What the Giss?

Yes, I’m coining a new “term of art”. “What the Giss?” is in essence “What the Technical Foobar Did You do?”

So in a prior posting we saw that the Arctic temperatures were strongly impacted by The Great Pacific Shift:


And that maybe the PDO had started a shift back and THAT prior shift to a warm cycle would account for a lot of the Blood Red Hot Hot HOTT! in the Arctic.

Yet there is more…

So I started wandering along the Canadian Arctic Coast at NASA GISS and thinking I’d take a look at some stations there to see if they, too, like Barrow, had cooling trends with a Great Pacific Shift signature in them.

I stumbled on a couple pretty quickly (not hard as you just click on the map) that had Something Strange about them.

An example:

Baker Lake NASA GISS May 2011

Baker Lake NASA GISS May 2011

Well, it’s got “Regime Change” written all over it. Just look at that “Step function higher” in 1995 or so.

But what kind of Regime Change is that? Not in sync with the PDO / Great Pacific Shift. Not in sync with the 1998 Great Stagnation when the world stopped warming.

Just what IS that?

Wolfram Alpha Says?

Baker Lake WA May 2011

Baker Lake WA May 2011

But But But… that doesn’t show much of anything at all happening!

There is a dive in 1966 or so that contributes to a statistical warming, but the line itself does NOTHING of merit in the 1995 era. WTF? Or, rather WTG?


Oddly, when I went to Wundergound to try doing some “custom” long duration graphs, their site now gives you a “One Year” graph. Even if they have data for many years and even if you can pick out individual years and even if you select specifically longer term graph options. One is left to wonder “who got to whom” at Wunderground…

At any rate, I’ll leave it to others to find similar examples for Canada. Clearly someone Up North is buggering the data and doing it in particularly obvious ways.

In Conclusion

Something was done to the Canadian data about 1997. There isn’t much other possible interpretation.

We have not had a raft of “Record Ever” high temperatures out of Canada. It’s been rather cold and frozen up there lately. We’ve got some record rain and snow events all over North America. Whatever is going on, it is not a warm balmy sunny Global Warming world.

The yield of wheat and barley are not reaching record highs. The planting schedule is not moving earlier in the year.

This GHCN / GISS Data is buggered. Canada is in the thick of it. Sharpen your pitchforks and light your lanterns and head North To Canada…


If anyone can explain why with ENSO as it is:



From this ENSO page.

And the rest of North America having a cold fit, that Canada might be claiming such “warmth”: Please, feel free to offer sugestions. All I see is “Rum, Buggery and The Lash” at work…

Update, just a bit later

Curiouser and Curiouser…

From Wunderground, I’ve gotten two images of temperatures. One from 1973-74, and one from 2010-11. Per the GISS graph, the later one ought to be “way hotter”… But when I look at it, I mostly see more “volatility” in the spring when sun hits tarmac now, and not a whole lot of “hotter summers”…

Canada Baker Lake 1974 Wunderground

Canada Baker Lake 1974 Wunderground

Compare that to the present, and I’m just not seeing all that added heat:

Canada Baker Lake 2010-11 Wunderground

Canada Baker Lake 2010-11 Wunderground

Summers are cooler (10 C vs 16 C) but the late winter is a touch warmer. Perhaps all that added heating in winter and all the growth they had growing from a few dozen to over 1700 people.


In 1946 the population was 32 of which 25 were Inuit. By the 2006 census, the population of 1,728 represented an increase of 14.7% from the 2001 census There are also roughly 1,000 miners that work in nearby mines. There is also potential for a uranium mine, called the Kiggavik Project, which is being proposed by AREVA Resources Canada.

Looks to me like winter UHI and spring Tarmac Solar Heat at the expanded airport, coupled with an overall cooling trend visible in the summer months.

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 AGW Science and Background and tagged , , . Bookmark the permalink.

19 Responses to What the Giss?

  1. Level_Head says:

    The TOPIX satellite record was grafted onto an earlier one with a similar step displacement around that same time, as I recall.

    But here, the discrepancy between two records of the same underlying data — can that be true — seems astounding.

    What adjustments are done at the Wolfram display level?

    ===|==============/ Level Head

  2. E.M.Smith says:


    It’s a bit hard to know what W.A. have as data. They have a “submit request” screen for details and I just don’t have the time for that much followup.

