The flyover file for the compressor station arrives with a site-level measurement: a rate for a few hours in the afternoon. The source-level inventory for that station, built from equipment counts, run hours, and emission factors, implies a much lower average rate for the reporting period. The flyover rate and that source-level inventory are on the table in the same meeting. Someone asks which one is right.
That is the wrong question to open with. The useful question is whether the site-level measurement and the source-level inventory describe the same place, the same hours, and the same operating state closely enough that the difference is worth investigating, and whether the annual inventory the program already accepted should change. A difference between the flyover rate and that inventory is easy to see. It is not yet evidence that the inventory is wrong.
In oil and gas, emissions reconciliation compares a site-level emissions measurement with a source-level inventory for the same facility and period, then investigates meaningful differences between them. The flyover rate and the station’s inventory are that pair. People in the room may call the measurement top-down and the inventory bottom-up. Those labels do not say whether they can be compared, or what should happen to the inventory if they cannot.
A month-end tie-out between two workbooks, or a roll-up from facilities into a corporate total, is sometimes called reconciliation as well. Those jobs start from totals that were already meant to describe the same account. This comparison starts with a measurement and an inventory derived in different ways, often over different time periods or operating conditions, and asks what their difference actually says about the site.
Teams that carry measurements into an inventory sometimes call the result a measurement-informed inventory. What that term refers to, and how one event is attached to an asset and either updates a value the program has already accepted or leaves a reason it did not, is covered in from methane measurement to inventory. At the station, the flyover rate and the source-level inventory are already in hand.
The flyover rate and the source-level inventory can look like they disagree before anyone has established that they describe the same thing. Several common mismatches produce that picture, and more than one can be true at once.
The boundary of the site-level measurement is often not the facility in the source-level inventory. The area covered by the flyover can include inlet piping, a rental compressor, or the edge of the next lease, while the inventory still names the facility “Station 12,” as finance and environmental named it three years ago. The flyover rate then includes gas the inventory never had, or the inventory includes equipment the flyover did not cover. Until those two maps are the same piece of ground, the argument is about what “the station” includes.
The measurement period is another mismatch. The flyover lasts a few hours. The inventory may represent a month of activity or an annual estimate built from emission factors and activity data across the reporting period. Put the flyover rate next to an annual inventory that was never an hourly rate, and the difference looks large even when both records are doing what they were built to do. Averaging the flyover up to a year, or slicing the annual inventory down to an afternoon, does not make them the same observation.
Operating state can create the same kind of mismatch even when the boundary is right. A planned blowdown, a liquids unloading, or an upset during the flyover will lift the rate for those hours above the average rate implied by the annual inventory. The measurement can be right about those hours. The annual inventory can still be right about the year, and it can already include blowdowns and other episodic events when the activity data records them. The mistake is treating the afternoon rate as a replacement for that annual estimate.
The source-level inventory and the site-level measurement also do not have to resolve the same equipment. Pneumatics, connectors, and other small sources can sit in the inventory because the count and the factor say they emit, while this flyover would not separate them within the measurement. The reverse happens too. A compressor or tank that was operating can be missing from the source list, while its emissions can still contribute to the site-level measurement. In the first case, the inventory resolves sources the measurement cannot attribute individually. In the second, the measurement is showing emissions from a source the calculation never included.
Units can create another apparent disagreement. A flyover rate in kilograms per hour and an annual inventory expressed in tonnes per year are not automatically comparable emissions estimates. A rate averaged over the flyover and a factor applied to run hours for the month are not the same thing either. A spreadsheet can show both in one column. That column does not mean the measurement and the inventory describe the same conditions.
After those mismatches are accounted for, the remaining difference can still be an emission the factors do not represent. A source that has started leaking, or a factor carried forward from a different operating condition, will not show up as a boundary mismatch or a unit mismatch. It shows up only once the measurement and the inventory describe the same site, the same period, the same operating state, the same units, and the same sources. Most of the first argument in the meeting happens before anyone has earned that conclusion.
A visible difference does not mean the site-level measurement and the source-level inventory are comparable. The station team can treat that difference as evidence about the inventory only after the site boundary, measurement period, operating state, units, and source coverage describe the same operating conditions. Until those conditions are understood, the difference cannot yet be interpreted as evidence that the inventory is wrong.
