Carbon data gaps: missing primary data, proxies and the limits of a footprint figure

Carbon data gaps are missing or incomplete data in a carbon footprint calculation. Gaps occur in primary data, in secondary data, in the expected service life assumption, and in the electricity accounting basis. A footprint figure calculated with gaps is not necessarily wrong, but the figure is only as strong as the treatment of the gaps. The JRC method rules provide for proxies, defaults and data quality assessment, so that gaps are handled transparently rather than hidden.

A calculation that presents a single precise figure, with no statement of gaps, proxies or data quality, overstates certainty and prevents a reader from judging whether the figure can be compared with another figure. Gap transparency is a method requirement, not a presentational choice.

Where do carbon data gaps occur most often?

Gaps occur most often in upstream material data, in supplier energy data, and in the expected service life assumption. Upstream gaps arise where a material supplier cannot provide primary data for extraction and processing, and the calculation must rely on secondary datasets that describe a generic process rather than the actual supplier process. Supplier energy gaps arise where a plant cannot provide metered energy data by process, and the calculation must allocate site energy by a proxy. Service life gaps arise where the expected service life, which determines the total energy delivered in the functional unit, is based on a test assumption rather than on observed life for the model in its application. Each gap has a different effect: upstream gaps affect the materials stage, energy gaps affect the manufacturing stage, and service life gaps affect the functional unit itself, and therefore affect the whole figure. The calculation record should locate each gap by life cycle stage, so that the effect can be understood.

How are proxies and defaults used?

A proxy is a substitute dataset or value used where the required data is unavailable, chosen for similarity to the missing process and stated as a proxy. A default is a value prescribed by the method for a defined situation, used where the method allows it and stated as a default. Proxies and defaults must be identified in the calculation record, with the reason for use, the source, and the data quality assessment. A proxy chosen without a reason, or a default used where primary data was available, fails the method discipline. The effect of a proxy or default should be assessed where the method requires a sensitivity check: what happens to the result if the proxy is varied within a plausible range. A result that changes materially with a plausible variation in a proxy is a result that depends on the proxy, and the declaration should allow a reader to see that dependence. A result that is stable across plausible variations is stronger, and the stability should be stated as a finding, not assumed.

How is data quality assessed and shown?

Data quality is assessed on criteria the method provides: technological representativeness, geographical representativeness, time representativeness, completeness, and precision. Each dataset is scored or rated against those criteria, and the overall data quality for the calculation is summarised in the calculation record. The declaration itself does not need to reproduce the full data quality matrix, but the passport record and the declaration reference should allow a reader to reach the calculation record where data quality is assessed. A footprint figure with no data quality information cannot be judged for reliability. A footprint figure with a data quality statement that shows weak upstream data is honest about its limits, and a reader can decide whether the figure is sufficient for the decision at hand. Data quality assessment is not a judgement that a figure is good or bad. It is a statement of the basis on which the figure was produced.

What does the gap table illustrate?

How are gaps closed over time?

Gaps are closed by obtaining primary data, by improving supplier data collection, by metering energy by process, and by updating the expected service life assumption as observed data becomes available. Each closure creates a new calculation version, with a date, and the declaration should be updated where the method and the delegated act require an update. A gap closed silently, without a version note, leaves a reader unable to tell whether a change in the footprint reflects a real change in the battery or a change in data. Version discipline for carbon calculations mirrors version discipline for model specifications: current value visible, history retained, change explained. A footprint that improves because data improved is a different fact from a footprint that improves because manufacturing improved, and the record should distinguish them.

What limits apply to a footprint with gaps?

A footprint with gaps is a figure with stated limits. It can support a declaration where the method allows the proxies and defaults used, and where verification accepts the calculation. It cannot support a comparative claim, a threshold claim, or a marketing claim that ignores the gaps. This guide gives general educational information only and does not give legal advice for a particular battery, calculation or declaration.