Spencer shows compelling evidence of UHI in CRUTem3 data

Above graph showing UHI by county population in California, from Goodridge 1996, published in the Bulletin of the American Meteorological Society.

McKitrick & Michaels Were Right: More Evidence of Spurious Warming in the IPCC Surface Temperature Dataset

Guest post by Roy W. Spencer, Ph. D.

The supposed gold standard in surface temperature data is that produced by Univ. of East Anglia, the so-called CRUTem3 dataset. There has always been a lingering suspicion among skeptics that some portion of this IPCC official temperature record contains some level of residual spurious warming due to the urban heat island effect. Several published papers over the years have supported that suspicion.

The Urban Heat Island (UHI) effect is familiar to most people: towns and cities are typically warmer than surrounding rural areas due to the replacement of natural vegetation with manmade structures. If that effect increases over time at thermometer sites, there will be a spurious warming component to regional or global temperature trends computed from the data.

Here I will show based upon unadjusted International Surface Hourly (ISH) data archived at NCDC that the warming trend over the Northern Hemisphere, where virtually all of the thermometer data exist, is a function of population density at the thermometer site.

Depending upon how low in population density one extends the results, the level of spurious warming in the CRUTem3 dataset ranges from 14% to 30% when 3 population density classes are considered, and even 60% with 5 population classes.

DATA & METHOD

Analysis of the raw station data is not for the faint of heart. For the period 1973 through 2011, there are hundreds of thousands of data files in the NCDC ISH archive, each file representing one station of data from one year. The data volume is many gigabytes.

From these files I computed daily average temperatures at each station which had records extending back at least to 1973, the year of a large increase in the number of global stations included in the ISH database. The daily average temperature was computed from the 4 standard synoptic times (00, 06, 12, 18 UTC) which are the most commonly reported times from stations around the world.

At least 20 days of complete data were required for a monthly average temperature to be computed, and the 1973-2011 period of record had to be at least 80% complete for a station to be included in the analysis.

I then stratified the stations based upon the 2000 census population density at each station; the population dataset I used has a spatial resolution of 1 km.

I then accepted all 5×5 deg lat/lon grid boxes (the same ones that Phil Jones uses in constructing the CRUTem3 dataset) which had all of the following present: a CRUTem3 temperature, and at least 1 station from each of 3 population classes, with class boundaries at 0, 15, 500, and 30,000 persons per sq. km.

By requiring all three population classes to be present for grids to be used in the analysis, we get the best ‘apples-to-apples’ comparison between stations of different population densities. The downside is that there is less geographic coverage than that provided in the Jones dataset, since relatively few grids meet such a requirement.

But the intent here is not to get a best estimate of temperature trends for the 1973-2011 period; it is instead to get an estimate of the level of spurious warming in the CRUTem3 dataset. The resulting number of 5×5 deg grids with stations from all three population classes averaged around 100 per month during 1973 through 2011.

RESULTS

The results are shown in the following figure, which indicates that the lower the population density surrounding a temperature station, the lower the average linear warming trend for the 1973-2011 period. Note that the CRUTem3 trend is a little higher than simply averaging all of the accepted ISH stations together, but not as high as when only the highest population stations were used.

The CRUTem3 and lowest population density temperature anomaly time series which go into computing these trends are shown in the next plot, along with polynomial fits to the data:

Again, the above plot is not meant to necessarily be estimates for the entire Northern Hemispheric land area, but only those 5×5 deg grids where there are temperature reporting stations representing all three population classes.

The difference between these two temperature traces is shown next:

From this last plot, we see in recent years there appears to be a growing bias in the CRUTem3 temperatures versus the temperatures from the lowest population class.

The CRUTem3 temperature linear trend is about 15% warmer than the lowest population class temperature trend. But if we extrapolate the results in the first plot above to near-zero population density (0.1 persons per sq. km), we get a 30% overestimate of temperature trends from CRUTem3.

If I increase the number of population classes from 3 to 5, the CRUTem3 trend is overestimated by 60% at 0.1 persons per sq. km, but the number of grids which have stations representing all 5 population classes averages only 10 to 15 per month, instead of 100 per month. So, I suspect those results are less reliable.

