Assault on Mr. Hansen
Sadly the sense filter over at Climate Audit appears working again and excluding some of the bunny's words of wisdom, so although a bit out of place here is what got filtered for those coming over from Climate Audit (welcome):
In your anger, you have missed the entire point about the picture and US Climate Reference Network, which was posted
The stations are designed to be optimal, with respect to location, instrumentation and operation. More information can be found at the web site. While the USCRN is just being set up it can be used to quantitatively check measurements made in the past at other stationsIn other words, here is a sensible way of checking the accuracy of older climate networks in the past and calibrating them in the future. Tamino had an interesting post on station adjustment using Switzerland as an example with much detail filtering into the comments. There is a complete (although in German) description of the station history and adjustments available.
In order to assess the performance of the network in addressing this goal a performance measure (PM) was developed. This PM is an assessment of how closely the current and past configuration of the network captures the "true" national temperature and precipitation signal as defined by an area-averaged time series of annual temperature and precipitation derived from 4000 U.S. Cooperative (CO-OP) Network stations scattered across the continental U.S. The configuration of the CRN for a given point in time is used to select stations from the 4000 CO-OP station network, one station for each operating CRN site (the one physically closest in location), and the time series derived from these stations is compared, statistically, to the time series derived from all 4000 stations. The result is a "variance explained" that measures how closely the "CRN" time series follows the "true" time series
Assault on Mt. Hansen
Steve McIntyre fired the first gun in the assault on the GISS global temperature record, when, he rather amateurishly tried to down load the station data records in a way which looked like a denial of service attack to the webmaster. The back and forth over whether this was a DOS attack, or the GISS servers sucked was hilarious. Good summaries can be found at Deltoid and Climate Audit, much epistimology about the meaning of the word robot. The upshot was that Jim Hansen said welcome to it.
After a short meeting with Dr. Hansen, we were advised to let you download whatever you want as long as generally accepted protocols are observed. Please try to do so at a time that does not impact other users, i.e. late nights, weekends.Now Ho Chi Pielke Sr. is providing reinforcements by getting his irregulars to go out there and take pictures of stations in the Global Historic Climate Network (GHCN). Anthony Watts is setting up a web site for such pictures. The goal, of course is to falsify GISSTEMP
UPDATE: Following the crumbs left by the mice (Dano and Chuck) in the comments below, Eli observes that this is exactly what Roger's survey is designed to do with its bias, nay more than bias, prejudice for photographing sites close to people, e.g. in developed areas. Folk are going to take pictures of sites near them, so they are going to get a sample heavily tilted towards sites near them. It will be fun to correlate the locations of sites photographed with voting patterns. Of course we have the American speaking bias on top of that.
UPDATE: Eli would strongly encourage folks from Climate Audit and Climate Science who are visiting to make a side trip to the US Climate Reference Network site to see the care and thought that has to go into any serious test of the existing climate data system.
Thomas Karl has a better idea. He has taken the lead is setting up the US Climate Reference Network,
Its primary goal is to provide future long-term homogeneous observations of temperature and precipitation that can be coupled to long-term historical observations for the detection and attribution of present and future climate change. Data from the USCRN will be used in operational climate monitoring activities and for placing current climate anomalies into an historical perspective. The USCRN will also provide the United States with a reference network that meets the requirements of the Global Climate Observing System (GCOS). If fully implemented, the network will consist of about 110 stations nationwide. Implementation of the USCRN is contingent on the availability of funding.Pictures of stations in the network can be found on the web site. Roger can hike up to the Boulder station. The stations are designed to be optimal, with respect to location, instrumentation and operation. More information can be found at the web site. While the USCRN is just being set up it can be used to quantitatively check measurements made in the past at other stations
In order to assess the performance of the network in addressing this goal a performance measure (PM) was developed. This PM is an assessment of how closely the current and past configuration of the network captures the "true" national temperature and precipitation signal as defined by an area-averaged time series of annual temperature and precipitation derived from 4000 U.S. Cooperative (CO-OP) Network stations scattered across the continental U.S. The configuration of the CRN for a given point in time is used to select stations from the 4000 CO-OP station network, one station for each operating CRN site (the one physically closest in location), and the time series derived from these stations is compared, statistically, to the time series derived from all 4000 stations. The result is a "variance explained" that measures how closely the "CRN" time series follows the "true" time series.1221 of the best CO-OP stations with long records form the US Historical Climatology Network. USCRN is a part of GCOS whose stations can be found here.
Posts
87 comments:
A digression to make a point: as an undergrad I helped collect temp transect data in the CA Central Valley for an ag dissertation on a productivity theory. Easy and fun and one of the girls was cute and charming. I compared the transect data with an ag weather data network called CIMIS as one component of data verification. The point?
Temp of cities drops off as soon as you leave the pavement. The chimps at RP Sr's place can take real purty photos all they want, but so what. Until they do temp work and upload that too, it's just a big wanking exercise. I can see the snickers now.
Best,
D
I believe it is called sample selection bias when one looks at only a small subset of the data and attempts to draw conclusions about he validity of the full dataset.
