4 ms·

It's interesting that there is capability to locate methane leaks that doesn't seem to be directly tied to regulatory monitoring? Between remote sensing and mo

by dv_dt 6y ago

It's interesting that there is capability to locate methane leaks that doesn't seem to be directly tied to regulatory monitoring? Between remote sensing and more localized lower cost handheld and IOT sensors, this seems like an area which could be strongly improved for fighting climate change. https://hackaday.io/project/12568-portable-ecological-laboratory-suite-pels https://hackaday.io/project/12568-portable-ecological-labora...

ip26 6y ago
The capability is still developing. Europe has satellites capable of detecting methane sending back new data every day now. Processing that data to identify leaks is still a challenge, not something we have been able to batch in a nightly computer job. We can get there, but the very first satellite methane dataset is still only a few years old.
mturmon 6y ago
It is tied to regulatory monitoring, in CA: https://ww2.arb.ca.gov/our-work/programs/greenhouse-gas-inventory https://ww2.arb.ca.gov/our-work/programs/greenhouse-gas-inve... The ability to check the self-reported emissions with remote sensing is new, and has led to a recalibration (ahem) of self-reported emissions. The technology for remote sensing of CH4 is very exciting and it's being pushed forward in a lot of different ways. As you pointed out, you need sensors at various scales (in situ, airborne, space), calibration of retrieved concentrations (it's not a direct measurement), and inversions to find out what the sources are, given that you observe a certain set of concentrations. It's a complex problem!
- NotSammyHagar 6y ago
  So now we know that in at least two places in California, power companies didn't report leaks. This most recent one is much smaller, but as you joke about, having independent studies is the only thing we have. There should be a criminal policy with levels that require reporting. There should be financial and criminal penalties for not reporting.
  - mturmon 6y ago
    The viewpoint I have heard is that most enterprises want to fix the leaks if they are informed about where they are to a high degree of specificity (e.g., ~20 meters). Then they can send a crew with detection equipment. They're trying the carrot rather than the stick for now. Anecdotally, this has worked. It's not clear, to me at least, that there should be criminal penalties for medium sized leaks like in TFA. I don't know at all what the regulations are.
- dv_dt 6y ago
  I did a really quick skim of the program and didn't really see remote sensing. It looks like the program does collect data reports and map them - which is always nice. I did pick up this "For reporters subject to the California Cap-and-Trade Program, submitted data are verified by a CARB-accredited independent third-party verifier." Is it the remote sensing satellite that is a third-party verifier?
  - mturmon 6y ago
    I'm not sure what's done for Cap-and-Trade, when actual money might be at stake. Here's a link, which you may have seen, about how individual facility monitoring can work, when you need quantitative flux estimates: https://ww2.arb.ca.gov/our-work/programs/methane/ab1496-research https://ww2.arb.ca.gov/our-work/programs/methane/ab1496-rese... See at the bottom -- a third-party will fly a plane in circles around the site to measure all concentrations in a surface enclosing the site. The concentrations, plus wind speed, give the site emissions. (I think that's intuitive, but it's also the divergence theorem from vector calculus.) That's a lot of work to get just one day's emissions. And there are technical issues already because you want to get the whole surface concentrations+windspeed at the same time, but you can't, because the plane has to circle the site over many hours. Despite those cautionary footnotes, this approach is used in calibration/validation of the concentration/emission relationship. Getting it all quantitatively right, as opposed to just seeing a plume, is a complex problem. People are still exploring how it can all work together (space, airborne, in situ, plus modeling).
    - dv_dt 6y ago
      That's very interesting, thanks!

by dv_dt 6y ago

ip26 6y ago

The capability is still developing. Europe has satellites capable of detecting methane sending back new data every day now. Processing that data to identify leaks is still a challenge, not something we have been able to batch in a nightly computer job. We can get there, but the very first satellite methane dataset is still only a few years old.

mturmon 6y ago

It is tied to regulatory monitoring, in CA: https://ww2.arb.ca.gov/our-work/programs/greenhouse-gas-inventory https://ww2.arb.ca.gov/our-work/programs/greenhouse-gas-inve... The ability to check the self-reported emissions with remote sensing is new, and has led to a recalibration (ahem) of self-reported emissions. The technology for remote sensing of CH4 is very exciting and it's being pushed forward in a lot of different ways. As you pointed out, you need sensors at various scales (in situ, airborne, space), calibration of retrieved concentrations (it's not a direct measurement), and inversions to find out what the sources are, given that you observe a certain set of concentrations. It's a complex problem!

NotSammyHagar 6y ago
So now we know that in at least two places in California, power companies didn't report leaks. This most recent one is much smaller, but as you joke about, having independent studies is the only thing we have. There should be a criminal policy with levels that require reporting. There should be financial and criminal penalties for not reporting.
- mturmon 6y ago
  The viewpoint I have heard is that most enterprises want to fix the leaks if they are informed about where they are to a high degree of specificity (e.g., ~20 meters). Then they can send a crew with detection equipment. They're trying the carrot rather than the stick for now. Anecdotally, this has worked. It's not clear, to me at least, that there should be criminal penalties for medium sized leaks like in TFA. I don't know at all what the regulations are.
dv_dt 6y ago
I did a really quick skim of the program and didn't really see remote sensing. It looks like the program does collect data reports and map them - which is always nice. I did pick up this "For reporters subject to the California Cap-and-Trade Program, submitted data are verified by a CARB-accredited independent third-party verifier." Is it the remote sensing satellite that is a third-party verifier?
- mturmon 6y ago
  I'm not sure what's done for Cap-and-Trade, when actual money might be at stake. Here's a link, which you may have seen, about how individual facility monitoring can work, when you need quantitative flux estimates: https://ww2.arb.ca.gov/our-work/programs/methane/ab1496-research https://ww2.arb.ca.gov/our-work/programs/methane/ab1496-rese... See at the bottom -- a third-party will fly a plane in circles around the site to measure all concentrations in a surface enclosing the site. The concentrations, plus wind speed, give the site emissions. (I think that's intuitive, but it's also the divergence theorem from vector calculus.) That's a lot of work to get just one day's emissions. And there are technical issues already because you want to get the whole surface concentrations+windspeed at the same time, but you can't, because the plane has to circle the site over many hours. Despite those cautionary footnotes, this approach is used in calibration/validation of the concentration/emission relationship. Getting it all quantitatively right, as opposed to just seeing a plume, is a complex problem. People are still exploring how it can all work together (space, airborne, in situ, plus modeling).
  - dv_dt 6y ago
    That's very interesting, thanks!