Note: See update below, new graph added.
There’s a new paper out by Dr. Edward Long that does some interesting comparisons to NCDC’s raw data (prior to adjustments) that compares rural and urban station data, both raw and adjusted in the CONUS.
The paper is titled Contiguous U.S. Temperature Trends Using NCDC Raw and Adjusted Data for One-Per-State Rural and Urban Station Sets. In it, Dr. Edward Long states:
“The problem would seem to be the methodologies engendered in treatment for a mix of urban and rural locations; that the ‘adjustment’ protocol appears to accent to a warming effect rather than eliminate it. This, if correct, leaves serious doubt for whether the rate of increase in temperature found from the adjusted data is due to natural warming trends or warming because of another reason, such as erroneous consideration of the effects of urban warming.”
Here is the comparison of raw rural and urban data:
And here is the comparison of adjusted rural and urban data:
Note that even adjusted urban data has as much as a 0.2 offset from adjusted rural data.
Dr. Long suggests that NCDC’s adjustments eradicated the difference between rural and urban environments, thus hiding urban heating. The consequence:
“…is a five-fold increase in the rural temperature rate of increase and a slight decrease in the rate of increase of the urban temperature.”
The analysis concludes that NCDC “…has taken liberty to alter the actual rural measured values”.
Thus the adjusted rural values are a systematic increase from the raw values, more and more back into time and a decrease for the more current years. At the same time the urban temperatures were little, or not, adjusted from their raw values. The results is an implication of warming that has not occurred in nature, but indeed has occurred in urban surroundings as people gathered more into cities and cities grew in size and became more industrial in nature. So, in recognizing this aspect, one has to say there has been warming due to man, but it is an urban warming. The temperatures due to nature itself, at least within the Contiguous U. S., have increased at a non-significant rate and do not appear to have any correspondence to the presence or lack of presence of carbon dioxide.
The paper’s summary reads:
Both raw and adjusted data from the NCDC has been examined for a selected Contiguous U. S. set of rural and urban stations, 48 each or one per State. The raw data provides 0.13 and 0.79 oC/century temperature increase for the rural and urban environments. The adjusted data provides 0.64 and 0.77 oC/century respectively. The rates for the raw data appear to correspond to the historical change of rural and urban U. S. populations and indicate warming is due to urban warming. Comparison of the adjusted data for the rural set to that of the raw data shows a systematic treatment that causes the rural adjusted set’s temperature rate of increase to be 5-fold more than that of the raw data. The adjusted urban data set’s and raw urban data set’s rates of temperature increase are the same. This suggests the consequence of the NCDC’s protocol for adjusting the data is to cause historical data to take on the time-line characteristics of urban data. The consequence intended or not, is to report a false rate of temperature increase for the Contiguous U. S.
The full paper may be found here: Contiguous U.S. Temperature Trends Using NCDC Raw and Adjusted Data for One-Per-State Rural and Urban Station Sets (PDF) and is freely available for viewing and distribution.
Dr. Long also recently wrote a column for The American Thinker titled: A Pending American Temperaturegate
As he points out in that column, Joe D’Aleo and I raised similar concerns in: Surface Temperature Records: Policy Driven Deception? (PDF)
UPDATE: A reader asked why divergence started in 1960. Urban growth could be one factor, but given that the paper is about NCDC adjustments, this graph from NOAA is likely germane:
http://www.ncdc.noaa.gov/img/climate/research/ushcn/ts.ushcn_anom25_diffs_urb-raw_pg.gif
Sponsored IT training links:
Pass 1z0-051 exam fast to save best on your investment. Join today for complete set of 642-972 dumps and 650-251 practice exam.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
surely even the UN realises it needs to put some of its vast resources into a rebuttal of this and Anthony’s paper. The silence is deafening. I suppose that if they concede any of this a chain reaction starts, with assumptions on peer-reviewed literature being discredited. Is there any way of identifying all of the papers that would be discredited? Is there any precedent for this? It could be quite amusing to have a web page with all of the AGW papers on one side and the same rebuttal on another.