    I suspect it may be the PRISM data they are using, but again, have not the time to “dig here”.

    But given the “reality on the ground” is that in California I’m cold and not runnig the AC, with tomatoes a fond memory and with even Peas slow to germinate; with northern rivers just now melting enough to not have ice jams. With growing schedules running late, not early….

    You add it all up and it is the GHCN / GISS “data” that are out of sync, not the W.A. set…

  3. Level_Head says:

    I’m further south in California, and thirty to forty degrees cooler than a few years ago. That is quite delicious, but it has the effects on growing seasons that you describe, only partially offset by increases in CO2.

    However, all of our personal experiences are not Baker City, Nunavit — and we can anecdote on them all we want and still not know who the baker was for the different recipes in your graphs.

    It’s not your responsibility to find out why the data from that spot produces two wildly disparate stories, but you certainly have piqued curiosity on the topic.

    ===|==============/ Level Head

  4. Pascvaks says:

    A Prayer:

    “Dear Lord, whoever it was you sent to the University of East Anglia’s Climate Research Unit not too long ago, could you please send them to the National Aeronautics and Space Administration’s Ultra Secret Goddard Institute for Space Studies to see what they can find in backchannel emails and file cabinets and trash cans? Something’s rotten in Canada. Their address is 2880 Broadway New York, NY 10025 USA. Thank You Lord!”

  5. boballab says:


    Wunderground is run by none other then one Jeff Masters a confirmed, died in the wool, The end is near!…Alarmist. Maybe he noticed you using his longterm graphs to debunk his religion? Don’t know.


  6. Verity Jones says:

    The WolframAlpha data gives a trend of 0.079 degF/year – which is huge – you need to look at the scales.

    Here’s where GHCNv3 has Baker lake:

    Again huge trend.

    Glad you’re starting to look at climate stuff again. I’m away from home and have a very dodgy web connection so looking at the surrounding stations will have to wait until I’m home.

    IIRC there was some funny stuff between GISS and Env Canada I looked at a while back e.g. this one –

  7. R. de Haan says:

    OT, 5.3 quake in Spain topples buildings

  8. E.M.Smith says:

    @Verty Jones:

    The trend line is a “fit” line to a block of data that has a large gap and “dip” at the start.

    IMHO, a less “misleading” technique (very useful in stock charts where various spurious events can cause “odd excursions”) is to look at the peaks and where the bulk of the data items lay on the chart.

    In this case, they are essentially on top of the 10 degree line.

    They are at the start. In the middle, and at the end.

    For that 0.079 trend to have been valid over the 60 years of the graph, their ought to be a 4.75 degree difference between the beginning and the end. I see a strange plunge in the begining, and a block of data on each side of it that is substantially on the 10 degree line.

    Even as late as 2007 or so, the ‘non-peak’ parts lay on the 10 degree line.

    The “slope” of the fit line is, IMHO, a statistical artifact and not “real”.

    I’ve not figured out a decent way to turn this technique of “seeing the data” into a mathematical process. It needs some kind of “center of the data initial trial fit” line then you do a kind of “refit” that deweights the data items furthest from that line.

    Perhaps a “least square roots” fit ;-)

    At any rate, I find the “least squares” line fit very misleading in a lot of circumstances, and especially when the data are a bit “dirty” with outlier segments toward the ends (and most of the temperature data is very much like that in the early years…)

    So, for example, on this chart of SPY, you find a LSF line and it says it’s just in a wopping big trend up. But add “volume by price” and they point out the structure of it is really 3 segments. A flat high volume, a rise on lower volume, and a flat on high volume again.


    As there isn’t any “volume” on temperatures, it’s harder to define how to find the same thing. But, as we saw in the “Regime Change” charts in Alaska, it is very easy for the eye to spot the “step change” higher and the flat after it, and very hard for a LSF line to do anything other than shout “Consistent warming trend” that simply isn’t there…

    So, much as I’d like to have a better way to describe why I’m saying “the data look flat” when the LSF line is showing a rise; that is the best I can do because that is exactly what it is. The “look” of the data is flat. It rests on the 10 line through the mass of the data points +/- a little bit.

