Playing around with my hometown data, I was horrified when I found what NASA had done to it. Even producing GISTEMP Ver 2 was counterfactual.
Guest essay by Philip Lloyd
The raw data that is fed to NASA in order to develop the global temperature series is subjected to “homogenization” to ensure that it does not suffer from such things as the changes in the method of measuring the mean temperature, or changes in readings because of changes in location. However, while the process is supposed to be supported by metadata – i.e. the homogenizers are supposed to provide the basis for any modification of the raw data.
For example, the raw data for my home city, Cape Town, goes back to 1880:
http://data.giss.nasa.gov/tmp/gistemp/STATIONS/tmp_141688160000_0_0/station.txt
The warmest years were in the 1930’s, as they were in many other parts of the globe. There was then a fairly steep decline into the 1970’s before the temperature recovered to today’s levels, close to the hottest years of the 1930’s.
In NASA’s hands, the data pre-1909 was discarded; the 1910 to 1939 data was adjusted downwards by 1.1deg C; the 1940 to 1959 data was adjusted downwards by about 0.8 deg C on average; the 1969 to 1995 data was adjusted upwards by about 0.2 deg C, with the end result that GISS Ver 2 was:-
Being curious, I asked for the metadata. Eventually I got a single line, most of which was obvious, latitude, longitude, height above mean sea level, followed by four or five alphanumerics. This was no basis for the “adjustments” to the raw data.
Which should I believe? The raw data showed a marked drop from the 1940’s to the 1970’s, which echoed similar drops elsewhere. Time magazine covers showed the 1970’s were indeed cold.
The raw data is probably accurate. The homogenized data is certainly not. It is difficult to avoid the conclusion that “homogenization” means “revise the story line” and “anthropogenic global warming” really means “humans changed the figures”.
Prof Philip Lloyd, Energy Institute, CPUT, SARETEC, Sacks Circle, Bellville
I am of the opinion that if the data requires that much adjustment then the data is not good and needs to be remeasured. I have to take measurements all the time and if I had to apply that much adjustments to it then I would not be able to make and accurate assessment of object.
These measurements are from the past. They can’t be remeasured. They can only be adjusted to fit the objective …er… object.
The past is gone. It is past. Whatever numbers they wrote down is all we’ve got.
I seriously doubt if the all the changes in the local temperature logs that went into GISS have behind them the “boots on the ground” kind of examination that changed the hottest temp ever from Libya to Death Valley.
(Might that be the only change that wasn’t made with a keyboard?)
But you can retrofit the station with exactly the same type of LIG thermometer and can take measurements in exact accordance with the same standards and procedures of the past.
We have the raw data for the 1930s and 1940s. Why not replicate the experiment using the same equipment and methodology and we will then know how much it has warmed.
The reason for the warming can then be examined, eg., encroachment of urbanisation/uhi,etc.
If the station has moved then a retrofit is unlikely to tell us that much, but we could examine nearby stations that have not moved and retrofit those.
When you can take a single temperature site in Northern Canada, called the Garden Spot of the Arctic because it is a pocket that tends to be the warmest place up there most of the time, and then use the Garden Spot data to extrapolate the entire Arctic area, it’s for sure that the results are not worth the water that should be used to flush it away.
Richard, the fundamental problem goes back to measurement theory itself; extrapolation from empirical data simply doesn’t ever work. It is never valid. Ever. It’s a rule. Never ever, ever works. Never. Not at all. Under no circumstances. Verboten. Here Be Dragons. Closed course, do not attempt this at home.
Please keep this i mind before apologizing for the few who don’t grasp that concept.
Richard: I neglected to point out the statement I took issue with. My apologies for that. Specifically:
“but we could examine nearby stations that have not moved and retrofit those”
This is the argument for infilling and “Kriging” (which in my mind still sounds like a minority sexual fetish). We can’t generalize accurately. As Higley points out, the variability of our environment can’t be captured by statistical moments with the proposed resolution. Billions have been made by folks who bet on the idea their valley would produce better Zinfandel than the valley 20 miles down the road, and they were right (at least some of the time).
The entire idea of infilling is bogus.
Vive La Différence!
Thanks your comments.
i am not suggesting any infilling of data.
I am not suggesting compiling a global data set.
I am merely suggesting at looking at each individual station where there is concern as to the adjustments made, simply to have a reality check on whether those adjustments to raw data at that particular station are obviously wrong. This is the sort of check which any other science would have adopted.
My view is that the land based thermometer record is worthless, and is employing a system that was never intended to provide information on global temperatures. When cAGW first took hold, they ought to have conducted an audit of all stations, just like the surfacestation project, and selected only the best sited stations, free of any station moves/changes, free of encroachment of urbanisation or other nearby land use changes, those that had the best systems and practices of observation, and had the longest uninterupted record.
One would only need 10 to 20 such stations in each country, or even say just 10 countries across the Northern Hemisphere. These stations could have been retrofited with the same equipment as used in the 1930s/1940s and observations carried out in the same manner as that used in the 1930s/1940s.
One would then have two sets of raw data against which no adjustments need be made. One would simply compare each station individually, to see what if any change had occurred at that particular station. If the vast majority of the individual stations showed little warming, then one would reasonably conclude that there has been little warming in fact. It is not necessary to make a reconstruction of the entire Northern Hemisphere, and try and create a Northern Hemisphere wide anomaly data set.
Why pick the 1930s/1940s as the comparison? First, we know that this period was warm. Second, this is when manmade CO2 emissions took off; some 95% of all manmade CO2 emissions have taken place since 1940. This will therefore shed considerable light on what impact manmade CO2 emissions have made.
Most of us on this site consider that there is a problem with the data sets, but none of us know for sure. My objection is that why use data that we know is bad and then seek to adjust it to make it acceptable/good. I consider that we should have weeded out everything, and only used pristine data. But there are still two problems even with pristine sites. First, changes in equipment. Second TOB. Those two problems can largely be overcome by retrofit with the same LIG thermometers (calibrated in Fahrenheit where appropriate) and taking observations in accordance with the practice and methodology used in the 1930s/1940s. This would give us a sanity check. We have the satellite data after that (although that has its own issues).
Science is about experiment and observation. Why not do this?
There are significant metadata breakpoints for the long term Cape Town temperature series, for instance a station move in 1961:
http://berkeleyearth.lbl.gov/stations/158976
However, you don’t have to adjust historical data to construct regional and global averages.
You only have to cut the series at significant breakpoints, and let the neighbor stations carry the trend over the break…
By the way, GISS don’t even use Cape Town prior to 1942. The data is discarded in that period, probably because there are no nearby rural stations that can be used for the UHI adjustment.
https://data.giss.nasa.gov/cgi-bin/gistemp/stdata_show.cgi?id=141688160000&dt=1&ds=5
@ richard verney – January 29, 2017 at 1:47 am
Right you are Richard V, ….. the much touted per se, US Temperature Record that is in the possession of the National Weather Service, …… is in fact, ….. not a “US near-surface temperature record” even in the wildest of one’s imagination.
The source documents that were used to CREATE the per se, bogus US Temperature Record is nothing more than a humongous collection of archived “weather reports” from the mid to late 1800’s with the data therein accepted “as is” without question or verification of its factuality,
And given the fact that 98% of those pre-1940 “daily weather report” documents were manually recorded ….. and then manually transmitted via the telegraph, the factuality of their data content is highly questionable to say the least. But 100% validity of the “data” was really not that important simply because its only purpose being to generate “one (1) to five (5) day weather forecasts”.
Read more @ – History of the National Weather Service @ http://www.weather.gov/timeline
Anyone that wastes their time, energy and OPM on trying to “squeeze” or ”wring” something of scientific importance out of the per se, US Historical Temperature Record is either miseducated, a nurtured delusional or a happy “troughfeeder” just doing what he/she is being paid to do.
@OR – LANGGEWENS and KIRSTENBOSCH have a similar half degree drop between 1960 and 1970. Not so obvious because of the scaling used. That drop at Capetown after 1960 is used to adjust temperatures by half a degree because of the difference with its neighbors. The whole homogenisation process is a joke.
Used to drive through the bucolic little town in Libya that had the record high. From the first time I drove through I wondered how this place had the highest temp, at one time the town had a British military barracks, always thought those guys rigged that temp.
Bartley, kriging was developed by a South African mining engineer(I believe his name was Krig) around 80yrs ago to estimate a mineral deposit’s tonnage and grade from drill hole and surface sample assays. It is used to this day and there is a large number of mines that are now mined out and proved to have verified the procedure. A priori reasoning (the kind a clever teenager who lacks empirical knowledge uses in arguing with his parents) is rarely a reliable contribution to scientific argument.
This is not said to support the most felonious scientific work the world has ever known – the hopelessly damaged temperature record. This I compare to the Taliban blowing up the thousand yr old hundred foot statue of Buddha.
“It is used to this day and there is a large number of mines that are now mined out and proved to have verified the procedure.”
The problem is that concentration is an intrinsic property. The mean of max and min temps at a particular station is not.
Verifying the procedure is getting a result that was within 5% of the final yield. We have over a third of a degree more warming from adjustments since just 2000. That is 0.1% of the average (absolute) temperature. Its 0.5% of the difference in temperatures experienced in many parts of the world.
I remember when Enron had to make some significant data ajustments…but impact of changing past temperature measurements doesn’t raise the same kind of flags…but I bet my old p-chem professor would be pissed.
Good luck with remeasuring the 1990 temperature…
I am an engineer and if I made those sorts of adjustments to raw data without complete reconciliation and explanation, I would be censured by my professional body.
I have a rule of thumb: If your adjustments are greater than the signal you claim to have found, then you haven’t found anything.
Agreed 110% ;-D
I’m sick of proclamations of “temperature has increased by [fill in the blank] degrees C SINCE [fill in the blank], as if it were (a) factual, (b) meaningful, and (c) something that humans are the cause of, and/or can do anything to stop, when (1) the so-called “data” which indicates such temperature increase is CRAP, (2) the so-called “data” ISN’T EVEN DATA any more, since it has been “adjusted,” “corrected,” “homogenized,” and [insert additional euphemisms for “manipulated with an eye towards gargantuan confirmation bias” here], (3) there is NO empirical evidence to support CAUSATION of the supposed changes, just lots of unsupported or poorly supported assumptions that (i) rising CO2 levels are due to human fossil fuel burning, (ii) rising CO2 levels are the cause of warming, when no such causal link exists in the paleoclimate studies of Earth’s history AND a significant portion of such supposed temperature change can be attributed to other natural climate drivers with even our current limited knowledge (with the remainder probably being DATA ERRORS), and (iii) the temperature rise which is small and beneficial will somehow become runaway and harmful, when (again) there is no such occurrence CAUSED BY CO2 in the paleoclimate studies of Earth’s history.
/rant
Having lived there in the past, it is cool to have a Capetonian look at the data this way – well played
Well I can tell you the weather has not changed….. every year is still the hottest, every year is still the windiest, every year is still the wetest, every year is still the worst storm,……. just for good measure every year the property prices still rocket!
Just shows what a pile of Dingo’s kidneys the historical adjustment of the temperature recoed is.
You folks don’t really eat dingo kidneys? I had to live in England for a few years and they eat beef kidney there, they call it “steak and kidney pie”. Never could understand (or appreciate) it from a culinary perspective.
But wild dog kidneys? I’m not sure, but I’m starting to think the UN should have some culinary criteria before admitting members? You folks still eat lizards?
🙂
Hey, if the French get to eat frogs (their legs, anyway), the Aussies can chow down on lizards if they wanna.
Sorry to disappoint you but from the Urban Dictionary.
“A Douglas Adams original, was used multiple times in his book “The Hitchhiker’s Guide to the Galaxy”. Usually (but not necessarily) following “a load of”, dingo’s kidneys means rubbish, bollucks, crap or bullshit. ”
som
No, its lamb kidneys. You wouldn’t fit a cows kidney in a pie!
beefsteak and lambs kidneys when I make it.
Thank you Tim, I figured they just cut them up. I sit corrected. 🙂
SOM: I rarely find truly educational material on the web. Thanks for the derivation of dingo kidneys. I’ll use that in the future. 🙂
I wonder ??? History ? historical adjustments,? what fun changing history,
In the year1466 (23 May ) Uw Ki Woon ( of North Korea ) flew a modified XY43 super, supersonic, Korean designed and built astro- jet around the world and discovered America.
And that is a fact !!
But dang it , if he would have landed and planted a flag we’d all be North Korean !
It’s good to get down to specific locations where the leftist climate activists have manipulated the data … and show the exact data with links etc where possible. So, awesome!

And realize that those manipulations were done for locations all over the world!
This graph shows the stark divergence between the highly homogenized & pasteurized NASA data and the uncorrupted satellite data, worldwide:
Yes, it’s always interesting to get down to specific sites. Remember Rutherglen, Victoria? Adjusted from a downwards trend to an upwards trend. http://jennifermarohasy.com/temperatures/rutherglen/
Do you look at the USCRN data?
Sorry, but Dr John and Roy have done some nefarious things, they have adjusted non-drifting AMSU satellites that don’t need adjustments.
It’s better to use UAH 5.6 in the AMSU era since it actually relies on those nondrifting satellites that don’t need adjustments..
Surface and upper air together in perfect harmony…
http://postmyimage.com/img2/945_image.png
No, O R, it is you who has done nefarious things by arbitrarily introducing an offset of 0.4 and 0.51 to the UAH plots to force them to match the GISTEMP plot. You give no scientific basis for your adjustment other than the specious claim that the UAH data didn’t need adjustments for drift. If this is true, then why does the RSS satellite series, which is not processed by “Dr. John (Christy) and Roy (Spencer)” show almost exactly the same thing as the UAH plots?
http://woodfortrees.org/plot/gistemp/from:1979/to:2017/mean:12/plot/uah5/from:1979/to:2017/mean:12/plot/uah6/from:1979/to:2017/mean:12/plot/rss/from:1979/to:2017/mean:12
Just for fun I also added HadCRUT4 to debunk the statement by Rick Mears of RSS that the terrestrial datasets are more accurate because more groups analyze them and come to the same conclusions. In fact UAH and RSS are much closer to each other than GISTEMP and HadCRUT4 are.
http://woodfortrees.org/plot/gistemp/from:1979/to:2017/mean:12/plot/uah5/from:1979/to:2017/mean:12/plot/uah6/from:1979/to:2017/mean:12/plot/rss/from:1979/to:2017/mean:12/plot/hadcrut4gl/from:1979/to:2017/mean:12
Intuitively, which is likely to be more correct: the terrestrial data which has vast swaths of infilled (fake) data to cover the oceans and the poles where there are very few measurements, or the satellite data measures 97% of the surface?
#stopdatatorture
Its nothing nefarious with my chart. Giss and UAH 6 are aligned in 1979, the first satellite year.
UAH v6 and v5.6 are spliced in 1999, the first full year with AMSU:s
You have also missed the fact that RSS no longer recommends the use of RSS TLT 3.3:
http://www.remss.com/node/5166
“The lower tropospheric (TLT) temperatures have not yet been updated at this time and remain V3.3. The V3.3 TLT data suffer from the same problems with the adjustment for drifting measurement times that led us to update the TMT dataset. V3.3 TLT data should be used with caution.”
Thus, the faulty RSS 3.3 corroborates the faults of UAH v6 in the AMSU-era 1999-2016
The AMSU drift correction in UAH v6 is NOT corroborated by nondrifting satellite concept used in UAH 5.6
The AMSU drift correction in RSS v4 is corroborated by the nondrifting UAH 5.6, and RSS own similar concepts, REF_SAT and MIN_DRIFT, that were used in the development of v4
http://journals.ametsoc.org/doi/full/10.1175/JCLI-D-15-0744.1
RSS has yet not updated the TLT to version 4.
But in the mean time it is easy to make an RSS TLTv4 with the UAH v6 formula:
TLT=1.538*TMT-0.548*TTP+0.01*TLS.
That trend would be 0.21 C/decade, a little more than UAH’s 0.12 C/decade, isn’t it? 😉
I’m wondering about a missing comment here: my answer to stinkerp, who manifestly ignores how to properly plot, within a chart, anomaly-based time series having different baseline periods.
Maybe somebody at WUWT doesn’t know as well how to do, and thinks my comment is rubbish?
Here is the correction of his second WFT chart:
http://woodfortrees.org/plot/gistemp/from:1979/to:2017/mean:12/offset:-0.431/plot/uah5/from:1979/to:2017/mean:12/plot/uah6/from:1979/to:2017/mean:12/plot/rss/from:1979/to:2017/mean:12/offset:-0.091/plot/hadcrut4gl/from:1979/to:2017/mean:12/offset:-0.293
If stinkerp had any even basic knowledge of UAH and RSS, then he would himself wonder about the strange difference between UAH6.0 and RSS3.3 in his WFT charts!
Nothing in spam folder. Sometimes comments just don’t post. It even happens to me, technology isn’t perfect, browsers sometime fail to send the data.
But, your attitude reveals the smugness behind your fake name.
Olof, you’re at it again, I see.



UAHv5.6 is a flawed dataset. Period. It’s got nothing to do with Christy and Spencer somehow not knowing or caring about MSU/AMSU issues, as you’re continually implying. Its sudden lift in mid 2005 is all due to a spurious disconnect in the data between the oceanic and land portions of it:
Christy and Spencer realised this, investigated the problem, and corrected for it. And with that, version 5.6 appropriately ended up in the dustbin, and the much improved version 6.0 took its place. As it rightfully should.
Further, you cannot compare the tropospheric temperatures with surface datasets like GISTEMP LOTI, Olof. GISTEMP LOTI gl mean is deeply flawed in that it carries within it an out-of-this-world exaggeration of Arctic warming based on brain-dead methods of extrapolating already inflated heat.
You will have to compare it to rather more sober (and reality oriented) series, like HadCRUt3 (adjusted down 0.064 K from Jan’98 due to an obvious artificial, but never noted, much less corrected (!), 0.09 K jump in the HadSST2 dataset, a result of a calibration error occurring when the UKMO changed the source of its SST data going from 1997 to 1998):
There is no discrepancy here between the troposphere and the surface, Olof.
Further, since there is no HadCRUt3 data beyond May 2014, we can extend this particular surface series by simply latching on HadCRUt4 data from this month, carefully calibrating the two datasets to fit up front:





To form what we could call the “Real HadCRUt” dataset:
This series compares very well with JMA gl:
Also with the 3rd generation “Reanalysis Mean” (NASA MERRA + ERA Interim + JRA55), gl T_2m:





While the GISTEMP LOTI gl mean series clearly doesn’t:
You could also compare the UAHv6.0 gl TLT with UAHv6.0 gl TMT and with NOAA/STAR v3.0 TMT, specifically from 1997 onwards, when temperatures plateaued, both at the surface, in the troposphere and in the lower stratosphere:
Not to say the least, compare it with the CERES gl LW flux data (OLR) at the ToA, from March 2000:





CERES also provide a surface temp product called MOA skin temperature, and calculate their gl surface upwelling LW flux (UWLWIR) based on this:
stinkerp on January 29, 2017 at 1:32 am
No, O R, it is you who has done nefarious things by arbitrarily introducing an offset of 0.4 and 0.51 to the UAH plots to force them to match the GISTEMP plot.
Sorry stinkerp, but you are simply wrong here. The offsets used by O R certainly are correct. These datasets have different baseline periods: 1951-1980 for GISTEMP, 1961-1990 for HadCRUT, 1979-1998 for RSS, and 1981-2010 for UAH.
Thus to show plots of all them correctly in a chart, you have to shift them by their baseline differences to a common baseline.
I prefer to use UAH’s baseline as reference, and to shift all others, by computing their mean anomaly value for 1981-2010, which becomes the (minus signed) offset in WFT:
GISTEMP LOTI: 0.431
RSS: 0.091
HadCRUT4.5: 0.293
That gives the following plot:
http://fs5.directupload.net/images/170129/er4sxi6x.png
http://www.woodfortrees.org/plot/uah6/from:1979/mean:60/plot/rss/from:1979/mean:60/offset:-0.091/plot/gistemp/from:1979/mean:60/offset:-0.431/plot/hadcrut4gl/from:1979/mean:60/offset:-0.293
http://woodfortrees.org/graph/uah5/mean:12/plot/uah6/mean:12/plot/gistemp/from:1978/offset:-0.3/mean:12
There has clearly been a pea-and-thimble trick by O R
Robert B on January 29, 2017 at 4:50 pm
There has clearly been a pea-and-thimble trick by O R
Strange indeed: O R specified a much higher offset for UAH5.6 than for UAH6.0, though both have the same baseline period.
But sorry: your WFT graph nevertheless is not quite correct: GISTEMP’s baseline offset from 1951-1980 to 1981-2010 isn’t 0.3 but… 0.43; that difference is similar to OR’s mistake, or trick if you think there was such intention 🙂
http://woodfortrees.org/plot/uah5/mean:12/plot/uah6/mean:12/plot/gistemp/from:1979/offset:-0.43/mean:12
Even more appalling is that no one can even trust NOAA to report raw data accurately. It’s criminal conspirators now put their thumbs on daily temperature readings, before their bosses cook the books further with unwarranted adjustments.
Even when they report data accurately, the criminals cannot do the math correctly.
But NOAA’s written data sure doesn’t add up right.
(1) The Climate of 1997 – Annual Global Temperature Index “The global average temperature of 62.45 degrees Fahrenheit for 1997″ = 16.92°C.
http://www.ncdc.noaa.gov/sotc/global/1997/13
(2) http://www.ncdc.noaa.gov/sotc/global/199813
Global Analysis – Annual 1998 – Does not give any “Annual Temperature” but the 2015 report does state – The annual temperature anomalies for 1997 and 1998 were 0.51°C (0.92°F) and 0.63°C (1.13°F), respectively, above the 20th century average, So 1998 was 0.63°C – 0.51°C = 0.12°C warmer than 1997
62.45 degrees Fahrenheit for 1997″ = 16.92°C + 0.12°C = for 1998 = 17.04°C
(3) For 2010, the combined global land and ocean surface temperature tied with 2005 as the warmest such period on record, at 0.62°C (1.12°F) above the 20th century average of 13.9°C (57.0°F).
0.62°C + 13.9°C = 14.52°C
http://www.ncdc.noaa.gov/sotc/global/201013
(4) 2013 ties with 2003 as the fourth warmest year globally since records began in 1880. The annual global combined land and ocean surface temperature was 0.62°C (1.12°F) above the 20th century average of 13.9°C (57.0°F). Only one year during the 20th century—1998—was warmer than 2013.
0.62°C + 13.9°C = 14.52°C
http://www.ncdc.noaa.gov/sotc/global/201313
(5) 2014 annual global land and ocean surfaces temperature “The annually-averaged temperature was 0.69°C (1.24°F) above the 20th century average of 13.9°C (57.0°F)= 0.69°C above 13.9°C => 0.69°C + 13.9°C = 14.59°C
http://www.ncdc.noaa.gov/sotc/global/2014/13
(6) 2015 – the average global temperature across land and ocean surface areas for 2015 was 0.90°C (1.62°F) above the 20th century average of 13.9°C (57.0°F)
=> 0.90°C + 13.9°C = 14.80°C
http://www.ncdc.noaa.gov/sotc/global/201513
Now for 2016 and they report average temperature across the world’s land and ocean surfaces was 58.69 Fahrenheit >> 58.69 F = 14.83°C
So the results are 16.92 or 17.04 << 14.52 or 14.52 or 14.59 or 14.80 or 14.83 using data written at the time.
Which number do you think NCDC/NOAA thinks is the record high. Failure at 3rd grade math or failure to scrub all the past. (See the ‘Ministry of Truth’ 1984).
We have no idea what the temperature of this planet is. All we can say is that the temperature of the planet lies somewhere between the warmest place on Earth and the coolest place on Earth.
I have seen papers where it has been suggested that the planet is as cool as about 12 degrees, and warm as about 18 degrees, Perhaps we can say that it is 14degC +/- 5 degC.
The 1997 and 1998 reports come with a caveat:
“Please note: the estimate for the baseline global temperature used in this study differed, and was warmer than, the baseline estimate (Jones et al., 1999) used currently. This report has been superseded by subsequent analyses. However, as with all climate monitoring reports, it is left online as it was written at the time.”
They are stupid to use a baseline estimate at all. It is very hard to get a global estimate of absolute temperature, and so it is not surprising that it changed. But the average anomaly is properly calculated.
I have raised this item on many websites many times.
It was the greatest mistake ever made by the NASA “So called” Scientists. They published for all the world to see the ACTUAL TEMPERATURE instead of just the anomalies based on their calculations and left it there.
It can’t be removed now because too many people have screenshots of the data.
They think, like NIck tried to suggest, that by adding the caveat we would be fooled in to ignoring the data and it works with anyone of the warmist persuasion, but not for anyone with a few brain cells still left to do the calculations as DD More has done.
Well done for bringing it up again.
The Berkeley Earth team publish the absolute temperature for their base period 1951-1980, both monthly and annual values.
http://berkeleyearth.lbl.gov/auto/Global/Land_and_Ocean_complete.txt
The annual average for the base period is 14.76 C, in the standard global land/ocean dataset.
The anomaly +0.938 for 2016 means that the absolute global temp was 15.698 C
“It was the greatest mistake ever made by the NASA “So called” Scientists.”
Can’t people ever get anything right? This wasn’t NASA. But it was dumb.
Forrest,
“Nick, baseline + anomaly = actual value. Your explanation is bogus.”
No, that is true only for stations. There is no baseline average normally computed – I explained here why it shouldn’t be.
the link under the first graph doesn’t work for me. Is it correct?
NASA Giss and the warmist fraud squad worldwide have taken a blow touch to the raw data. They haven’t met a raw data temp they cannot Homogenize down in the early years and UP, UP, UP in the later years. These people have no decency or even a a hint of morality and certainly not an honest bone in their body’s!!!
“Your all fired”
The “adjustments™” ALWAYS seem to do exactly the same thing..
get rid of the 1930/40 warming, create a warming trend.
Gees, its almost like they had some sort of INSTRUCTION or AGENDA from somewhere.
Or maybe its something to do with regional expectations.. (opps, that’s a different set of scammers, isn’t it.)
When you look at when the “homogenisation™” routines were invented, its almost as though they were designed specifically for this AGENDA driven purpose.
Yes, just Re-read the ClimateGate emails for the AGENDA & the INSTRUCTIONS.
I smell a rat. NIWA of New Zealand discards pre-1909 data too. What a strange coincidence.
NIWA insists that NZ’s temperature rose by 0.98degC between 1909 and 2015.
https://www.niwa.co.nz/climate/information-and-resources/nz-temperature-record
The evidence for this has not been published in a peer-reviewed scientific paper.
We learned in 2010 that NIWA had somehow lost the source data for the Temperature record.
In 2010 The Minister of Science told Parliament that NIWA would carry out a review of its work on the NZ Temperature Record and during the 2010/2011 financial year would submit the work as a paper to a scientific journal.
Well, we are coming up to 7 years since that work was meant to see the light of day. I could have completed 2 Ph.D. theses in that time.
Maggy – Have you tried to get temp data off their web site? I am not strong in IT but all I get is the name and location e.t.c of each station. No temp data. Am I not looking correctly? I recently requested the 2016 data ( 7- Stn av and anom for the year) and got no reply
@Michael Carter
You could try to access temperature data off the National Climate Database.
https://cliflo.niwa.co.nz/
Don’t hold your breath Maggy, I contracted for NIWA in and around 2001 and I have seen, first-hand, how utterly shonky their work, practices and data is.
A similar thing with Australia.
The 1880s were quite possibly the warmest period in Australia, so the 1880 data is excluded. The reason given is that there are some doubts as to the screens being used, but there is evidence to suggest that most stations had the appropriate screens.
No, screens were rare in Australia in the 1880’s. The big period for screen updating was after Federation in 1901, prior to the BoM taking over state functions.
I was hoping Nick, or anyone, would explain the Capetown adjustments.
I’ve given the NOAA sheet plots below. There were just two major adjustments. One about 1960, which was the year the new airport opened, and you see the big dive in the unadjusted graph. The new airport is inland at about 48m altitude. The other is in 1888, which is a plausible date for the introduction of a Stevenson Screen.
Nick Stokes writes
When much of the warming comes from changing where the temperature is read from …to somewhere else nearby and adjusting for the fact they may have tended to read it in the morning and now evening then I for one get a sense of just how small a change its been over the last hundred or so years. And how uncertain our measurements are.
Its just crazy how scientists can suggest that proxy measurements come within a bulls roar of uncertainty of actual thermometers when the actual thermometers are so sensitive to simple location and timing of readings.
Sounds a bit like what I was trying to say here.
https://wattsupwiththat.com/2017/01/26/warmest-ten-years-on-record-now-includes-all-december-data/#comment-2410547
This is going to be great fun to watch. Popcorn strong buy. The futures just got through the roof.
Raw Temps.



Adjusted (note the scale on the temp axis changed)
From climateexplorer (all) is the raw. (adjusted) is the adjusted.
http://climexp.knmi.nl/selectstation.cgi?id=someone@somewhere
The NOAA sheet for Capetown in here. The plots which show unadjusted, adjustments and difference are below:

Forrest,
“Nick, are the pieces of paper used to record temperatures still in existence?”
Yes. NOAA has facsimiles here.
What, pray tell, is the reason for using anomalies from an average that changes every 30 years and just using the absolute temperatures and graphing them from a degree or so below the lowest reading on record? We’re only talking about a spread of 6 degrees or so. There is some point in recentering graphs to the same starting point in an analysis or presentation. It makes the differences easier to see, the point of showing a graph.
Using broad brush formula corrections to measurements and then using the modified numbers should never be done. There is no way to show that the formula corrections produce more accurate results since there is no way to replicate the original temperature measurements with modern equipment. If the data you have is noisy, erratic,or sporadic that it what it is.
Deal with it. We do not have accurate, pristine, NASA standards data, calibrated and collected every day in a uniform manner. Before about 1980 we have messy data. That should be reflected in the analysis with using error bands instead of imaginary lines.
WMO-
“Because the data with respect to in-situ surface air temperature across Africa is sparse, a one year regional assessment for Africa could not be based on any of the three standard global surface air temperature data sets from NOAANCDC, NASA-GISS orHadCRUT4 Instead, the combination of the Global Historical Climatology Network and the Climate Anomaly Monitoring System (CAMS GHCN) by NOAA’s Earth System
Research Laboratory was used to estimate surface air temperature”
Estimations , you just can’t make it up what NOAA and Giss get up to.
“Estimations , you just can’t make it up what NOAA and Giss get up to.”
This is neither NASA nor GISS. But there is no context. What was being estimated, and why?
actually, contrary to AGW I found minimum temperatures here dropping




I also found rainfall in Potch unchanged over the past 92 years.
both results brought me to wonder why it is that the ice on the north pole is melting and NH temperatures are still rising
until I found the elephant in our room
when I went down 1km into a gold mine here
…..
maybe Philip and I can meet over coffee?
Philip: the geothermal heat flux is fairly well known and it’s not the “elephant in the room” It’s just not enough to affect climate, except locally where large masses of magma are erupted on surface. However, particulates and SO2 from large volcanic eruptions can cause global cooling for periods that may be several years.
HenryP
Can you get to the DUE conference in April at CPUT?
We will meet then.
Help me out here DUE and CPUT
Stands for?
Here’s a paradox: at specific location after location we are finding that … the 1930s was hotter than today, so perhaps everywhere locally it’s colder now than the ’30s, but globally it’s hotter now.
And there’s another paradox: wherever we are in the world it seems it’s colder or just the same as before; it’s just “globally” that’s its hotter. Like it’s colder in the US, in Australia, Europe, wherever. Except in the Arctic and “the south pacific” where no one lives they say it’s hot as h3ll. It’s always somewhere else that it’s hotter.
Another thing that CLEARLY contradicts the idea that we’ve just gone through a century of runaway warming:
Umm…do you have high and low reversed ?
Link shows 6 of 7 set cold more recently ?
YES. My bad. It should read:
By continent, all but one set their all-time cold temperature record more recently than their all-time high temperature records.
You know, I know what the problem is, you are not a climate scientist. And since you’re not, it is beyond your scope of understanding…. that explains it all. See if you were a climate scientist you wouldn’t question it at all. You’d see the temperature going up and just know that that’s right. (sarc, sarc )
Evidence of “anthropogenic” warming? ie caused by humans?
The link to the GISS data doesn’t work. There are actually four GISS iterations of raw data (two stations here and here), after removing “suspicious stations” (one station here) and after homogenization here.
However, the GISS post homogenization data is very different. For example, it starts in 1940 …
So I fear I can’t replicate any of your graphs.
Finally, the Berkeley Earth data is here. It starts in 1865.
Berkeley Earth maintains all of the raw data. If you want raw data I’d use them, although I don’t like what they do to the data after that …
w.
I’d be curious to see what happens if averages were built from the raw data by random sampling rather than adjustments/gridding/infilling. It has been my experience that the more you try and massage the data, the less it reflects reality.
w. ==> Philip Lloyd’s “raw data” graph appears to be this one , the second link you supplied, which is for station ID 141688160007






(or very nearly, re-scaled, of course, and made over into a five year moving average).
His graph (for comparison):
So that settles the first issue — missing link, raw data. It appears that he is correct (MOL-ish).
The second graph Lloyd supplies, appears to be some version of the GISS graph for station ID 141688160000:
Both stations are listed as Capetown South Africa, at the same lat-long co-ordinates.
They are dissimilar, wildly so, so something is amiss somewhere.
Addition: Berkley Earth is somewhat useless, as it fails to identify the Station ID number of its graphs. It uses the lat-long, fails to mention that there are two very divergent station records, and lists only the one that “looks right”?




Perhaps the best overall picture for Capetown is this, from Berkley Earth (though still wildly different from GISS Station id=141688160007):
which shows how unchanging the surface temperature has been in Capetown for the last century.
Compare this Berkeley graph:
Willis, did you ever get a satisfactory answer to the issue posed June 28, 2014 in Problems with the Scalpel Method
To summarize the issue presented:
For clarity: the origin of the sawtooth is a thought experiment
“In any kind of sawtooth-shaped wave of a temperature record subject to periodic or episodic maintenance or change, e.g. painting a Stephenson screen, the most accurate measurements are those immediately following the change. Following that, there is a gradual drift in the temperature until the following maintenance.” Same link: https://wattsupwiththat.com/2014/06/28/problems-with-the-scalpel-method/
The tooth of the saw is the instrument drift. The discontinuity between teeth are the points of maintenance and recalibration. The scalpel process throws away the recalibration information and retains the instrument drift as climate signal.
Sadly, Stephen, that question still isn’t answered. I saw Zeke Hausfather at the recent AGU meeting and he said they were looking at the issue … however, given that that has been the answer since June 2014, I have to confess that I figured his statement would sell at a significant discount from full retail price …
It’s too bad, because both Zeke and Mosher are good smart guys … does make a man wonder.
w.
“The first principle is that you must not fool yourself and you are the easiest person to fool.”
— Richard P. Feynman – 1974 Caltech Commencement address.
There is lots of supporting evidence for this everywhere in the world. Lots of examples in essay When Data Isn’t in ebook Blowing Smoke. Taking a step back from land temp fiddles, there are four bigger issues. 1. UHI and microsite issues (US Aurface Stations.org). 2. Lattitude creep innstation mix toward the equator. 3. Massive land regions with no quality data, like central Africa that mysteriously get infilled as warming when surrounding refions aren’t. (homewood and Heller post these illustrations frequently.) 4. Woefully inadequate SST prior to Argo, enabling Karlization.
Global warming? Probably; the last Thames ice fair was in 1814 at the end of the LIA. How much? Really hard to say. How much of whatever that warming is caused by AGW? Impossible to say because of the attribution problem.
Ruud
Your points !, 2, 3, 4 are all accepted.
Note, however, that the Thames gained two embankments – The Victoria Embankment [N Shore] and the Albert Embankment [S shore] – in the middle of the Nineteenth Century. Per Google [the peerless Google] built in the 1860s.
These dramatically narrowed the river [so increasing flow speed, and, thereby, lessening the likelihood of freezing, even in winters like 1947 and 1963]. The thoroughfare The Strand – formerly the edge of the marshy bit of the Thames is – at its closest – about 100 metres/yards inland from the Victoria Embankment, where the Thames, today is probably only 200 metres/yards wide.
But: “Global warming? Probably” – no dissent, although perhaps the 1930s were warmer than today, despite various folk with skin in the game proclaiming 2014/2015/2016 the warmest ever – EVVVAH!! (sorry, couldn’t resist) – not mentioning that – at best – these are [if the fudged data are actually accepted] only the warmest of the instrumental years.
MWP?
Roman Warm Period?
No – nor the rest.
Pernicious cherry picking from some of the watermelons.
Auto – wrapped in a blanket here in S London.
The removal of old London Bridge did at least as much – it reduced flow massively.
Rud
The Thames frost fairs were held in a variety of locations in and around London.
I attended one in the winter of 1962/3 close to the centre of london
http://www.itv.com/news/2013-01-17/walking-on-the-thames-1963s-big-freeze-in-pictures/
Mind you frost fair would be too grand a term. There were vane parked on the ice selling things like hot dogs, chips pies and sandwiches though.
The party piece were the local young bravados driving their cars across the ice from one bank to the other. To the disappointment of the crowd they all succeeded.
I also walked on the frozen river Thames some time in the early 1990’s but that was further upstream at pangbourne ‘in wind in the willows ‘country.
Tonyb
Nothing like offering a fake Time magazine cover to make a point in the original post …
http://science.time.com/2013/06/06/sorry-a-time-magazine-cover-did-not-predict-a-coming-ice-age/
Well there is “something” like it…like all of the fakes used to make a point about sea level rise, ice melting, etc.
So Time didn’t have a cover story, just an internal one. Yawn. http://www.nationalcenter.org/Time-Ice-Age-06-24-1974-Sm.jpg
Yes I’ll take note of those wise words on your link namely-
“The reality is that scientists in the 1970s were just beginning to understand how climate change and aerosol pollution might impact global temperatures. Add in the media-hype cycle — which was true then as it is now — and you have some coverage that turned out to be wrong. But thanks to the Internet, those stories stay undead, recycled by notorious climate skeptics like George Will. Pay no attention to the Photoshop. It’s the science we should heed — and the science says man-made climate change is real and very, very worrying.”
Kinda kills the very point the senior editor at Time is making when he refers the reader to another panic front page of Time to finish up with-
http://content.time.com/time/covers/0,16641,20060403,00.html
John@ef, yop. I caught that photoshop fr@ud for essay Fire and Ice, and supplied the second real Time cover proclaiming same. Plus links to non-cover Time stories. Plus links to other newspapers for all themother years.. And to counter the notion it was media, not science, links to Holdren’s book contribution on same, CIA’s brief on same, and Nixon’s WH EO setting up a government commission to study same.
Sorry, I was there, it was what we were being taught in schools.
Plenty of evidence, even if one cover is a fake.
And one fake cover doesn’t invalidate the rest in any way. how could it?
must say that my data for CT do fit the original
The fiddling with figures has been apparent at least since 2007.
https://climateaudit.org/2007/08/08/a-new-leaderboard-at-the-us-open/
Yet we are too polite, to a degree that we must spell the word fr@ud in a strange manner or risk being snipped.
I have a hope the New President may put a stop to this nonsense. Harry Truman is said to have had the motto on his desk, “The Buck Stops Here.”
I think a good new motto might be, “The Bull Stops Here.”
Fiddling is just part of the dirty tricks, which started with Hansen’s “testimony” in 1988, if not before.
IOW. a conspir@cy, sc@m, ho@x and fr@ud from the git-go.
@Caleb
“I think a good new motto might be, “The Bull Stops Here.””
I’m certain you don’t appreciate the irony of this motto …
Caleb says, “I think a good new motto might be, “The Bull Stops Here.””
Better trademark that quickly. If you don’t President Trump will.
So this chart below shows Cape Town, South Africa before the manipulators got their hands on the data, and guess what, the 1930’s are hotter than subsequent years, just like unaltered charts from all over the world that show the very same temperature profile.

Isn’t Cape Town in the Southern Hemisphere? I guess that means the heat in the 1930’s was global. The Climate Change Gurus knew this, that’s why they decided to get together internationally to skew these temperature records so they conform to CO2 emissions and to their CAGW narrative. They have been pretty thorough, haven’t they, but now they are going to get caught out on their tricks, their costly TRILLION dollar tricks.
Thanks for that chart. I’ll add it to my collection of charts that show the very same temperature profile.
That’s why they ‘homogenise’.
If the DOE announced $100 million for studies showing that global warming was NOT caused by CO2, academia would be falling all over itself to prove CO2 did not cause warming.
And 97% of them would agree that it did not….
More of the same. Let’s face it these achingly clever NASA/GISS guys NEVER thought they’d be busted adjusting temperatures. Not even if they pulled these stunts for 130 years. They probably thought it the perfect, uncatchable ‘crime’, but here we are with example after example from all around the world. Busted. It should be the most brutal public sector takedown in history.
Mr Trump, get medieval on their arssses with a blow torch and some pliers.
“They probably thought it the perfect, uncatchable ‘crime’…”
I doubt that they thought of it as criminal at all. I think it more likely that they thought that was the minimum changed needed to create the narrative they wanted. They’re “saving the world”, after all.
Holy heavens! I didn’t realize that the data was manipulated to that degree! I am floored.
So…..what is anyone going to do about this?
What CAN anyone do about it…?
“Follow the money” is another way of saying “Follow the Power”.
So far, Trump is doing well in doing something about it.
Trump isn’t god, but so far, in the context of “GAGW”, he’s doing very well.
Keep an eye those, Democrat or Republican, who are opposing or hinting at opposing his nominations.
They are the ones who want Big Government, even if the goals of what they would do with it differ.
DG,
It would be helpful to have a global study that produced say 5 sets of data, starting with least homogenised (raw if you have it), going to a 1st step like removal of outliers and wrong transcriptions, then to a 3rd set with deletions of data where there has been a recorded station shift (instead of trying to correct for the shift), a 4th step where all but ant final GISS etc type post-fact adjustments are made, then the 5th set being the polished turd.
Let the fun begin. By doing a blink comparitor, one might see that the overall direction of change is one way only (cooler past, greater warming with time) whereas the change expected by experienced observers of earth science data would be neutral, but with wide confidence limits.
Oh and to bang a current drum, let us insist on formal, proper, approved calculation of error bounds (including bias as well as stats type precision) in place of the optional extra ways that errors are currently treated.
Geoff
Geoff, nice post. But one never removes outliers unless there is a sound, physical reason, such as a known miscalibration or something. If that’s what was measured and recorded, that is what should be used. Sometimes data is noiser than we want.
DG,
Obvious outliers like 10 times too big, I displaced decimals, etc.
It is obvious why the pre-1909 data had to be deleted. If they had kept it, they would have to at least apply the same corrections to the data pre-1909 as they did to the 1910-1940 data. If they had, that would have resulted in about a 2.5 C temperature rise since 1890 to present. NO WAY did they want to have to explain that.
In my mind, when I see a temp record with the obvious amp or pdo signature, become adjusted and the ocean signal removed, I can’t imagine a more clear cut case for either fraud or absolute incompetence. Science has become political.
The Cape Town raw data has a large temperature drop within only a few years in the early 1960s. I doubt such a large downward step was in the actual temperature at any given location.
Indeed. 1960 was the year the airport opened. Inland, 48m altitude. The station would have moved there from seaside Cape Town. As the NOAA sheet shows, that was one of the adjustments made.
There is no need to make any adjustment. It is a new data point. Like here in Sydney where the airport is reported to show the highest temperature evah since records began. The BoM took over record keeping in 1901. Sydney had an airport then too! /sarc
Pathetic defense of shonky “fake science”.
Inland made it cooler? At an airport? And 48m is supposed to be that much higher so cooler?
I believe we have the same issue here in Oz. IIRC, the raw data show no warming, and slight (statistically irrelevant) cooloing.
If you were to remove 90% of the rural data, and homogenise the rest with urban data, then reduce old data by 0.5C and increase later data by 0.5C, you would get some significant warming.
But I can’t imagine that responsible governments employees would do anything as obviously dishonest as that, would they?
Have you been looking at the “homogenized” data?
The only dishonesty apparent here is that a fake magazine cover is being used as some sort of evidence of something – god knows what since TIME is hardly a scientific resource. One of the obligations of a researcher is to consult source material whenever possible. Obviously Philip Lloyd did not. Or, alternatively, he *knew* the TIME magazine cover was a photo-shopped fake and used it anyways.
The image was actually used on a TIME cover on April 9, 2007 with the caption: The Global Warming Survival Guide
http://img.timeinc.net/time/magazine/archive/covers/2007/1101070409_400.jpg
The cover on the left is a con too. It isn’t about climate, or weather. It is from Dec 3 1973 and is about the oil crisis.
Nick Stokes… the master of the con !!
So the magazine covers are fakes – so what? How does using a fake cover mean the temperature changes made are correct?
There is abundant evidence that climate scientists were warning of cooling in the 1970s.
Your “point” makes no sense whatsoever.
“Your “point” makes no sense whatsoever.”
So what was the point of the covers at all, if being fake doesn’t matter?
There’s a lot of effort being made by warmists to assert that the 1970s concern about global cooling was either non-existent or a trivial part of public discourse.
Well, I do remember it, and yes, there were a few articles about cooling and what it might mean, but it never came to dominate public opinion in the way that global warming has for the last 20-plus years. Two reasons for the difference, based on my armchair analysis:
1. 1970s cooling was not attributed to human activity, so it wasn’t “our fault” and so there was nothing we could do about it. Hence the powerful guilt component wasn’t there to exploit. Plus there were other much more threatening things going on like the prospect of large scale nuclear war.
2. It’s not that the 1970’s cooling was under-reported, but that post-1970s warming is over-reported to the point of insanity(IMHO). It’s been taken over by a loosely coordinated movement with well defined political goals. Well, everyone knows about the IPCC and governments being subverted, and funding for research about “climate”, and the mainstream media being dominated by warmists, so there’s no point in rehashing all that stuff (Tim Ball has several really good posts about the underlying management of the AGW theme). But it seems to me that, if the IPCC and all its spin-offs weren’t hammering the message about global warming/climate change and impending disaster, day in and day out, it would just be another non-event that just possibly might inconvenience our grandchildren, but fades in significance when compared with real and current issues like Isl a mic Ter r o rism.
One possible reason why warmists might want to downplay the 1970s cooling scare is the way it was impressed on us at the time and how that would affect subsequent thought patterns, like this:
a) it was getting colder
b) that (if it had continued) would have been a Bad Thing
c) if cooling is a Bad Thing, then it followswarming must be a Good Thing
d) hence the 1979 to 1998 warming (and a few blips since 1998) has been beneficial
Well, we can’t have thoughts like that circulating, can we? Never mind record crop yields and desert greening….. Warming is a Bad Thing, it’s all Our Fault and so we must pay carbon taxes to try and stop it, and build windmills and buy Teslas and blah blah blah
Adding /sarc just in case anyone thinks the last paragraph is actually my opinion.
The covers are fakes — Time Magazine has all its covers available as (searchable) images at
Time Cover Search.
The author had the responsibility to verify that images before use – and should have done so.
The post should be CORRECTED — clearly stating the\at the images have been found to be fakes.
MODERATOR take note.
Yes, try a search and instead of the covers it takes you to a “latest news” (“The Brief”) page, nothing to do with your cover search.
AGW == > The Time Mag cover search page is broken, in all three popular browsers. (in different ways). If you know the date of the cover you are looking for, you can use the Birthday Covers search on the right of the page, it will give yo the cover closest to the date you enter.
A few topics back Dr. Roy Spencer suggested that we should be studying physical geography in relation to climate change. I agree. We can take this a step further by studying an even more responsive proxy: commercial agriculture – in particular fruit production.
Profit margins of products from the land have always been marginal. As a consequence we find enterprises that survive are almost always clustered into environments that best suit the crop being grown. This comes down to soil type, shelter, sun hours, rainfall, and temperature – in some cases both hot and cold. The difference between economic regions and uneconomic regions often comes down to a few degrees temperature or days of winter frost. In many cases winter chilling is essential. Should climate – in particular temperature – have changed, then specific crops would have migrated.
My own country, New Zealand, is slightly larger than the UK and 63% the size of California. It extends 1600 km (990 m) approx Nth to Sth. The far north is subtropical and the deep south, temperate. Along with an obvious Nth-Sth temperature gradient there is a notable climate variation E-W. The prevailing W-SW winds bring more rain and cooler air temperatures to the west coast. Central ranges make westerlies warmer and dryer in the E.
This results in a great variety of growing environments and localised production centres e.g:
Northland (warm with few frosts): Sweet potato, citrus
Bay of Plenty (warm summer, good sun hours, some frosts): Kiwi fruit
Hawkes Bay: (hot, dry summers, strong frosts in winter) Wine viticulture, stone and pip fruit
Nelson: (warm summer, good sun hours, some frosts) pip and stone fruit, hops, tobacco
Marlborough: (Hot dry simmer, severe frosts in winter) Wine viticulture
Otago: (Very hot dry summer, severe winters) Apricots, cherries
For all of these crops a slight change of temperature, rainfall, or wind, can mean economic make-or-break e.g. too warm in winter and Otago could not produce the cherries and apricots it is famous for.
I was born where I now farm 66 years later. I have visited most of New Zealand and have a good understanding of its geography. Furthermore, growing up we picked up things from parents going back further into history. Guess what? NOTHING HAS CHANGED. There has been no migration of crop locations and production has not dropped.
I remember our family or neighbours planting oranges, figs, Kiwifuit, grapes. The fruit set but never get big and sweet. We are just a fraction shy on sun hours and temperature. This has not changed!
There is another example involving a perennial grass, Kikuyu (origin E Africa). It is very frost sensitive but invasive of most other pasture species that have higher food value. Dairy farming in almost an entire province in NZ (Northland) is now dependent on it as the prime pasture species. It is cheaper to work with it than against it. Production/hectare is probably 30 % less than on top pasture.
Aside from in coastal strips of KM-scale width, Kikuyu runs out south of Auckland City where frost is more common. In my living memory the kikuya boundary has remained remarkably static. A coastal strip within 20 minutes from where I write has not migrated by any more than ½ KM
And only 7 thermometers used by NIWA to make up an average for such variance.
actually i noticed that we know bullocks about the global surface temperature record….
just an example: june 2016 was at our RMI the wettest june ever recorded… this because of an unusual stalling thunderstorm system in a low wind environment. I live some 80 km away near the sea, we had a sea breeze that protected us and i was sitting outside in the sun while inland everything was flooding.
Then the last two weeks the weather was interesting: inversion with no wind.
this is where the coast always has warmer temperatures then inland ‘SST induced heat convection keeps the coast from freezes)
well both urban stations recorded warmer temperatures then even the rural station on a pier at the coast surrounded by a SST of 7-8°C!
that’s odd imho Well not odd when you look at the study made by the university of ghent: they concluded that in our weather patterns UHI can sometimes give an anomaly of +8°C compared to a nearby rural station.
I get a monthly update from NIWA. The language used relates to “average” (“above”, “below”, “about”). – no numbers or indication of what the average base line is. This how National Radio reports too. I asked NIWA for the adjusted data for December 2016 2 weeks ago. I got no answer. We pay their salaries FGS
The entire idea of using a computed temperature average for the entire globe and then using that to compute a climate change has me laughing at the shear incompetence of academia and government.
One simply cannot average temperature readings from a bone dry Arctic/Antarctic region with the high humidity regions over the tropical oceans (remember that 70%?). What is the actual energy change if 10000 square kilometers of polar regions goes up by 1°C and 10000 square kilometers of oceanic tropics declines by 1°C. Dear James Hanson, it is not zero.
If one wants to detect a global trend, then compute the trend from each temperature recording site on its own. There is no need for any homogenizing or “infilling” of data to do so. One can compute a trend at any one site even with lapses in the record. Now perhaps one can compute a global climate change by averaging trends over the globe. Even this leaves too much room for mischief in the averaging method.
However, everyone is aware of the data that tells us any “corrections” that are real must reduce the computed trend. The first thing everybody knows is that the world wide population has increased greatly since 1880. The second item we all know is that the down town temperatures are always higher than out in the sticks (Urban heat island effect). If the corrections do not reduce the apparent temperature increase from the raw data, the “corrections” are incompetent.
Gary Palmgren
“One simply cannot average temperature readings from a bone dry Arctic/Antarctic region with the high humidity regions over the tropical oceans…”
_____________________
Nor does anyone do that. They average area-weighted temperature ‘anomalies’.
Except when they don’t use anomalies. 1997 – 62.45ºF; 2016 – 58.69ºF ( 0.94°C (1.69°F) above the 20th century average of 13.9°C (57.0°F))
No, they do use anomalies. Sometimes there are parts of NOAA dumb enough to add an uncertainly estimated average global temperature. Here they explain why you shouldn’t.
Nick, “Sometimes there are parts of NOAA dumb enough to add an uncertainly estimated average global temperature.”
The climate science part.
The Berkeley Earth team also analysed Cape Town temperatures and came to pretty much the same conclusions as NASA: http://berkeleyearth.lbl.gov/auto/Local/TAVG/Figures/32.95S-18.19E-TAVG-Trend.pdf
That’s a pretty good graph, DWR54. Now if they would plot the actual data, not averages, and then overlay the standard deviation of the data it would be totally honest. Although the 12 month moving average and 95% uncertainty range would tell about the same story. Comparing(it has to be a comparison on a graph) standard deviation in a 10 year moving average to a 12month moving average is a deceitful practice. They aren’t comparable at all.
Substitute “Climate Warming Doomsayers” for “Party”, and the following quote of Winston Smith in “1984” is eerily accurate…
I think it is impressive how people can say stuff like
“the past is falsified, but it would never be possible for me to prove it”
and then show plots (using current data) of how it once was to prove it.
I showed above how NOAA publishes for each station plots of data before and after adjustment.
Let’s hope that Trump has read “1984”.
I’d suggest he start with Animal Farm.
Actually, maybe you should read both to understand what they are about and where “CAGW” and UN Agenda 21 means, if implemented, and what that means for everyone.
Are you in it , or something, McClod?
Looking for yet another selfie. !
NASA-GISS = fr@ud. Drain that swamp.
I would be gad if the author would tell us about his knowledge of changes to the siting of thermometers. BEST shows two station moves at Capetown coinciding with shifts in temperature. Do the alphanumerics mean something to anyone else? I’ve commented on these graphs in good faith on my own website, and found a better GISS one for raw temperatures at Capetown, but I need some reassurance about the rules regarding homogenisation in South Africa.
It’s just too easy to see problems that may not be there.
The adjustment (shown by NOAA here) came in two stages, one in about 1887 and one in about 1960. The first I don’t know, though it sounds like the time for change to Stevenson screen. The second coincides with the opening of the new Cape Town airport in 1960. That is the current site.
A site at an airport doesn’t just suffer from the Urban Heat Island Effect, but from the Jet Engine Effect, as was discovered by the UK MET office last year.
“Capetown South Africa”
As distinct from all the other Capetowns.
Well, Cape Town is.
But no-one needs to be told where Cape Town is. Capetown, on the other hand, …
Not just Capetown. GISS has been altering it’s temperatures for many years. It occurs on a monthly basis. Around 2008 I started saving some of the monthly text files of a few Antarctic stations from their website.
Then, in 2012 I extended that by saving the monthly data files of about 48 stations. It was a random sample of all stations with long continuous temperature records that remained active. Then watching on a monthly basis in 2012, I noticed that about 10 percent of the stations had obvious changes of old data. Most station data from the past remained identical from month to month. Over time, random stations did show significant alterations. In December 2012 a much more radical and extensive alteration of data occurred when compared to the January 2013 data. Some major error occurred to many stations that caused losses of data for all years after 2007, but that was corrected a few months later.
Now with December 2016 saved, all but a couple of those 48 stations have changed old data by various amounts since 2012. Some changes to pre-2012 data show much cooling of data before 1980, resulting in increased warming trends over the decades. There is also a substantial amount of monthly data that have been replaced with “999.9” to indicate data loss. In many cases, old years with missing data have been resurrected with apparently good temperatures. Much of the change looks suspicious in that almost every month has been changed by an exact amount for every record, such as 0.5 degrees lower. Then in later years, that constant change vanishes in one month. Some stations show much larger changes that seem impossible, such as 2 or more degrees colder in the past.
Although most stations show some amount of cooling in the early years, some changes in some cases do show the opposite trend to various stations with no pattern that is obvious to me. And changes can occur at any time for a few months and then a few months later those changes are reversed. It’s very suspicious, and especially obvious when placing a december 2012 station record next to a december 2016 record.
The stations for which december 2012 data saved on my hard drive are:
Akureyri, Amundsen-Scott, Anthony, Bartow, Beaver City, Bridgehampton, Byrd, Calvinia, Concordia, Crete, Davis, Ellsworth, Franklin, Geneva, Gothenburg, Halley, Hanford, Hilo, Honolulu, Jan Mayen, Kodia, Kwajalein, La Serena, Lakin, Lamar, Lebanon, MO, Loup City, Marysville, Mina, Minden, Murteshwar, Nantes, Nome, Norfolk Island, Nuuk, Orland, Red Cloud, Scott Base, St. Helena, St. Paul, Steffenville, Talkeetna, Thiruvanantha, Truk, Valladolid, Vostok, Wakeeny, Yakutat and Yamba.
bw on January 28, 2017 at 11:27 pm
bw, I understand your point.
But I saw lots of comments published at WUWT threads complaining about GHCN adjusted station data being “far higher” than the unadjusted variant.
Within many of them you see graphs comparing the two variants for carefully selected GHCN stations, e.g. Reykjavik, Santiago de Chile, Darwin etc.
That sort of repeated insisting became so boring to me that I computed, for all 7,280 stations, their linear trend in both records, built a list of the trend differences and plotted the sorted data after having eliminated nonsense ( about 0.5 %) due to stations with e.g. exceedingly short lifetime (for example, Tucumen with over 14 °C / century of trend difference, or Elliott with -12, etc).
Here is a plot of the remaining about 7,170 trend differences:
http://fs5.directupload.net/images/170129/u4cx6rim.jpg
(over 4,200 of them are less than ± 0.1 °C /century, and thus not so very significant I guess).
Nobody speaks about the blue part of the trend distribution line.
So your data certainly is accurate, bw, but imho you shoud consider all stations instead of such a little group of them.
Unfortunately, while the complete GHCN station data is available in text form, the corresponding GISTEMP records are in NetCDF if I well remember, and extending a hobby line to that format is too much work. Otherwise, I would have made the same comparison for GISTEMP.
Maybe Nick Stokes has such data…
“Maybe Nick Stokes has such data…”
I do have GHCN data in various forms. That is where the adjustment happens. There are histograms of trends here. Or here you can see the trend effects of adjustment laid out on a Google map with colored markers that you can play with.
Thanks Nick… but I explicitely meant GISTEMP data showing for all GHCN stations, in one continuous dataset, the difference between NOAA adjustment anf GISTEMP homogenisation.
Without extracting the data out of a NetCDF database, we can access the data only station by station using the web link
https://data.giss.nasa.gov/cgi-bin/gistemp/stdata_show.cgi?id=station id&ds=7&dt=1
Nice tool, but.
Nick, could you comment on this? Why should we be repeatedly changing the temperature readings for a period a hundred years or so ago? It makes no sense. If we have new data which casts doubt on a particular reading, change it, and explain why. But repeatedly changing the readings with no explanation cannot be science, surely? You do not treat observations like this if you are doing science. The right way to do it is plot the observations as they are, then in the argument on the paper or assessment, explain why there may be biases in the record. But you cannot just change observations all the time with neither reason nor notification.
The alternative conclusion would be that we simply don’t know what past temperatures were from the instrumental record, which may be reasonable, but in that case we have to stop plotting them as if we did know.
The Capetown case as graphed, assuming the author has done his work correctly, is really weird isn’t it? There can be no justification for altering observations from the 1900s.
To add a bit. There is a well known thing in perfectly respectable science, when observations don’t fit a theory. The observations will always be to some extent theory laden, so it may be reasonable to question whether they are right if they are different from what a well confirmed theory says they should be.
This is quite reasonable. This is, for instance, the French paradox on heart disease and saturated fat in the diet, and we ask, are we sure they are recording all the heart disease that there is? Maybe they are calling it mal au foie?
But in this case there is no theory that gives any indication that the observations are out of line. There’s no reason to think that the temperatures really were a bit higher or lower in Capetown on some day in 1908 than those which were written down.
It just makes no sense to me. If this is the foundation of the view that modern warming is unprecedented, one is inclined to conclude the whole thing is rubbish.
There is no French “paradox” on heart disease and saturated fat. The studies used to attempt to demonize fats did not separate saturated fats from trans-fats, and it is trans-fats that are extremely unhealthy. Saturated fat is good for you, and is a staple of the human diet. The “fat is bad” science conflated the effects of certain facts as being applicable to all fats, and was junk science – just like AGW is.
Should be “applicable to certain FATS as being applicable…”
” There can be no justification for altering observations from the 1900s.”
On the contrary – if there is evidence of a change at a station, you have to take it into account. And I suspect people here would be all over them if they didn’t.
As I showed in my comment here, there are really only two adjustment made to cape Town data, one in about 1878 and one in 1960. I am not sure about the earlier one, but it is not very significant anyway. The one in 1960 removed the massive dive you see in the unadjusted data that year. So they look around. Here is Berkeley’s plot of temperatures nearby relative to Cape Town
http://berkeleyearth.lbl.gov/auto/Local/TAVG/Figures/32.95S-18.19E-TAVG-Counts.png
As you see, looking at stations within any of a variety of radii, there is a massive dip. That is, the change at Cape Town did not appear in the neighboring sites. And you can measure the difference.
Then a bit more research shows that the new Cape Town airport opened in 1960. It was then quite a bit out of town. That’s a substantial move. Adjustment is required.
There is no indication that Cape Town is “repeatedly changed”. There seem to be just a few major ones, probably done only once. It’s true that the global average changes frequently, but that is because thete are many thousands of stations, and changing any one will change the average.
NO.
A change of siting is a new station, not a continuation. Hence,
You have one set of data up to the change.
Another set of data post the change.
No splicing of the two together. They are simply different stations and should be handled as such.
The same with equipment change. as soon as there is equipment change, you have a new data point. One should not splice the two together, not unless there has been at least 10 years of overlap with both types of equipment installed, wherein a proper assessment of bias introduced by the equipment change can be assessed.
michel on January 29, 2017 at 12:40 am
There can be no justification for altering observations from the 1900s.
As Nick wrote: “If there is evidence of a change at a station, you have to take it into account”.
I allow me to add that anomaly based temperature records (without which we see was is biggest but don’t detect what differs the most) have a fundamental drawback: the fact that any data change in the reference period, called climatology (here: 1951-1980) automatically results in a modification of all the record’s anomalies, as these all are constructed month by month (or even day by day) by computing the difference between an absolute value and the monthly (daily) mean of the reference period.
“allow me to add that anomaly based temperature records (without which we see was is biggest but don’t detect what differs the most) have a fundamental drawback: the fact that any data change in the reference period, called climatology (here: 1951-1980) automatically results in a modification of all the record’s anomalies,”
So a change because of a legitimate reason in one part of the record may cause an illlegitimate change in other parts of the record when using anomalies.
Exactly. Better is to look at the average rate of change per annum.
TA on January 29, 2017 at 9:45 am
So a change because of a legitimate reason in one part of the record may cause an illlegitimate change in other parts of the record when using anomalies.
lllegitimate? No, TA. Because the absolute data the anomalies are originating from was not modified.
If you can’t live with that, so please use that absolute data instead. We all are laypersons here, with no binding to career or superiors’ meaning.
For me there is no way back since I learned that these “anomalies” are no simple deltas wrt some overall mean: if you have a monthly record, your baseline is a 12 month vector; the same holds for daily records, the baseline then having 366 units.
And that’s the difference you best see when looking e.g. at sea ice extent.
If you sort ice extent e.g. in the Arctic by increasing surface using absolute values published by colorado.edu, you will see at top, as expected, lots of septembers, then a mix of august/september/october. Far below at position 78 (of 456) you detect a timid july 2012. The first july (2016) appears at position 167, the first may (2016) at position 216, etc. The first winter month is at position 289.
But when you now sort the stuff by anomalies, the list gets quite different. The first july (2011) appears at position 14, may (2016) at position 31, and the first… february (2016) at position 42. A winter month!
The same holds of course for the Antarctic.
Thus I keep this for me: anomalies, i.e. positive or negative departures from an average, aren’t a tool fabricated by warmistas to scare people. Their use is in removing annual cycles in time series. That’s all I see in them.
ftp://sidads.colorado.edu/DATASETS/NOAA/G02135/north/daily/data/
ftp://sidads.colorado.edu/DATASETS/NOAA/G02135/south/daily/data/
Forrest Gardener
to understand movement of global T and weather/rainfall you must understand that there are two sources of the energy here, namely the energy coming from outside in and then there is the energy from inside earth to out. The strength of the latter becomes more evident when you go down 1km into a gold mine here…It seems that everyone has assumed that energy inside to out is more or less constant, and I think that is true – when measured over a short period [e.g. the Holocene] -, but, like I said, my finding here for the past 40 years is that most of the SH is cooling whilst the NH is warming. I could not figure that one out. A simple theory [that now makes sense to me] is that earth’s inner core is aligning itself with the magnetic force from the sun – very much like a magnetic stirrer, if you like, [if you know what that is?]. Indeed the evidence clearly shows that Earth’s magnetic North Pole has been moving northwards, and that movement is causing said melting of ice at the north pole and relative more warming in the NH.
As far as rainfall goes, I think this is largely influenced by the solar cycles, i.e. the energy coming from outside to earth. My finding is that William Arnold’s report, back in 1985, before they started with the CO2 nonsense, on the solar cycles is largely correct, and I have subsequently identified that one complete solar cycle (Hale-Nicholson) consists of two succesive Schwabe solar cycles. 4 Hale cycles makes up for one Gleissberg cycle 87 years, of which the first 43,5 years is the mirror of the next 43,5 years.
To try to explain, I give you another example.
http://oi66.tinypic.com/einoz6.jpg
Clearly you can see that the result from the first few decades of the 20th century falls off from the curve? It is because the Gleissberg cycle causes the pendulum to fall down for 43.5 years and then it goes up by 43.5 years.
I have to take a break now, but feel free to ask me more questions.
The top link to Cape Town Safr station 141688160000 is wrong (maybe other commenters have seen it):

… and this link doesn’t work too, even right now in my own browser. Probably the NASA web engine produces short-living links only like do many others.
I hope this one lives a bit longer:
https://data.giss.nasa.gov/cgi-bin/gistemp/stdata_show.cgi?id=141688160000&ds=7&dt=1
Capetown’s measurements are based at the airport, which like all airports, has grown in leaps and bounds over the years.
Any adjustments should be the other way round, increasing historical temps
I suppose the only response is: God damn, wat ‘n vrag van kak!
So lets review the changes and make some obvious assumptions:
1) the data pre-1909 was discarded;
– On Jan 1 1910 the station data became good. Obviously it got upgraded.
2) the 1910 to 1939 data was adjusted downwards by 1.1deg C;
– On Jan 1 1940 they replaced the old station for some reason. We know this because the data was changed differently on Dec 31 as it was on Jan 1st.
3) the 1940 to 1959 data was adjusted downwards by about 0.8 deg C on average.
– On Jan 1 1960, exactly 20 years after they changed the station, it was replaced once again and we know this because on Dec 31st the results were different.
4) the 1969 to 1995 data was adjusted upwards by about 0.2 deg C.
– Then one again on Jan 1st, but this tome 1970, they again changed the station.
My powers of observation are quite clear. To see what is happening to the data, we only need to observe the station in Dec 31st to Jan 1st of each decade to see who is going in to change the station.
I am a little confused as to how a computer process was able to know these things happened but the people that programmed it do not. But computers are really smart right?
Never underestimate a computer. They are wonderous things used by really smart guys and dolls..
For those confused by the lack of information about ‘The Garden Spot of the Arctic’, it is Eureka, Nunavut.
Retroactive homogenization by seems to be one of the most dubious things that can be done if objective decision making is the goal. Now if selling a new program is the goal, then retroactive homogenization is a preferred tool of choice.
Up thread there was a fascinating idea: Recreate the *devices used at the time*, and make side-by-side measurements with modern instruments and see what the difference might be in current climate. Perhaps that would clarify the question as to whether or not the steady adjustments (oddly preferentially down) of the past are justified.
Nick Stokes January 29, 2017 at 3:12 am
Nick, sometimes you are so right, and other time … that is not “Berkeley’s plot of temperatures nearby relative to Cape Town”. LOOK AT THE Y AXIS LABEL. It is a plot of the NUMBER of stations within a certain distance. As a result, all of your lovely conclusions about the plot and about Cape Town based on the plot are … well …
Priceless.
w.
Willis,
Yes, indeed, you are right and I misread the diagram.
Willis,
I have now found that the old Cape town station, CSIR, that moved to the airport, also continued on as a separate GHCN station. So we can do direct comparison of changed and continued. It’s in my reply to Phili[ below.
Thanks, Nick, I saw that. Nice piece of detective work.
w.
Errare humanum est 🙂
The data manipulation goes way beyond just the temperature data. They manipulate the peer review process, they only give the public 1/2 the information, and they adjust their data to “hide the decline” so their models give them the answers that they want.
Climate “Science” on Trial; Data Chiropractioners “Adjust” Data
https://co2islife.wordpress.com/2017/01/29/climate-science-on-trial-data-chiropractioners-manipulate-data/
I have in an Excel spreadsheet the record highs and lows reported for my little spot on the globe from 2002, 2009, 2012 (twice) and a more recent ones. The numbers have been changed, some as much as 5 degrees F.
After all the manipulations, I think it’s safe to say that we don’t really KNOW for sure what the past temperature of the globe has been. At best we have educated guesses (along with a few political assertions).
Bet a few trillion in a game where a few cards are missing from the deck?
From NOAA temperature station maintenance teams:
Ocean temperature measuring devices, buoys, ARGO, JASON, etc. are similarly affected, only by different organisms; e.g. algae, barnacles, mollusks, bird droppings, etc.
Each time the instrument used is changed introduces error.
The size of the error is unknown unless there were side by side exacting measurements for several cycles (years). Crude changes to high/low temperatures are equivalent to fresh manure of the foulest origins
Each time a temperature instrument is changed, that is the end of temperature measurement for the previous instrument and the beginning of a new temperature measurement series from the new instrument.
As mentioned in other articles, splicing is improper.
Showing the various temperature series on the same graph is possible so long as each end/begin is clearly demarked.
Each time a temperature station is infested/inhabited by any creatures, that contaminates the temperature record.
Whether it is wasps/bees cooling and warming their nests or chipmunk squatters, the temperatures taken during the infestation are contaminated. There is no possible adjustment!
Wasps/bees cool their nests during the day by helping the structure ventilate and then turn around to warm the nest at night, by vibrating/buzzing to generate warmth.
This brings up the classic comment, “good enough for government work”. But seriously suspect for declaring anomalies/records by fractions of a degree.
Siting temperature stations in poor locations or allowing the local land use to change again, seriously compromises temperature records.
Stations record day/night differences anywhere from a few degrees to over thirty Fahrenheit degrees on any day.
Changes in land use affect these daily and seasonal fluctuations; and as people notice in their car thermometers, UHI can change measured temperatures by multiple degrees.
Adjustments must be based on long term dual measurements tracking the changes by minutes. Again over multiple full cycles.
Our previous discussions, involving Engineers and discussing how error rates are properly carried forward from initial measurement through presentation, included a number of other issues. But reached the same conclusions.
Spliced records are improper and usually invalid.
Adjustments to records should never be performed without full disclosure of all adjustments and reasons for every adjustment. Showing the historical record and then following with a graph of the proposed adjustments is best, so that analysts understand the full impact.
In the real world, changing or adjusting data is wrong and often illegal. When the bank makes an adjustment, they have all metadata, every record and reason for every adjustment. Even then, adjustments are clearly marked in any charts.
When architects design a structure, but the owner or contractors involved decide to change numbers so they can utilize ‘shortcuts’; when discovered, charges and/or fines are applied.
When structures are repeatedly invaded by pests, it is time to devise as impregnable a structure as possible; or perhaps completely revise how temperature is measured.
Strategy “search and destroy”:
“Being curious, I asked for the metadata. Eventually I got a single line, most of which was obvious, latitude, longitude, height above mean sea level, followed by four or five alphanumerics. This was no basis for the “adjustments” to the raw data.”
__________________________________________
Escalates any following ‘civilized’ debate.
Here we can see that 2016 wasn’t everywhere the hottest evah:
http://ds.data.jma.go.jp/tcc/tcc/products/gwp/temp/fig/dec_wld.png
But yes it’s true, we must take that with some caution: the JMA is known to be highly underrepresented in the Arctic regions.
Exactly where even UAH6.0 shows, in its 2.5° grid data, a warming trend of over 4 °C / century, namely in the latitudes 80N – 82.5N.
Anyway, it’s nevertheless nice to read that not all people tell you the same refrain.
Here we can see that 2016 wasn’t everywhere the hottest evah
That’s the month of December. JMA ranks 2016 as the warmest calendar year.
http://ds.data.jma.go.jp/tcc/tcc/products/gwp/temp/fig/an_wld.png
http://ds.data.jma.go.jp/tcc/tcc/products/gwp/temp/ann_wld.html
This time you’re right barry! I clicked on the wrong line in the browser’s history.
Thank you.
That could be a reason for a ‘full stop now’ to any further data access by members of the former Obama administration
until the data is shaped into a reasonable form
of the original data till now.
Thank you, Nick Stokes and others for providing the missing metadata. It, however, does not provide a very satisfactory explanation for changing the raw temperature data. Nick’s Berkeley Earth glitch lets me draw attention to the fact that several other stations along the South African coast saw a marked drop in temperatures over the 1950 – 1980 period. Being coastal, I would have expected an oceanic impact to moderate any nearby land temperature shift.
I am interested that more did not comment on the cooling. There had to be some reason for the Time covers. There is an excellent explanation of this hiding the decline over at Climate Depot http://www.climatedepot.com/2016/09/13/83-consensus-285-papers-from-1960s-80s-reveal-robust-global-cooling-scientific-consensus/. Then the northern hemisphere land temperature drop was reported in the literature at around -0.4 deg C over 30 years. It doesn’t seem to be in modern records.
specifically the last 30 years, 27 stations NH and 27 stations SH [balanced to zero latitude]
my data show NH warming by about 0.026K/annum whilst in the SH it is 0.0014K/annum
IOW the warming is basically nothing in the SH
giving me an overall average of about of 0.013K/annum, globally, the past 30 years,
which btw is close to what UAH and RSS is reporting.
\
my problem was
how to explain the difference?
bw
Henry
Thank you, Nick Stokes and others for providing the missing metadata. It, however, does not provide a very satisfactory explanation for changing the raw temperature data. Nick’s Berkeley Earth glitch lets me draw attention to the fact that several other stations along the South African coast indeed saw a marked drop in temperatures over the 1950 – 1980 period. Being coastal, I would have expected an oceanic impact to moderate any nearby land temperature shift.
I am interested that more did not comment on the cooling. There had to be some reason for the Time covers. There is an excellent explanation of this hiding of the decline over at Climate Depot http://www.climatedepot.com/2016/09/13/83-consensus-285-papers-from-1960s-80s-reveal-robust-global-cooling-scientific-consensus/. In the early 1980’s the northern hemisphere land temperature drop was reported in the literature at around -0.4 deg C over 30 years. It doesn’t seem to be in modern records.
I regret the link to the data no longer works. The link changed twice while I was playing with the data. Could it be a policy to do so, to frustrate those asking simple questions?
” Nick’s Berkeley Earth glitch lets me draw attention to the fact that several other stations along the South African coast indeed saw a marked drop in temperatures”




