Skip to content

Temperature Adjustments In The Canadian Arctic

February 8, 2015

By Paul Homewood


I have had a quick look at temperature adjustments in the part of the Arctic which I previously looked at, and where they were so dominant.

The area already covered was Greenland east to Siberia.



Using Nick Stokes handy tool, we can look at Siberia through to Canada. (Pinks mark stations where a warming trend has been added, cyan the opposite, and yellow unaltered).




As can be seen, the yellow stations predominate, meaning that any adjustments made at the rest will have little overall effect.

Otherwise pinks and blues seem largely to cancel each other out. Most importantly of all, though, is that there seems to be little pattern to the adjustments, unlike the ones we found in the rest of the Arctic. For instance, Barter Island, Alaska crops up as blue, but the adjustment cools then warms up to about 1990. Then there is a long gap before we get a handful of recent readings that may have no real relevance to what has gone before.




Meanwhile, across the Bering Strait, we have the example of Markovo, which again shows as a blue. Here, temperatures prior to 1920 were adjusted up, but since then there has been no change.






Overall, it seems reasonable to conclude that adjustments either way have had little effect on this part of the Arctic.

  1. Rud Istvan permalink
    February 8, 2015 5:59 pm

    Paul, I was curious about extreme NE Canada. Above, you show Nick Stokes’ 6 stations there. I went to his tool to pull up the GHCN station names. Of the six, Nick shows one cooling adjustment (Mould Bay), one warming adjustment (Alert), and four ‘no adjustment’ (Isachsen, Eureka, Resolute, and Sachs Harbor).
    GISS uses GHCN. Went to the GISS site just as you do, and pulled up all six station comparisons. Mould Bay was cooled a bit. Alert was warmed a bit. But of the four supposed no adjustments, GISS warmed three. Resolute and Sachs Harbor were strongly warmed. Eureka was warmed some. Only Isachsen is visually unchanged.

    Now, this is NASA GISS v3 adjustment to NOAA GHCN v2 raw. But Nicks tool appears not accurate for the GHCN raw to adjusted, because I am quite certain his general conclusion that there is no GHCN homogenization warming bias (cools and warms balance out, most are neutral) is wrong. See the 2012 EGUA paper available at Steirou and Koutsoyiannis deserves much wider attention than it has gotten. It pulled a geographically diversified random sample of 125 GHCN homogenizations and statistically checked. Strong general warming bias with near statistical certainty, and almost no neutrals. They apparently could not get it published for the usual warmunist reasons, so read it at session HS-7.4 and then posted it. Examples DeBilt and Sulina were included in essay When Data Isn’t in the new ebook.
    If you want the v2 ‘raw’ and v3 adjusted for the six NE Canada stations for a post, email and they are on the way, already archived. Regards.

    • February 8, 2015 6:32 pm

      I think Nick did admit that his data was not up to date. If you click on a current site, it says “To 2011”, so I would guess that was after GHCN V3 was introduced in Nov 2011.

      The other problem I have found with Nick’s is that if, for instance a station ends in 1970, and has the typical Icelandic adjustment cooling the 1940’s, it actually appears as a “cooling adj” (Because the period before 1940 was left unadjusted.)

      This of course in reality a warming adj as far as 2015 is concerned.

  2. Anything is possible permalink
    February 8, 2015 7:08 pm

    Speaking of Alert, check out the data for March 2014, available here :

    It claims that about half of the daily maximum temperatures are missing, but when you check against the hourly recordings, they are nearly all intact and the only missing data is actually under “wind direction.” This appears to happen (shock horror) more often on colder days – notably 11th-12th and 15th-19th.

    By my calculations the average maximum temperature for the month based on the hourly data should be -25.6C, as opposed to the claimed figure of -22.2C.

    What’s betting that our friend Gavin has taken the higher figure on trust, and joyously spread 3.4C of apparently spurious warmth 1200Km in every direction?

    I have no idea how widespread this sort of nonsense is, but if that is the best they can do, the whole darn lot needs to be thrown in the dustbin.

    • A C Osborn permalink
      February 8, 2015 7:32 pm

      Very common. They use “E” estimated data all the time.