I wonder how they propose to cast doubt on the satellite data that shows warming that is consistent with the station data. Perhaps they will take pictures of the satellites and show how they are located above blacktop.
A point Eli occasionally makes when he wants to twit the twisted is that central city density has fallen over the past century.
If anything, I can see this project making a worthy contribution to the world of Boring Postcards.
And hey, if it gets people off the Internet and into the outdoors, all the better!
What these bozos are doing is absolutely ridiculous.
There are literally thousands of these weather stations across the US so for them to take pictures of a few of them and attempt to show something (anything) is the height of stupidity.
We know McIntyre et al are not the sharpest knives in the drawer, but I thought Pielke was a real climate scientist.
Does Pielke think this is going to help his reputation when he associates himself with such crap?
Here's the real goal of the project (that Peilke has apparently put his seal of approval on, even if unintentionally)
"If you have a digital camera, a portable handheld GPS device with accuracy within 100 feet or better, and the ability to follow simple instructions, you can help us demonstrate that many of the assumptions about climate change based on the surface temperature record may in fact be due to faulty data!"
Got that? They apparently already know the conclusions are wrong and are just out to demonstrate as much.
They've got a great start -- pictures of 3 stations!
What a bunch of morons.
Your claim in an earlier post that my downloading GISS temperature data "resulted in denial of service to everyone else" appears to be a total fabrication on your part and was made without any evidence that service to any person was denied or even affected. As you aware from reading the thread at climateaudit.org, a test showed that there was ready access to GISS temperature data while the download program was running.
Your claim in this post that it "looked like a denial of service attack to the webmaster" is also without any evidence that the webmaster had this belief. No such claim was made by the GISS webmaster; he merely claimed that downloading of station data was a violation of their robots.txt policy. This became a moot point, when GISS decided that the downloading was not problematic and permitted me to continue downloading data in exactly the same way that was doing in the first place.
Who really cares?
You yourself blew the whole data access thing all out of proportion.
Most people would merely have worked things out with the people at NASA GISS without making a public issue of it on a blog as you did.
NASA agreed fairly quickly to allow you to download the data.
Your entire post on the subject was childish.
There are some stations outside the USA too. KNMI has some pictures available for the Dutch stations; the article is in Dutch, but you can click on a station name to see a picture.
But perhaps these aren't the pictures they are looking for ...
By the way, the station in my area (Groningen/Eelde) was located downtown in the city of Groningen before WWII, now it is located on an airfield in a more rural area where the dominating SW winds are not coming from the city. However, the warming is the same as all over the country (higher than the average global warming).
If the temperature records are so good, I see no reason to ridicule photographic documentation made of any stations, nor, if temperature records are so good, the necessity for setting up new stations under the US Climate Reference Network.
That stations do not appear to have undergone any independent auditing and because the issue of AGW has become so political, honest scrutiny of any type of data should be welcomed.
It seems ludicrous to me, a layman, that individual stations do not appear to have undergone photographic documentation as of yet, as one step in their validation.
Is this the type of quality control that we the general public can expect in the data being supplied to us on the issue of AGW?
I'm not sure if you guys are aware just how sideways this entire climate things has gone from an outsider's (engineer) perspective. That data is lost, made unavailable, source code withheld, etc, really makes me question the claims. That the ClimateAudit guy can't just contact the NASA guy and get the data is a joke.
I think some valid points have been made when it comes to questioning the adjustments of the data. We're looking for very small changes here, and the adjustments that I've see made seem to have been handed down from on high. That just doesn't seem like science. If, in the end, we see 100 folks go photograph their local weather station and it looks fine, then great. If, on the other hand, they go photograph it and 90% of them are sitting next to a parking lot and BBQ grill, then I think you'd also have to admit there is a real problem.
NOAA thinks that documenting weather station characteristics is important enough to maintain this site.
http://www4.ncdc.noaa.gov/cgi-win/wwcgi.dll?wwdi~ASOSPhotos
The ASOS stations look very standardized, a testament to their critical role in aviation safety.
If you ran the data on BOS PVD TAN and OWD you could probably tease out the effect of having a thermometer in a 24/7 blast furnace (BOS), a smaller furnace (PVD), a thermometer on the bottom of the Great Neponset Swamp with a touch of small jet traffic (OWD) and a well sited thermometer at a fairly rural prop job airport (TAN). The data and site documentation exists to do this. Can the same be said of other thermometer networks? If it cannot, then why not?
The real issue is Jones' UHI adjustment. Jones et al. (1990) determined that the impact of urbanization on hemispheric temperature time series was, at most, 0.058 deg C century. This value defies common sense and the experience of anyone living in a city. Hey, why have that house on the lake or send your kid to summer camp in the country. Its only 0.058 C cooler, per century.
Each thermometer stands on its own and can only be normalized in its local context. You might build out some accurate model factors eventually but not if you are performing one local normalization and using it to normalize another set of independent thermometers. Even highly local normalization will have some effects as a statistical filter. In this case piling one filter on top of another will not yield purity as the UHI signal will drop out quickly. With the 100 year temperature rise at .6C, you need to do some careful science to really figure out where the delta comes from. Until the science is performed with properly calibrated tools consistently deployed according to well accepted conventions, I don't think we can know or learn very much. Good science will require rigourous evaluation of any datasources and the discipline to just say no to poorly measured and collected data. You cannot turn S**T into Shinola with a statistical washing machine.
