Producers exporting covered goods to the European Union may already operate stack analyzers and assume that these readings can be transferred directly into a product carbon figure. That approach can fail because an instrument measures an emission source, while CBAM reporting ultimately concerns emissions embedded in specific goods. A properly designed monitoring and attribution system is needed to connect the two.

Yes. CBAM CEMS monitoring can support actual-emissions reporting through the measurement-based approach recognised in current European Commission guidance. It is not mandatory for every installation, because calculation-based and combined approaches are also allowed. Measured emissions must still be complete, verified and correctly attributed to the relevant production processes and goods.

The practical decision is therefore broader than purchasing a CO₂ analyzer. Operators must first define which sources belong inside the reporting boundary, how gas concentration becomes a mass-emission result and how installation-level totals will be allocated to production.

What Does the 2026 CBAM Framework Say About Direct Measurement?

Treating CEMS as either universally required or unsuitable can lead to the wrong monitoring investment. A CBAM monitoring plan should instead select the method that produces accurate and reliable results for the installation, subject to sector-specific requirements.

For the definitive period, the European Commission’s August 2026 guidance describes three possible arrangements:

• A calculation-based standard method using activity data and calculation factors.
• A calculation-based mass-balance method.
• A measurement-based methodology using CEMS at emission sources.
• A combination of calculation and measurement methods, provided that it creates neither gaps nor double counting.

The measurement approach is especially relevant where a facility already has a suitable emissions system or where direct monitoring better captures variable process emissions. It does not remove the need for production data, boundary definitions or verification.

When Can CEMS Be More Useful Than a Calculation-Based Method?

A plant may have excellent fuel-metering and laboratory data, making calculation the most proportionate choice. At another plant, changing fuel mixtures or significant process emissions may make direct measurement more representative.

A comparison can help:

At ESEGAS, we recommend comparing both methods before specifying instruments. Existing CEMS data should only be reused after confirming that its range, reference conditions, quality controls and boundaries meet the selected method.

What Must the CEMS Measurement Chain Determine?

A concentration trend alone cannot establish tonnes of greenhouse gas released. If production or exhaust flow changes, the same concentration may correspond to a different mass emission.

A measurement chain may need:

• CO₂ or another in-scope greenhouse-gas concentration.
• Flue-gas flow measurement or another accepted method of determining corresponding gas quantity.
• Temperature and pressure for reference-condition conversion.
• Moisture data or a documented dry/wet basis.
• Oxygen where it supports correction or plausibility assessment.
• Operating status for each included source.
• Time alignment between concentration, flow and production.
• Validity and quality-assurance flags.
• Calculation and reporting software.

All measurements must use compatible units and averaging periods. The data system should retain the source readings used to calculate the final mass result rather than storing only a rounded annual total.

How Should Installation and Production Boundaries Be Defined?

A reliable stack total can still produce the wrong product result if it covers the wrong sources. Installation emission boundaries must be established before analyzers, meters and production records are combined.

Operators should map:

1. Every included emission source.
2. Every calculation-based source stream.
3. Production processes that make covered goods.
4. Shared utilities and heat flows.
5. Waste-gas transfers.
6. Internal and purchased precursor materials.
7. Production quantities and functional units.
8. Sources or activities explicitly outside the applicable boundary.

The embedded emissions calculation begins with installation-level emissions, attributes relevant emissions to production processes and then incorporates applicable precursor information. The resulting value is divided by the relevant activity level to obtain a product-specific result.

CEMS therefore supplies one important data layer. It does not automatically perform allocation between several production lines or identify the embedded emissions of purchased precursors.

Which Greenhouse Gases May Need to Be Covered?

Assuming that every covered product requires the same gas list can create unnecessary instrumentation or leave a sector-specific component unmeasured. The applicable gas depends on the goods and production process.

Current European Commission guidance identifies:

• CO₂ for all CBAM sectors.
• N₂O additionally for the fertiliser sector.
• Perfluorocarbons additionally for aluminium.

