Pacific Basin – The Australian Hockey League

OK, Splitting Graphs from Analysis

What I’m going to do is put up the graphs for a region, like the Pacific Basin, and then have follow on posting that link back to selected sets of the graphs for looking at particular issues. This will let folks who want to “cruise the graphs” do it fast, but also avoid turning a regional “graph rich posting” that is already large into a completely unwieldy giant thing.

So this posting is “substantially complete”. There are one or two minor countries to add, but they often are just a reprise of one of the ones below. So a couple of “Singapore Hockey Sticks” and a couple of “Flats” and a few that are truncated (one has only 9 years data…). They will be added “for completion” as I’m looking for something to do with the hands while the brain works on contemplating another question ;-) They won’t add much to what you can ‘figure out’ by inspection of the charts that are here.

As an “issue” is observed, I’ll put up a “smaller bite” posting about that issue which will reference this archive of graphs. So, for example, you can expect at some point a Singapore Hockey Stick vs (something else nearby not a hockey stick) posting.

Feel free to post comments here about anything you see in these graphs, but don’t be surprised if, over time, “issues” get covered in another posting as well.

Pacific Basin Overall

Pacific Basin - Flat, then a Hockey Stick

Pacific Basin - Flat, then a Hockey Stick

Here we have a nice sleepy Pacific, then we get a Hockey Stick right as The Great Dying of Thermometers happens. When we were looking at the raw temperature averages, there was not much going on in the average of the smaller islands, but there was a pretty good “warming signal” in Australia and New Zealand. We also saw a lot of change “by Latitude” and “by Altitude” and “by Airport Flag”. So how much you want to bet we find The Australian Hockey League? And maybe a B League in New Zealand?

Australia, New Zealand, Belau


Australia Hair Graph monthly anomalies and cumulative change

Australia Hair Graph monthly anomalies and cumulative change

New Zealand too?

New Zealand Hair Graph by thermometer count segments

New Zealand Hair Graph by thermometer count segments

Yeah, I ought to have put in a separate segment for that 1940’s Pivot when thermometers start changing, then you would see the cooling early segment, then the rise with thermometer change, the Step Up and then The Ramp in the next two segments, but by now you folks ought to be figuring this stuff out on your own ;-)

And a medium sized island further north?

Belau Hair Graph monthly anomalies and cumulative

Belau Hair Graph monthly anomalies and cumulative

My God! The Pacific is Just On Fire!!!

Singapore – Splicing a Blade on The Shaft

OK, what do you do when your stick is just got no blade? Splice on a new one!

Singapore Splicing The Shaft monthly anomalies and cumulative

Singapore Splicing The Shaft monthly anomalies and cumulative

Notice when the thermometer count doubles? At that time the monthly anomaly volatility goes WAY down (as the pluses of one thermometer offset the minuses of the other), then when the cooling one ‘goes away’ we have The Reveal and, PRESTO! A Stellar Hockey Stick! (Though even then we get a double dip in 1990 with the Magic Sauce that is fairly consistently applied to the data then. Just look at how clean and sharp the heel of that Pivot is!

Playing With Your Instruments In Fiji

When you team isn’t winning, you can always change the players on the field…

Fiji Playing WIth The Instruments monthly anomaly and cumulative

Fiji Playing WIth The Instruments monthly anomaly and cumulative

This is just fascinating. Look at how much the thermometer count changes. Bobbing and weaving all over the place. That first segment really ought to be divided into 2 or 3. Notice the later part of that first segment where it pivots down just at the start of the 1951 – 1980 GIStemp baseline period? Then a bit later we have something that I’ve seen as hints before. A 1970 era thermometer change (there is a hint of a ‘bullseye” where the monthly dT lines converge at near zero volatility in about 1970). This happens right when we get a small pop up as the baseline interval is coming to an end (but with a negative slope… can’t keep that for long). Then we have the 1980 Jump (with both some adds and some drops) followed by the 1990 ramp, but ending in more thermometer loss.

Now the second thing that’s interesting about this is that Fiji is smack dab in the middle of a bunch of other Island groups. Surrounded by Samoa, Tonga, Tuvalu, Vanuatu, New Caledonia, Solomon Norfolk and Cook Islands are a bit further out, beyond the typical ‘reach’ of the ‘homogenizing” process, but perhaps reachable via an “in-fill” to nearby that then gets used for a grid / box adjustment later. It’s just about exactly where you would want to be for the maximum number of other island groups to influence. And all around it are Islands not warming that could “use a little help”. So just drop some data or truncate a record and you will get “in-fill” giving a lift… Nicely done, very nicely done. (But would have been better without so much time playing with the instruments. I know, it’s hard to make a hockey stick out of nothing, and as we saw for French Polynesia, you had nothing to work with…)

French Polynesia

Talk about your flat stability (At least after that first plunge on adding a second thermometer.)