I find the above results to be quite compelling evidence for what Anthony Watts, Pat Michaels, Ross McKitrick, et al., have been emphasizing for years: that poor thermometer siting has likely led to spurious warming trends, which has then inflated the official IPCC estimates of warming. These results are roughly consistent with the McKitrick and Michaels (2007) study which suggested as much as 50% of the reported surface warming since 1980 could be spurious.

I would love to write this work up and submit it for publication, but I am growing weary of the IPCC gatekeepers killing my papers; the more damaging any conclusions are to the IPCC narrative, the less likely they are to be published. That’s the world we live in.

UPDATE: I’ve appended the results for the U.S. only, which shows evidence that CRUTem3 has overstated U.S. warming trends during 1973-2011 by at least 50%.

I’ve computed results for just the United States, and these are a little more specific. The ISH stations were once again stratified by local population density. Temperature trends were computed for each station individually, and the upper and lower 5% trend ‘outliers’ in each of the 3 population classes were excluded from the analysis. For each population class, I also computed the ‘official’ CRUTem3 trends, and averaged those just like I averaged the ISH station data.

The results in the following plot show that for the 87 stations in the lowest population class, the average CRUTem3 temperature trend was 57% warmer than the trend computed from the ISH station data.

These are apples-to-apples comparisons…for each station trend included in the averaging for each population class, a corresponding, nearest-neighbor CRUTem3 trend was also included in the averaging for that population class.

How can one explain such results, other than to conclude that there is spurious warming in the CRUTem3 dataset? I already see in the comments, below, that there are a few attempts to divert attention from this central issue. I would like to hear an alternative explanation for such results.

Advertisements

  Subscribe  
newest oldest most voted
Notify of

Problem is that he only uses NH sites so the bias estimate has a spatial error.
UHI varies with Latitude.
It’s higher in the NH than in the SH.
That said, Roy’s results are pretty close to the results we showed at AGU. which was .04C
per decade from 1979 to 2011.
extrapolating down to lower populations is suspect because of the heavy modelling used to derive
the 1km resolution population density data. To understand that you have to read the papers behind the GRUMPV1 population density data.

Victor Barney

[SNIP – Mr. Barney, take this sort of ranting about women, religious issues, etc elsewhere. There is no place for this sort of ugliness here. You’ve been snipped several times before. Final warning. – Anthony Watts]

richard

I would love to write this work up and submit it for publication, but I am growing weary of the IPCC gatekeepers killing my papers; the more damaging any conclusions are to the IPCC narrative, the less likely they are to be published. That’s the world we live in.
then set up another avenue of peer review, open to everyone, with a press release of why.

I tried including SH sites, Steve, but there were none that met the inclusion criteria. Also, see the update to my post…even without extrapolation, the results over the U.S. show a 57% (!) warmer CRUTem3 trend versus the low population station data.
-Roy

Hugh Pepper

This question was dealt with by the BEST study. They concluded that, since only 0.5% of the world is urbanized, even a 2 degree rise in urban temperature would contribute negligibly to the global average. Were they too part of the great conspiracy?

Hector M.

The study is very valuable. However, it does show differences in trends between places with different levels of population density in 2000, but it does not show that increasing population density correlates with increasing warming over time. That would require having population data not only for year 2000, but also for other years (at least for 1970, near the beginning of the series of temps studied). Passing from cross-section data to longitudinal analysis may be tricky.
Besides, residential population density is not the whole story about UHI. Some downtown areas have little residential population but high density of offices and other structures capturing and generating heat, and also intense vehicle circulation. Industrial areas are similar in this regard, with little population living there but lots of factories, refineries, machinery and trucks going around. Also, from the survey of US stations led by Anthony some time ago we learned that many isolated rural stations are now on top of a tin or concrete roof, or in a concrete or asphalt parking lot, but were probably in greener surroundings in the past.
What should be shown is that an increasing trend in pop density and/or other relevant features (buildings, vehicles, engines, generators, highways, airports and the like) correlates with a higher trend in temperature.
However, this analysis is a very useful addition to the literature on UHI.

RockyRoad

Does this mean I’m gonna have to move to a densly-populated area to stay warm for the next several decades. Dreadful is the thought.