Mr. Watts…. it looks like Tamino is challenging you to a throw down.
http://tamino.wordpress.com/2010/02/26/thanks/#respond
BBk said: “The trick is that their “correction” isn’t reducing the UHI.. it’s applying the UHI EVERYWHERE! Look at the absolute temp of the recent data of the adjusted graphs.
I think you are correct with this. Halfway down the NCDC/NOAA details of the adjustments it says this:
Application of the Station History Adjustment Procedure (yellow line) resulted in an average increase in US temperatures, especially from 1950 to 1980. During this time, many sites were relocated from city locations to airports and from roof tops to grassy areas. This often resulted in cooler readings than were observed at the previous sites. When adjustments were applied to correct for these artificial changes, average US temperature anomalies were cooler in the first half of the 20th century and effectively warmed throughout the later half.
http://www.ncdc.noaa.gov/oa/climate/research/ushcn/ushcn.html
Urban stations are moved to ‘rural’ areas and the temperatures go down. If I have understood this passage correctly those cooler temperatures are adjusted upwards to maintain the continuity of the station record. This exports UHI to ‘rurals’ sites. There is then nothing to prevent exported UHI from infecting the entire set through the homogenization process.
I cannot also see why earlier temps should be adjusted downwards and more recent ones adjusted upwards. Unless… They know an urban station is running hot. They know a station removed from a UHI bubble will run cooler than before. A bit of cognative dissonance and you could justify the pre-move adjustments downwards (as it was running too hot) and the post-move adjustments upwards (as it is now running cool). Logically they should only do the former.
A general question regarding satellite data. How can we discern between actual temperature changes and simply getting more coverage and more accurate at recording? Before satellites we relied on surface stations, sea surface records and weather balloons. Accuracy has been improving since the 1850s and so too has coverage.
Could the general hump in temps in the middle of the 20th century be a result of more complete coverage than ever before, or since (until satellites) due to WWII activities putting people and equipment in more places to make more observations? ie it wasn’t actually appreciably warmer but our observations were better and more numerous.
It bothers me that the urban/rural alignment is so close pre-1960. Yes, I can imagine the trend separation accelerating over the entire period as urbanisation increased, but there were quite a few sizeable cities (certainly from a UHI perspective) in this country prior to 1960 that I’d expect to see *some* separation –what’s shown here is practically overlaying each other.
Peter Dunford,
I’m of the mindset that UHI should not be corrected for – cities are hotter than countryside, no reason not to show them as little red dots on the maps. I just wish we were in a scientific climate where that could be done without someone saying “ZOMG look that’s CO2!!!1!!ELEVEN!!”
John Holland,
I’m pretty sure I’ve corrected you on this more than once without acknowledgement – Dr. Spencer was not testing UHI, the accuracy of the “adjusted” record, etc – he was *only* testing station drop out. You are reading more into it than what was said.
And for those who are worried about cherry picking or holding up to RC or Tamino’s standard, the guy who posted “witness for the prosecution” earlier had it right. If support for the Endangerment challenge is the aim, there is no need to “disprove AGW”. Demonstrating to a judge or jury that the IPCC link to CO2 (which is the basis of the EPA Endangerment) was unreliable from a burden of evidence standpoint – which by itself this is probably adequate to do – could get the EPA finding thrown out.
Think about the sampling method used in the GHCN v2_mean database. Of the 130-odd stations for the U.S., 92% are airports. In other words, the U.S. sample is urban. This is an undisputed fact. However, only a very small percentage of land in the U.S. is actually urban. (If anyone can find an estimate of the percentage, please post it.) This is yet another undisputed fact.
The scientists who created the GHCN made the decision to over-sample the urban areas and to under-sample the rural areas in the U.S. This decision was most certainly not accidental. And it was not due to lack of data from rural stations.
It appears that the major divergence started in the early to mid 60’s. Why is that time frame significant? That is a major clue to what’s going on. If you can isolate that to just urban growth, station moves or other changes we’ve got it nailed. I’m thinking airports (jet age) and station moves.