French Polynesia "monthly changes of temperature" by years

French Polynesia (click for larger version). dT by month year/over-year.

Somehow I think the French have it right… Though I do have to note that we plunge right into the GIStemp baseline period, rising out of it, if only a little, at the end. Maybe 1/3 a degree C? Though a peek at Christmas Island leaves me hoping for a Christmas Present…

Not So Much – The Other Islands


Nauru Hair Graph monthly anomalies and cumulative

Nauru Hair Graph monthly anomalies and cumulative

Solomon Islands

Solomon Islands Hair Graph monthly anomalies and cumulative

Solomon Islands Hair Graph monthly anomalies and cumulative

Christmas Island

Somehow I think Instrument Issues matter…

Christmas Island Hair Graph monthly anoamlies and cumulative

Christmas Island Hair Graph monthly anoamlies and cumulative

Cocos Islands

Cocos Islands Hair Graph monthly anomalies and cumulative

Cocos Islands Hair Graph monthly anomalies and cumulative

Cook Islands

Cook Islands Hair Graph monthly anomalies and cumulative

Cook Islands Hair Graph monthly anomalies and cumulative

Federated States of Micronesia

Micronesia Hair Graph monthly anoamalies and cumulative

Micronesia Hair Graph monthly anoamalies and cumulative

Marshal Islands

Marshall Islands Hair Graph monthly anomalies and cumulative

Marshall Islands Hair Graph monthly anomalies and cumulative

Norfolk Island

Norfolk Island Hair Graph monthly anomalies and cumulative

Norfolk Island Hair Graph monthly anomalies and cumulative

Pitcairn Island

Hey Hey! Ho Ho! Cooling Islands Have To GO!

Pitcairn Island Hair Graph monthly anomalies and cumulative

Pitcairn Island Hair Graph monthly anomalies and cumulative

Wake Island

Quite a dip. No wonder the record is so short…

Wake Island Hair Graph monthly anomalies and cumulative

Wake Island Hair Graph monthly anomalies and cumulative

Wallis and Futuna (France)

Go FRANCE! Look at that dip.

Wallis and Futuna Hair Graph monthly anomalies and cumulative

Wallis and Futuna Hair Graph monthly anomalies and cumulative

Some Interesting Cases


Now that’s interesting…

Samoa Hair Graph segments by Baseline Status

Samoa Hair Graph segments by Baseline Status

Looking at Samoa by segments relative to the baseline is interesting. We do get the mid-70’s dip, but it isn’t a very deep one. We then exit, and go not very far, then drop Samoa from the record. To be filled in from Fiji, with a very different shape…

Fiji by Baseline Segments change in temperature

Fiji by Baseline Segments change in temperature


On the other side of Fiji is Vanuatu, also rather a bit flatter… despite thermometer change, and it too will benefit from a bit of “lift” via homogenizing…

Vanuatu Hair Graph monthly anomalies and cumulative change

Vanuatu Hair Graph monthly anomalies and cumulative change

That leaves Tuvalu, Tonga, and New Caledonia as ‘near Fiji” to look at. One of these is a single decade of data ( Tonga with 9 years):

Tonga Short Hair Graph

Tonga Short Hair Graph

1972	-0.83	0.11	1	0	0	0	0	0	0	0.2	0	-0.2	0.2	0.4	0.7
1973	-0.73	0.19	1	1.3	0.5	0.6	0	0.9	0.2	0.1	0.5	-0.6	0.5	-0.5	-1.2
1974	-0.53	-0.47	1	-1.6	-0.5	-0.8	-0.8	-0.9	-0.3	0.3	0.4	-0.5	-1	-0.4	0.5
1975	-1	0.04	1	0.1	0	0.7	0.4	-0.3	0.1	0.1	0.1	1	-0.5	-0.3	-0.9
1976	-0.96	-0.16	1	-0.5	-0.8	-2	-0.5	0.1	-0.6	0	-0.3	0	0.8	0.7	1.2
1977	-1.12	0.75	1	1.7	2.2	2.1	0.4	0.8	1.2	0.4	-0.2	-0.2	0.6	-0.2	0.2
1978	-0.37	-0.19	1	-0.3	-0.5	0.4	0.5	-0.2	-0.4	-1.2	-0.5	0	-0.1	0	0
1979	-0.56	-0.23	1	-0.8	-0.2	-1	-0.7	-0.3	-0.2	0.2	0.9	-0.5	-0.2	0.3	-0.3
1980	-0.79	0.79	1	0.5	0.2	1	1.2	1.4	1.1	0.5	0	1.6	0.8	0.6	0.6

So 9 years of going from nowhere to nowhere with a lot of negative anomalies

While Tuvalu gets a 1990 “Pivot” that’s a beauty:

Tuvalu Hair Graph with 1990 Pivot Segment

Tuvalu Hair Graph with 1990 Pivot Segment

And even The French in New Caledonia can not resist forever… and get a 1 – 3 – 2 thermometer whammy. Starting with one, dropping, adding 2 to neutralize the drop, then removing one for a nice warming rise and a 1990 bit of a pivot.