MangoChutney

Am I missing something here?
Spencer shows above a bias in CRUtem due to UHI, but in his satellite data he shows a nice sinusoidal curve displaying the natural rhythm of temperature.
Or is it simply that CRUtem disagrees with the satellite data?

David Schofield

Listening to the weather forecast in the UK the weather guy always says the cities will be a few degrees higher. Usually 4+ degrees C.
In response to Hugh pepper surely it’s the amount of thermometers that are urbanised?

RockyRoad

Hugh Pepper says:
March 30, 2012 at 12:46 pm

This question was dealt with by the BEST study. They concluded that, since only 0.5% of the world is urbanized, even a 2 degree rise in urban temperature would contribute negligibly to the global average. Were they too part of the great conspiracy?

Maybe it’s all that additional CO2 they find in the homes of MSA’s–isn’t that what keeps everything warm, that CO2?
Based on your statement, however, the answer to your question is obvious.

Alan S. Blue

Dear Dr. Spencer,
Does the lower troposphere satellite data have the resolution to make this same sort of observation?

Jeff Westcott

So if 50% is from UHI, and up to 50% is from natural cycles (LIA recovery), then the gold standard temperature record will never actually decline, but only level off if “true” temperatures are in fact in long term natural decline. As for CO2, well, never mind.

chemman

Mr Pepper,
If the majority of temperature stations world wide are in urban areas then the fact that urban areas are only 0.5% of the surface is meaningless. The temperatures used to plot an average worldwide temperature is coming from predominately urban areas. So actually BEST didn’t really deal with the issue at hand.

Peter Miller

Hugh Pepper says “This question was dealt with by the BEST study. They concluded that, since only 0.5% of the world is urbanized, even a 2 degree rise in urban temperature would contribute negligibly to the global average. Were they too part of the great conspiracy?”
A classic goofy, alarmist half truth: Obviously only a tiny part of the world is urbanised, but equally obviously most of the world’s temperature monitoring stations are located in that tiny part. These stations then bias the results from the other circa 99.5% of the world.

oldgamer56

Would it be worthwhile to focus on some long term stations that meet the Cat 1 or 2 standard and have experienced transition from rural to urban if they can be linked to some nearby long term Cat 1 or 2 station that has stayed rural? Would suggest that the light density photos from satellite would be the best way to define urban/rural, as it is more infrastructure specific than population.
Would seem this approach would get away from models and be strictly observation driven.

Urederra

Hugh Pepper says:
March 30, 2012 at 12:46 pm
This question was dealt with by the BEST study. They concluded that, since only 0.5% of the world is urbanized, even a 2 degree rise in urban temperature would contribute negligibly to the global average. Were they too part of the great conspiracy?

That is faulty logic, IMHO.
If only 0.5% of the world is urbanized, then only 0.5% of the weather stations used to calculate global temps should be in urbanized areas. It is more logical, IMHO.

Mooloo

This question was dealt with by the BEST study. They concluded that, since only 0.5% of the world is urbanized, even a 2 degree rise in urban temperature would contribute negligibly to the global average. Were they too part of the great conspiracy?
BEST were part of the “avoid the real question and pretend to answer it with another one”.
Yes UHI contributes negligibly to global temperatures.
But what is of concern is how many of the temperature readings are contaminated by UHI.
If we were trying to get an actual “global temperature”, whatever that might mean, we would not want to use a data based known to be heavily contaminated. Unless, of course, we want the answer to be heavily contaminated.

I know that it goes without saying, but do note that Ross and I were only (obviously) looking at land data, and adjusting for the percent of land in each hemisphere gave us a reduction in total global warming of about 18%. That 50% figure that people often quote is within the land data only. Interestingly, our 2007 result changed the distribution of warming to look very much like the distribution of the satellite warming records, cutting off the extreme tails in the thermometric record.
Also, correcting the land record for our nonthermometric effect gave us the same rate of warming as in the Spencer and Christy MSU.
Roy–remember that the Climategaters went after deFreitas for publishing our paper. I have since found out that the situation is now worse, having manuscripts just being rejected out of hand that clearly merit at least a review.

Thanks for this article and the professionalism of the site. I would be very interested in hearing a comment from the author or another expert in response to Hugh Pepper’s point about the BEST study. Is David Schofield’s suggestion correct? It seems distribution of measuring locations would be taken into account by any well constructed study.