    • Rud Istvan permalink
      February 8, 2015 9:16 pm

      Went and looked. Seems to be another example of the regional expectations adjustments problem. These almost always reject/adjust only ‘cold’ data. Best two ‘pristine’ examples found so far are BEST 166900, Amundsen Scott (Mosher violently disagrees, but doesn’t argue facts, only conclusions), and Aus BOM Acorn station 82039 (Rutherglen, a rural agricultural research station with proper Class 1 siting and maintenance, solid metadata, and eye witnesses). GHCN Reykjavik (Iceland), Sulina (Rumania), and Valencia Observatory (Ireland) are probably good examples. De Bilt, (Netherlands), maybe. Many remote sites are not pristine because located at airports with tarmac (a Greenland problem). Some famous long record ‘observatory’ sites are not pristine because of UHI over time (Norway and New Zealand for sure). I am trying to accumulate as many corroborated additional examples as possible of adjustments to known ‘pristine’ longish record sites around the world. The problem is ‘known pristine’. Any help greatly appreciated. Plan is to get a sufficiently large inventory and then do some aggregate adjustment analysis, trying to reconstruct local ‘true’ trends and compare to ‘regional expectations’. Sort of a historical USCRN for the world. That is why went and looked at the 6 remote NE Canada sites. Rural/airport is not pristine. Eureka, however, looks very promising for these purposes.

  3. Ralph Bullis permalink
    February 8, 2015 9:52 pm

    Paul – I am in the process of compiling temperature data from the Canadian Arctic for the purposes of completing a paper. Here are a couple of notes from that paper:

    “Weather stations
    The longest continuous records of temperature data come from a very few weather stations often based at locations that were originally trading posts operating in the late 1800s. These include Fort McPherson (from 1892), Fort Simpson (from 1895) and Fort Good Hope (from1908). These sites at old trading posts provide over 100 years of fairly continuous temperature records, with some notable discontinuities of data; however, it must be noted that only one of these stations actually lies within the “Arctic” (commonly accepted as that part of Canada located north of the Arctic Circle, latitude 66 degrees, 32 minutes North). Other weather stations date back to the 1930s when the Canadian north began to be explored for mineral resources and began to “open up” to modern means of transportation. These stations include Norman Wells, Yellowknife (1943), Coppermine/Kugluktuk (1930), Aklavik (1926), Cambridge Bay (1929) and Wrigley (1943). Later still, weather stations were added in the late 1940s and into the 1950s and these include Sachs Harbour (1955), Alert (1950) and Resolute Bay (1947). It is of some importance to note here that some of the longer term stations do have extensive periods (up to several years) when data were not recorded. Environment Canada’s explanation for these gaps in data is simply that there were times when the stations were not manned and data were not collected. These gaps in data therefore only add to the uncertainty in interpreting long-term climate data.
    All temperature data used for this paper can be found on the Environment Canada website. Historical data sets available from Environment Canada for the stations studied in this paper often do not continue beyond 2007. The explanation for this from Environment Canada is that “the most recent data available …. is from 2007 as this is when it was last updated. We are currently looking into other options to provide large amounts of data. Unfortunately we do not yet have a timeline for when data may be available in a different format”. Therefore, trends discussed here generally terminate in 2007. When available, later temperature data have been added.
    Because lengths of time shorter than about 30 years are not considered sufficient to support any valid conclusions regarding long-term temperature changes, this paper will use only those stations having 50 years or more of temperature data.

    Weather stations with more than 100 years of data

    Weather stations in the Canadian north with the longest continuous records are located at sites of old trading posts operated predominantly by the Hudson’s Bay Company. The earliest data available from the Environment Canada website are from Fort Good Hope, Fort McPherson and Fort Simpson. Each of these stations provides temperature data for periods in excess of 100 years with the earliest data going back to 1892 (Ft. McPherson).”

    Some observations are:

    “Observations for Fort Good Hope are that the current warming trend is not statistically different from that observed during 1940/1950 and that one of the coldest periods in the record occurred during 1980/1990. Any long-term (i.e. >100 years) persistent increase in temperature is not obvious in the station data. These trends mimic the regional trends seen in the HadCRU graph shown above.”

    “Observations for Fort McPherson are that there appears to have been an increase in temperature from 1908 through 1918 at which time temperature appears to have stabilized through to about 1985 when temperatures appear to have increased to about 1995 at which point they stabilized once again. Any long-term (i.e. >100 years) persistent increase in temperature is not obvious in the station data.”

    Ft. Simpson shows a modest upward trend from 1855 through 2009; however, mean monthly temperatures for that station are trending down from 1997 to 2009.

  4. eric permalink
    February 9, 2015 10:08 am

    “Ice conditions were exceptional. In fact, so little ice has never before been noted. The expedition all but established a record, sailing as far north as 81º 29′ in ice-free water.”

    “Formerly the waters around Spitzbergen held an even summer temperature of about 3º Celsius; this year recorded temperatures up to 15º, and last winter the ocean did not freeze over on the north coast of Spitzbergen.”