New Caledonia Segments by Count

New Caledonia Segments by Count


Clearly the Philippines suffer from a lot of thermometer change. I especially like the way it stabilizes right after all the change stops…

Philippines Hair Graph monthly anomalies anc cumulative change

Philippines Hair Graph monthly anomalies anc cumulative change

The Rest of the Pacific Basin


Indonesia Hair Graph

Indonesia Hair Graph


Kiribati Hair Graph

Kiribati Hair Graph


Malaysia Hair Graph

Malaysia Hair Graph

American Samoa

American Samoa Hair Graph

American Samoa Hair Graph

Coral Sea Islands

Coral Sea Islands Hair Graph

Coral Sea Islands Hair Graph


This is the Agana WSMO Station at 13.55 N 144.83 W

Guam Hair Graph

Guam Hair Graph

Including the OTHER Agana station, the NWSO at 13.48 N 144.80 W

Agana NWSO Hair Graph

Agana NWSO Hair Graph

Now look at those two Agana Stations for just a moment. Gee, different….

And the surrounding North Mariana Islands

Northern Mariana Islands Hair Graph

Northern Mariana Islands Hair Graph

Johnston Atoll

Johnston Atoll Hair Graph

Johnston Atoll Hair Graph


Niue Hair Graph

Niue Hair Graph


Tokelau Hair Graph

Tokelau Hair Graph

Subscribe to feed

About E.M.Smith

A technical managerial sort interested in things from Stonehenge to computer science. My present "hot buttons' are the mythology of Climate Change and ancient metrology; but things change...
This entry was posted in dT/dt and tagged , . Bookmark the permalink.

18 Responses to Pacific Basin – The Australian Hockey League

  1. E.M.Smith says:

    OK, That’s about 3/5 ths of the global surface area between North & South America and the Pacific Basin.

    I’ve got to add some text to this posting (but the other two are mostly done with the additions of today), add the South Georgia Islands to the South America posting, fill in a couple of missing island groups in the Pacific Basin, and then go back to North America and add the Central American and Island countries that I was working on before I decided to “jump ahead” and knock out the South American and Pacific Basin sets.

    Probably about Sunday to get that all tidied up…

    Then I’m going to “take a break” and spend some quality time with the TV and a beer…

    Shortly after that, I’ll come back to Asia. Then Europe (though I may put up a European stub with some key country graphs in it when I put up the Asia posting). Finally I’ll “do Africa” after that. Africa has a LOT of countries in it and they have some “odd characteristics” so it will likely take a fairly long time to work through. Thus leaving it for last. (Though, frankly, I suspect that the pattern is pretty well already demonstrated and that by the time we get to Africa it will be more for esthetics of completion than really learning anything new…)

    Unless, of course, I find myself sitting up late thinking thoughts of Asia…

    But I think as you look over these “by Country” graphs of the “Change of temperature over time” it’s pretty clear that:

    1) It is “by country” and not a regional process nor a global one.

    2) CO2 does not work “by country”.

    3) The processes of “adjusting”, “in-fill” and “homogenizing” hide too many of the interesting bits.

    4) SOMETHING happened to the data in 1990. It is very important, it causes “warming”, and it isn’t CO2. It compresses peaks in both directions, but compresses cooling excursions more than warming.

    5) Lesser things happened in 1980 and 2006 (more or less) and can be seen in some countries more than others.

    6) Thermometer Change Matters. It’s a bad idea to screw around with the instrumentation and change the thermometers in the middle of a calorimetry experiment.

  2. Alan Davidson says:

    These country analyses are incredibly informative and interesting. I think the current lack of comments is because most people haven’t found your site yet, so some cross-posting would be good e.g. to WUWT, Climate Audit, Climate Depot.

    I’m in Ottawa Canada. I’m going to look into the Canada and Arctic data and your analysis some more to try to figure out what’s being done, why and by whom. Looks like widespread manipulation to me…….

    In the major countries when there has been sudden large dropouts of thermometers, do you see a consistent pattern or characteristics of the dropouts? Are they often the ones that are higher altitude or consistently cooler or something else?

    Europe, UK and particularly the other EEC countries would be interesting to see next as this is where much of the AGW hoax is supported.

    Thanks for all your outstanding efforts on this!

    REPLY: [ There are a whole series of “by altitude” and “by latitude” and other reports of thermometer change over time. A good starting point is:

    which is a little ways down under the “GIStemp” tab up top. ( I probably need to re-do that page, it’s too GIStemp oriented and reflects a time when I was neck deep in “code”. It probably needs a more Joe Sixpack friendly ‘index’ kind of character… )

    But yes, there is a lot of “locational bias” to the station drops. Probably the most obviously biased is that there are now 92% Airports in the USA GHCN data. The Pacific is headed for 100% (there is a NOAA link in one of the “airport” postings comments where they state the goal of being 100% airports fairly soon). Well, 10,000 foot jet runways surrounded by tarmac taxiways and parking areas are a mite warmer than the grass shack and “seaplane port” that was common in the days of Pan Am running Clippers (boat planes) all over the Pacific…

    Per “order of doing”: Europe also has a large number of countries, so doing it “by country” takes a lot of work (yet yields the most insight). Asia actually has fairly few countries that account for almost all the area. That makes it fairly “quick”. (Russia, China, Mongolia, India and you’ve got a good feel for the place. A lot of the Middle East is counted as Europe [though some is Asia – Israel is Europe, Saudi is Asia ;-) ] also European are Ukraine, Belarus and everything west of the Urals.) So what I’ll most likely do is actually two “stub postings” with a few of the “big lumps” in Asia and Europe, then fill in the smaller countries as interesting things surface. FWIW, a cursory exam of the reports for Europe show what looks like some “early efforts” and some experimentation. Yes, that “projection” and the data will have to speak for themselves… but there are some ‘odd bits’ that will take a while to fully sort out… It will likey be just a couple of days in any case. The data is already made into reports, so it’s just crank out the charts (and you can now see how many I can make in 3 intense days ;-) The biggest “issues” in the way are just my desire to make the existing three pages (N&S America, Pacific) fully complete and my need to “take a break” from constantly clicking on graph options ( 12 months lines per graph, 5 clicks each to change the width to fine so the graph is readable, 60 ‘clicks per graph’ for just that change… 20 graphs…) Lets just say I’m a little tired of clicking “change line width… on the other hand, I’m getting really fast at it ;-)

    In any cae, it’s going to be less than a week to get everything but Africa up (and probably even get a start on Africa…)
    -E.M.Smith ]

  3. oldtimer says:

    In (4) above you say “SOMETHING happened to the data in 1990”.

    No doubt, we shall be told, that it was just a coincidence that the IPCC was set up in 1989.

    Nevertheless someone, somewhere arranged for the great dying of the thermometers. It seems to me a good possibility it was done under the auspices and (guidance?) of the IPCC. Sir John Houghton was i/c, with the backing of Margaret Thatcher, in those early days. Is this a case for more FOI requests?

  4. KevinM says:

    I joined a carpool, and am forced to listen to NPR two hours a day. You would never know there was a wave of new thinking about AGW. There is still concensus in that world. The stories focus on what to do about it, not whether it exists.

    Folksy story about farmers suffering from falling carbon credit trades ran twice yesterday. The crash in that market was attributed entirely to fears of corrupt selling of duplicate shares – no mention of the recent exposure of CRU, failure in Copenhagen, and significant shift in poll numbers around belief or importance of AGW.

    I’ve been riding with the NPR gang for two months, and never heard mention of Phil Jones and CRU, or Mann under investigation. Its all pet political issues all the time.

    Sorry, just venting. I love the charts and analysis here.

  5. whbabcock says:

    I have been lurking around your site for some time and find your work invaluable. While I am not a computer programmer, I have worked most of my career with large energy models, so I have experienced many long days trying to figure out why a model produces the results it does. Hence, I appreciate your focus “on the data” and on “exactly what the computer program (GISTemp) actually does” (rather than what it should do, or what people think the program does).

    Towards that end, I seem to recollect that you have done an analysis of temperature trends from “long-lived” individual stations, but I haven’t been able to find it. While I find your work intriguing and important for identifying and characterizing the implications of “adjusting,” “in-fill,” and “homogenization,” I believe it would be equally powerful to characterize exactly what the “best temperature data” from individual stations have to tell us. (Note: I am defining “best” as the raw (before adjustment) temperature data from those long-lived stations with few missing data points. I don’t understand the data like you do, so I’m not sure how to operationalize “best.”)

    From my perspective, focusing on raw individual station temperature data allows one to see the original trends in the data without having to figure out whether the trends are being created by the underlying data or by the weighting and “adjustment” processes. I am aware that these raw data may embody UHI and/or the impact of instrumentation changes, etc. However, by focusing on long-lived stations, one would hope to minimize biases and be able to observe the potential impacts of UHI over time. While individual long-lived station temperature data will not be “representative” of global trends, global trends must necessarily be rationalized with such data to be believable.

    REPLY: [ A good entry point is hiding under that GIStemp tab up top or you can scroll back through the “AGW and GIStemp issues” and the “GHCN” categories on the right margin. FWIW, I think the one you are looking for is:

    that was one of the earliest hints of The Reveal based on the Pivot in 1990. Though earlier evidence came here:

    It’s probably time for me to divide those categories into more organized groups. Split out the general AGW stuff from the GIStemp and GHCN stuff. Maybe after Europe is graphed and up, as a break from graphs before tackling Africa…

    FWIW, a couple of other folks have focused on “long lived station trends” and found that they tell a much more stable story than the “Sausage Temperature Products” do. I think TonyB has a nice series on it, but I’m not sure when the link is ( I think it’s in comments somewhere… maybe he can post a link here? …) TonyB did a good job in this posting:

    that has some nice links in comments, IIRC.