Once you’ve nailed the reason for the divergence you can then see how valid the “adjustments” really are.
This looks like a code glitch in the string used to adjust Urban to Rural to control for UHI. We all know that the code is a tangled mess. But with the above graphs, someone like our EMSmith could hunt down the code string by searching for certain parameters that would be limited to the years the discrepancy occurs in the graphs. Is the entire code available and is it in searchable format?
By the way, an interesting piece of trivia, anyone ever see the code for Windows? That is the mother of all tangled messes.
Senator Boxer and the EPA’s Lisa Jackson this week said they did not rely on the IPCC for the endangerment finding about CO2, but instead looked to the US.
Where can they look, now?
The raw rural data shows no drift in temperature over the whole period. Sure, you can put a line through all of those points and it will have a slight positive gradient, but that depends on where you put the endpoints of the line.
The data has an undulation, the amplitude and duration of that undulation is perhaps an interesting feature in that data. But to measure a “linear drift” in the temperature you will need to measure for several cycles of the undulation. I.e. the dataset needs to be at least 3 to 4 times as long before you can start talking about a trend – especially one as small as 0.13C/century.
I any case, what is all the fuss about? I mean, are you kidding? 0.13C/Century. Is that a joke? Am I supposed to be worried about 0.13C/Century? When I look at the temperature record, that is not significant.
Even if there was a 0.79C/century increase, the system is an equilibrium system – it will not continue to go in the same direction because other processes will provide an opposite and balancing force. Why can’t people understand this simple facet of well designed non linear systems?
I am surprised that they have not forced us to shift to mC so that they can say “130 mC’s”, becuase 130 is a bigger number than 0.13 and sounds much more siugnificant.
kzb (04:58:59) :
Is the terrestrial record now of any relevance to the issue? We now have satellite data.
There is clearly a lot of doubt about terrestrial measurements, and you have got to wonder why they are given any weight by either camp in the satellite age.
Having said that, I am a little concerned that no-one seems to seriously question the satellite data. I personally found the “warmest January in the northern hemisphere on record” measurement outrageous. I am sure there must be a “mistake” somewhere.
————————
I’ve been wondering what exactly is the basis for claiming that this past November and January were the warmest ever. Judging from comments by Australians at WUWT, this summer has not been anything exceptional. I have not heard anything about Africa or S. America (nor could I find any reports of exceptional weather in several minutes of doing a search online), but did see reports about abnormally cold weather in northern India in December, which carried on into mid-January. China and N.E. Asia have been in a deep-freeze, as has Northern Europe. The US has been experiencing an unusually cold winter, but Eastern Canada has been around or slightly above average with lower than normal snowfall. Is this claim based on SSTs and satellite data, or is there also some excessive warmth documented at a sufficient number of surface stations that would justify listing the past two months as the hottest on record? Can the satellite data be trusted?
Roger Pielke, Sr has documented the warming bias in the NCDC records, and how NCDC director Tom Karl has actively surpressed that information.
http://wattsupwiththat.com/2009/08/13/pielke-sr-on-warm-bias-in-the-surface-temperature-trend/
This is the same Tom Karl selected by Obama to head a new climate organization. When will the MSM ever report the deviousness and deception?
This, if confirmed, is going to have major implications, not only for AGW crowd, but for some people on “our side” as well, for example Spencer and Christy. Their satellite record show the warming of 0.22 degrees C per decade since 1979 for the USA 48. If this paper is correct, the rural stations in the USA 48 would have at least 3 times lower warming rate than that of UAH! What confidence we can have in the UAH data set after that? I suppose that rural trend is as close as possible to the real climatic trend. It is somewhat ironic and very worrisome in the same time, that UAH satellite data, allegedly free of any non-climatic bias, are much, much closer to the USA urban and rural-adjusted data, than to the real, unadjusted rural record!
