Cross-sectional flux vs. integrated mass enhancement for emission rate
Every methane emission rate that comes out of a plume image is a modeling choice wearing a number's clothes. Two methods do almost all of that work right now: cross-sectional flux (CSF) and integrated mass enhancement (IME). They start from the same retrieved concentration field and land on different numbers, sometimes by a factor of two, because they handle wind and plume geometry differently. If you're reconciling a vendor-reported rate against a facility's own leak detection and repair log, knowing which method produced the number tells you where the uncertainty actually lives.
Cross-sectional flux: one line, one wind speed
CSF draws a transect across the plume, perpendicular to the prevailing wind, and sums the column-integrated methane enhancement along that line. Multiply that linear mass density by the wind speed crossing the transect and you get a flux, in kilograms per hour, through that one slice.
The method is clean when the plume is well-formed and the wind at the transect is known with some confidence. Aircraft campaigns that fly repeated passes downwind of a source lean on CSF because each pass gives an independent transect, and averaging several tightens the estimate. The weak point is exactly that dependence: pick the transect too close to the source and the plume hasn't mixed enough to give a stable reading; pick it too far and dilution and meandering start eating the signal. And the whole number scales linearly with whatever wind speed you plug in, so a wind product that's off by 20% moves your reported rate by the same 20%, even if the imagery itself is perfect.
Integrated mass enhancement: the whole plume, once
IME instead sums the total excess methane mass across the entire detected plume footprint, not a single slice. That integrated mass gets divided by an effective residence time, usually built from a characteristic plume length divided by an effective wind speed, to convert total mass into a rate.
Because IME averages over an area rather than a line, it's less sensitive to local turbulence at any one point in the plume. That's the appeal for single-overpass satellite and high-res imaging spectrometer data, where you get one snapshot and no second pass to average against. The tradeoff moves into how you define the plume's edge and its effective length. A looser detection threshold pulls in more background noise as "plume," a tighter one clips real mass at the edges, and both decisions shift the denominator in ways that compound with wind uncertainty.
What this means when you're comparing plume quantification methods
Neither method is wrong. They're just doing different division problems with the same input. CSF divides a line integral by a point wind measurement. IME divides an area integral by an estimated residence time. Both put wind speed in the denominator, which is why cross-sectional flux vs. integrated mass enhancement comparisons in the literature so often converge on the same conclusion: the dominant source of disagreement between two emission rate estimates for the same plume is almost never the imagery. It's the wind field each method assumed.
For an MRV analyst pulling a rate into a facility-level ledger, that means the method tag on a reported number matters as much as the number itself. A CSF estimate from a single aircraft pass carries different uncertainty than an IME estimate averaged over three overpasses of the same site. When you're stacking numbers from different providers into one reconciliation, treat the method as metadata you keep, not a detail you discard once you have the kilogram-per-hour figure.
None of this replaces the harder operational question, which is simpler than the math: is a given pad emitting today, and has it been emitting for a week or just this afternoon. A daily, per-site plume map attributes the SWIR absorption signature back to a facility boundary so you're not just averaging rates, you're watching whether the plume is there at all on a given pass.
If your team spends more time arguing about which quantification method to trust than deciding what to do about a confirmed plume, it may be worth seeing what a site-attributed daily map adds to that workflow.