CARMA v3.0: Known issues (ongoing updates…)
With the CARMA v3.0 database now out in the wild, it’s time to begin documenting known issues, problems, errors, omissions, loopholes, mistakes, gaffs, laughers, and anything else generally erroneous.
If you find anything you think should be added – or, even better, a may to correct one of this issues (with new data, for example), please contact us.
I expect this list will grow rather than shrink with time, as my ability to address errors quickly is limited. In any event, here goes:
Last updated: August 24, 2012
- Hubei Province (China) is missing. It appears the fuzzy string geocoding algorithm incorrectly classified Hubei plants as being in Hebei province. The largest affected plant is the Three Gorges Dam (it is, indeed, in Hubei province – not Hebei). Province and city totals will be wrong for these two provinces as well. Hope to fix in future.
- “Future” intensity values might be inflated. Model estimates for future carbon intensity of large coal plants are almost certainly overestimated (e.g. http://carma.org/plant/detail/8238). Looking into for future update…
- CO2 intensity for combined heat and power (CHP) plants are misleading. Since CHP units provide heat as well as electricity, not all of the emissions are strictly due to electricity generation. This leads to overestimating CO2 emissions per MWh. Hoping to devise a way to address this in future version of CARMA.
- Some country totals may be slightly off if disclosed data available. For European countries, Canada, India, and South Africa, the country totals may be slightly off. This is because CARMA does not (yet) re-adjust the national totals after adding disclosed data. The error is likely to be small. Plan to fixed in future version.
- Electricity generation for dams especially uncertain. No way around this, I’m afraid. It’s hard enough predicting electricity generation for any plant. When it’s dependent upon rainfall in a specific catchment and downstream uses in a specific year, etc. it’s just not possible to do better than a guess. Interpret data accordingly.
- Non-hydro renewable electricity totals are debatable. It’s pretty clear that national wind and (especially) solar electricity totals are rough estimates – even in the U.S. (see this NREL article on the subject). Elsewhere in the world it’s unlike to be any better. For example, the national figures almost never capture distributed PV generation, nor does CARMA know about every little solar panel on someone’s roof. Really need more data to get better figures. If you have any ideas, contact me (please!).
- Some U.S. plants have MWh=0 but CO2>0. Yes, this can happen if the net electricity generation remains below zero, but fuel is still being consumed and emissions being produced.
- U.S. plants don’t reflect planned capacity expansions. While existing, non-U.S. plants have “Future” values that reflect expected capacity expansions/retirements, the same has not yet been done for U.S. plants. Hope to do so in future.