Eli Rabbet wrote: A point Eli occasionally makes when he wants to twit the twisted is that central city density has fallen over the past century.
I'm not sure this point really helps you in your want--unless you can show that tarmac, buildings, etc are removed in proportion to the drop in density.
"Ho Chi Pielke Sr."? This is scientific comment? Looks more like an ad hom. What's the problem with QC anyway?
Fun with Mr. McIntyre:
http://bigcitylib.blogspot.com/2007/06/auditing-climate-audit.html#links
Sorry, a shameless plug.
Bozos? Ridiculous? Selection bias?
Commenters might stop to think about this for a moment.
First, these folks are recording every site they can find, not just the ridiculous ones. So they're happily helping confirm the integrity of the surface record, if that's possible.
What's shocking is that things are so bad. Let's be very conservative. Suppose only ten percent of stations have experienced compromising microsite issues. Suppose the average compromised site has seen only a two degree rise vs good sites. If so, this little experiment will have demonstrated a 0.2 degree error in the assumptions of the models. If so, that would be huge.
What do you think they'll find? Only one percent with anomalies? What do you think they have found already?
Why be a defender of ignorance?
Anon 9:38 said: "That the ClimateAudit guy can't just contact the NASA guy and get the data is a joke."
Yes, you are right. It is a joke that the Climate Audit guy (Steve McIntyre) did not just contact the NASA guy (GISS group) to get the data in the first place. McIntyre could have, but he chose not to -- not initially.
Instead, he tried to download all the data (with thousands of requests) with an automated script without contacting the NASA GISS webmaster first.
The NASA webmaster had a filter (robots.txt file) in place to disallow such repeated requests from the same address on the possibility that such repeated (scripted) requests might be malicious.
That's how a denial of service attack is carried out and without any additional information, there is no way (on the NASA webmaster's part) of distinguishing what McIntyre was doing from such an attack.
When McIntyre finally did contact the webmaster, the webmaster promptly informed him that he needed to contact Hansen's group. After McIntyre did so, there was no problem. He got the data he wanted.
You obviously have not informed yourself of what transpired.
You also need to inform yourself about sampling methodology, since the statement you made above makes it clear that you do not understand random sampling:
"If, in the end, we see 100 folks go photograph their local weather station and it looks fine, then great. If, on the other hand, they go photograph it and 90% of them are sitting next to a parking lot and BBQ grill, then I think you'd also have to admit there is a real problem."
That's not random sampling and a small number of samples chosen in this way is not sufficient to draw any conclusions about the data.
I suggest you take a couple intro classes on statistics -- including random sampling. This stuff is very basic and the way McIntyre et al are going about the whole thing ins amateurish -- stupid, really.
"Suppose only ten percent of stations have experienced compromising microsite issues."
Suppose the world were flat.
You are engaging in pure speculation. There is a scientific problem with that.
Welllll...the bots have spread their ignorance to Eli's little hutch, their comment spam much like a DOS attack in that the thread is made tedious.
None of the chimps have proposed how to measure the temp change from the central city/pavement to the ICAO/WMO station.
Until the chimps start talking about this, funny little comments about "scientific comments" are just joke lines.
Best,
D
Steve, I strongly suggest you stop acting like a victim bully and get on with your life. Eli is not nearly as much interested in what people claim as motive as what they do in reality.
Sampling bias is indeed the issue. Pielke's experimental design is clear evidence of card forcing or ignorance. However, let us get beyond that. There is some possibility that over years all of the sites in the US (maybe even including Alaska) will be photographed. Some of these photo's exist, however, as single photos they are useless, because what is measured are changes in average temperature, anomalies, not absolute temperatures, so what is needed is an historical archive of photos, e.g. a continuing series that shows changes at the sites.
OTOH, Tom Karl's project seeks to bootstrap information about past and future measurements at a large number of sites by setting up a small, optimal network which can be used as a yardstick to check the larger USHCN and the US CO-OP networks. By implication this is then a check on the Global Networks, AND on the various corrections that have been applied to the raw data.
The bit about engineers always keeping their data available is a sad joke to anyone who has ever had to fix an old piece of equipment that has been orphaned. Get off it.
Finally if you look at old pictures of center cities it is clear that the density of people and vehicles has decreased, if for no other reason that industry has moved to the suburbs. Take a look at Pittsburgh, for example, where the steel mills lined the rivers close to central city
Eli, Could you confirm for me that the Boulder 14 W site went live in September 2003.
Don't have much time now, but if you go to the NOAA web site, there are station histories for each station. I'll look it up tonight and post the detail
But if McIntyre was not whining about not being able to get data, what would people have to cheer-lead about on his blog?
That's not random sampling and a small number of samples chosen in this way is not sufficient to draw any conclusions about the data.
I suggest you take a couple intro classes on statistics -- including random sampling. This stuff is very basic and the way McIntyre et al are going about the whole thing ins amateurish -- stupid, really.