Even with extra stations being used in the north-eastern part of the United States, this is an interesting study. I think that possibly even the rural stations in the north east show some UHI, just because population density is higher. (just a guess of course)
Here, researchers, your daily portion of freshly adjusted data, go write your papers. What do you want? Unadjusted data? Do you even know how complicated it was to get it adjusted right? We are doing this for you. How do you think would you detect an anthropogenic signal in that noisy trendless mess of unadjusted time series? So, now shut up and analyze the fine data, we have hidden a correlation in there and the first one of you who finds it and gets it peer-reviewed gets an extra grant.
My questions are these:
1. Are the effects seen in this paper the results of homogenization?
2. Do we know the algorithm used by the NCDC to produce the adjusted data?
3. The NCDC makes the assumption that stations at similar latitude bands have the same climate, is this confirmed across their grids?
It appears that homogenization within the grid was used by the NCDC as a method to produce an average. While this may be appropriate in some instances, any method of averaging must be performed with the utmost care. While urban areas may represent a significant number of stations, their weight as far as surface area represented within the grid should be proportionally small. If this is not corrected for in the NCDC correction algorithm it will result in the effects seen here.
If we do not know what methods are used at the NCDC how can we even evaluate their efficacy? The graphs presented show that there are issues with this system that need to be explained. To adjust data in this way is fraught with hazards, and should only be done to correct known offsets from an individual measuring device. This appears to be a very wide paintbrush saying that it is warmer at this station just because it is hot downtown at the airport. This is contrary to experience and measurements.
So according to the NCDC a station to the west of Chicago will have the same temperature as one buried in the middle of downtown? Or from a station on the other side of Lake Michigan? This seems highly improbable. These assumptions are part of the issue here, and the NCDC should have to prove that they are correct. However, personal experience and measurements know that this is not true.
In summary, this paper calls into question the methodology of the NCDC averaging method. Given the ability of computers today to manipulate data why are we using such coarse grids? Why are not looking at the area represented by each station as part of a sample of the grid? To give a station embedded in a city with less than 1% of the represented grid area characteristics the same weight at a rural station that represents the other 99% (an approximate analogy) would obviously skew the results. Also the unsupported assumptions made further add to the poor results of this product.
This is just a first start in the path of true science of replication of results, that I was taught many years ago. Hopefully the NCDC will respond appropriately and perform some self assessment of the methods used.
JonesII (05:06:33) :
What happened in 1965, when curves began diverging?
————————-
Much larger cities with growing population increasing the heat-island effect as one approaches the centre. Widespread introduction of air-conditioning to commercial and domestic applications. I experienced pre-air-conditioned life in Philadelphia as a child: life changed considerably as air-conditioning became wide-spread and those a/c units do emit a lot of heat. Also, the rise of 2-car families in the 1970s, and the subsequent increase in the number of automobiles more recently. Larger cities mean longer commutes – cars do emit heat. The use of more electronic technology, including microwaves and computers, which were absent in the 1960s. All of these would contribute to warming.
I think most of you are missing the point here. Who cares if the rural was adjusted up and the urban left to follow UHI?
Remember, it is the “anomoly” that matters to this GW argument. So, as has been seen for the past 12 years or so, the vaunted “anomoly” has flattened out.
Why? Because all the crappy temp stations that are by runways or air conditioning units, etc. have now come to stasis within their enviroment!
The GW crowd can keep on saying that the last decade was the warmist all they want, cause for the next decade, the temperature “anomoly” is going to be flat, then the next, then the next……, the graphs for the next thousand years will go back to being a hockey stick handle. 🙂
They have skewered themselves upon their own corrections!
I don’t know about you, but all I see is the beauty of this irony!
Mooloo (01:17:01) :
As for stations selected, they seemed to have picked 48 and one from each state. I would have expected something like one in each of the 5×5 grids that was mapped out. 5×5 grids of course are poor representations of geography and climate zones.
So your “correction” for allegedly poor methodology is … wait for it … a methodology you acknowledge to be poor.
The 5×5 grids are a meaningless abstraction. They should be ignored IMO.
There is only one answer to this paper. To show that urban stations and rural stations do not differ in the manner shown. I doubt they can actually do that, even with cherry picking.