NoAstronomer

As David Schofield says at 12:52pm the TV, and radio, forecasts always note that the temperatures will be higher in the cities. Especially for overnight lows. UHI is assumed. Built into the forecast.
It also occurs to do that UHI is not only increasing due to growth and sprawl but on a per capita basis we’re also increasing our energy usage so UHI is probably increasing even where population levels are static.

scarletmacaw

Hugh Pepper says:
March 30, 2012 at 12:46 pm
This question was dealt with by the BEST study. They concluded that, since only 0.5% of the world is urbanized, even a 2 degree rise in urban temperature would contribute negligibly to the global average.

If that is their conclusion then they are idiots. What matters is the percentage of the THERMOMETERS that are urbanized. (David Schofield beat me to it).
BTW, in the BEST study 1/3 of the sites showed cooling, and those were mixed well among the sites that showed warming, so the differences were not due to regional effects but are what one would expect with UHI and micro-climate warming.

Victor Barney

[snip – you’ve been warned, see upthread, and now you are no longer welcome here for this hate speech towards women and religion you are spewing. Mods- delete this poster at will. – Anthony]

AndyG55

Well done sirs.. Now we have what looks like reasonably solid evidence of UHI affecting the calculated global land temp.
Then you think of the loss of all those remote stations which would be in non-dense population areas thus placing more emphasis on those in denser populations, and the constant “adjustments” to make the past colder, and you really have to wonder if there has actually been much warming at all !!
Certainly the records and calculations done by the AGW priesthood (Hansen, Jones, CRU) are NOT going to give us a reliable answer. !!

Joachim Seifert

Lets accept, Had CRUT3 is boosted upward by UHI to some degree, and
HadCRUT4 is boosted further by warm spot chasing in the Arctic…..
Plain sneakiness in statistics, we know this type of people…..
But…..this cover-up of the temp decline is futile, because we have reached
the top temp plateau already from which it cannot get any warmer…. they
achieve only to buy time for a couple more years until the full truth of temp
decline will globally be evident…..
JS

Peter Miller says:
March 30, 2012 at 1:08 pm
“A classic goofy, alarmist half truth: Obviously only a tiny part of the world is urbanised, but equally obviously most of the world’s temperature monitoring stations are located in that tiny part. These stations then bias the results from the other circa 99.5% of the world.”
————————————-
Well Said! Mr Miller 🙂

Dr. Spencer,
Because UHI warming depends on an increase over time, what your study here has shown is that the UHI warming increases faster in higher population density areas. Doesn’t this disagree with your previous results?

Doug Proctor

Hansen and Jones have dismissed the UHIE influence on the global temp profile by saying that the seriously urban part of the record is a small part of the record. However, if in their adjustments they adjust rural sites to reflect urban sites, and not the other way around, then a greater part of the record is compromised.
The main thing is, when the urban centres only are considered, there should be a REDUCTION to the adjustment effect. If isolation of serious urban data does not show the reduction, then there has been no net UHIE adjustments. Then the effect is both real and, for normalization purposes, pushed into the general dataset.

Roy
“I tried including SH sites, Steve, but there were none that met the inclusion criteria. Also, see the update to my post…even without extrapolation, the results over the U.S. show a 57% (!) warmer CRUTem3 trend versus the low population station data.
-Roy”
Yes, the SH stations in ISH and GHCN daily are rather sparse. So, you need to add in other data sources IF you want to understand the UHI bias in the complete record. If I wanted to show the highest bias possible I would just pick a NH dataset. That’s well known.
What you should find ( consistent with UHI literature) is that UHI has a statistically significant relationship to latitude. That is, UHI is lower in the SH than it is in the NH. If you want to establish a bias in the whole record ( spatially complete) then you have to find data in the SH.
Further, I’m not sure you weighted your data the same way CRU weights its data per grid cell.
That’s a minor detail but important. Its also unclear if you looked at CRU grids on a monthly basis.
Further, I would not use GRUMPv1 to do an analysis of the US. There are much better datasets, especially for the US. basically with GRUMPv1 your 1km data is the result of a model.
Again, one of the things you should do is calibrate GRUMPv1 against some known quantities.
That’s pretty easy. If the producers of GRUMPv1 had done a producers accuracy test you could
just cite that, but I havent seen one. Basically, you need to audit GRUMPv1 before just using it.