This distinction matters because the analytical methods are different. N₂O is infrared-active and may be measured by a validated FTIR or dedicated infrared system. PFC measurement requires methods and calibration suited to the relevant compounds and aluminium process.

We would not add N₂O or PFC channels merely because the installation produces a CBAM good. The sector guidance, production route and approved monitoring approach should determine the final configuration.

How Can Operators Avoid Double Counting and Data Gaps?

Mixing measured and calculated data without a clear source map can count one emission twice or omit it completely. The risk is highest where several stacks serve shared equipment or when a monitored source also appears in the fuel calculation.

A combined method should document:

• Which emissions are measured directly.
• Which source streams use calculations.
• How common utilities are assigned.
• How overlapping sources are excluded.
• How bypasses and abnormal routing are handled.
• What happens during analyzer or flow-meter failure.
• How startup and shutdown are treated.
• Which data-gap substitution procedure applies.
• How method changes are approved and recorded.

Operators should test the combined result against fuel use, production and historical performance. A large unexplained difference should trigger investigation rather than an undocumented manual correction.

What Makes CEMS Data Suitable for Verification?

An annual spreadsheet cannot demonstrate whether individual readings were valid. Verification-ready emissions data requires evidence covering the entire measurement and calculation chain.

A quality system should retain:

• Original analyzer signals.
• Raw and corrected flow values.
• Calibration and maintenance records.
• Zero and span results.
• Instrument status and alarm flags.
• Correction factors and reference conditions.
• Data-validation decisions.
• Missing-data calculations.
• Software and configuration changes.
• Production and batch records used for attribution.
• Responsible-person approvals and audit trails.

Calibration gases, reference methods and test procedures should be appropriate to the gas, range and matrix. Where several instruments feed one calculation, their clocks and averaging periods must be aligned.

At ESEGAS, we also recommend preserving raw data when possible. If a verifier questions a correction or production allocation, the plant can then reconstruct the calculation without relying on an overwritten monthly total.

How Should Biomass and Mixed Fuels Be Addressed?

A stack analyzer measures total gas leaving the source; it does not by itself distinguish the origin of carbon atoms. This becomes important where biomass and fossil fuels are used together.

If the applicable rules allow qualifying biomass emissions to be treated differently, the plant needs supporting information beyond the CO₂ concentration. Depending on the approved method, this may involve:

• Fuel records and sustainability evidence.
• Biomass and fossil fractions.
• Laboratory analysis.
• Energy or material balances.
• A recognised method for determining the origin of emitted carbon.
• Clear linkage between supporting data and the monitored period.

Operators should not subtract a presumed biomass percentage directly from CEMS totals. The method and evidence must follow the applicable CBAM provisions and be available for verification.

Should Every Non-EU Producer Install CEMS for CBAM?

Installing a new system solely because CBAM has entered its definitive period may create cost without improving the result. Conversely, dismissing existing CEMS data may waste a valuable source of actual-emissions information.

CEMS is more likely to be appropriate when:

• Direct measurement is required for a relevant sector or process.
• Emissions vary in ways that calculations do not represent well.
• Suitable, representative emission points exist.
• Exhaust flow can be determined reliably.
• The plant already maintains compliant emissions infrastructure.
• Time-resolved data also supports process or energy improvement.
• The operator can maintain the required quality and availability.

A calculation-based approach may be more proportionate where source streams are clearly measured and accepted factors provide a reliable result. A hybrid approach may suit installations with both combustion and process emissions.

CBAM CEMS monitoring can support actual-emissions reporting, but the analyzer is only the first part of the evidence chain. At ESEGAS, we recommend defining the applicable method and reporting boundary before configuring the gas analyzer, flow measurement and data system. When measured emissions are complete, traceable and connected to production records, CEMS can provide both a defensible CBAM input and useful insight into plant performance.