    So anyone with pointers to all those nice “long lived” station reports, feel free to post a link…

    -E.M.Smith ]

  6. Pingback: Cloud Studies « TWAWKI

  7. Ripper says:

    “4) SOMETHING happened to the data in 1990. It is very important, it causes “warming”, and it isn’t CO2. It compresses peaks in both directions, but compresses cooling excursions more than warming.”

    The answer may be in the Cru2010 files at the met office that they released after climate gate.

    “Number= 943150
    Name= MARBLE BAR
    Country= AUSTRALIA
    Lat= -21.2
    Long= -119.8
    Height= 182
    Start year= 1901
    End year= 2007
    First Good year= 1901
    Source ID= 30
    Source file= Jones+Anders
    Jones data to= 2000
    Normals source= Data
    Normals source start year= 1961
    Normals source end year= 1990
    Normals= 33.4 32.4 31.8 28.7 23.6 20.4 19.6 21.5 25.4 29.2 32 33.6
    Standard deviations source= Data
    Standard deviations source start year= 1941
    Standard deviations source end year= 1990
    Standard deviations= 1.1 1.5 1.3 1.2 1.2 1 1.1 1 1.3 1.3 1.3 0.9”

    I would just about bet my left nut that any months since then that exceed the SD would be dropped by “quality control “.

    Note the “normals” are from a different period to the SD’s.

    There is some interesting stuff in there from Australia at least.

    I have discovered that in a lot of cases these figures are actually higher in the early years than the pre Torak and Nichols adjusted Aust Bom data.

    E.G. Wiluna is .4C hotter pre 1934 than the Bom data.’

    The graphs say it all for Halls Creek

    I am still working my way slowly through the stations but in the ones I have I have found most are warmer in the past and 2007,2008,2009 are ~0.2C cooler than the Bom data

    It may pay to run those figures through your software and see what they say.

    REPLY: [ I”m coming to a similar conclusion that the “recomputing” is an error and that the “peak clipping” is a result of applying a symmetrical filter to asymmetrical data. They need to just leave the data alone…

    BTW, I can send code or you can send data if you want to see what the result looks like (or you can re-implement the dT/dt method. It’s pretty easy. Month vs Month delta goes into a new file. ( So Jan(2005)-Jan(2004)=Delta(2005) and goes into the JAN field in the file, skipping over any -9999 missing data flags and if 2005 was -9999 you would put zero in 2005 and look to 2006 and put the result of 2006-2004 in 2006 instead) Average them by year. Sum that up over the years. Only “wrinkle” is that the first value has zero delta to itself (and it is computed ‘past to present’) while the sum is computed present to past. Though I’ve been thinking of doing the “monthly anomaly creation” step present-to-past (though that would make most of 2009 ‘zeros’ right now… so I’ve been reluctant to loose the visibility into the present …)
    -E.M.Smith ]

  8. ditmar says:

    This is all good stuff. What do the agw attack dogs make of your analysis? You should also add gin and tonic to your essential household goods list :-)

    REPLY: [ Oooh! Good Call! Tonic being, of course, essential for the quinine to prevent a variety of protozoan infections such as malaria and Gin needed to make the tonic drinkable ;-) “For Queen and Empire! – Down the hatch lads!” {8-0)

    The AGW True Believers are kind of amusing to watch some times. They seem to have trouble with the notions of “Be Polite” and “conversation at a lawn party with friends”. Their preferred mode seems to be “attack and insult” and I’m just not going to play that game. Especially not the “Assassinate The Messenger” variation. So…

    There is an odd quirk in their behaviour that has been very useful to me (though I fear I’ve dampened that use lately). Typically if you have a weakness or oversight in an argument you will get slapped around about it. If you have a strong point, you get “dead air”. So I’ve used “complete silence and rising ‘hit rate'” as a 100% accurate indicator of “hot stuff”. Unfortunately, they recently (after I was on TV) had some folks assigned to badger me ( at least, that’s what it looked like – could have been ’emergent behaviour’ from individual actions). When I put a few of them in the SPAM filter, I started getting ‘Friendly Posting of links to people who think you are an idiot and your stuff stinks’ with a ‘have you seen this’ or a ‘thought you might want to know’. I let the first few through, as they were being polite, but when the pattern was clearly a 100% “hit” on every new posting within hours, often the first comment, well, those now go to the bit bucket too. I’m sure a few really were “concerned friends” but they were inextricably contaminated with the masquerading trolls.

    Somehow they also seem to have problems understanding that it isn’t about ME at all. It’s all about the DATA and what they say when you ask them politely and listen carefully. So they wanted to play “attack the messenger games” and I refused to play…

    It’s kind of a Zen thing for me. I am but an empty vessel. I bring to the data some listening skills and an open mind devoid of preconceptions. If I’d found linear upward trends, that’s what would be in the graphs… but I found something else… and who I am or what I would like to find are just not relevant. For that matter, “liking” is really a mu! moment. I “like” being asleep. I “like” gin & tonic. What I “like” is just not relevant and so is not brought to the data. I most “enjoy” being that empty vessel and standing on the cusp for the first results of a new program run to come in. That is the moment of discovery (whatever it may bring… even a rising 45 degree rocket ride of heating – which would be truly an interesting find) and that moment can ONLY happen if you are at that moment “an empty vessel”. How can you have that tingle of anticipation hanging on the cusp of “discovery and change” if you already KNOW that you will find what you “wanted” to find? “Did you find what you were expecting?” is a MU! (The question is ill formed!) approach. I “expect” to find something unexpected, and that can only be done if you are “the empty vessel”… but I digress.

    So I basically shut down the ‘attack the messenger’ comments. And now they’ve all run away. Odd, really. If they can’t play “smear and insult” they don’t want to play? So this has made the “Silence is Golden” indicator less useful as they are now silent anyway. (It’s not me censoring. I clamped down on about 3 or 4 trollish folks for about a week, and then on the “friendly ‘idiot’ links” for about another week. Now ALL the AGW True Believers have just evaporated. Odd. Guess they can’t confine themselves to ‘the data and what they say’…) So now I only get 1/2 an indicator. The rising hit rate.

    I know, a 2 page answer to a 2 line comment ;-)

    But it’s an interesting “social meta-discussion” and you DID ask ;-)

    It’s not something I spend a lot of time noticing, but it is something that’s basically not possible to avoid noticing. I do have to manage the comment queues. (For example, I’ve noticed that the Harvard IP address is the source of many of the AGW Troll postings… don’t know what it is about Harvard, but they did also provide a large number of “participants” for a show on the Looney Left Liberal “The Link” channel that was slamming the corporate form ( “The Corporation” was the title I think. Basically something you would expect to be straight out of the communist world of 1970’s but sprinkled with some Green Wash and a bit of “white Europeans are evil” reverse racism, though subtle, via the ‘oppressed people of color’ meme – ignoring that in many Latin American nations the oppressors are the same ethnicity as the oppressed.) And YES, I do watch shows from The Looney Left, just as I watch those from The Radical Right. It’s hard to know where the middle is if you don’t measure against the ends… But at the end of the day I see that the data says Harvard is highly biased to one end. It’s not ME saying that, it is them. Their IP address. Their participation on Link TV. etc.

    So I’m mostly just “keeping on keeping on” finding new tools to see patterns (whatever they may be) and using the old tools to work through all the countries; and NOT caring at all what the AGW True Believers think of it, or of me, as “I am not important. I am but an empty vessel.” But noticing patterns also applies to their behaviour, so I observe it. One of the patterns I’ve observed is that The AGW-TB just don’t “get it” that I really don’t care what they think about me. Really. Not a quiver. It is just TOO much a Mu! topic. Not only am I completely irrelevant, and not only is my personal goal to be completely irrelevant to the process (the “perfect” empty vessel); but their evaluation of me is even more completely irrelevant. The data are what they are and say what they say. One might just as well ask “How do you feel about the USGS measuring earthquakes?” (I “feel” nothing about it. They do it well. I’m thankful they do it. I feel nothing about how they measure them when a Mag. 7 hits Haiti.) Or how do you “feel” about the computer that made the charts? And that disconnect with who I am, how I see the world, and how they react to what the data say; that is humorous to watch ;-)

    Hopefully this isn’t too much being a Zen Head and folks will find it an interesting insight. I’m not ‘hard core’ into it, but there are truths in Buddhism that are worth knowing… though I think they could teach it more directly and with less “Mu! (whack with a stick) and hours spent staring at nothing…”

  9. Love it, thanks so much for remembering us in the Sth Hemisphere. Its sad when all we get is the top half.
    I found Bom graphs showing hotter in the early 1900s a few days later the page link changed and the data vanished! their excuse? housekeeping.
    yeah sweeping the TRUTH! under the Rug!

    REPLY:: [ Yeah, I ran into one of those. Had a link to a NOAA site that listed the data collection procedures and specifically stated that the observer could make up a missing value (i.e. guess) if needed. After a while it “dissapeared”. The fact that it was (and may still be for all I have found lately) “proper” procedure to just “guess” is now swept away… FWIW, having found that both Uruguay and Fiji have what look like rather manufactured “dips” in the baseline making “little dippers” and given that it’s just ODD for everywhere to have a ‘dip’ at the same time (notice the USA has a rise in 1934, other parts of the world not so much…) I’m wondering if the “dip” is made by selective listening skills applied to thermometers. So it would be VERY interesting to find records for places with a “Little Dipper” to either confirm or question the “Dip”. Especially Fiji as the nearby islands have very different profiles.