    @R. de Haan:

    They are in the 5.x range and I usually don’t put up a new posting just for that size. OTOH, there were a pair of 5.x in the Philippines as well, so perhaps it’s the start of something new after the “pause that refreshes”…

    So I’ll likely put up a new quake posting some time tonight.


    It has gotten steadily more “buggered” over the last year or two… Useful things removed. Things that used to work, broken… Sigh. I wish folks could just do “Data, -> truth” instead of “Agenda -> bugger data -> SEE I’m RIGHT!”

  9. R. de Haan says:

    @E.M., I know on this scale they are hadly worth mentioning but quakes are very rare at this location, there is considerable damage to buildings and infra structure and 7 people got killed.

    We had quite a big quake some time ago in the South of Spain but it was very deep.

    We also had the Koblenz quake in Germany, regular quakes occuring in Poland and recently Rumania.

    Italy, Greece, the Balkan’s, Turkey and Crete are always shaking.
    The seismic at the EU Continent however is exceptional.

  10. E.M.Smith says:

    @R. de Haan:

    Thanks for the explanation.

    I’m still “busy with other things” until later, so a posting focusing on them will have to wait a bit. Until then, folks can find maps at the “Earth Sciences” category on the right (prior quake and volcano postings with ‘live’ maps) and you’ve posted the “heads up” here.

    Such as this posting:


    I’m not doubting the “interest” in the quakes (they are on CNN and Fox both), it just doesn’t need an “immediacy” like an 8.0 does… Frankly, I’m expecting they may just be foreshocks…

    USGS closeup of Spain here:



    Magnitude 5.3 – SPAIN
    2011 May 11 16:47:26 UTC

    This event has been reviewed by a seismologist.Magnitude 5.3
    Date-Time Wednesday, May 11, 2011 at 16:47:26 UTC
    Wednesday, May 11, 2011 at 06:47:26 PM at epicenter
    Time of Earthquake in other Time Zones
    Location 37.697°N, 1.556°W
    Depth 1 km (~0.6 mile) (poorly constrained)
    Region SPAIN
    Distances 50 km (31 miles) SW of Murcia, Spain
    118 km (73 miles) SW of Alicante, Spain
    124 km (77 miles) NE of Almeria, Spain
    352 km (218 miles) SSE of MADRID, Spain
    Location Uncertainty horizontal +/- 14 km (8.7 miles); depth +/- 6.2 km (3.9 miles)
    Parameters NST=292, Nph=297, Dmin=301.7 km, Rmss=1.24 sec, Gp= 43°,
    M-type=body wave magnitude (Mb), Version=7
    Source USGS NEIC (WDCS-D)
    Event ID usc0003c5s

  11. R. de Haan says:

    @E.M., Thanks.
    One hour before the 5.3 they had a 4.5 shock.

    The news is stating this was the worst quake in 55 years but a few months ago they had a 6.9 in the South but very, very deep which caused no damages.

    At this link you find some pictures that show the damages.

  12. R. de Haan says:

    Will we have a second Maunder Minimum (1740) in 2020?

  13. boballab says:


    That graph from Wolfram Alpha is for Baker Lake, Washington USA not Baker Lake, Nunavut Canada:



    Look at the High on the Wolfram Alpha graph: 92°F in 1991.

    92°F just off the Arctic Circle? Egads the Warmists were right!!!

  14. E.M.Smith says:


    GAK. OK, I’ll try again.

    WolframAlpha has this one very annoying thing where you can state “Baker Lake Canada” and it will happily swap to either the center of a state or province or sometimes some semi-random other place. If you don’t check the station IDs each time, it’s a bit of a crap shoot…

    For example, putting in Baker Nunuvut gave me Baker Lousiana with an option for Baker Montana… Sigh…

    OK, after re-checking I get the same graph from this station: CYBK

    That Wunderground says is the same place


    Baker Lake, Nunavut
    8:28 PM CDT on May 11, 2011 (GMT -0500)
    Elev: 59 ft
    Lon: 96.1° W
    Lat: 64
    Baker Lake (CYBK)

    For July, 1991, they have this high temp on the 26th:
    Jul high avg low
    26 91 75 59

    So unless you have some further evidence that it’s not BOTH W.A. and Wunderground both wrong in the same way…

  15. E.M.Smith says:

    @R. de Haan:


    Victor Herrara

    “The New Solar Minimum and The Mini-Ice Age of the Twenty First Century”

    This was a wonderful presentation showing the incredibly strong correlation between solar activity and climate. Victor used wavelet analysis to show that high solar activity matches warm periods and low solar activity matches cold periods. He also identified the periodicity in this function and matched it to the Solar System Barycenter orbital changes of the sun.

