To Tell A Tale of Tyler Texas

First off, where or what is Tyler Texas?

It’s a modest sized town to the south east of Dallas an hour plus drive. I’m familiar with it as it is where my Uncle Ken lives. (Well, not IN Tyler, but close enough that it’s a nice little “puddle jumper” flight from Dallas to where he can come pick me up in his Cummins Diesel Pick-em-up Truck…)

Tyler has not changed a whole lot in the last 1/2 century (and not even THAT much in the last century). It’s growed a little. The airport turned from grass to concrete and now has some commercial jets. But mostly it’s still a sleepy college town with some agriculture around it and some decent restaurants for getting Bar B Que.

Even has a wiki page:,_Texas

Now the fact that it is also the county seat of Smith County has NOTHING to do with my thinking it’s a fine place to live… nothing at all! ;-)

But the reason I picked Tyler all comes down to this map from GISS via GIStemp:

GISS Anomaly Map for USA in March 2011

GISS Anomaly Map for USA in May 2011

Notice that Texas is shown as being Very Red.

Here, let me give you a close up of it:

USA on the GISS GIStemp May 2011 Anomaly Map

USA on the GISS GIStemp May 2011 Anomaly Map

Notice that per the legend all of Texas is about 2 to 4 C over heated. That’s like 3.6 to 7.2 F Hot Hot HOT!

Now I’m not always the sharpest tool in the shed, but I figure that if it was that hot, folks would have notice and my Uncle Ken woulda said something, him being prone to talking about the weather and all, and he’s not said a thing about it being hot enough to scald a scorpion in Texas, so I figure either he has completely lost that Texas Gift for the Met-A-phore, or it just isn’t hot enough to really notice. And GIStemp is a-lyin’ to us agin…

So I set out t’ see what I could see…

First up, I checked in at Wunderground. Does it think things are “way hot” in Texas?

Tyler Tx March 2010 to March 2011 Wunderground

Tyler Tx March 2010 to March 2011 Wunderground

Well now, as I reads that, it looks to me like the tempur-A-ture line is stayin’ pretty much inside them average lines. It was a mite cold in the middle o’ winter when it colored outside the lines to the downside jus’ a bit, but nothin’ here is lookin’ A-Tall like it’s 4 to 7 degees hot.

Does Wolfram Alpha have a different idea?

Tyler Texas 10 year data Wolfram Alpha

Tyler Texas 10 year data Wolfram Alpha

Well, doesn’t look out’a line with the last ten years. Winters of 2010 and 2011 were a mite cold, but nothin’ spectacular to report.

Tyler Texas All Data Wolframalpha

Tyler Texas All Data Wolframalpha

Well, even the “All Data” graph just does not show a whole lota warming goin’ on… Basically all that trend comes out of that initial sample being a bit low and then a long gap. The recent data just doesn’t show things getting warmer.

Maybe we can look at some parts o’ Texas that have a mite longer record?

Austin Texas, All Data, Wolframalpha

Austin Texas, All Data, Wolframalpha

Well, looks like things are getting colder in Austin!

Guess all that UHI from the growth of the town and adding the tech center just didn’t add up to much!

Maybe some other Texas towns can be warming?

Curtis Field, central Texas, All Data Wolframalpha

Curtis Field, central Texas, All Data Wolframalpha

Well, that’s a tiny bit better. We’ve got 0.015 F / year or 1.5 F in 100 years. IFF the trend continued… but it sure looks to me like like that hot time in the 1940s is not much different from the hot time in the 1990s and maybe the “warming” is just a statistical ar-T-fact…

Maybe it’s all President Bush’s fault? He lives near Waco, did he do it?

Waco Texas All Data Wolframalpha

Waco Texas All Data Wolframalpha

Oh Hell, it’s dropping in Waco. Looks like the “Gore Effect” frosts anything Mr. Bush might be doing…

So What’s The Deal?

Well, we can tell GIStemp to “only” smear data around by 250 km instead of by 1200 km and get a bit of an idea. On this ‘lesser smeared’ map we can see that there is a tiny little hot area toward the South West of Texas. (Not surprising, really, as it has been in the news for being basically on fire for the last many months. No, not a met-e-phore, actually On Fire. As in uncontrolled scrub fires.)

GISS GIStemp 250 km "smoothing" Mar 2011

GISS GIStemp 250 km "smoothing" Mar 2011

So does anyone ELSE have a more precise record of the temperatures? From the same “pest management” site that we saw in this prior posting:

we have a nice little map that makes things MUCH clearer.

DegreeDays May 20011 Deviation NV_41us

DegreeDays May 20011 Deviation NV_41us

Now we can see that there IS a hot spot. Down in South West Texas. While the entire northern half of the country is freezing it’s butt off with below average “degree days” of heat accumulation, there is a “hot spot” in the low side of Texas (even accepting that this is based on valid data, which has some doubts about it given all the adjustments and such done to it…)

So now we can see what the net-net effect of GIStemp is, simply and clearly.