    At any rate, I’ve learned there is a very active group of Langoliers that erase the past should you discover it, so my habit is now to SAVE a copy of everything interesting IMMEDIATELY upon discovery. It’s also why I’ll keep on using the copy of data I have NOW for future study and not go with GHCN Version 3. In “my world” data doesn’t need to come in “Versions”… -E.M.Smith ]

  10. E.M.Smith says:


    I’d wanted to just filter out the “non-zero non-one non-two” Duplicate Numbers and figured there might be some 6 or 7 numbers but figured by 8 and certainly by 9 there would be two files with the same byte count as there would be no place that had been changed 10 times…. I was wrong.

    This is the size of the starting file where I have the anomaly data, then as I successively removed one Duplicate Number (Mod Flag) at a time:

    [chiefio@Hummer data]$ ls -l v2.mean.inv11.M.dt
    -rw-rw-r– 1 chiefio chiefio 76528453 Feb 24 22:57 v2.mean.inv11.M.dt
    [chiefio@Hummer data]$ ls -l EMS*

    -rw-rw-r– 1 chiefio chiefio 75317626 Mar 20 09:00 EMS.THREE
    -rw-rw-r– 1 chiefio chiefio 74751743 Mar 20 09:01 EMS.FOUR
    -rw-rw-r– 1 chiefio chiefio 74498125 Mar 20 09:02 EMS.FIVE
    -rw-rw-r– 1 chiefio chiefio 74367856 Mar 20 09:02 EMS.SIX
    -rw-rw-r– 1 chiefio chiefio 74275474 Mar 20 09:03 EMS.SEVEN
    -rw-rw-r– 1 chiefio chiefio 74246583 Mar 20 09:03 EMS.EIGHT
    -rw-rw-r– 1 chiefio chiefio 74243296 Mar 20 09:04 EMS.NINE

    So there are some places that will need a new ‘thermometer number’ to get a new change history… So watch that minor number in the StationID. ( 3 digits country code, 5 digits major location, 3 digits minor number, 1 digit of “duplicate number” or “modification history flag”)

  11. frank says:

    Comparing the rate of change of temperature for different periods of time with confidence intervals for the slopes doesn’t mean anything.

    If station drop out selected stations with an artificial bias towards warming (such as those with UHIs), then comparing the warming (with confidence intervals) at drop-out vs. non-drop-out stations before the great dropout would expose the bias.

    REPLY: [ Unfortunately comparing A before to B before says nothing about A-dropped after the drop no matter what B does. Thus all the A-early vs B-early therefor when A-dropped it must have been the same as B-kept arguments are just trash. THE major issue is that B-kept is a different “Duplicate Number” (or Modification History Flag) and has a profoundly different slope from B-early. The only valid conclusion is that IFF A were kept AND treated to the same new “modification process” it would match B-kept; but A-if-kept and treated exactly as before is most likely to diverge (as the “modifications” are different) from B-kept (with new dup number). IMHO, it’s the “knee” at 1990 in B-early vs B-kept that shows the broken parts. -E.M.Smith ]

  12. Keith Hill says:

    Re “Something happened to the data in 1990” and further to my posting on cooling in Tasmania from 1988 to 1992 at “Assume a Spherical Cow”. Cooling averaged approximately 1.24 degrees Celsius across all the 19 stations with continuous records.

    1985 had a low base temperature in Tasmania and rose steadily to a high in 1988 , the year James Hansen appeared before a US Senate Committee and made his catastrophic claims about AGW.

    On checking many other Australian stations, four year cooling was evident at most of them. Chief, as you found out for me, in 1992 401 Australian thermometers “died”, leaving only 45 for the entire continent in 1993, the main of them at Airports.

    I have just rechecked the late John L Daly’s excellent article
    on “What’s Wrong With The Surface Record” at
    and found a ‘World Climate Report’ graph of Annual Global Temperature Departures 1979 -1999. This shows Surface, Satellite and Weather Balloon (radio sonde) Temperatures. It seems there was a cooling everywhere at some time in that period 1988 to 1992.

    Between ’79 and ’91 there is not much divergence between the three records, almost matching in warming periods but the Surface record diverging more from the others when temperatures cooled. Between 1991 and 1992, the Satellite and Radio Sondes show a sharp cooling, with Surface stations showing less.

    What is even more interesting is that the first sharp divergence between the three records appeared at that time and became entrenched until the El Nino year 1998 when the three records closely matched but then again diverged sharply the next year.