Perhaps you should describe exactly how this work should be done. Or are you suggesting that the data is irrelevant and should not be collected at all?
"None of the chimps have proposed how to measure the temp change from the central city/pavement to the ICAO/WMO station."
Not needed. These are official stations with official records. The challenge is to validate both realities and assumptions about the surface record.
Isn't it at least slightly interesting that jumps in the station temp record are being found to correlate with historical microsite impacts, while sites with no local physical change do not see such jumps?
You are engaging in pure speculation. There is a scientific problem with that.
So how do we eliminate scientific speculation? Do we stop the inquiry, and stop attempting to falsify assertions?
Or, as the http://www.surfacestations.org group is doing, do we attempt to learn more, gain more understanding, develop better insights?
Speculation is best eliminated through inquiry and interaction, not suppression of all dissent.
Science is all about skepticism. The hackles raised here are absolutely astonishing for those who value real science.
Mr. Pete,
How are you and the gang at surfacestations.org being "suppressed"?
Sorry, but this does not sound like science to me:
"If you have a digital camera, a portable handheld GPS device with accuracy within 100 feet or better, and the ability to follow simple instructions, you can help us demonstrate that many of the assumptions about climate change based on the surface temperature record may in fact be due to faulty data!"
http://www.surfacestations.org/get_involved.htm
It sounds like they are looking to "prove" their position.
There is a right way of verifying the information obtained from such stations and lots of wrong ways.
What they are doing is just silly.
If you can't see that, i suggest you take some very basic science classes at the local community college rather than jabber on about scepticism and hackles and the like.
Some of us who are actually trained in science cringe when we see this kind of crap.
That's not random sampling and a small number of samples chosen in this way is not sufficient to draw any conclusions about the data.
Okay then, let's see modern (like in the last year) pictures of all the monitoring sites. Oh wait there are none that are easily available.
Is your complaint that not every single one of the monitoring stations is photographed makes this an unworthy project?
No, that was not it at all. You missed the whole point.
If you really want to draw conclusions about an entire data set (ie, from all the stations) from a subset of the stations, then you have to do proper random sampling of many of those stations (the greater the number, the higher the confidence in the result, of course).
But that is not all. As Eli pointed out above, it's not enough to just look at the stations now, but you have to look at them over time because some are not even in the same locations as they were in in the past. Also, you also have to actually demonstrate that the purported "issues" seen in the photo have had an actual effect on the measured temperatures over the years.
The "take a photo of the weather station" junior high project may seem really scientific to some, but it is really rather pathetic.
I have worked on many instrument QC projects myself and I certainly understand the importance of doing QC, but the way they are going about this is just idiotic.
Don't look now, Eli, but McIntyre is saying you pulled a "bait and switch" with the image you provided -- even though it is clear to anyone with a brain that you were not saying or even implying that the pictured station was part of GHCN, but rather that it would be part of the proposed USCRN.
Also, here's how McIntyre
"his [Eli's] fabricated story that I had caused a denial-of-service to other users at GISS)"
twists your words
"he [McIntyre] rather amateurishly tried to down load the station data records in a way which looked like a denial of service attack to the webmaster."
What an imagination McIntyre has.
I have worked on many instrument QC projects myself and I certainly understand the importance of doing QC, but the way they are going about this is just idiotic.
At least one paper written by the apparantly controversial Dr. Pielke does demonstrate that at least some of the purported "issues" seen in the photos has an actual effect on the measured temperatures. And it is clearly true, as Eli points out, "it's not enough to just look at the stations now, but you have to look at them over time. The station closest to my home was originally located in a grassy field near a municipal airport with a grass runway. Now, thirty-five years later, that runway is much larger and made of asphalt, and an enourmous mall is located adjacent to the property.
It is my understanding from reading past posts by Anthony Watts on various blogs that his goal is not to obtain photographic information from random sites, but from all sites. Therefore, at some point in time in the (hopefully near) future, his documented sample size will be sufficient enough to do some real science with.
But all that aside, if what they are trying to do is "idiotic", how do you suggest they go about it? What would the design of your experiment look like?
... and my understanding -- from what Watts states on his site -- is that his goal is this:
"help us demonstrate that many of the assumptions about climate change based on the surface temperature record may in fact be due to faulty data!"
So, the first -- critical -- step would be to eliminate the bias:
take Anthony Watts out off the experiment.
As a layman, it astounds me some of the comments here criticizing any attempt to photographically document stations.
Even more astounding is that this photographic and historical site documentation appears not to have been done already by scientists.
Has solid statistical quality control been absent from the data gathering from these stations? If so, why?
And why is it falling onto the shoulders of amateurs to provide some visual documentation that climate scientists appear in large part to have failed to provide?
As a layman, it astounds me some of the comments here criticizing any attempt to photographically document stations."
Who is criticizing "any attempt to photographically document stations"?
You are making things up.
=="Who is criticizing "any attempt to photographically document stations"?
You are making things up."==
BBQ's and cell towers next to temperature stations used to reconstruct global temperatures?
I assure you, no one can make stuff like that up.