If the AGW crowd cannot show that rural stations do not show warming correlating to CO2, then the allegation that the USA is warming due to CO2 is going to be hard to maintain. By “hard” I mean, of course, near impossible.
>>>>>
The one per state sampling is idiotic methodology – like temperatures have anything to do with political entities and imaginary lines drawn in the sand. At least, with 5×5 grids, there is near uniform coverage of area. It’s an abstraction and a somewhat bad one, but it’s not meaningless in trying to compute average temperate.
When doing science, it is not what your opposing intellectual team can show, but the robustness of the science of your own side. Everyone gets held to a high scientific standard, right? The AGW crowd has some of the shoddiest methodology in science, but stooping below their level makes it hard to mount a credible critique and replace their unfounded conclusions with yours.
Small sample size, 48, is also a problem in the paper, but the results are a bit too strong to be coincidental. One thing that might be explored is the difference between the urbanization of the early 20th century and the urbanization of the late 20th century. There’s a remarkable difference in UHI which suggests that something special about the late 20th century that created a large UHI.
I am shocked at how many have failed to realize that Steveta_uk’s comment was sarcasm.
Get a sense of humor, people.
I could never understand how UHI was minimized. If you look at New York City as an example.
Area, including water 468.9 sq mi ( 2,590,000 sq m)
Power used (2008) 54,869 GW-hr (http://www.nyc.gov/html/planyc2030/downloads/pdf/progress_2008_energy.pdf)
Watts/sq m = 2,416 total. The Mayor says 80 percent is used by buildings and therefore 100 percent ends up as heat loss.
So the forcing is 1,933 W/Sq M
The file also remarks that the city has seen a 23 percent increase in the last 10 years, which is close to the increase showing up in the charts.
I understand that many in the AGW community have rejected UHI as having an effect. Is there any reasoning behind why this could not be? The physics behind it would seem to make sense; more buildings, concrete, people = increase in heat. What is the AGW reasoning why this mechanism would not work? That, to me, would be a very interesting scientific explanation.
Remember its an anomoly chart, taken from different means. The author had to choose a reference year for zero. he seems to have chosen 1960. That would answer a few “where was UHI before 1960” questions.
I had my hopes up on this paper, but took a quick look at a couple of example stations. I can’t say the entire paper is suspect based on this small sampling, but when I find two without trying it leads me to question it. I looked at Beatrice NE and Bedford MA because I’m at least a bit familiar with both areas.
Beatrice 1N, Nebraska is classified as “urban”. However, Beatrice is a isolated town (not part of a larger urban area) of about 12,000 and the station is sited at or very near the airport a half mile from the north edge of town. This is a typical small-town airport with no commercial service – from airnav.com, <30 flights per day, ~25 of them general aviation (think Cessna). If there was any UHI effect on the record, I think it would be very weak. If it's anything, it's a rural station.
Bedford, MA is classified as "rural" and while it is in a more lightly developed area near a park and the relatively open areas north of Hanscom AFB, I would classify it more closely as "urban", or remove it from the candidate list as not being clearly either one. This area has a lot of trees, but a lot of houses and asphalt too. It made the criteria probably because the surrounding towns (Lexington, Bedford, etc) are indeed small but they abut one another in typical New England suburbia.
New Bedford, MA is solidly urban, so I guess you could say one is more urban than the other. The question here is whether (not New) Bedford is rural enough.
I think the author was honestly making an attempt to objectively come up with a list of "rural" vs "urban", but I don't think his classification criteria was good enough. A few hours of recon by Google Maps would be necessary to verify the classification of the other stations.
It'd be interesting to see a similar analysis on a better/broader/more r-word list of stations. Agree that the CRN rating might be a better classification criteria.
Why did it start in 1965? I remember the migration to the cities. The Great Society” benefits were available in cities, Suburbs built up near cities, not in rural areas. Small farms were left vacant and converted to pastureland. Many small communities dwindled away with the young people going to the cities.
Only an occasional small town actually grew in population. I think the “urban” curve mirrors that chabge.