For this flyover, the boundary can be settled from the file. The flyover boundary follows the station fence, and that fence is the facility the source-level inventory uses. The inlet the team worried about is outside that boundary. On the boundary, the flyover and the inventory are about the same ground.
Units can be put on a common basis. The flyover rate and the annual inventory can be expressed in the same unit so they can be looked at together. That does not turn the annual inventory into an estimate of those afternoon hours. It does not make them agree, and it does not make the afternoon the year the inventory describes.
Operating state and source coverage are not settled. Nobody has opened the operations log for those hours, so the team does not know whether the station’s activity during the flyover matches the activity the annual inventory used. Nobody has checked the equipment and emissions sources in the inventory against what was actually at the station. The inventory also includes small sources this flyover would not have resolved on their own. Until those two checks are done, the remaining difference cannot yet be read as evidence that the annual inventory is wrong.
Even after those conditions line up, both the measurement and the inventory carry uncertainty. A remaining difference is not automatically evidence that the inventory is wrong. The question is whether that difference is meaningful given what each method can support.
A site-level estimate derived from the same underlying factors and assumptions as the source-level inventory does not provide an independent check on that inventory. Where a program requires the site-level method to be independent of the source-level methods, that requirement sits with OGMP 2.0. It is a property of how the measurement was designed.
You cannot judge a flyover rate against a source-level inventory that is not the equipment at that station. For this station, the question is whether the equipment listed in the inventory was the equipment at the station during the flyover.
Once the boundary and the units are on a common basis, each record answers one question about the station during those hours.
The measurement file answers where and when. The flyover boundary, the start time, and the end time are what the team already used to decide the flyover covered the station and not the neighboring lease. The file also records the reported result: a flyover rate for that window. It is not a source-by-source split, and it is not the annual inventory. If the measurement record cannot establish the boundary, measurement window, and reported rate, the team does not yet have a defensible basis for the comparison.
The source-level inventory shows which emissions sources the calculation included and how their emissions were estimated. The inventory rows for each piece of equipment or emissions source, the activity they used, and the method on each row (a count and a factor, an engineering estimate, a metered volume) are what the flyover is being held against. If a row does not say how the estimate for that source was calculated, the team cannot tell whether that source could explain part of the difference between the flyover rate and the inventory.
The operations log answers what the station was doing. For these hours it shows a planned blowdown of one unit, started before the flyover and finished during it. Because the blowdown occurred during the flyover, the team cannot assume that the flyover rate represents the reporting-period average. The operations log provides the evidence that establishes that overlap. How a blowdown is quantified and carried as venting is covered in venting and flaring in oil and gas. The investigation does not need that quantification redone.
An unresolved leak identified in an earlier survey would matter if it were large enough, and tied to equipment at the station, to explain the flyover rate. The earlier survey findings do not provide enough evidence to explain the flyover rate, so the team does not treat them as the explanation for the difference.
Someone still has to make the call. The measurement report can show the flyover rate. It cannot decide whether those hours stand in for the reporting period, or whether the annual inventory should be left alone. That call sits with the environmental lead, after operations confirms the log is the blowdown that overlapped the window. A meeting where everyone agrees the numbers “look off” has not made the call.
The flyover overlapped a planned blowdown, so the team cannot assume that those hours represent the reporting-period average. The team does not have evidence from this comparison that the annual source-level estimate is wrong, so the annual inventory stays unchanged. The flyover remains evidence about emissions during that measurement window.
The decision is usable later only if the comparison can be reconstructed. What has to travel with it is specific. Keep the flyover rate and the source-level inventory used in the comparison. Keep the basis: the station fence, the start and end of the flyover, the unit both were expressed in, and a statement that the inventory side was the annual estimate for the reporting period, not a claim that the flyover measured the year. Keep what was ruled out, including the neighboring lease. Keep the earlier survey findings, and the note that the team did not treat them as the explanation for the flyover rate. Keep the blowdown log. Keep the decision, which is that the annual inventory does not change. Keep the name of the person who accepted that.
Whether that package is good enough for assurance or for a verifier is a later question. What has to travel with an emissions total, and how an operator’s own review uses it, is covered in emissions assurance and verification in oil and gas. The investigation is finished when the decision and its basis are in the record. A flyover rate and an inventory forced into one cell are not that record.