The sloppiness and carelessness at some of the sites recently photographed suggests poor quality control oversight by professional climate scientists.
That it is falling to amateurs to do some of the most basic legwork that climate professionals neglected to do astounds me.
I thought it would be clear that my comment "making stuff up" had to do with your claim of "criticizing any attempt to photographically document stations", not with the content of photographs.
I am astounded that I have to point this out.
"Sadly the sense filter over at Climate Audit appears working again and excluding some of the bunny's words of wisdom,"
I was going to ask you about your comment at Climate Audit that began "In addition...", since that usually means something came before it and it seemed odd for a first comment.
But then I remembered who monitors/censors comments at Climate Audit and it made perfect sense.
Thanks Henk L.
I updated my De Bilt page accordingly.
Eli, Why are you calling the weatherhut of De Bilt a "Climate Observatory" ? And did you notice the GISS "correction" to it?
Population of De Bilt rose from 3000 in 1901 to 32000 in 1999.
Well, Hans, I could have chosen any one of the Dutch stations, but de Bilt is of such historic importance that I chose it. As to climate observatory, I just thought it a useful description.
So, the first -- critical -- step would be to eliminate the bias:
take Anthony Watts out off the experiment.
Oh come on, spare me. How is the mechanism he has created for posting this data biased? Is he filtering the data somehow? Are his site survey instructions biased? If so, try suggesting changes to make them un-biased. His personal bias aside, he is at least posting data. He may have an express purpose / use / interpretation of that data, but at least he is attempting to collect it and make it available.
I am trying to understand from the dialogue here which one of the following is true:
1) The attempt to collect station siting data is scientifically irrelevant. If so, please explain why.
2) The process and type of data being collected is flawed. If so, suggest how it should be done.
3) The collectors of the data are the problem because of their bias toward a pre-determined outcome that does not fall into line with your conclusions. If this is the case then we really aren't interested in science, are we?
You hit the nail on the head. No, you are not not.
Anon 12:53
The question is: why is this data being collected in the first place? And the answer is bleeding obvious: McIntyre et al are convinced it will lead to the conclusion that the results gathered from these stations are bogus. There isn't even an attempt on CA that that isn't what its being done for.
Ah, but bigcitylib you are missing the point...
If you contribute to Steve's work, you might receive an authorship credit on a real scientific paper, albeit one which can only be published in E&E.
So the conclusion I must draw is that the only ones interested in collecting this information have a clear bias, those who are uninterested in the data have no bias, and therefore no one should be collecting the data. Just wanted to be sure before I put my head back into the sand and did what I was told.
I genuinely hope that the posters to this blog are not working climate scientists. It would mean that climate scientists were a pack of jabbering fools
Eli said; "In your [Steve McIntyre] anger, you have missed the entire point about the picture and US Climate Reference Network, which was posted"
Probably a combination of things, but the failure of the recent article by von Storch to even mention him in this article can not have helped in that regard.
=bigcitylib says:=
=="The question is: why is this data being collected in the first place?"==
No, the question is: Why have climate professionals failed to collect this data systematically in the first place?
Instead of setting up a new, expensive temperature network or ploughing barrelfulls of money into fancier computers, how about the climate professionals do the proper site verification they were obligated to do, and professionally neglected to do, in the first place.
how about the climate professionals do the proper site verification they were obligated to do, and professionally neglected to do, in the first place."
There you go again. Drawing conclusions based on just a few photographs.
Anon 3:04, the information is available, but is not centralized. For example, follow the links in the update to the Swiss Meteo Bureau's reports.
Anon 3:04 wrote:
"Instead of setting up a new, expensive temperature network or ploughing barrelfulls of money into fancier computers, how about the climate professionals do the proper site verification they were obligated to do, and professionally neglected to do, in the first place."
Is that what Steve's up to? How noble!
Except that the tactic so far has been to show pictures of climate observatories--which most of Steve's readers wouldn't have been able to identify if the observatories had run up to them and bit them in the ass, screaming "I am a climate observatory"--and saying "omygawd what a scandal! Its all a scam! The stats the IPCC are using are crap!" I mean, don't you even read his blog, or don't you think other people do? This whole exercise is to gather "evidence" of "sloppiness" on the part of the climate science community, except that the people who are doing the "data collection" have very little understanding of what they are looking at.
anon 3:04 is not interested in data. he is interested in attacking climate scientists with unsubstantiated claims (much like the brotherhood at Climate Audit).
Yes, you are right. It is a joke that the Climate Audit guy (Steve McIntyre) did not just contact the NASA guy (GISS group) to get the data in the first place. McIntyre could have, but he chose not to -- not initially.
Google the phrase "Why should I make the data available to you, when your aim is to try and find something wrong with it?" and report back.
And while you are at it, take a look at how many climate scientists refuse to release source code and data.
It's an outrage.
Instead, he tried to download all the data (with thousands of requests) with an automated script without contacting the NASA GISS webmaster first.
Hardly an outrage. I've done this many times. It's often quicker to code something up than go through the request process. If you get shut down, then you ask. That is what makes the web wonderful. Nobody was denied service.
That's not random sampling and a small number of samples chosen in this way is not sufficient to draw any conclusions about the data.
