Field notes: the pass that caught a plume the LDAR crew walked right past

There's a specific kind of frustration that shows up in MRV meetings this time of year, right after a quarterly LDAR cycle closes out. Someone pulls up the OGI log for a pad, points to a clean result, and says some version of: "the crew was there Tuesday. Nothing. And then the aerial survey four days later flags the same pad for a plume big enough to see from a Twin Otter."

Nobody did anything wrong. That's the part that takes a minute to sit with.

OGI surveys are a snapshot. A tech walks the fence line, scans flanges, valves, the separator, the tank battery, and logs whatever the camera sees in that window. A clean pass only tells you the camera saw nothing during that specific walk. A stuck-open thief hatch, a pneumatic controller cycling on a schedule, or a liquids unloading event keeps its own schedule, independent of the quarterly walk. A pad can be clean at 10:40 a.m. on a Tuesday and venting by Thursday night, and both of those things are true without anyone's paperwork being wrong.

This is the gap that shows up over and over in the comparisons between ground-based LDAR and aerial or satellite overpasses covering the same basin: the emitters that get missed aren't usually the ones hiding in hard-to-reach corners. They're the ones that are intermittent, episodic, tied to a process step rather than a constant leak. A well-timed OGI pass catches a constant seep every time. It has no better odds than a coin flip at catching something that only runs a few hours a week, unless the timing happens to line up.

More looks from above are what change the math here. A single annual flyover just moves the timing problem to a different calendar; it doesn't fix it. You still get one roll of the dice. The pattern that closes the gap is repeat coverage of the same site, often enough that an intermittent source eventually gets caught doing its thing, and specific enough that when it does, you know which pad it was, not which quarter-section it was in.

That specificity matters more than it sounds like it should. A regional methane advisory naming "elevated emissions in the basin" is useful for understanding scale, but it doesn't tell an MRV analyst which of the forty pads on their asset list to call about. The useful version of this data answers a narrower question: on this specific day, was this specific site boundary emitting, yes or no. Closing the loop between ground survey and remote detection means answering exactly that question, and it's why attribution to a site boundary, rather than a basin-wide hotspot, matters for running imaging spectrometry over operating assets. Methane Plume Map is built around exactly that per-site question.

None of this is an argument against LDAR. The ground crew is still the only one who can put eyes on a specific valve and fix it. But LDAR answers "what does this pad look like right now," and that's a different question than "has this pad emitted at any point since our last visit." Operators who are starting to treat those as two separate problems, not one program, tend to be the ones whose MRV numbers stop surprising them at year-end reconciliation.

The crew that walked past the plume didn't miss anything. The plume just hadn't started yet. That's worth remembering the next time a clean OGI result and a flagged overpass land on the same pad in the same week; it's not a contradiction, it's a timestamp problem, and more frequent looks are the only real fix for a timestamp problem.

If your reconciliation meetings keep running into that same gap between a clean walk and a flagged overpass, it's worth seeing what daily per-site coverage looks like for your pads.