    It is hard to ignore the link between Hansen’s predictions in his Senate appearance in 1988, the setting up of the IPCC in 1989, the apparent sharp cooling in 1990-1992, the Great Dying of Thermometers in 1992 and the divergence of the three records from then on.

    REPLY: [ Bingo! Give that man a cookie… -E.M.Smith ]

  13. vjones says:


    I got a chance to put up China yesterday:
    I hope you don’t mind. Thanks for sending me the data. I still hope you’ll treat us to some hair graphs sometime.

    REPLY: [ Don’t mind at all, it’s a feature! As I’ve said before, this is sort of a “Y’all Come! Barn Raising approach to science”. All hands welcome. Make lemonade or put up rafters, every bit helps. Heck, if you want I’ll send you all of Europe, Asia, or Africa and you can “have at it”! -E.M.Smith ]

    REPLY2: [ I tried posting this as a comment on your page, but for some reason it didn’t love me ;-{ so I’m posting it here:

    I see that you, too, are comfortable with having “time run both ways” in graphs… But I’d caution that I “got a bit of flack” (AKA feedback ;-) that it was confusing to some folks… so you might want to have the graphs have “time run the same way” or at least placard them with a notice when “time runs backwards” with 2009 on the left…

    I like the trend analysis approach. I’m not set up for it, but it is a very intriguing thing… Which way was a station headed when it was added or dropped. Might I suggest Fiji as a candidate? It has a very complex set of adds/drops that look to be “trying” to make a “little dipper” in the baseline. But without the trend data I’m just speculating.

    Oh, and Singapore has a great Hockey Stick. Would be interesting to see the “trend data” for the station change there too… Might help sort out the “thermometer change” vs “Duplicate Number / Mod Flag change” as causal agent for the hockey stick blade…
    -E.M.Smith ]

  14. vjones says:

    I only said “I hope you don’t mind” because having sent you my graphs I thought you might have planned to post them, but no harm in two different interpretations anyway.

    Yes blog time is a bit short at present, but please fire any data my way and I’ll queue it. If I don’t have time to write anything on it I’ll email the graphs back to you and you can add them here.

    Oh and sorry about the comments. I think quite a few people have problems. I have signed up with WordPress and Kevin has offered to host. Hope to move as soon as time permits.

  15. E.M.Smith says:

    @vjones: No worries. Just two folks trying to save the world being a bit overrun some times ;-) I’ll save South America and the Pacific Basin. I’m whacking on Asia now. Any interest in Europe? Or Africa? You could save them ;-) North America may be hopeless 8-)

    BTW, my comment did make it through when I did “name / url” instead of “wordpress” as authentication. So it looks like it’s the “automatic authenticate vs worldpress ID” that’s having issues. Might not be “your issue” and might well be a wordpress issue…

    And I guess that “having sent” means I need to check my email again 8-} Ok, Sunday we have a “standing event” at my home and I don’t check email on the weekends much due to “Family Stuff” but I promise I’ll feel guilty about it from time to time ;-)


  16. vjones says:

    Er, no the China stuff you’ve already seen and replied to. But I’ve just sent a ‘useful’ (?) file just now.

  17. E.M.Smith says:

    You know, looking back over all those graphs, there is one thing that stands out to me:

    No 2 are the same.

    Some are similar. Many are dramatically different. There is no “one single trend”.

    CO2 can’t do that…

  18. boballab says:

    The Sharp cooling between 1991 and 1992 was the June 12th and June 14th/15th eruptions from Mt. Pinatubo which dropped temps worldwide.

    [blockquote]The volcano’s ultra-Plinian eruption in June 1991 produced the second largest terrestrial eruption of the 20th century (after the 1912 eruption of Novarupta) and the largest eruption in living memory.[4] [/blockquote]

    [blockquote]The effects of the eruption were felt worldwide. It ejected roughly 10 billion metric tonnes (10 cubic kilometres) of magma, and 20 million tons of SO2, bringing vast quantities of minerals and metals to the surface environment. It injected large amounts of aerosols into the stratosphere—more than any eruption since that of Krakatoa in 1883. Over the following months, the aerosols formed a global layer of sulfuric acid haze. Global temperatures dropped by about 0.5 °C (0.9 °F), and ozone depletion temporarily increased substantially.[6] [/blockquote]

    [blockquote]This very large stratospheric injection resulted in a reduction in the normal amount of sunlight reaching the Earth’s surface by roughly 10% (see figure). This led to a decrease in northern hemisphere average temperatures of 0.5–0.6 °C (0.9–1.1 °F), and a global fall of about 0.4 °C (0.7 °F). At the same time, the temperature in the stratosphere rose to several degrees higher than normal, due to absorption of radiation by the aerosols. The stratospheric cloud from the eruption persisted in the atmosphere for three years after the eruption.[/blockquote]

    I was in Subic Bay when that all happened and the eruption that started on June 14th blotted out the sun for an entire day. At high noon it was as dark as midnight outside.

Comments are closed.