Data mustn't always be randomly sampled. When an experiment is first set up, processes are always checked in a non-random fashion. That is how scientists and engineers work. Once the process is understood, you move to a random sample. What the folks are doing with the cameras and temp stations makes perfect sense to me. If you are right that there's nothing going on, they will quickly get bored and move on.
I suggest you take a couple intro classes on statistics -- including random sampling
Oh please. If you only knew.
Anon 4:13 wrote:
"Data mustn't always be randomly sampled. When an experiment is first set up, processes are always checked in a non-random fashion."
Presumably, though, you do not conclude from this checking that climate scientists are communists out to hoax the planet. Which makes your methodology different from Mr. McIntyres.
==bigcitylib said:==
==This whole exercise is to gather "evidence" of "sloppiness" on the part of the climate science community, except that the people who are doing the "data collection" have very little understanding of what they are looking at.==
Chronicling sloppiness is bad?
And yes, as a layman, I wouldn't understand the scientific merits of a trash burn barrel being located within 5 feet of a temperature station.
As to "understanding what they are looking at", it is better then some climate professionals who appear to never have inspected these stations in any way, shape or form.
Presumably, though, you do not conclude from this checking that climate scientists are communists out to hoax the planet. Which makes your methodology different from Mr. McIntyres.
Manufacturing and process engineers are the nemesis of product development engineers. And actually, when you first turning on a manufactauring line you ARE assuming the process is VERY suspect.
And if the first few non-random inspections yield surprises, then you revert into 100% sampling until you've collected enough data to convince yourself that the first few surprises were anomolies. Modern manufacturing is a highly statistical process and I suspect that because trillions of dollars are on the line, the processes companies have put into place are quite a bit more rigorous than that in the academic world.
I worked as an intern for the USGS, and while there were some very smart scientists there, they would have had their lunch eaten by the private sector.
When I look at global temperature anomaly maps put out by NASA, I see that the most dramatic warming is occurring in the high northern latitudes (in places like Alaska, Siberia, and Greenland).
Then I see brilliant production engineer types trying to attribute warming to localized urban heat-island effects.
Perhaps said brilliant engineers ought to consult maps.google.com to investigate urbanization trends in Siberia and Greenland before they put too much more effort into cherry-picking temperature-station data in the continental USA.
Are you suggesting that temperature-station data from the continental USA is not important?
Perhaps said brilliant engineers ought to consult maps.google.com to investigate urbanization trends in Siberia and Greenland before they put too much more effort into cherry-picking temperature-station data in the continental USA.
Perhaps said anonymous poster should be less concerned about the motivations of said brillant engineers and go out to said stations and help document them a little more.
This whole exercise is to gather "evidence" of "sloppiness" on the part of the climate science community, except that the people who are doing the "data collection" have very little understanding of what they are looking at.
What understanding do you need to know that putting an AC unit next to a thermostat might skew the results?
Then I see brilliant production engineer types trying to attribute warming to localized urban heat-island effects.
I think you are getting all your anonymous posters mixed up.
Read this if you think Hansen et al have confidence in the underlying data: http://www.climateaudit.org/?p=1648#more-1648
Is it any wonder that "climate scientists" are losing credibility at an accelerating rate.
They are trying to defend the indefensible. Any credible scientist is passionate about the FACTS. They don't engage in ad hominems. They don't challenge efforts to verify data/methods. They understand that independent work that verifies methods and conclusions helps endorse the merit of the work.
As one individual observer, I am appalled at the lack of commitment to sound scientists by the "climate scientists" and their apologists gavin(sigh), Eli, Dano et al.
"Is it any wonder that "climate scientists" are losing credibility at an accelerating rate."
Where is your evidence of that?
That's just a silly statement.
Anon 9:19,
Steve quotes statements made by Hansen et al about 8 or 9 years ago in documents he "neglects" to link to, and implies that Hansen made them in response to work by surfacestations.org. That's almost hilariously misleading!
As Alan Moore might ask: "Who will audit the auditers?"
Let's get one thing straight here.
There is a lot of unsubstantiated hogwash being claimed by some posters here.
Notwithstanding claims to the contrary, the notion that all of us are somehow "opposed to verification" of the temperature data is just silly.
Eli's post was primarily about a system of stations that would do just that -- acting as a check on the data.
Of course, there is nothing wrong with people taking photographs. That can only be beneficial if, after further investigation, it ends up weeding out bad sites. The sites that are shown to be poor should be eliminated from the record. It is really that simple.
The problem as I see it is that some people have an agenda that has little if anything to do with the science and everything to do with politics. We have seen this many times before. They try to make political hay of cherry picked pictures and the like. It does not matter to this crowd if the vast majority of the stations are perfectly fine if a few of the stations appear to have problems.
There is an implicit assumption that everything the contrarians do benefits the science. That is clearly not the case.
Others have commented on this.