David, UK

I’d love to read “Hugh Pepper’s” response to the much repeated criticism of his comment.

Victor Barney says:
March 30, 2012 at 1:39 pm
And I heard that women were the thinkers!
————————–
We are…just ask us 🙂

Owen in GA

The fact of papers being automatically rejected is the worst thing I have read here today. We knew they applied pressure, but the idea that the publishers have now lost all sense of propriety means basically science is dead within their pages. If these blogs can’t make up the difference in at least keeping the ideas out there, then we are headed for a darker age then after the fall of Rome.

rgbatduke

Hi Roy,
Suggestions: I know you are doing apples to apples, but 5×5 degree gridding is absurd as it builds in a horrendous projective correction near the poles, and one that is reasonably accurate only near the equator. I’d strongly suggest building an icosahedral tiling of the sphere at a scale-adustable granularity. That way you can scale down to tile sizes that permit fine-grained assignment of stations and populations, and can apply simple affinity/range rules for building the local estimates of UHI warming, without having to worry about whether the grids are in CA, Maine, Alaska, or Ecuador (all of which have very different areas in a 5×5 grid cell no matter how you try to correct for that areal difference. The problem is that the gridding over the sphere is not uniform in the first place, and just accounting for the spherical jacobean cannot “fix” this in the statistics (at least not without a lot of expertise that I strongly suspect is not being correctly applied). A uniform tiling lets you do straight-up statistics with no fanciness.
Aside from the increased ease in doing the stats and integrals right with a good tiling (which can be made entirely automatic in computer code, right, so it is only “difficult” once and then is easy forever after) I really suspect that looking at the actual structure of the data on moderately fine grained icosahedral grid would provide a rather big hit of pure insight, especially if you are seeking to uncover either deliberate if occult bias (of the sort that makes all “adjustments” make the past cooler) or Anthony’s hypothesized poor-siting kind of bias.
As for publishing it, an icosahedral gridding might well help. This is something that might interest a reviewer enough to publish your paper for the method, even if they (want to) disagree with your conclusion. For one thing, it permits lots of folks to save face (if nothing else) if a scalable icosahedral tiling reveals problems or general structures that weren’t seen with lat/lon tiles. It “explains” why they were wrong without making them deliberately wrong, as it were. Indeed, you might well consider calling the paper “Comparing of icosahedral versus lat/lon tiling strategies in Climate Science” and only incidentally point out the resulting 57% underestimation of the UHI effect in at least some of CRUTem4.
It actually paves the way for taking the same data and doing a much better straight up estimate of global temperature, in a defensible way, and one that ALSO badly needs to be applied to the incoming ARGO data.
rgb

adolfogiurfa

No kidding: That´s really anthropogenic: It´s the long wave radiation we emit after eating trash food rich in saturated and unsaturated greases, carbohydrates and proteins….

Jaye Bass

Hugh Pepper was peppered by logic. He won’t be back.

More Soylent Green!

Hugh Pepper says:
March 30, 2012 at 12:46 pm
This question was dealt with by the BEST study. They concluded that, since only 0.5% of the world is urbanized, even a 2 degree rise in urban temperature would contribute negligibly to the global average. Were they too part of the great conspiracy?

That was quite specious logic when proposed by BEST as it has nothing to do with where the temperature records were recorded. Are 99.5% of the temperature records used in BEST from non-urbanized sites?
This was completely debunked months ago. I’m surprised anybody is still trying to get by with repeating it.

Dave

“At least 20 days of complete data were required for a monthly average temperature to be computed, and the 1973-2011 period of record had to be at least 80% complete for a station to be included in the analysis.”
Why? I’d like to hear more justification for this, because these type of decisions are often the bit which decides the results. Did you try with other figures to see what results you got?

Hugh Pepper

Several comments assert that discrepancies can occur because thermometer readings are inaccurate or are calculated disproportionately. But be aware that the BEST studies were based, not only on land-based instruments, but also on satellite calculations.

