EPA Counts Carbon
Well, it’s official – greenhouse gases are now under EPA regulation. There’s been talk for the last year or so, and battles over whether carbon falls under their jurisdiction. At the same time, many states (including my state North Carolina) have enacted legislation to perform the same function on a more local level.
The feds require reporting from any source over 25,000 tons. That’s more than your average small business will create, but small enough to capture a pretty large portion of US emissions.
During the debate over the EPA’s role in carbon accounting, there was much discussion of just where to set that reporting limit. Set it too low, and you place a large burden on small business owners, not to mention create a bureaucratic nightmare for EPA employees who have to collect and verify data, enforce compliance, etc. for a large number of entities. Set it too high, and you only track a small amount of the total US emissions.
I spotted the following graph during that debate, and added it to my junk-charts file (yes, I keep a file).
Along the X-axis, it shows the threshold for reporting, from 1,000 tons to 100,000 tons. Green bars show the number of facilities under that threshold who would be required to submit reports. You’ll notice that increasing the threshold from 1,000 to 10,000 drops the number of reporting facilitiies by two thirds, going to 100,000 decreases the number of reporters 90%.
So far so good. But now look at the pink line, showing the percentage of emissions that will be accounted for in the plan. It looks like there is a steep drop in coverage as you go from 1,000 tons to 100,000 tons. In fact, it LOOKS like a reduction of about 60 percent. But if you check the scale on the right hand side, you’ll see that the actual shift is just 3.6 percentage points, a proportional difference of 6.5 percent. 60… 6.5… what’s the difference?
I suspect this is not a benign oversight, but an attempt to convince the audience to set a lower threshold and increase the number of reporting facilities. When you change the categorical threshold axis to a continuous scale, and adjust the coverage axis to show the entire range, you get a rather different picture.
But, alas, that’s not as dramatic. Perhaps a threshold of 25,000 strikes the right balance, but I’d like to see data extending past 100,000 tons. At what point do you get a significant drop in the coverage line? And on the other side, how low must you set the threshold to capture 80 or 90 percent of the emissions?
Can't get enough? Try these related posts:
