Aviation & SAF
View this service
Trust, but Verify: Ensuring Fuel Data Quality
As an aircraft operator, have you ever been curious about what exactly happens between sharing raw fuel data with verifiers and receiving the opinion statement? This article provides a verifier’s perspective on fuel data quality checks, outlining the key checks performed by us. We try to brief the most time-consuming step during the verification cycle in this short article. Let us get started!
We expect to receive operators’ raw flight & fuel data in a single, organised Excel file after sharing the Audit Preparation Letter (APL), and we will transfer these data to Normec Verifavia’s Emissions Data Analysis System for review, which allows us to evaluate the accuracy and consistency of flight & fuel values. Among several checks that we perform on fuel values, let us begin with a few of them that are straightforward to understand:
As verifiers, we also assess whether the aircraft operators apply the Fuel Use Monitoring Method (Method A, Method B, or Block-off / Block-on) correctly. That said, the fuel consumption of every flight reported by the operator must be calculated by the method recorded in the approved Monitoring Plan. Our data analysis system will identify and flag any flight that deviates significantly from the expected value, and the fuel values (fuel uplift, block-off value, block-on value) are then examined more closely.
Benchmarking is also a useful method for us to gain reasonable assurance that operators’ fuel data is duly recorded, transferred, and reported. We compare the reported consumption versus the estimation given by SET (Small Emitters Tool developed by EUROCONTROL) or CERT (CORSIA CO2 Estimation & Reporting Tool developed by ICAO). However, these tools are always ideal and conservative, since real-world inefficiencies in aircraft operation (diversions due to airspace closure, long holding procedures due to weather) are not taken into consideration. In addition to that, air cargo operators may find these tools less convincing for their fleet because they were developed using the aircraft statistics drawn from commercial passenger operations. Therefore, this check can also be reinforced further through a mean statistical check, measuring each flight against the average value for one specific aircraft for one specific leg. For instance, among all flights of one operator between Paris and London with A320 aircraft during 2025, two or three flights might record fuel consumption that is 52% above the average. Then we would like to investigate what leads to these exceptions. Is it because of the extremely long holding pattern above Heathrow on that day, or a simple data entry mistake?
All findings during the data verification could constitute misstatements or non-conformities, and that is the reason why they would be communicated to the operator through the issue log for clarification. Since the objective of verification is to mitigate the risk of misstatements or non-conformities in the reported data, we expect the airlines to correct those data entry mistakes before generating the Emission Report. On the other hand, we also examine the justifications on the qualitative aspect from the operator for anomalies. As mentioned above, diversions, positioning flights, headwinds, and long holding patterns are genuine occurrences in real-world aircraft operations that could materially affect fuel consumption; consequently, the associated data points do not necessarily represent misstatements.
Alongside checking the full flight dataset in our data analysis system, we use risk-based sampling to confirm the sources of reported data and assess whether the operator’s monitoring procedures and internal controls work effectively. The sample size, selection, and checks are determined by the complexity of the data flow, the strength of the controls, and any specific risks identified.
For selected flights, we compare reported fuel data with supporting source records mentioned in the approved Monitoring Plan, such as technical logs, fuel slips, or ACARS messages. If a requested record cannot be provided, we investigate the reason and seek appropriate alternative evidence. Where necessary, we extend the sample or perform additional checks to assess the risks, whether the issue is isolated or more widespread, and accordingly obtain sufficient evidence to support reasonable assurance that the data were properly recorded and transmitted in compliance with the Monitoring Plan. Where supporting evidence for a sampled flight is unavailable, we investigate the cause and assess whether alternative records can substantiate the reported values. Additional sampling may be performed to establish whether the issue affects other flights. Any unresolved lack of evidence is evaluated for its potential impact on reported emissions and the overall verification conclusion. Any changes to the sampling plan are documented and justified.
A data gap occurs when the source document for the fuel/flight data mentioned in the approved monitoring plan goes missing, rather than simply because a document cannot be provided during sampling. The aeroplane operator must fill the gap using the applicable method mentioned in the approved Monitoring Plan. Under EU ETS and UK ETS, this involves the approved alternative method or, where that cannot be applied, an eligible estimation tool such as EUROCONTROL’s SET tool can be used. We assess whether the method has been applied correctly as per the approved monitoring plan. If data gaps exceed 5% of annual reported flights (EU ETS), the operator must inform the competent authority and take corrective action.
Under CORSIA (Non-EEA), operators using a Fuel Use Monitoring Method must use ICAO CERT to fill qualifying data gaps. For 2021–2035, the 5% threshold relates to international flights subject to offsetting requirements, not all reportable flights. Exceeding the applicable threshold requires engagement with the State to address the issue. These thresholds concern the number of flights, not the percentage of emissions, and do not permit gaps to remain unfilled.
The potential effect is assessed individually and together with other misstatements against the applicable materiality threshold: 5% for aeroplane operators reporting up to 500 kilotonnes (1 kt = 1000 t) of fossil CO₂ and 2% for those reporting more than 500 kilotonnes. Materiality is the mean used by the verifier to assess the impact of identified misstatements, non-conformities, and non-compliances on the reported emissions. This assessment considers all errors, omissions, and misrepresentations identified by comparing the emissions data in the report with the actual data shown in supporting documents. It also includes deviations from the approved Monitoring Plan (MP) or the applicable monitoring and reporting requirements. The verifier evaluates these issues individually and collectively to determine their effect on the reported data. The total impact of the misstatements is compared with the applicable materiality threshold, while non-conformities and non-compliance are also assessed to determine whether they affect the reported emissions and lead to a material misstatement.
Data verification goes beyond checking calculations. Through the analysis, sampling, and the investigation of discrepancies, we assess whether reported emissions are accurate, complete, consistent, and traceable. We also review how data gaps and other findings have been addressed. For aeroplane operators, maintaining reliable records, providing clear explanations, and correcting identified issues promptly helps verification run smoothly. Together, these activities provide the evidence needed to assess compliance with the approved monitoring approach and support a well-founded verification conclusion.