If finds a hot spot in a part of Texas, one that’s not a whole lot, just a touch warm, and turns that into a 4 to 7 F HOT HOT HOT over almost the entire USA, even though on a very fine grain look, the USA is cold in half, and a touch warm in half, and overall about normal.

Even though the individual stations in Texas don’t show a warming trend to speak of, and many of them even show significant cooling.

How does it do that?

Well, a good Magician never divulges the “Trick”…

Added “May to May” Wunderground history

Tyler Texas May 2010 to May 2011

Tyler Texas May 2010 to May 2011

Still staying inside the ‘normal lines’ pretty much.

Here is the “one year” chart from Wolframalpha showing that their data do include a March group:

Tyler Texas 1 yr plot from Wolframalpha

Tyler Texas 1 yr plot from Wolframalpha

And in response to a comment, here is Missoula Montana

Missoula Montana

Missoula Montana

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 GIStemp Specific and tagged , , . Bookmark the permalink.

16 Responses to To Tell A Tale of Tyler Texas

  1. Thanks EM. Good work!

  2. Another Ian says:


    You have been GISSTIFIED!

    Use it if it works


    Ian Beale

  3. Eric Barnes says:

    Unless I’m missing something, pest management is badly off for my locality of Missoula.

    Missoula Weather Underground 2011

    It’s been cold and hasn’t ventured into above average for a solid 3 months.

  4. Eric Barnes says:

    OK, I’ve added a link to the website portion of my post.

    [ Don’t know what happened to your link. You don’t need to type the html, just past the url from the browser in. If you get the html off of what WordPress likes, it just deletes it… I’ve added a Missoula graph to the posting on the assumption that was what you had. -E.M.Smith ]

  5. Joel Heinrich says:

    Huh? Gismap says March 2011, your caption for it says May 2011, “wunderground” or Wolfram Alpha don’t have any data for March ’11. What are you trying to compare here?

    German National Weather Service says the same (although with a different baseline):

  6. E.M.Smith says:

    @Joel Heinrich:

    Sorry, mislabled the caption on the GISS map, it’s now corrected to March. The “Degree Day” map goes to May as that is the “now” end date on their data (why GISS runs 2 months late is anyones guess). For Wunderground, I ended the year in March so as to match GISS and Wolframalpha. They all ought to have the same ending date at MARCH except for the degree days map where it’s been running for overnight on my custom request and not completed yet, so I’ve just accepted the ‘current’ map.

    As degree days is a running total, it converges on ‘reality’ rather than diverges, so becomes less prone to trend error with longer duration.

    At any rate, I turned a “Mar” into a “May” when making the caption, nothing more. (It was right on the 250 km smoothing version, for comparison).

    Wunderground does have data for March. As does Wolframalpha. As you can see on the graphs above, they are long duration graphs that end “now” in March, so include the prior March, for Wolframalfpha. For Wunderground you can see that I set the “cutoff” at Mar on the far right to give a “Mar to Mar” year long graph.

    will give the current US “degree days” map

    will give you a wonderground history from May to May with March clearly visible.

    and hit the drop down for “year” will give you a graph ending in / with March data.

    Basically, everything is MARCH with the exception of the degree day map that is “now” and “now” is early May.

    So what I’m comparing here is pretty simple. March to March except for a ‘running total’ degree days that shows we’ve just not stacked up the degrees in most of the country to justify calling the place a burning hot zone…

    I’d have used a “Jan to Mar” GISS Map if they let you make one, but they don’t (you must take the prior Dec-Feb).

    To the extent there is a ‘mis-match’ it is only that Jan and Feb were a bit cool (not dramatically so, you can see it in the Tyler history up top from Wunderground) so the degree-day running total is going to be a bit cooler.

    So something in GIStemp is happy to take a temperature history that is staying pretty much ‘inside the lines’ of normal ranges, and is cooling over long times in a large part of the space, and that has been cold overall this year to date; and turn that into HOT HOT HOT!!! over most of the USA in the present default map. Highly misleading at least.

    The degree day map gives a much better idea what trends are actually going on.

    What Germany is doing is something I’ve done no looking at at all. But if they get their data from the same NCDC thermometers as GISS and use the same methods from the “peer reviewed literature” and ….

    @Eric Barnes:

    Isn’t Missoula on that left edge of Montana? I place that on a part of the degree day map that says it’s between -14 and -134 degree days on the low side of normal. That’s saying it’s been cold mostly. You are saying it’s been cold mostly. Sounds like a match to me…

    I’ve added a Tyler Texas “year to date” map at the bottom so folks can see any month of the last twelve as they like it.

