Did the landfill intervention hold?
A fictional landfill operator tells Alder Climate Team it has completed a change at an identified methane source.
The question
Four satellite passes were usable before the change. A detection reports an estimated rate in kilograms of methane per hour for the sampled conditions. A non-detection gives no measured rate.
| Pass 1 | Pass 2 | Pass 3 | Pass 4 | |
|---|---|---|---|---|
| Detected? | Yes | Yes | Yes | No |
| Reported rate (kg CH₄/h) | 28 | 32 | 30 | No estimate |
These passes are observations at specific times.
“Decide what the observations establish and what needs more evidence.”
Not saved by this site. Your browser may restore text when you navigate back.
A strong answer, step by step
Open each step in turn. This is one strong answer, not the only one.
Step 1: What do the opening observations establish?
What would you do here? Open the move you’d make.
Count the non-detection as zero in the average
Tempting, but it misses something. The observation gives no measured rate.
Report the detection count and detected rates separately
The strong move. The reader can see how often the source was observed and which rates were estimated.
Multiply 30 kg/h by all hours in the year
Tempting, but it misses something. There is no evidence that the sampled rate persisted through the year.
See how a strong candidate says it
“Three of four usable passes found the source. The mean of the three detected rates is 30 kg/h. The other pass gives no rate estimate.”
Why it works: The detection count and the mean describe the observations. They do not estimate total methane over time.
Step 2: Does the first follow-up confirm the result?
Two follow-up passes were scheduled. Cloud made one unusable. The other was usable and had no detection. The approximate sensitivity under those conditions was 20 kg/h.
What would you do here? Open the move you’d make.
Report two valid passes with zero emissions
Tempting, but it misses something. The cloudy pass was unusable and the other did not measure zero.
Record one valid non-detection and keep follow-up open
The strong move. This preserves what was observed and the limit of that observation.
Ignore the follow-up because it found no plume
Tempting, but it misses something. A valid non-detection provides useful information about that pass.
See how a strong candidate says it
“There is one usable pass with no detection. The cloudy pass gives no observation. The usable result alone cannot show that emissions stopped.”
Why it works: The conditions can hide smaller or intermittent releases. The evidence for sustained change is still incomplete.
Step 3: What does the ground evidence add?
Independently reviewed ground observations use the same method at comparable loads. Readings were 26 and 29 kg/h before the work and 8 and 9 kg/h after.
What would you do here? Open the move you’d make.
Compare the ground readings from like conditions
The strong move. This gives a measured difference for those observations.
Compare the detected satellite mean with one ground reading
Tempting, but it misses something. That mixes methods and leaves out a comparable set of before and after measurements.
Use the 19 kg/h difference as a continuous rate
Tempting, but it misses something. Two samples per period cannot establish the rate across all times and operating states.
See how a strong candidate says it
“The ground readings averaged 27.5 kg/h before and 8.5 after. That is 19 kg/h lower in the sampled conditions. I would state which conditions the samples cover.”
Why it works: The comparison uses the same method and comparable operating loads, so it adds evidence beyond a non-detection.
Step 4: Do you close the case?
A later ground observation using the same method shows 27 kg/h at a similar reported load. A newly reviewed operator log shows an episodic operating pattern.
What would you do here? Open the move you’d make.
Close the case based on the two lower readings
Tempting, but it misses something. The later reading shows that the earlier observations do not cover every state.
Replace all past readings with 27 kg/h
Tempting, but it misses something. The earlier samples still describe their conditions. One later reading is not a full history.
Reopen and compare readings across logged operating states
The strong move. This tests when the change holds and when higher emissions recur.
See how a strong candidate says it
“I would reopen the finding. The later reading of 27 kg/h is close to the previous ground rates. I would match repeated readings to the operating log before reporting sustained impact.”
Why it works: An episodic source can have low rates in some states and higher rates in others.
The answer, in short
- Decision
- Report what the observations support and keep unresolved findings open.
- Because
- A usable pass, a detected rate and evidence of sustained change answer different questions.
- What would change it
- Repeated comparable measurements across operating states could support a stronger conclusion.
Common traps
- Keep unusable observations out of the denominator.
- Use only measured rates in an average of detections.
Check your answer
- I reported detections and rates separately.
- I checked which follow-up pass was usable.
- I compared the ground measurements.
- I updated the finding after the later observation.
Last reviewed Nov 11, 2026