The flyover overlapped a planned blowdown, so the team cannot assume that its rate for those hours represents the reporting-period average. The annual inventory can already include that blowdown when the activity data records it. The log shows the overlap. It does not, by itself, show that the blowdown equals the entire difference between the flyover rate and the annual estimate.
A missing source would end in a different place. If the inventory rows omitted a compressor that was running during the flyover, and the boundary, time, and operating state otherwise held, the missing count is an input to the source-level inventory. Correct the source list and rerun the calculation. This flyover did not show that.
If the area covered by the flyover had included the next lease, the comparison was never valid. Neither facility’s source-level inventory changes, and the note stays with the file so the next person does not treat the flyover as evidence against either one. An internal estimate kept beside the reported total can still change. A Subpart W reported total has to be calculated the way the rule requires, so the difference can sit next to that reported total without becoming the filed method. The lesson for a later flight can also be separate from this period’s filed total: schedule it outside a planned blowdown, or flag the window when the flight cannot be moved.
Where a decision does change the inventory the program has already accepted, the path from that event into the inventory is the one in from methane measurement to inventory. In this case, the decision is to leave the annual inventory unchanged, so no inventory update follows from the flyover.
Operators run this comparison because a site-level measurement and a source-level inventory are both in use. Reporting is one place the comparison has to be explained. The contexts are not the same kind of requirement.
|
Context |
How it bears on the comparison |
Where to read it |
|---|---|---|
|
Level 5 is reconciliation of a Level 4 source-level inventory with independent site-level measurements |
The OGMP guide, including what Level 5 requires |
|
|
The filed total still has to follow the calculation method the rule requires |
The reporting guides |
|
|
A site-level measurement used as evidence still has to be traceable to a method and a decision |
The MRV explainer |
Level 5 is the clearest voluntary statement of this comparison. Teams also face a site-level measurement and a source-level inventory that do not match when no framework requires the work.
A site-level measurement and a source-level inventory often do not describe the same boundary, the same period, the same operating state, or the same set of sources. A few hours over a station is not the same observation as an annual source-level inventory for the reporting period. That annual inventory can already include episodic events. Once the boundary, period, operating state, and sources do match, a remaining difference can still be an emission the factors did not represent. Uncertainty in each estimate still has to be considered before that difference is treated as evidence the source-level inventory is wrong.
The site boundary, measurement period, operating state, units, and source coverage need to be aligned well enough for the site-level measurement and the source-level inventory to describe the same operating conditions. Even then, uncertainty in the measurement and in the inventory still matters. Putting a site-level measurement and a source-level inventory in the same unit does not finish that test.
No. A higher or lower site-level measurement does not, by itself, show that the source-level inventory is wrong. First determine whether the measurement and the inventory represent the same boundary, period, operating state, and sources, and whether the remaining difference is meaningful given the uncertainty in each estimate. A missing compressor in the source list is one ending that does change the calculation: correct the input and rerun it.
Level 5 is one framework that requires reconciliation of a Level 4 source-level inventory with independent site-level measurements. Operators also do the work when they are not reporting under OGMP. What Level 5 asks for is in the OGMP 2.0 guide. Investigating why a particular site-level measurement and source-level inventory differ is the same kind of operational work either way.
No. The calculation method required by the rule still has to be followed for that report. The Subpart W reporting guide is where those methods live. The investigation can still change an internal estimate, a source list, or the plan for the next measurement. A more recent site-level measurement does not become the Subpart W or GHGRP reported total.
Keep the site-level measurement and the source-level inventory used in the comparison, the basis of that comparison, what was ruled out, the decision, and who accepted it. The basis includes the site boundary, the measurement period, the unit, and whether the source-level inventory was left unchanged. Whether the package is good enough to rely on in assurance is covered in emissions assurance and verification in oil and gas.
The flyover and the source-level inventory do not determine whether the inventory should change. Someone still has to interpret the difference and decide whether there is enough evidence to update the annual inventory. In this case, the flyover window included a blowdown, and the annual inventory stayed unchanged.
Validere’s emissions measurement and response work can connect the measurement, the inventory, the investigation record, and the resulting decision, while sitting beside the measurement and calculation systems an operator already runs. It does not replace that judgment. Validere’s Air & GHG emissions software connects that inventory and its supporting evidence to the calculations and reporting that follow.