    The basic point stays the same: It’s just not unusual in any way, though things have generally be a bit on the cool side most of the time. It’s certainly not been 4 -7 F overheated…

  7. Eric Barnes says:

    Went back to your post on the definition of degree days which says …

    This method calculates daily accumulation of GDDs as the average daily temperature (degrees F) minus 50. The “modification” refers to the limits imposed on the daily maximum and minimum temperatures allowed in the calculation. Daily maximums greater than 86 degrees F are set equal to 86 in the calculation of the daily average temperature. Similarly, daily minimums less than 50 degrees F are set equal to 50 in the calculation.

    So I guess the map makes sense, but would under-represent deviations of temps on the upper and lower end of the absolute temperature scale.
    If there was no lower limit for missoula (upper bridge of the nose on the western face) it would be much lower IMO, just by eyeballing. -134 would be less than a degree off average for the calendar year which doesn’t seem to make sense? Also Tyler seems off according to weatherundground and application of GDD.

  8. Roger Sowell says:

    E.M., this post agrees with what I found when looking at the HADCRUT3 data. Small towns such as Abiline, TX have had zero warming in the past 120 years.

    The J.D. Goodridge paper (1992) also found the same thing for rural counties in California – no warming. No population growth, either. However, urbanized counties with large population growth had significant warming.

    Conclusion: CO2 is very lazy where few people are found. Something about the human-CO2 interaction causes great warming. More research funds must be poured into study of this effect.

    (sarc off now.)


  9. E.M.Smith says:

    @Eric Barnes:

    There are several definitions of Degree Days. It is intended for use in planing planting and pest managment. Plants have particular needs for a certain quantity of heat, and the associated pests tend to track the same total heat so as to be bursting on the scene as the same time as their plants.

    For the charts on this page I accepted the “default” of 41 F rather than the 50 F. It is often a user configurable option what bounds you set; on the low side, and on the hot side.

    For those plants that have a ‘high temp cutoff’ on growth, the tops can be clipped at various points too. Often, but not always, in that 86 F to 95 F range. For tropical plants, they need a hotter low end before they will grow (over 50 F for most, over 60 F for some) and have higher ‘top ends’. For some cool crops, like kale, they continue some growth even at freezing. So in studies of particular plants, you will find the Degree Day limits chacterized for each species. (Sometimes even by variety. So there are “Siberian” tomatoes that will set fruit down at 40F while most cut out at 50 F). At the same time as some rice slows growth over 86 F, some varieties grow well at well over that. So when a farmer knows the ranges of their variety, they can set those into the degree day calculator and KNOW absolutely what day the crop has ripened or grown to it’s optimum. (As can the investor 1000 miles away watching the ‘degree days’ for Iowa corn, for example, and the approach of winter…)

    The Dedgree Day view tends to match best how plants and animals will react to any “trend”. Also, the maps use a much larger number of thermometers so you can have local granularity. (As it doesn’t do your plants any good to know what the average was over a 1200 km range, they are only going to grow in accord with your local micro-climate!) So it is, in many ways, much closer to how folks “feel it has been”.

    The major virtue that map had for me was the fine grain. I’d love to have a fine grained “anomaly map” using all the station of America (or the world) without all the adjusting and homogenizing but I don’t know where one is right now.

    The NCDC maps that use the USHCN are adjusted in strange and wonderous ways. The GHCN based GISS is similar in whacky adjustment / homogenizing but with a drastically reduced set of thermometers. The only folks who seem to start from the “straight poop” from detailed thermometer counts are the “degree day” folks (as they do NOT want the data UHI corrected or homogenized; the plants only care about the ACTUAL temperatures experienced…)

    At any rate, it’s the best I could find / remember in the time available; and the fact that it most accurately reflects how life in general is experiencing the temperatures is, IMHO, rather a feature.

  10. Jim says:

    @Another Ian-But E.M. GISSticulates well in the telling.

  11. Another Ian says:


    Look out – this could be a growth industry!

  12. Pascvaks says:

    (EM – just came here from GIS article, opened as “https”)

  13. Another Ian says:


    Re E.M.’s next post – would Hansen be GISStaken?

  14. boballab says:


    Here is the NOAA/NCDC Trend for the state of Texas at this link:

    What is funny they show a cooling trend for 1895-2010 and you see a nice sine wave in it as well.

  15. E.M.Smith says:


    Well, I’ve changed nothing at all. I think “it isn’t me”…


    Ooh! New Toy! And a cool one too ;-)

  16. Eric Barnes says:

    I like the higher resolution too. I’d think anyone would (outside of GISS at least :) ). I just wish it was higher resolution with full temp range. Thanks for the explanation. :)

Comments are closed.