An Exercise For Dull-Minded Folks

People who try to guess what temperatures should have been in the past are not thinking clearly. A change in elevation can cause temperatures to either increase or decrease. Along the Colorado Front Range, temperatures in the foothills are almost always higher than they are down at lower elevations. Cold air sinks.

As far as TOBS goes, you have no idea if a station operator in 1940 reset their thermometer before they went to bed, as anyone would do who wanted to get meaningful data would do. You don’t know when the road got paved in front of their house. You don’t know when the city started doing snow removal. You don’t know when they planted a lawn, painted their house a different color or installed an air conditioner. There are dozens of factors which affect temperature which can’t be guessed about the past.

Leave the data alone. Adjusting the data will almost certainly introduce confirmation bias, and the odds of getting the adjustments correct are slim to none.

About stevengoddard

Just having fun
This entry was posted in Uncategorized. Bookmark the permalink.

36 Responses to An Exercise For Dull-Minded Folks

  1. talldave2 says:

    I have an idea, maybe not a new one. It should be provable that the simple average you’re doing is more accurate than USHCN. Great Lakes ice is a proxy, the number of 90 and 100 degree days you’ve posted before, the “modern proxies problem” that led to Mike’s Nature trick… I’m sure there are far more.

    I think if you bound them all up together you’d have an almost irrefutable argument that the real truth lies closer to the simpler average than what they’re publishing. Heck, their entire adjustment probably goes the wrong way, given the UHI effects that McIntyre and others have found.

    • Eric Simpson says:

      That’s an outstanding idea!

      Perhaps people could ignore a couple of things that don’t fit the fear mongers’ narrative. But point to one thing after another thing after another thing that doesn’t fit, and it will be impossible to ignore. I’m sure there are many many types of records or data that doesn’t fit the warmist narrative, so dig them up and put them all together in an essay or presentation and make it clear as day that the notion that things are hotter than hell or even hotter than it was many decades ago, that just doesn’t fly in the face of all these things that shouldn’t be true if it really was hotter now.

      Yes, one thing that doesn’t make jive with the warmist picture is the number of hot days now and then. If it’s so much hotter now then there should be a lot more hot days now then before. It’s just flies in the face of reason that it is otherwise. Another is temperature records. Globally, we know that the hottest day ever was set in 1913. The coldest.. 1983. That doesn’t make sense. After a century of hockey stick warming, the coldest day should have been set way way back then, like in 1913, and the hottest day, that should have been set just in the last decade, in “the hottest decade on record.” And my understanding is that we see the same thing with the hottest and coldest day records for US states, with on average the coldest day records happening at a more recent time than the hottest days. Probably the same thing would be found if we look at individual countries, and continents.

      So, yes, put a ton of things together that scream: it’s not hotter now, it was hotter then! Add to that the evidence of data manipulations and they are sitting ugly. And top it all off with the point that there is in fact zero empirical evidence that CO2 causes warming:

    • Eric Simpson says:

      Holy jiminy cricket I’ve found something! I looked at the seven continents, and sure enough, lo and behold: 6 out of the 7 continents set their record for their coldest day after their record for the hottest day (only Asia was the exception). I used this source (which used data from the World Meteorological Association).

      Wait, don’t tell me, no, unbelievable, um, geez, that can’t be the case, we’ve supposedly been going through a century of crazy hockey stick warming, and this. And if anyone thinks the 6 out 7 continents thing, as well as the worldwide record, was a fluke, guess what is really going to blow your socks off? I also clicked over to the WMA site, and I find the Northern Hemisphere set it’s record for the coldest day AFTER the record for the hottest day. And, the Southern Hemisphere set it’s record for the coldest day AFTER the record for the hottest day. And, the Western Hemisphere set it’s record for the coldest day AFTER the record for the hottest day. And, the Eastern Hemisphere set it’s record for the coldest day AFTER the record for the hottest day. Every single possible hemisphere set it’s cold record after its hot record. 6 out of 7 continents set its cold record after its hot record. Same with the world record. Something doesn’t add up, as far as the Chicken Little Brigade and their temperature “data.” Global warming my rear. This is a total charade. A century of runaway out of control hockey stick warming? Nope.

