I thought I write a note about Watts et al. Evan Jones, one of the co-authors of the paper was kind enough to drop us a note. Emphasis mine.
Evan Jones says:
August 30, 2012 at 4:50 am (Edit)
I came up with an obvious solution to the TOBS. We simply dropped the stations where TOBS is an issue. Time of observation for each station is directly available from MMS.
Like the man said, “If you want to stay out of trouble, stay away from it.”
That takes our sample down from 790 stations to 592, but that is still plenty.
We also factor in the MMTS adjustment as well.
The result is that the gap between Class 1\2 stations and NOAA-adjusted data is narrowed somewhat (but is still dramatically wide), but the gap between Class 1\2 (“good”) stations and Class 3\4\5 (“bad”) is actually increased.
There is a prejudice that urban sites must obviously have worse micrositing than rural. Taking TOBS into effect was supposed to wash away the differences between good and bad stations because rural areas are more affected by the TOBS issue and therefore the “good” stations would be disproportionately affected.
But that is simply not so. Quite the opposite. 30% of urban sites are Class 1\2 compared with 20% of non-urban. Also a higher % of Class 3s than 4s.
Perhaps when stations are sited in cites, there is more conscious consideration of surroundings? I can’t tell you the “why”, but the “what” is what it is.
So not only does the paper live, but is as robust as all hell. I’ll be finishing up the current set of spreadsheets this weekend.
So, the Watts paper can stand alone without addressing TOBS issues. Hmm….. 😉
What a perfectly good bit of logic. You don’t have to go through obstacles all the time, sometimes it’s better to just go around them.
You know this is deliciously funny. I’ve another post coming which is entirely unrelated to this with the exception of one thing. The reactions. It’s like finding a sore toe. When a skeptic makes a posit and the howling becomes so intense that the entire focus becomes on that one subject, then the skeptic knows he’s on the right track. This works in politics with the left/right dynamic as well.
suyts space 
Does the update compare the results to the CRN?
-Scott
I wish I knew more. But, I don’t know.
I find it interesting that an urban station can be classified as ‘sited well’, when it is sitting smack in the middle of a heat island. IMHO only well sited rural stations, where there has been no infrastructure changes, should be used. Of course this is a tall order, but it would eliminate the majority of land use change noise.
I think the Watts paper goes a long way towards addressing that.
I’m sure he does his best, but we just don’t know what those adjustments should be. There is an enormous gulf that separates estimates of UHI, and until an accurate figure is known, it is not possible to crunch estimates to tenths of a degree.
I for one am glad Anthony has taken this on, but we need to be realistic about what comes out the other end.
PS – Really enjoying your posts of late, and sharing them. 🙂
Agreed, as I pointed out in a past post, I don’t think the ground temp readings mean much of anything, but yes, accuracy to a tenths of a degree is a rounding error.
And, thank you very much. It’s always nice to hear that about the posts.
UHI from a well-sited urban station is clearly lower than UHI from a poorly sited rural station.
UHI from a well-sited rural station is clearly lower than UHI from a poorly sited rural station.
Anthony’s work, if it pans out, would seem to indicate that well-sited urban stations suffer less than poorly sited rural ones…a useful result.
Now, theoretically a well-sited urban station should be worse than a well-sited rural one due to what one might can mid-distance effect, but the CAGWers disagree. But it looks like Roy Spencer has good evidence showing the mid-distance effect to be true…check out the Stillwater, OK data that Roy looks at here:
http://www.drroyspencer.com/2012/08/spurious-warmth-in-noaas-ushcn-from-comparison-to-uscrn/
0.6 C difference with well-sited stations with that small of a local population? Wow. The thing is, this is a very, very easy experiment to perform and could be done well beyond what’s needed in 2-3 years with a mid-level budget. Just choose several cities/towns of different sizes and at different latitudes/elevations and smatter them with 40 or so well-sited PRTs each…monitor for a few years and check out the results. Note that this will NOT allow one to make quantitative corrections to the temperature record but instead would just show that such a correction is needed. Has such a study been done?
-Scott
Scott, to my knowledge, it has not. To my knowledge, in the last couple of years at least, Spencer and team Watts are about the only ones actually doing much about these siting and adjustment problems.
And I agree, we can only guess as to why this hasn’t been done, but as much as we as a nation spend on the climate issue, one would think this would have been done.
0.6C is a huge discrepancy.
Hey Scott! The problem with urban stations is that no matter how well they are sited, they are still measuring UHI. Studies show that the effect is like concentric domes of ever increasing heat as one approaches the urban center. GIGO.
Does the update compare the results to the CRN?
Mmmm, no. CRN has only been online for a handful of years. And after the 1979 – 1998 warming period, at that. We’ll need a passle more years, preferably during a positive PDO phase to assess the problem. Unfortunately, that’s one of those “call me in 50 years” kind of deals. So we have to make do with what we have.
I find it interesting that an urban station can be classified as ‘sited well’, when it is sitting smack in the middle of a heat island. IMHO only well sited rural stations, where there has been no infrastructure changes, should be used. Of course this is a tall order, but it would eliminate the majority of land use change noise.
As it turns out, lousy mesosite by no means translates to lousy microsite. Same deal with airports. Horrible mesosite, but, on the whole, sublime microsite.
And, yes, we split up the data every which way from Sunday:
— All stations
— Airports removed
— Rural only
— Urban only
— Airports only (those are nearly all ASOS systems)
— Rural, No Airports
— MMTS only (majority of time as MMTS)
— CRS only (majority of time as CRS)
— “Pure” CRS (never converted to MMTS)
— Rural MMTS
— All the stations we dropped (For amusement value. And so we can’t be accused of cherrypicking.)
— Urban vs. Semi-urban vs. Rural
In all cases — ALL of them, the Class 1\2 (good) stations have lower trends than Class 3\4\5.
And, yes, if you are looking at urban sites, you are looking at UHI. But the question is not whether cities ARE WARMER than non-urban sites (both sides acknowledge that), but whether they WARM FASTER.
You can, however, look at the rural sites alone (with/without airports and with MMTS only) and judge the vaiance of the trends.
Well, anyway, you can once I have finalized the spreadsheets and remade the PowerPoint.
Thanks Evan, I can’t wait to see the polished work.
Yes, thank you Evan. I am not by any means belittling your work or results, just opining. Keep up the noble effort and the much needed perspective.
I have no problem at all with anything you have said. I am also happy to answer any questions i can at this point.
I remember reading a paper about UHI some years ago. I thought it was on WUWT, but I can’t find it now.
The author(s) tried to discover the changes (corrections) applied to temperature records to adjust for UHI. Because these corrections are kept secret, they (the authors) had to try and guess the method by looking at the before and after temperature data.
They noticed two things: 1) the temperature corrections appeared to be applied at random; and 2) the temperature corrections had no effect on the temperature average of the two data sets (before and after).
So the question is: “Why bother?” The answer is so they can claim that corrections are made, but nothing is really done to remove the UHI.
Their sycophants also get to claim that the UHI is handled properly.
Jim
I remember something similar,as well, but I can’t find it either.
Well that’s two of us. Winston Smith obviously threw the UHI paper down the memory hole.
Jim
For urban stations we find a greater trend than for rural — in spite of substantially better microsite in cities. We also find that well sited stations in urban areas have lower trends than well sited urban stations, but somewhat less of a discrepancy than in non-urban areas.
We also find a greater trend between well and poorly stations for stations which have been MMTS for a majority of the study period (even after we apply MMTS adjustment). It is generally agreed that MMTS is a superior system to CRS or, for that matter, ASOS.
“We also find that well sited stations in urban areas have lower trends than well sited urban stations, but somewhat less of a discrepancy than in non-urban areas.”
make that:
We also find that well sited stations in urban areas have lower trends than poorly sited urban stations, but somewhat less of a discrepancy than in non-urban areas.
As for UHI, we recommend using rural, non-airport stations (preferably MMTS) as the “true” signal for US warming.
Evan, is this new data going to be available this weekend?
Not yet. It still has to undergo more scrutiny by the co-authors, although I am entirely confident of it. No doubt Anthony will want to release the revisions all at once on WUWT.
But I am happy to discuss the paper in the mean time.
Timing is important on these things. Especially now. Were I you guys, I’d want to make sure the new release doesn’t occur during a big political event. Unless, of course, the big political event starts to blather about climate change in the U.S. (Dem convention)
Typical denial industry tactic, all this fuss about reading stations while the most significant warming is in the Arctic; the greatest ice melt and ecological changes, so the heat island has no bearing. Just another baseless smear by armchair trolls for the fossil fuel giants.
P.S. Jones get a real job or go back to school and earn a degree in science.
“Just another baseless smear by armchair trolls for the fossil fuel giants.”
A “baseless smear”? You think that when temperature readings refuse to cooperate with the warmist alarmist industry, that’s a “smear”? No; it’s only the falsification of the theory of CO2AGW.
Why do you take the falsification of a theory personal? What’s wrong with you?
More than that, we were having a discussion about “global” temps, not some local fluctuation in some remote 4% of the globe where there’s hardly any stations to read. This post is over a year old and he’s going to comment here? Typical alarmist trolling for the grant leeching industry.
P.S. Jones get a real job or go back to school and earn a degree in science.
Hmm. You don’t think I have a “real job”?
With such trolling abilities, maybe he can find that UHI paper we’re looking for.
Jim
>>
evanmjones says:
September 12, 2013 at 5:02 pm
Hmm. You don’t think I have a “real job”?
<<
He’s projecting (as most trolls do), because he doesn’t have a real job. Unlike our troll, many of us also have degrees in the hard sciences.
Jim
Unlike our troll, many of us also have degrees in the hard sciences.
He he he he
And some have hard science in degrees 🙂
Reblogged this on Climate Ponderings.