Crispin in Johannesburg

@Doug Proctor
” If isolation of serious urban data does not show the reduction, then there has been no net UHIE adjustments. Then the effect is both real and, for normalization purposes, pushed into the general dataset.”
++++++
Well argued. An additional effect is not just the population (which is a proxy for urban development) but the wealth that population comamnds. If one were to use municipal taxation as the proxy, one might find a similar correlation and that the increase in population (noted above to be at a lower rate) is not as good. One might use the reported income of the gridcell population rather than the number of people. UHI is really a development impact, not entirely a population impact.

Andrew

Hard data like this needs to be used to challenge the EPA in the courts!
Thank you Roy!

Hugh Pepper says, March 30, 2012 at 12:46 pm :
“This question was dealt with by the BEST study. They concluded that, since only 0.5% of the world is urbanized, even a 2 degree rise in urban temperature would contribute negligibly to the global average. Were they too part of the great conspiracy?”
But if thermometers used are placed in the “0,5%” of the world with urbanization, then UHI is suddenly dominating.
Ex: From 87 hadcrut stations used by Appinsys.com, the avg population is 1,3 mio people.
A typical Hadcrut (crutem3) station has over 100.000 inhabitants in the US:
http://hidethedecline.eu/media/ARUTI/USA/fig9.jpg
Ex, Turkey.
Systematically the 250 stations has been limited in data use so that the rural stations + stations from towns under 50-100.000 inhabitants are limited to years 1960-90. Somewhat larger cities are available from 1950-90 or 1960-2000, and then only the few very largest citites allows us to see the long trends.
http://hidethedecline.eu/pages/ruti/asia/turkey.php
Thus, UHI stations are forced to play a much more dominating role.
This and an endless row of other issues appears not to be considered in BEST.
Sceptics should in no way accept BEST. They did not do the needed.
K.R. Frank

The Other Tex

Correct me if I am wrong Dr. Spencer, but my impression was that you were not comparing your dataset to the overall HadCruT trend, but to the trend in their data on a grid square basis. So if you were only comparing the HadCruT trend for that grid to the trend at the stations in that grid, it doesn’t matter if you have SH grids involved or not, right? What you have established is that UHI does in fact affect the calculated trends. A much more detailed analysis would be necessary to discern the degree to which the UHI contamination affects the entire global dataset; but what this analysis has done is significant, because it has shown the UHI contamination to be real and significant.

RockyRoad

Face it ladies and gentlemen–surface temperatures are simply data points. They are NOT the actual temperature of the earth (or any significant part of it, for that matter). To arrive at the actual temperature in any given time or place, the data points must be projected to the remaining volume, which is almost infinitely larger than the samples. A highly questionable method called “polynominals” is currently being used to make this jump, but far superior methods exist.
For example, geostatistics is used in mining to project small sample values to much larger volumes and the reason is obvious–the profit motive. When climate science gets to the point where they’re using best available technology for this exercise, then we’ll have a much better estimate. Until then, much of what is discussed regarding this data is just conjecture. Interesting, but still conjecture.
(You might say we’re still at the stage where sample quality is a concern, as this paper indicates–there are serious “contamination” issues with our set of data points.)

Jaye Bass

…just answer the question Mr. Pepper.

Gary Pearse

Steven Mosher says:
March 30, 2012 at 1:40 pm
“That is, UHI is lower in the SH than it is in the NH. If you want to establish a bias in the whole record ( spatially complete) then you have to find data in the SH.”
Good point. Surely there is sufficient data in Argentina, Chile, South Africa and Australia/NZ to get some idea. If it showed a similar direction of trend, perhaps lesser slope, this would be good support for a global over estimate. I suspect, like the US/Western Europe, you are probably going to find clustering of thermometers in places like N.Z. and Australia but, since that is the best HADCRUT and company can do, it should relate to the global “record”.

juanslayton

Driving through Quartzsite, Arizona, one is impressed by the number of visitors fleeing the brisk weather of Montana, Alberta, North Dakota….. This annual influx of ‘snowbirds’ increases the local population by many thousands every winter.
Worldwide there are any number of other seasonal movements. Oshkosh, Wisconsin, comes to mind. And the Hajj, to Mecca, is probably the mother of all migrations, with an influx of millions.
I have been wondering for some time if it might be possible to detect a periodic temperature signal from these locations that would give a clue to the magnitude of UHIs.

