Friday, September 22, 2006

Inquiring minds wish to know......

The big bird has a question. He was innocently reading RPJr who made the following claim:

Taking a look at data from the US EIA (here):

It projects out to 2030 that the accumulated global carbon dioxide emissions will be 235 GtC. It also projects that of this total about 15 GtC will come from the use of petroleum in the United States. Let's assume all of this comes from cars. Lets further assume the EPA regulates carbon dioxide such that no emissions are allowed.
So he flew on over to find the home page of the US Department of Energy Energy Information Administration, with tons of links, and no data. Eli wants to know where Roger got those facts because a reference to the home page of a site that has thousands of data tables and studies is not exactly helpful for those of us who wish to RTFR. Indeed as far as we here at Rabett Run can tell, none of the commentors at Prometheus even thought to raise the question. This, of course breaks the first commandment, RTFR, but you need to know what shelf it is on, don't you.

Some time ago, Rabett Run posted some comments from the investigative committee at Colorado that dealt with Ward Churchill. Of course, eventually Colorado terminated Churchill with extreme prejudice The committee said of Churchill's footnoting:
p 24..Not only is his statement unsupported by his source, but also more significantly, he did not follow the referencing convention that a lawyer or historian citing a lengthy statute for a particular detail normally would follow, which is to pinpoint the precise section number of the multi-section statute that supported his claim. As one will see throughout this report, this general reference to an apparent independent source in its entirety constitutes an unconventional referencing style frequently employed by Professor Churchill to create the appearance of independent support for his claims, while simultaneously discouraging or, at least, making far more difficult, any effort by other researchers to check his claims by failing to pinpoint the precise location of his claimed support in an otherwise lengthy work.
We can't tell if the first sentence holds here without knowing where the data came from but RPJr has been known to send folk on snipe hunts and engage in data dumping.

from the Free On Line dictionary
snipe hunt n.
1. An elaborate practical joke in which an unsuspecting person takes part in a bogus hunt for a snipe, typically being left alone in the dark with instructions not to move until the snipe appears.
2. A futile search or endeavor.
from Changing Minds: Data Dump

Cover them in information, so much so that they’ll spend forever trying to wade through this endless detail. It helps to give them good quality information, but if in doubt, you can put low-grade rubbish at the bottom to fill out the pile.

If they ask you how you are, tell them, in great detail.

If they ask you what you want, give them a pile of differing and conflicting needs.

If they ask you how things happen, describe the process in great detail.


Hank Roberts said...

I think he posted that as a teaser, an invitation to a friendly game of "Peiser" --- you know how that's played, right?
Playing Peiser:

Player who challenges: Post a summary assertion, and challenge others to come up with an online search [engine+site+string] that includes the challenge among its results.

Players: a single hit is a win, with a score of 1, but "close" counts as a fraction of a point, 1 divided by the number of hits.

Players pledge to be honest in presenting their first result, and not improve their search strings by successive approximation before recording their result.

Nowadays mostly used grade school kids who have papers to write, who can't figure out how to get facts, so they make them up and challenge others to prove or disprove them.

It's an adaptation of the early Usenet advice that the best way to get good information is to post what you know and await correction.

Okay, I'll play --- My first play:

My first score is 1/9th of a point.
9 results for: 235 projection 2030 "carbon dioxide"

Anonymous said...

I won't speculate on the why, but I will second the notion that RP has a habit of doing this.

I went to the EIA and found a couple papers that showed projections for yearly CO2 emissions out to 2030 (see figure 108 in this reference

If you integrate the emissions under the curve labeled "reference" for the period from 1990-2030, the total is about 240 billion metric ton (ie Gt). I am assuming this is in tons of carbon, but that is not made explicit.

This value is pretty close to RP's value of 235gtC, but I can only assume (!) this is what he is referring to.

Actually, becasue he provided no explicit reference to a document and figure or table within that document, what he refers to in this case is about as clear as mud.

I'm with you on this Eli. I don't know how someone can even discuss things when it is not even clear precisely what they are discussing, but people seem to do it all the time on blogs. It's a waste of time as far as I can see and contributes nothing.

But hey, when has that ever discouraged anyone before?

Anonymous said...

What I posted above is for US emissions, not total world emissions.

Here's a document with the international emissions:

From fig 66, it looks like the total accumulated for the period 1990-2030 is closer to about 1000 gt.

If this is in ton CO2 instead of ton of carbon, that would make the equivalent number for GtC about 270, not too far off from RP's number of 235 (given that my "integration" is by eye and very crude, I would guess that this might be the case).

Anonymous said...

IF you look again at the document I referenced in my first post above
you can see from fig 107 that in the US, oil accounts for about 1/4 the yearly contribution to CO2 emissions.

Assuming the 240gt number I gave above (obtained from fig 108 in the same document) actually refers to Carbon dioxide, not carbon, the total accumulated emissions from 1990-2030 in ton carbon would be about 240 x .27 = 65 gtC.

If oil accounts for about 1/4 of this, that would be about 16GtC for the period 1990-2030.

That is very close to the number given for oil by RP (15 GtC).

So assuming DOE gives their numbers in ton Carbon dioxide rather than ton carbon, I get numbers for the period 1990-2030 that are consistent with RP's.

But, I still can not be sure that this is what RP referred to, which gets us back to your main point, Eli: the reference should be explicit.

EliRabett said...

What Prof. Pielke is doing is comparing US emissions in one sector with total worldwide emissions. He is trying to force the card that the US making changes in one sector/one fuel is futile as if that was what was being asked. He shows us US emissions from petroleum and worldwide total emissions. This is more convincing if the reader is not show the other two numbers, total US emissions from all sources and worldwide emissions from petroleum (see below). It also does not allow the reader to see the relative percentages of the petroleum based emissions. The assumption being forced is that only the US will be required to change.

In addition to sending us out on a Peiser, Pielke uses a unit, tons of C that the relevent tables at the site he points us to, do not use. Readers who are not aware that conversion between tons of CO2 and tons of C emitted requires mutiplying by 12/44 will be lost. (this is the ratio between the molecular masses of carbon atoms and carbon dioxide)

The eia site has a table which shows the estimates from 2005 up to 2030 for CO2 total emissions in 5 year intervals. If you interpolate you get about 860 gT CO2 and 235 gT C. There is also a table for CO2 emissions from petroleum. If you interpolate that you get 520 gT CO2 and 141 gT carbon between 2005 and 2030 (about 60% from burning oil).

Finally there is the table of US emission estimates, which gives you about 20.5 gT C for petroleum use in the US (2005-2030) and an emission total of about 48.0 gT C (about 40% due to oil).