    • Eric Simpson says:

      Remember, the one thing that the warmist deceivers can’t mess with are the records for high and low temperatures. They cannot “adjust” these records away. So Steven finds evidence of wholesale data manipulation. And you have these low level trolls coming on here tying to quibble with details of Steven’s work, but the one thing that the data manipulators can’t manipulate flat out contradicts all of their manipulations. Steven is on to something. The Chicken Little Brigade has been messin’ with the temperature data.
      “It doesn’t matter what is true, it only matters what people believe is true.” -Paul Watson, Greenpeace

      • Jason Calley says:

        “Remember, the one thing that the warmist deceivers can’t mess with are the records for high and low temperatures.”

        Exactly! If there really were a long term rising trend in world temperatures there ought to be some trace of it in the records of extremes. Remember the CAGW cultists argument that warming was “loading the dice” toward extreme heat? If the dice are loaded, why are they not rolling like loaded dice?

      • Andy DC says:

        I am not sure you are correct. National Airport near DC is constantly 3 degrees warmer than surrounding stations. On June 17th this year, National set a high temperature record of 97. Old record was 95 Among a large number of surrounding stations, the highest was 94. Dulles was 91. Absolutely no meteorological reason for such a spread. So it seems apparent that the cheaters stoled a record for their team.

  2. Bill S says:

    I can’t quite buy your argument yet. Assuming that errors average to zero just means to me that you did not have the time to review the data carefully. What I can say accurately is that I have never learned how to make a computer tell the difference between an outlier and corrupt data.

    e.g. I believe you could make your case stronger if you simply told us what is the base period NOAA uses that skews the results. Otherwise, I just have you versus Watts and no time to look into it myself.

    • Unless you can travel back in time, you can’t possibly tell what what was going on at the station in Salina Kansas in 1932.

      • Sundance says:

        Steve FYI interesting exchange by Reto Reudy Gavin and team obtained by judicial watch. It’s from 2010 but new to me and thought you may be interested.

      • Anthony S says:

        Unfortunately coverage is a bit(very) spotty, but there is imagery from 1938 for Salina.

        • That picture provides no where near enough information to be able to judge what influences might contaminate the temperature record from that location. I

      • It seems to me that if scientists were truly interested in creating a more accurate temperature record, they would be comparing the original thermometer record, their adjusted version, and numerous other proxies. If the beginning of the last century really was colder than the thermometer record shows, then other proxies (lake bed sediment, tree rings, stalactite growth, you name it) should show a stronger correlation to the adjusted data than the original data. It seems far more common for proxies to be replaced by the corrected data, rather than compared to it, however.

  3. If only people documented stuff like time of observation, it would make it a lot easier to correct. Oh wait, they do:

    • DirkH says:

      Why do all the errors the stupid operators in the past made go in the same direction, and why is it the direction that today’s warmist movement needs for its agenda?
      What are the odds?

      You’ll be running out of such stupid mistakes by the operators of the past in the future, and you will be unable to cool the past down further (also; for simple logical reasons – you can’t cool the past down to ice age conditions.)

      The current plateau (since 26 years) is probably already a sign that the warmist movement ran out of revisionist steam a while ago. Reality has already caught up with the number torturers.

    • Anyone who operates a min/max thermometer becomes aware of the TOBS problem within a week. It is not a subtle problem. I figured out when I was seven years old that I had to reset the thermometer at night in order to get any sort of meaningful results.

      The fact that they took their observation at noon, does not preclude them from resetting the thermometer at night.

    • talldave2 says:

      Zeke, how do you know the TOBS observations get you closer to reality, as opposed to farther from it? How on Earth can we have record-late Great Lakes ice and every economist saying the “harsh winter” crushed Q1 GDP when USHCN is saying the Great Lakes region and the US had one of their warmer years/quarters on record?