Gavin Schmidt replied to Steven Mosher over at Real Climate with the following comment which is also applicable in this case:
"It's generally not the contrarians who drive better understandings of the science, because most of the contrarian points are completely irrelevant and are used as rhetorical, not scientific, points. If, however, you go to scientific meetings like AGU and you sit in on a session where there is some conflict (real or apparent), what you'll see is not contrarians vs establishment, but a whole bunch of skeptical individuals trying all sorts of ways to reconcile the different data. This is certainly the case for the ocean heat content discussion, and the GW/hurricane links, the aerosol issues etc. The 'contrarians' as you see them generally play the role of Greek chorus - just adding to the background noise. They are (with a couple of exceptions, Lindzen having been one in the past) completely irrelevant to the actual practice of science."
'Google the phrase "Why should I make the data available to you, when your aim is to try and find something wrong with it?" and report back.
And while you are at it, take a look at how many climate scientists refuse to release source code and data. "
I am really puzzled by your apparent defense of McIntyre in this case because had he done what most people would have done -- made a simple request for the data, he would have got it.
Are you implying here that there is a conspiracy to keep data from McIntyre and perhaps that McIntyre is paranoid as a result?
The fact is, for whatever reason (quite frankly, I don't care what he thinks) McIntyre initially did not ask for the data (which would have been very easy) but when he did, he got it -- and promptly!
Your points are quite irrelevant to what I said above. Perhaps you might actually read that because you are simply putting words into my mouth .
I never claimed his script was an "outrage".
I simply explained that there was a reason he did not get the data when he used the script and that reason had to do with the fact that the webmaster had no way of knowing a priori that the large number of requests coming quickly in succession from the same address were not part of a denial of service attack.
Perhaps you believe the NASA GISS webmaster has a crystal ball that allows him to look at the motivations of those who are making scripted data requests, eh?
As far as the random sampling comments, I was referring to drawing conclusions about an entire data set based on a subset thereof. For that, one does need random sampling.
I never claimed that taking photos of individual sites yielded no relevant information. It clearly can. It's just that one can not draw conclusions about the data set as a whole from such information. And clearly, it is advantageous to take photos of (and do other QC for) every site in which case one need not do sampling at all.
And by the way, I believe that the sites that are shown to be bad should be removed form the data record.
GHCN QC summary:
http://www.ncdc.noaa.gov/oa/climate/ghcn-monthly/index.php
Also worth noting that GCOS only uses stations that are classified as "rural".
==one of many anononymous posters said:==
="Eli's post was primarily about a system of stations that would do just that -- acting as a check on the data."=
Before funding a completely new and expensive station setup, it would be wiser, and less expensive, to do a proper audit of existing stations.
That climate professionals appear disdainful of doing a proper audit of existing stations, exhibits an arrogance that diminishes them.
"That climate professionals appear disdainful of doing a proper audit of existing stations, exhibits an arrogance that diminishes them.'
Who said that?
You are simply making stuff up here and I for one have gown tired of your BS.
"That climate professionals appear disdainful of doing a proper audit of existing stations, exhibits an arrogance that diminishes them."
That that's posted immediately below a link to the work on how the data is QC'd says it all really.
== anonymous above said: ==
="That that's posted immediately below a link to the work on how the data is QC'd says it all really."=
So a trash incinerator located next to a temperature sensor has been QC'ed?
Thanks for the reassurance!
Come on you anonymous posters, at least sign your posts with something, whether it be an amusing comment or a made up name. That way we can play whack a mole properly.
As for the checking climate stations thing, I welcome it because it will take some denialists away from the internet and hopefully get them some exercise in the real world.
anonymous 8:19, you didn't answer anonymous 8:16 question, you handwaved it away.
"So a trash incinerator located next to a temperature sensor has been QC'ed?"
Go on, read the papers, you might learn something.
The Climate Science Community and the Warmer Community anxiously await the Amateur Auditor Angel Community's scientifically rigorous data collection and empirical, neutral analysis of said collected data.
The Climate Science Community and the Warmer Community anxiously await the Amateur Auditor Angel Community's scientifically rigorous empirical paper detailing the neutral analysis of said collected data. Presumably it will be presented to Journ Clim or PNAS or some such prestigious journal where people who do such work for a living can Audit the empiricism and rigor of the Amateur Auditor Angel Community. The rest of the world, presumably, will read the companion paper in Galileo: The CA Journal of NewScience on the prestigious, groundbreaking CA website.
IOW,
Put up or STFU.
Best,
D
A progress report on Thomas Karl's US Climate Reference Network project would be useful.
The site doesn't appear to post such - one has to ask for it.
Eli do you have it on hand?
Who cares why they want it, it could still be of some use. I'd think everyone would be interested in getting rid of bad data regardless. As far as the stated goals:
1. To provide a standardized method for site survey and reporting so that interested individuals can gather site survey data, pictures, and anecdotal history of climate recording sites worldwide, and upload to a publicly searchable photographic database
2. To provide a repository for screened and approved qualitative and quantitative site survey data, pictures, and anecdotal history
To provide a publicly searchable database of such information for USHCN and GHCN climate station sites
3. To photographically document sites that have been well preserved and maintained through their history
4. To photographically demonstrate examples of sites that may introduce biases and errors through faulty siting, encroachments, or maintenance issues, and to identify specific issues when possible
Find the bad and good stations. That is oh so biased? Sounds like a bias; to make the data more reliable. Good.