Bill Illis

The question becomes “how do we take the UHI out and how do we remove the bias of station selection”.
Just use the rural stations that are not in or have been dropped from the GHCN and Crutemp3 station inventory lists.

Kev-in-UK

Gregory Prinsze says:
March 30, 2012 at 1:25 pm
The spatial distribution should be taken into account, especially in terms of assessing UHI – but the only real way is to ‘cut down’ the number of stations used in the total analysis from the known biased areas (i.e. UHI). But I still don’t accept there can ever be a realistic global temp analysis or trend due to all the discrepencies no matter how well it’s bloody modelled!
As far as I can see, only rural, well sited, well documented, top class and top maintained sites can be used for any analysis. All urban sites should be discounted IMO. I cannot see there being many of those rural sites around, but those are the ones that should provide a ‘base trend’ if you like, as in theory, they are the ones most likely to be realistic (mind you, those next to airports and highways must be suspect too – so even for rural sites, a full appraisal is required before use!)
It’s not difficult to see that from several thousand stations – perhaps only a small percentage could actually be of real use.
I don’t buy into the UHI biased station ‘corrective’ adjustments either – no way can such subjective and temporal aspects such as building vents, reflective radiation (from glass or tarmac, etc), be fully considered within a multi decade dataset. Just think of how much your local town/city has changed over the last few years! What about the holiday rush hour type traffic, traffic jams caused by accidents on an urban road near a station, large event parking (football matches), etc, etc – basically, think of all the temporal ‘local’ changes that could occur next to an urban station! How many are recorded and adjusted for! Not many, I’ll wager…..

According to Richard Muller
“Urban areas are heavily overrepresented in the siting of temperature stations: less than 1% of the globe is urban but 27% of the Global Historical Climatology Network Monthly stations are located in cities with a population greater than 50000.”>
Rather says it all.
http://notalotofpeopleknowthat.wordpress.com/2011/10/23/mullers-problem-with-uhi/

Bernie McCune

I am not sure where BEST got the 0.5% of the world as being urbanized but the CIA Factbook indicates that 1.5% of the approximately 30% of land surface is urbanized (I would guess this is out of date). The real issue is to define urbanized. There is a lot of controversy over what this really means or how to determine it. Using light pixels from night time satellite imagery (minus perhaps North Korea) and/or surface disruption features from day time imagery has been considered. Do suburban populations or areas where there are less than 5 people per square kilometer not fit the profile? In other words should there be an urban and rural definition. Some researchers have noted that as much as 3.5% of the land surface has been occupied as of recent times. The majority of studies seem to indicate the “real” value is probably closer to 2.4%. Bounds and definitions are very controversial. An ongoing effort on this subject indicates that with more sophisticated satellite imagery and monitoring methods a much more “firm” figure will be available to us in the next 5 years.
I do not think there is any question that there are a large number of stations in the past 30 years that were originally “rural” and are now “urban”. Many airports with stations that were far from towns 80 years ago are now mostly paved and generally surrounded by urban sprawl. The big question that I have is – How many of the present stations are found in forests and woodlands which make up 32% of the land surface? Or arable land (agricultural) – 10%, pastures 26% – deserts 14%. Or in other land types such as permanent ice, tundra, wild grasslands, steppes, mountains, rocky coasts, coral atolls, salty marshes (or other unihabitable land) that make up the remaining 15%?
The Goodbridge chart at the beginning of this post indicates that all the California station’s results, even the rural ones, are likely to be skewed by population effects. Only in last 40 years or so have we been interested in gathering climatology information from surface stations that were never meant to collect this sort of data. Can surface station data be “tortured” to determine this human climatology effect? I am skeptical.
Bernie

Bernie McCune

I should say that almost 30% of the planet is land and water covers the remaining 70+%.

cd_uk

To Paul
Can I ask you regarding the BEST data. Did they use Kriging to produce global temperature coverage? And then were these interpolation maps used to give global average temperature? If so then there are huge issues here. They need to publish the Krigiing variance maps also – as a proxy of accuracy. We may find that the confidence levels are in the order of the spatial variation in which case one cannot place any confidence in time series data accrued in this way. We need the Krige Variance maps! This will resolve the issue you’re discussing.