Well, they didn’t show that 1960 drop. Here is the Berkeley plot I should have shown. It shows the sharp 1960 drop relative to neighbors. That is what the homogenization saw as a non-climate change that needed correcting, and it did match the timing of the move to the new airport.
But I can now wrap this one up. The Cape Town station was formerly known as CAPE TOWN-SAAO:CSIR. In January 1961 it moved to Cape Town airport. The NOAA sheet is here. But the CSIR station continued as a separate GHCN station, with an 002 appended (to show it was related). Its sheet is here. It has a small overlapping record. Here is a table, from unadjusted GHCN V3 data, of the ongoing Cape Town record as shown in the head post, and the CSIR record, which is the continuation of the old station – ie CT before 1961. You can see that the records are virtually identical in 1960, but CT drops by 1.6 in 1961 and stays about that level.
Here is a graph, continuing to about 1975. You can see why the algorithm adjusted pre-1961 figures down by about 1.2°C.
Sorry, I didn’t show the Berkeley plot I said I would. It is here. But I think it is superseded by finding the old station continued.


Oops

Given that the station move to the airport in 1961 was supposedly made inland, the sharp DOWNWARD adjustment of earlier temperatures in GHCN3 doesn’t make any physical sense. All of the Matroosfontein district is still within the sea-breeze regime, despite some sheltering by Table Mountain. If anything, one should expect HIGHER temperatures at the airport than at the coast.
1sky1
” If anything, one should expect HIGHER temperatures”
Then one would be disappointed. What I have shown in the table are the actual measured, unadjusted temperatures. The airport is measured to be cooler than the Observatory.
What’s truly disappointing is the cavalier disregard of physical basics in interpreting field data by blind number crunchers. That “the airport is measured to be cooler than the Observatory,” which is ~4km closer to the coast, should alert one that something other than climatic factors are at play. To geophysicists somewhat familiar with Cape Town, the reason for this is quite apparent: the Observatory is also that much closer–in fact, adjacent–to the urban center of a city that had grown enormously prior to the 1961 opening of the new airport. The “homogenization” of the airport record to conform to that of the UHI-corrupted Observatory is a poster child of how a bogus trend gets baked in by geophysical amateurs while manufacturing ostensibly “better” up-to-date station records.
1sky1

“The “homogenization” of the airport record to conform to that of the UHI-corrupted Observatory”
The airport record is not modified to conform; the observatory part of the record is realigned to match the airport near 1960.
I have plotted together here all the temperature records – CT combined (as shown in head post, pink), CT adjusted (green), and the observatory record as it is given from 1960 (blue). It is annual average data. The continuous record at the Observatory is seen by following the pink to 1960, then the blue. Since 1900, at least, the Observatory record runs parallel to the CT adjusted. The offset does not affect any climate analysis, since the first step is to subtract a mean to get anomalies.
This is sheer semantic casuistry! Since there was no record at the airport prior to 1961, the realignment of the UHI-corrupted observatory record prior to that date to match the cooler airport temperatures constitutes a major modification of the 20th century record for Cape Town.
In the absence of strong UHI at the Observatory, the in corpore realignment near 1960 splice would have been in the OPPOSITE direction, obviously affecting the 20th century trend at Cape Town. As it stands, even the mean required to obtain anomalies is corrupted by such blind “realignment.”
“In the absence of strong UHI at the Observatory, the in corpore realignment near 1960 splice would have been in the OPPOSITE direction”

That is your opinion. But the duty of the various people here does not include injecting such opinions. It is the duty of the people reporting measurements to simply do so, and not opine about UHI. Then the people homogenising, seeing a clear inhomogeneity, have to allow for it. There is no choice.
Then comes the question of UHI. GISS does allow for that. Here is their station plot for Observatory post-1960:
As you see, they raise past values – ie lower the trend. It would be good to see what they did before 1960, but there is a surprise. GISS only gives UHI adjusted (which is what they then use) back to 1942. Since they have GHCN adjusted back to 1857 or so, that means that they are discarding the old part because they could not do a UHI adjustment, presumably for lack of stations to compare.
So to get back to the topic of this thread, that is why homogenisation did what it did. There was a clear break, with cause.
HADCRUT took the alternative approach. They treated the Observatory record as one entity, and the airport as a new record starting in 1961. The effect is much the same.
The characterization of well-established geophysical expectations in the coastal zone as mere personal “opinion” speaks volumes. By-products of such willful ignorance on the part of “people homogenising” are yearly Cape Town temperatures incredibly down in the low 14s Celsius during the earliest part of the putative record and the equivocation of the downward slope of temperatures during the 1960s.
The production of bogus trends by people who have scant concept of temperature variability in situ and mistakenly think that simple offsets are adequate compensation for UHI is commonplace throughout GHCN 3. The truly sad part of that venture into pseudo-science is that far-more-skillful combinations of neighboring records into long continuous series, such as done by the South African Weather Service (see GHCN Ver2: https://data.giss.nasa.gov/cgi-bin/gistemp/show_station.cgi?id=141688160000&dt=1&ds=1) are discarded in favor of fragmented fictions that support the AGW narrative in the minds of climatological amateurs.
Philip Lloyd January 29, 2017 at 6:27 pm
I am interested that more did not comment on the cooling. There had to be some reason for the Time covers.
The one with the penguin on it was about Global warming and dates from 2007, the fake version you showed is apparently a photoshopped version of that one.
Yes, others have drawn my attention to misinterpretation of both quoted covers. I used them for illustrative purposes without checking their provenance. They were to draw attention to the widespread documentation of global cooling between 1950 and 1980, a cooling that was recorded in the raw data and is no longer apparent in the homogenized version. I do not believe the error to be material.
Philip Lloyd on January 30, 2017 at 12:47 pm
They were to draw attention to the widespread documentation of global cooling between 1950 and 1980, a cooling that was recorded in the raw data and is no longer apparent in the homogenized version.
A widespread mistake. There is no global cooling documented by GISS between 1950 and 1980, even not in the 2011 archive. The cooling period was between 1940 and 1975, as is represented by WFT
http://fs5.directupload.net/images/170130/n9xwi9f2.png
and calculated by Cowtan’s trend calculator:
1940-1975: -0.023 ± 0.045 °C / decade (2σ)
1950-1980:+0.046 ± 0.056 °C / decade (2σ)
” I do not believe the error to be material.”
But neither had anything to do with the alleged cooling. If you had a point to make, how can this not be material.
The Old Capetown Station continues to have a record.
What happened in that GHCN made the mistake in their adjustment.
There are two stations in the GHCN database.
The original station Capetown WMO identifier 68816
http://climexp.knmi.nl/data/ta68816.png
And the new airport station Cape_Town WMO identifier 68816.2
http://climexp.knmi.nl/data/ta68816.2.png
It is the NCDC and Nick who have made the mistake here. The adjustment should be reversed.
Bill Illis,
“And the new airport station Cape_Town WMO identifier 68816.2”
No, it is your mistake, as clearly shown on your second graph. The identifier SAAO:CSIR means South African Astronomical Observatory (located in Observatory, Cape Town), and CSIR is the organisation which ran the station. That is not the airport. Another clue is that the second, observatory record ends in 2002, while the top one goes to present. Here is December’s CLIMAT form, marked DF Malan, the old name for the airport.
I have been reading WUWT for many years but never knew Mr Watts was once a Capey. Me too. Excellent blog Mr Watts, keep it up.
I have a suggestion for assessing the quality of the century plus temperature records from around the world. Capetown’s record looks pretty much like USA’s, Greenland ‘s, Iceland’s, Siberia’ s Paraguay’s,….. in terms of the thirties being hottest, a notable decline between 1945-1975, and many other details of the up’s and downs of these records. These widely spaced similarities are powerful support for the raw records as useful data for assessing the state of the global climate.
If global warming is going to be significant, then the raw data of even a handful of such scattered records is early warning enough. Let’s say that when the the raw temperature record in the future surpasses those of the 1930s temperatures at 50% of these scattered global sites by 1C, then we will be on “first” alert. Until then give this issue a well deserved rest. Similarly If the sea could rise a metre or more in a century, stop rushing down to the shore with a micrometer gage!!
Ferdinand
to quote from your quote
On a global scale, therefore, the magnitude of the effect of biological drawdown on surface water pCO is similar in magnitude to the effect of temperature, but the two effects are often compensating. Accordingly, the distribution of pCO in surface waters in space and time, and therefore the oceanic uptake and release of CO , is governed by a balance between the changes in seawater temperature, net biological utilization of CO and the upwelling flux of subsurface waters rich in CO.
sic [there is an important word missing, ‘differs’, and CO2 not CO? ]
Henry says
I figure that all of what you say is theoretical, not empirical,
iow
show me your data,
man, method, machine, sampling procedure and all that jazz?
show me in your closed vessel how 70 ppms of CO2 difference above the water makes any difference to any parameter measured in the water of your vessel??
ooohh
you did not do the test?
I am so sorry
this comment should have been posted on a different blogpost
[getting old and confused…]
guys



@Bindidon and them
you cannot really compare data from before 1970 with those after 1970. It is like comparing apples with pears.
Since 1970 we have seen the introduction of thermocouples and continuous recorders with measurements taken every second or minute. Before that time we had wind up recorders that measured T and RH and had to be filled with ink…
Before the 1950’s they did not even re-calibrate the thermometers used. In those days they had to rely on people taking a measurement 4 x a day…..
Do you honestly want me to believe that before 1970 is the same as after 1970?
Hence looking from 1976
there has been no warming here in South Africa, whatsoever,
The average warming in the SH has been zero since 1976, if you are prepared to believe my results.
Thanks for putting together this informative data.
Interesting how the “Big Chill” cover is not about global warming and the “Surviving the Coming ice Age” is a fake, but despite this being pointed out my several commenters it has not been corrected.
The author, Philip LLoyd, has been made aware of these errors and has responded, but he apparently doesn’t feel the need to ask for a correction either.
One might raw the conclusion that these covers are just “alternative facts” and that real facts have no place here.