      How can you justify ignoring UHI given the work of Watts and McIntyre?

  4. Bill S says:

    Horse manure. I can look at the data and at a minimum determine if there is an obvious bias. Easiest example is a case of simple dyslexia. A value of 7.3 for a parameter when physics says the parameter has a range of 3.0 to 4.5.

  5. Eliza says:

    Sorry SG 1 Zeke 0 just joking LOL. I dont buy it for a second!

  6. There Is No Substitute for Victory. says:

    A good example of this type of error is the weather station at Needles, California. For a year or two Needles reported temperatures that seemed more like the surface of the Sun than the lawn of city hall. When the weather weenies went to investigate they found that Needles had built a new city hall and fire station and that the new blacktop fire station parking lot was now only 15 feet away from the weather station thermometer.

    This is just one example of what Steven is mentioning, there must be many others.

  7. bkivey says:

    ‘. . . what temperatures should have been in the past . . .”
    People who engage in this activity are practicing magical thinking. The concept is nonsensical.

  8. Bob Koss says:

    I was going to put this on the well known skeptic/Mann thread, but it is so far down I figure it might get lost. Hope you don’t mind me putting it here.
    It appears GHCN adjusted uses the USHCN FLs.52i.tavg file without the estimated values. They are set to -9999. The raw files seem to match.

    I calculated the average data-point latitude from 1901-2013 for both the USHCN raw and adjusted data found in the GHCN database. The network’s average latitude hasn’t moved very much since 1901 even though quite a few stations are now missing or unrecorded. Only 921 raw and 736 adjusted stations remain.

    The USHCN network has been very stable location-wise when averaging the location of all the yearly data-points. I think it is a good candidate for the type of analysis you are doing. It would be a problem doing it for the whole world though. There, there is a several degree shift in latitude over time.

    Here are the USHCN latitude figures.
    Average Raw=39.55, StDev=0.06, min.=39.29, max.=39.69.
    Adjusted=39.67, StDdev=0.1, min.=39.38, max.=39.98.

    And the USHCN longitude figures.
    Average Raw=-95.56, StDev=0.55, min.=-95.92, max.=-92.95.
    Adjusted=-95.44, StDev=0.61, min.=-95.92, max.=-92.7.

    If I had started with 1921 after the network was more than 90% complete the StDevs would drop considerably.

    I agree showing only anomalies is a poor way to go. The base period anomalies don’t show the actual temperature difference between raw and adjusted data-sets since by definition the anomaly period for each data-set will average out to zero and show virtually no difference during that time. This also reduces the actual temperature spread in all other years when both sets are seen together in a graph.

    I have a couple comparison graphs I’ll put up tomorrow when I’m not so tired.

    Keep up the good work, Steven.

    • Bob Koss says:

      HEre the graphs I promised last night. They are just to show the visual difference between an anomaly graph and a graph showing actual temperature. My graphs are in Celsius while Steve works in Fahrenheit, so in mine the visual difference isn’t quite as stark as in Steve’s graphs would be of the same data.

      Here is actual temperature.

      Here is the same data shown as an anomaly graph.

    • vicgallus says:

      I might have to add that Darwin was a large town of 47 000 people in very low density housing. The cyclone didn’t rip up roads or replant trees so it wasn’t a reverse UHI effect. It just shows that large changes in environment don’t necessarily show up as a discontinuity in the data. Best not to assume that something not related to local climate is being corrected for when adjusting data.

  9. Stephen Richards says:

    Correct. Mosher’s dog has returned.

  10. Stephen Richards says:

    Zeke Hausfather says:

    June 25, 2014 at 3:09 am

    Even if you document tob that tells you nothing about what the temp was when it should have been measured.

  11. Robertv says:

    It seems that snow is also a thing of the past down under.

  12. Andy DC says:

    Wait until this weekend down in Aussieland, you ain’t seen nothing yet as an even much stronger cold front moves in!

  13. Brian H says:

    The Master Scam is composed of many Little Scams. The TOBS retro-adjustments are one of the more blatant of the latter.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s