    There were several charts of correlation of fit between the solar output, long term weather changes, and the Maunder, Dalton, etc. solar minimums and cold periods. And the correlation with the solar system barycenter to sun position AND with cosmic ray intensity. A wavelet squared transform was used to predict the solar output going forward and found a roughly 120 year period.

    The expectation, for a 120 year period, results in “solar secular minimums” in 2030-2040 and 2160-2170. The 2010 drop in solar output ought to continue for about 60 – 80 years. (though it was unclear to me if this meant the present ‘sleeping sun’ regime or just a return to normal cyclical values that were lower than the last 1/2 century of very high values).


    Habibullo Abdussamatov

    His presentation was titled simply “The Sun Dictates the Climate”. And that does more or less sum it up.

    A wonderful man with a quick broad smile. Yet he can scowl at the assertion that CO2 matters in a most effective way. I took an instant liking to him. There is much that American and British “climate scientists” can learn from this man. He impressed me as a very “old school” classical scientist. A “Mr. McGuire” type (for those who’ve read my stuff for a while).

    IMHO, he has the science exactly right.

    The downside? He is stating flat out that we are headed for a Little Ice Age. The solar changes dictate cooling. The ocean mass delays it for about 40 years. And we’re headed for a lot of cold. There is a 200 year periodic decent of Total Solar Irradiance, that causes a Little Ice Age, and we’re due. The mechanism he asserts is a 250 km decrease in solar radius and that changes solar interior dynamics and processes. A plot of phase and amplitude for both sun spot number and solar radius showed a near perfect match, with the onset of the solar quieting in 1999.

    Not the type to be bashful about making clear predictions (no wussy “projections” for this man!) he stated flat out the “New Little Ice Age begins in 2014.” Though with various lags from things such as ocean heat content and climate cooling rates, the depth of the NLIA is not reached until 2042 +/- 11 years for the solar minimum and then 2065 +/- 11 years for the temperature minimum.

    Mark your calendar accordingly…

  16. boballab says:


    I looked at the ID info the provided and they list it as Baker Lake Canada by WMO and NCDC station numbers, however I went to the source information, Environment Canada, and got the Monthly Means from them. After making them into Annual Means I compared that to GISS’s Graph and they are almost identical.

    Here is a link to the comparison Graph:

    There is no way that Wolfram Alpha is pulling up the right data for that station. Both GISS and EC do not show that big dip that they do between 1965 and 1970. Both EC and GISS show a big dip in 1972 and EC has another in 1976.

  17. E.M.Smith says:


    Got a link to the Environment Canada data? I’d like to do a more complete cross check of all sites against each other if possible.

    If there is a generic “aw shit” sloppy in W.A. I’d really like to know; as I’m using them more these days.

    BTW, if they have it, so does Wunderground. As noted above, the Wunder site had the same “high” in the data record.

    At any rate, once you have a “He said, She said” it’s very hard to figure out which one is truthful…

  18. boballab says:


    Here is the Link to the EC climate info:


    When you set the little green box’s search parameters use Monthly not Almanac, they haven’t got the Almanac all the way done (Almanac = Annual) for every year. Example if you do Almanac search for Baker Lake you only get back to about 2007, whereas when you do monthly you get back to the 1946 time frame.

    From there tell it to download a CSV for XML file from the Navigation Options box in the lower right hand corner.

    The funny thing about the EC site is when you look for Climate Data from their Homepage there is no listing using the word climate. They list it as “Historical Weather”. Check what I mean from the Homepage link:

  19. E.M.Smith says:


    I’ll take a look a bit later. For now, I’ve checked a “nearby” alternative site (Fort Smith) as an alternate and found the same general “look” to the data and same general “issue”.

    I’ve tossed the results into a new posting, so please take a look at it.

    Whatever “issue” is here is a much more generic one…


Comments are closed.