Enhancing Greenhouse Gas Emission Monitoring in the European Emission Trading Scheme:
A Focus on ISO 10723 Performance Evaluation of Gas Chromatographs in Natural Gas Plants

The European Commission has adopted legislative proposals aimed at achieving a 55% carbon emission reduction by 2030 and the climate neutrality by 2050. Within this framework, industrial installations under the Emissions Trading Scheme are mandated to develop approved plans for monitoring and reporting annual carbon emissions, as an operating permit requirement. This article provides a comparative analysis of various methods available for monitoring greenhouse gas emissions, focusing particularly on ISO 10723 Natural gas – Performance evaluation for analytical systems for validating process gas chromatographs. This standard ensures a highly accurate approach for determining calorific values and carbon emission factors required to compute site-specific carbon emissions.
The analytical equipment performance assessment is commonly advocated across the industry and, in certain instances, is mandated by legal or contractual obligations. This practice aims to guarantee the traceability and precision of gas measurements, with defined uncertainty levels, thereby substantiating adherence to established legal or contractual standards (e.g. fiscal transfer measurements, safety requirements, emissions monitoring).

European Emission Trading Scheme

The European Emission Trading Scheme (EU-ETS)1  has proven effective in reducing greenhouse gas (GHG) emissions, with historical data indicating a 40% reduction from 2007 to 2021 (figure 1). However, additional measures are essential to meet long-term climate objectives, as outlined in the European Green Deal, which introduces new requirements such as the European Climate Law and the Fit for 55 Package.
Figure 1 - EU-28 GHG emission trends – EEA Emission Inventory Report 2007–2021

The EU-ETS, established in 2005 as the world’s first carbon trading scheme, operates on “the polluter pays” principle, placing a price on carbon emissions and using pollution allowances companies purchase and trade based on their CO2 emissions. The EU-ETS sets caps on annual emissions, which are reduced regularly over time, contributing significantly to emission reduction. Participation is mandatory for energy-intensive industry sectors, like energy, manufacturing and other industrial processes which all together represent approximately 45% of the total emissions in the EU (figure 2).

 Figure 2 – GHG emissions in the EU by sector in 2021 – Eurostat

The Monitoring and Reporting Regulation (MRR)2  of the EU-ETS mandates a robust, transparent, consistent and accurate monitoring and reporting system. The MRR provides information on the utilisation of process gas chromatographs (GC) and the ISO 170253  prerequisites for calibration gases, while specifying the performance requirements for the utilised instruments and calibration frequency. Industrial installations must have an approved monitoring plan, which is part of the permit to operate, and they must annually submit emission reports – verified by accredited verifiers – before surrendering the corresponding number of allowances.
A revision of the EU-ETS is underway, which will reduce the quantity of available permits, and eliminate the provision of free allowances. The scope of the scheme is set to expand to encompass other high-emission sectors, including shipping, road transport, aviation and buildings. This revision also entails increased funding allocated to innovation and modernisation initiatives, accelerating the pace of the green transition.

Greenhouse Gas Emissions Monitoring

For combustion plants using natural gas, relevant CO2 emissions can be obtained through either a measurement-based method – using Continuous Emissions Monitoring Systems (CEMS) – or a calculation-based method. The calculation-based method is ten times more accurate than CEMS, which can have a mass CO2 uncertainty up to approximately 15% relative when measuring hot and wet flue gases.
Carbon emissions calculation according to the MRR:

CO2 Emission = Flow × NCV × CEF × Oxidation Factor

•     CO2 Emission can be expressed as tCO2/TJ (Energy), tCO2/t (Mass) or tCO2/Nm3 (Volumetric)
•     Flow of natural gas is representative of the activity data
•     Net Calorific Value (NCV) is the amount of heat released by the complete combustion with oxygen, under standard conditions, of a specified quantity of natural gas excluding the condensation heat of water
•     Carbon Emission Factor (CEF) is the quantity of equivalent CO2 emitted per kWh of natural gas used
•     Oxidation Factor is the ratio of carbon oxidised to CO2 because of combustion to the total carbon contained in the natural gas

While the flow is measured by a calibrated flow meter, there are three approaches for determining NCV and CEF through the analysis of natural gas composition. The first option, for sites not using a process GC, is to collect weekly on-site natural gas samples, which are sent to an ISO 17025 accredited laboratory for analysis. This method is costly, demands logistical coordination, and provides only a snapshot sample, rendering the singular outcome non-representative. The second method is to validate the on-site process GC results through a sample sent to an ISO 17025 accredited laboratory for analysis. This validates the GC’s performance at a single point, and it does not ascertain its effectiveness across various compositions of natural gas (table 1).

Table 1 – Example of typical natural gas composition ranges – Components content in %mol/mol

The third method is the performance evaluation of the process GC using ISO 10723. A process GC is often already in place on-site, incurring no additional capital costs.

Performance Evaluation Using ISO 10723

ISO 10723 provides a standardised approach to assess the process GC performance over a range of expected natural gas compositions. Conducted annually, these quantitative measurements involve injecting a suite of 7-10 ISO 17025 accredited secondary reference gas mixtures with different natural gas compositions to offer a comprehensive evaluation of the instrument’s performance, providing evidence of its usable range, linearity, accuracy, bias and uncertainty.

Performance Evaluation Process:

1. The data from each of the reference gases, injected individually, is used to establish the relationship between the instrument’s ‘actual response’ and the known reference compositions (figure 3).

Figure 3 – Non-linear response errors
2.    These results are then plotted, and a best fit line established for the ‘true response’ required for each individual component. The best fit line could be linear, quadratic or cubic. These co-efficients are known as the ‘analysis functions’ and can be used to measure the individual component amounts in a gas sample most accurately. Operating in this way is known as Type 1 mode (ISO 6974-14 ).
3.    A Monte Carlo simulation – which simulates 30,000 gas compositions – is initially performed with the instrument in its normal process operating mode, which could be Type 1, using the analysis functions from a previous evaluation, or Type 2, using single-point calibration. The results determine whether the instrument meets requirements for bias and measurement uncertainty, in terms of Maximum Permissible Bias (MPB) and Maximum Permissible Error (MPE). MPB and MPE would typically be specified by the gas network operator.  
4.    If the instrument is close to or outside the limits for MPB and MPE, an improved design of calibration gas can be modelled that reduces the errors by calibrating at a different point. Uncertainties can also be minimised by using calibration gas with smaller uncertainties. The Monte Carlo simulation would then be re-run and in many cases the MPB and MPE would be reduced to acceptable levels.
5.    The final, optional, stage – if the instrument can be operated in Type 1 mode – is to implement the analysis functions from step 2 (figure 4). Re-running the Monte Carlo simulation with these analysis functions can deliver a significant improvement in the MPB and MPE.
Figure 4 – Comparison of results when operating in Type 2 and Type 1 modes – Red markers show the spread of results in Type 2 mode with linear response function. Blue markers show the spread of results in Type 1 mode, with significant less spread of results and almost zero errors

Conclusions

In light of the ongoing climate crisis and the relevant evolving legislation, the European Emission Trading Scheme remains a crucial tool for driving carbon emissions reduction. Robust procedures – such as the ISO 10723 performance evaluation of process gas chromatographs – ensure accurate and reliable monitoring and reporting results. Implementing annual performance evaluations improves the accuracy of determining site-specific factors (i.e. NCV, CEF) for more precise carbon emissions calculation, providing operators within the EU-ETS with greater opportunities to save costs while trading carbon allowances.