Cheers
Ooops, missed the real 3. Sorry.
The people who can't see why it is important and of scientific interest to collect this data remind me of creationists. Their comments are personal, illogical and simply astounding. The aim is to collect this data of all stations. And hopefully not just in the US, but this important scientific process will spread around the world. People who oppose this, oppose scientific research. People who belittle and attack those who are doing this, are not interested in seeking the truth.
It is not the skeptic, but the alarmist who starts with the truth, and tries to make the data to fit it. And their commentary is becoming more and more personal and political. It's ad hominem after ad hominem after ad hominem.
There is nothing they can say in defense, there is no excuse. Just their process to supress, ignore and ridicule all research that might find holes in their climate scripture.
This systematic process was most evident in the WG1 data approval and review processes. All contradictory data and opinions were arbitrarily rejected.
The alarmists have long ago departed from the process of honest scientific inquiry.
It's more and more resembling a religion from top to the bottom.
Just to note that the now-canonical trash burning barrel in Tahoe is of no significance. Contained burning of any sort is banned in the Tahoe basin and the ban is rigorously enforced (across all jurisdictions). Very limited open-air burning of dry vegetation (no trash) is allowed on restricted days during the wet season.
Now, why was this obvious to me but not to a former weatherman living in Chico (much closer to Tahoe than Oakland)?
Also, notice the big deal made on the site about the proximity between the barrel and the station enclosure. Notice also that the barrel is sitting on a bare wooden pallet, which would be at severe risk of burning if the barrel were used. So whatever else is true, there is precisely zero useful information to be extracted from the presence of the barrel.
Lots of people, recently Buddenbrook, have made comments to the effect that someone posting here thinks QC data on climate stations should not be collected. I have not seen this position stated by anyone except those accusing others of holding it. So, for clarity, will anyone who is still reading this thread and thinks QC data should not be collected please speak up now? And, will anyone who thinks they've already seen posts arguing against the collection of QC data please quote those posts and explain how you arrive at that interpretation?
I've seen people say that QC is already being done, eg the comparison/verification network about which Eli originally posted, and the link to the GHCN QC summary. I've seen people express concern that the surfacestations.whatever people are likely to use biased sampling to create the illusion that what already want to believe is really the case. But I haven't seen anyone try to "suppress" anything.
I've seen people saying "if you don't like surfacestations.whatever's methods, how should it be done then?", and then acting like this question had been ignored when previously asked. But In response to those previous iterations of the questions, I've seen responses such as: 1. with random sampling (or using the total population if possible) and 2. without "researchers" who have already made up their mind about what the outcome is going to be, or 3. by comparing the historical systems output to an ideal reference system, such as the one Eli mentioned. And yet people act like their question is not being answered.
It really seems to me like the people claiming that QC efforts are being suppressed are not reading the responses that are being made to their posts. Not reading them carefully or thoughtfully, anyway. But maybe I'm wrong. So...anybody who thinks climate stations should not be subject to quality control (or think you've seen someone else expressing that opinion), please raise your hand.
It's not true that Anonymous is "criticizing any attempt to photographically document stations", he's only criticizing such attempts that don't include going back in time to document the same sites in the past.
Steve Bloom: "Just to note that the now-canonical trash burning barrel in Tahoe is of no significance. Contained burning of any sort is banned in the Tahoe basin and the ban is rigorously enforced (across all jurisdictions)."
And no doubt none of the air conditioners are ever used either. In either case, of course, the photo is useful to understanding the quality of the site. What was the harm in obtaining it?
scabrous k says the problem with Surfacestation is that it is not done 1. with random sampling (or using the total population if possible) and 2. without "researchers" who have already made up their mind about what the outcome is going to be, or 3. by comparing the historical systems output to an ideal reference system, such as the one Eli mentioned.
But: 1) They are aiming to cover all sites. It is not their fault they do not have a million dollars and 100 employees to do them all in a month.
2) The photos are of fairly high quality and as no one has made a claim of fakery in them the motivations of the photographer and web host are close to irrelevant. At any rate, you are free to find your own volunteers from among the many people climate alarmists.
3) A new reference system will tell us little about past biases, only about future ones. As such it has the same flaw as the non-time-traveling photographic survey. Worse, actually, because one can often figure out when elements in the photos were added.
I don't know how anyone could disagree with anonymous 7:03 when he said, "Of course, there is nothing wrong with people taking photographs. That can only be beneficial if, after further investigation, it ends up weeding out bad sites. The sites that are shown to be poor should be eliminated from the record."
posted on 21/01/10
Ofcourse the entertaining thing is that the suspicions of many of the posters above, that Watts et al would use the surface stations project to try to discredit the US temperature record and by implication climatologists everywhere and the IPCC, has turned out to be correct.
Notwithstanding the oh so nice posts by the likes of DWPitelli, it seems that politics was indeed the aim all along.
See this thread and its comments for various links and explanations of the problem with Watts et al.
http://scienceblogs.com/deltoid/2010/01/so_thats_why_surfacestationsor.php
Post a Comment