"One real difference between the ‘tribes’ of natural scientists and STS scholars may be the perception of ‘facts’: Ryghaug and Skjølsvold conclude that “scientific facts are made and not just discovered”. In contrast, I think most natural scientists feel that facts are facts, whether we know about them or not."
Well, if natural scientists really think this, then they are wrong, though I'm not sure they do really think it.
For example, lets take a global average annual temperature for year X. This figure seems to me to be massively "constructed", or "made". Nobody consults s single thermometer that gives you this figure. Instead, a thousand point measurements are adjusted, massaged, and otherwised fiddled with so that the one figure can be produced. Yet I would say it is nevertheless a fact.
That's all that the claim that facts are in some sense "made" need necessarily imply. There are sociologists who write as though there is something more sinister at play--that our broader political ideology drives the way we "make" our facts. Stephen Fuller comes to mind. But it needn't be taken to imply that; the word is meant to suggest the notion of LABOUR as much as INVENTION.
This might be of some importance, given the misunderstandings of a few science journalists like Fred Pearce, who thought that the fact that nobody ever asked Phil Jones for his "raw data" signified a breakdown in peer review. But why would anybody want to re-do the work Jones already put into his data set? Such a course of action would only be useful if Jones had been incompetent or a member of some kind of worldwide Marxist conspiracy. Which of course he wasn't, so why bother?
4 comments:
For example, lets take a global average annual temperature for year X. This figure seems to me to be massively "constructed", or "made". Nobody consults s single thermometer that gives you this figure. Instead, a thousand point measurements are adjusted, massaged, and otherwise fiddled with so that the one figure can be produced. Yet I would say it is nevertheless a fact.
Speaking of global-average temperature results, I recently took some time to play around with the GHCN temperature data (i.e. the land station data that NASA uses).
I wanted to see what the simplest possible global temperature averaging algorithm would produce, so I wrote a little routine that did the following:
1) Averaged temperature data from all GHCN temperature stations sharing a WMO id number into a single temperature time-series per WMO id (just a simple average).
2) Computed NASA 1951-1980 baseline temperatures for all the WMO station id's.
3) Computed temperature anomalies by subtracting the baseline values from the raw temperature data.
4) Did a straight dumb average of the temperature anomalies.
IOW, it was the dumbest, simplest, most plain-Jane averaging procedure that you could think of.
And as for the results?
My results matched NASA's "Northern Latitudes" temperature index surprisingly well (there are differences, especially early in the 20th-century, but overall, not a bad match). Given that the GHCN stations tend to be concentrated in the temperate latitudes of the Northern Hemisphere, this is a good indication that I didn't mess up the averaging procedure too much.
As for my "raw" vs. "adjusted" temperature results, the differences were very minor. The "raw" results were generated from the GHCN "raw" data (no homogenization, adjustments, etc.). My "adjusted" results were generated from the GHCN data that had been "fiddled with", "massaged", "homogenized" or whatever it is that they do with the data. And it turns out, that on a global-scale, the data adjustments nearly cancel each other out, especially for the last half of the 20th Century.
Temperature plots here:
My results vs. NASA's Northern Latitudes Index (my results in blue, NASA's in orange): http://img826.imageshack.us/img826/1117/rawvsgissnorthernlatitu.jpg
My "raw" vs. "ajusted" results: http://img228.imageshack.us/img228/7796/meannew.jpg
(11-year moving average applied to all results).
My intention wasn't to compute accurate global-average temperatures, but to see if a simple "high-school" averaging approach would get me "in the same ballpark" as NASA/CRU/etc, and to see what all those data-adjustments do (or don't do).
Well, my little project showed that a brain-dead simple procedure gets you in NASA's ballpark, and that all those temperature data adjustments have a pretty minor impact on the global-average results.
I'm keeping my results and code on-hand for the next time that a "skeptical" in-law starts going off about how climate-scientists have been "manipulating" the temperature data. Could be interesting (some of my in-laws are bit uncomfortable with data that conflicts with their worldviews) ;)
Just a quick followup: In the first figure (my results vs. NASA's Northern Latitudes index), I used GHCN "raw" data. The results generated from "adjusted" data matched NASA's even more closely.
While you may have used a formula to determine your "global average annual temperature for year X," it wasn't produced it actually was a fact that you identified the "global average annual temperature for year X."
caerbannog has a point about the method of arriving at a value for the "global average annual temperature for year X." But once that methodology is published the it is a fact that the "global average annual temperature for year X" using method A is what it is and could be no other.
2+2 = 4 can be said to be produced at a trivial level, but it is a fact and is so universally.
Also scientists are right in that facts are facts, whether we know them or not, just like a tree will emit waves of energy that are audible in the frequency range that is necessary for human hearing even if no human is there to receive it.
Yes, facts are facts, whether we know about them or not. While there is indeed a lot of room for human error and bias in our systems for collecting and interpreting scientific data, the real facts of global warming are becoming more and more undeniable as the ice melts and the oceans warm.
Post a Comment