  5. February 9, 2015 3:27 pm

    For the last couple years Environment Canada has listed the airports separate from the settlements/ stations. There is no trend in temperature, if you include wind chill the airport is usually the coldest because they are more open.

    You have to remember that places like Mould Bay and Eureka only get one flight of a Twin Otter a month in the winter, so it cannot have any effect on temperature.

    To access them go to link then click then select map and select Provincial Summary:

    ps This morning there are wind chill’s as low as -58 C in Baker Lake and Lupin.

  6. February 10, 2015 12:09 am

    Although this post considers GHCN adjustments rather than Gistemp results, the scarcity of current stations in the Canadian Arctic gives considerable influence to the few currently reporting. In this context I sent a comment to Reto Ruedy of GISS last autumn in regard to Eureka, mentioned in comments here. As that comment included images I have posted it, lightly edited, as Temperature Adjustments in The Canadian Arctic – some supplementary information for a comment” at my own blog.

    Also included in my blog post is a warning regarding recycling of some Canadian WMO IDs, which leads to amalgamation of data from two completely different locations into one single GHCN record, with the possibility that further records may be similarly amalgamated in future if reports from the newer locations make their way into GHCN data.

    Finally, note that the single raw data record for Eureka in GHCN v3 derives from seven separate scribal records in v2.mean in GHCN v2. Gistemp used to combine these into a single record using a version of the reference station method. GHCN v3 does this differently. Rud Istvan, previous comment (“Eureka, however, looks very promising for these purposes”), may need to look into this. I think I recall seeing a warning that Eureka data may need to be treated with care, but unfortunately I cannot recall where.

  7. bob miller permalink
    February 12, 2015 3:49 am

    please stay on top of this investigation and be unrelenting, the truth must be kept out in the public at all times, thanks so much for your work here.

  8. February 12, 2015 6:52 am

    We know that temperatures for the 1910 to 1940 period have been cooled, and the cooling between the 1940s and 1970s have been warmed, because we have a record of what they used to be prior to the “adjustments” in a 1981 paper published by James Hansen:


    Click to access Hansen81_CO2_Impact.pdf

    “The most sophisticated models suggest a mean warming of 2 to 3.5 C for doubling the CO2 concentration from 300 to 600 ppm. The major difficulty in accepting this theory has been the absence of observed warming coincident with the historic CO2 increase. In fact, the temperature in the Northern Hemisphere decreased by about 0.5 C between 1940 and 1970, a time of rapid CO2 buildup. The time history of the warming obviously does not follow the course of the CO2 increase (Fig. 1), indicating that other factors must affect global mean temperature.”

    “[R]ecent claims that climate models overestimate the impact of radiative perturbations by an order of magnitude have raised the issue of whether the greenhouse effect is well understood.”

    “Northern latitudes warmed ~ 0.8 C between the 1880s and 1940, then cooled ~ 0.5 C between 1940 and 1970, in agreement with other analyses (9, 43). Low latitudes warmed ~ 0.3 C between 1880 and 1930, with little change thereafter. The global mean temperature increased ~ 0.5 C between 1885 and 1940, with slight cooling thereafter.”

    “A remarkable conclusion from Fig. 3 is that the global mean temperature is almost as high today [1980] as it was in 1940.”


    Graph from the paper showing the global trends (notice 1940 was still about 0.05 C to 0.1 C warmer than 1980):


    Today, NASA’s graph of NH temperature shows only-0.2 C of cooling in the NH between 1940 and 1970 (instead of the -0.5 C NASA’s Hansen identified in 1981), only +0.4 C of warming between 1880 and 1940 (instead of the +0.8 C identified by NASA’s Hansen in 1981):


    And NASA’s graph of global temperatures now shows 1980 as +0.2 C warmer than 1940 rather than 0.05 C to 0.1 C colder on Hansen’s 1980 graph:


    So what is the justification for dramatically altering temperature data from many decades ago—and only in the direction that is more consistent with the models (cooling the early 20th century warming, warming the mid-20th century cooling)?

    And what is the justification for having land surface temperatures heat up about +0.5 C more (!) than sea surface temperatures since ~1980 (see graph below)? Why would “well-mixed greenhouse gases” heat up land surfaces almost 3 times as much as sea surfaces? Could the reason be that land surface temperatures are more…adjustable than sea surface temperatures?


  1. Temperature Adjustments In The Canadian Arctic – some supplementary information for a comment | Peter O'Neill's Blog

Comments are closed.

%d bloggers like this: