb infrared radiation at different spectral bands — CO₂ primarily around 1.6 µm and 2.0 µm, CH₄ around 1.65 µm and 2.3 µm — so a single hyperspectral instrument can in principle retrieve both, as demonstrated by GHGSat-D and the TROPOMI instrument on Sentinel-5P. However, the signal-to-noise requirements differ (methane enhancements from individual well pads are larger in relative terms than CO₂ from combustion), so optimised instruments differ in design. Many sovereign constellation architectures therefore plan for a combined CH₄/CO₂ payload to maximise science return per satellite.