Unur.com has a wonderful visualization of the data

Want to SEE a Station’s Temperature Data?

I’ve been doing graphs for each region and country, and thinking that, eventually, I ought to take on the work of doing graphs for each individual station… and stumbled on someone who has already done it.

http://www.unur.com/climate/ghcn-v2/

There is a “top level” with each “Region” laid out (Africa, Asia, South & North America, etc.). You “click a region” then get a list of countries. Pick one and you get ALL the stations listed. While you do need to know the station number to pick one out, that’s not so hard to find.

Two “nice bits”. Each graph starts as a constant scale. This lets you see how that record looks in the overall context of the entire temperature record. From start of the records to the end, from coldest to hottest. Just what is it’s context? The impact of seeing a “tiny little wiggly scrap of data” in a vast box of empty white puts things in perspective. The graphs start at 1850 (in keeping with the CRU / Hadley usage) so you get 30 years more than from GISS. It would be “nice to have” for the graphs to start time in 1720 when the first thermometers start, but that’s probably overkill for most folks. Ok, so you have your high level graph. What then?

You can then click on the graph proper and get a zoomed in version with much stronger ‘wiggle’, (but by now you have the context and know you are not seeing dragons, just small lizards under a very large magnifier…)

There is something about seeing a graph like this one:

Raoul Island New Zealand

Raoul Island New Zealand

From this page:

http://www.unur.com/climate/ghcn-v2/507/93994.html

to put things in perspective…

When you visit the actual pages you will get the “click through” version that gets to the “zoomed in” view. You ALSO will get the “meta data” for that station from two different sources. The NOAA meteorological station location information from NOAA and the GHCN Version 2 “inventory” data.

Nicely done. Just nicely done.

My only complaint would be that the GHCN data are described as “raw”, and as we’ve seen, even the “UN-adjusted” data are not “raw”. But many of us have fallen into that word trap, even me. (After all, GISS calls it “raw” at times, as do many other folks. But “unadjusted” is not raw and does have adjustments in it…)

Marble Bar

As another example, the data for “Marble Bar” Australia are graphed here:

http://www.unur.com/climate/ghcn-v2/501/94315.html

This is an interesting graph as you can see the “adjusted” values peaking out (bits of red) from behind the “unadjusted” values. At the far right, you can also see a tiny “lift” to the bottom of the data range as we reach recent years.

I recommend a visit. While I don’t know what else is on the site, the quality looks to be quite good. I expect to spend a while “browsing” around. ;-)

Oh, and the source code to do this yourself is also published:

http://www.unur.com/comp/ghcn-v2/

The site is run by Sinan Ünür and has some focus on Turkish culture and related that also looks like it might well be interesting. Given the quality work done by the Turkish Meteorologists who found cooling in Turkey (as we saw in this paper: http://www3.interscience.wiley.com/journal/114078036/abstract from comments in the Turkey Posting) I think I’m gaining a deeper respect for Turkey and the quality of the work being done there.

Advertisements

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 NCDC - GHCN Issues and tagged , . Bookmark the permalink.

24 Responses to Unur.com has a wonderful visualization of the data

  1. Michael M says:

    Chiefio,

    Thanks for the work you do on your site; I’m much indebted to it and to you, and I eagerly await each new update.

    I don’t comment much on these blogs because, frankly, I know I’m out of my depth – but I had one suggestion for a future project to add to your ever growing list. Simply for the visual impact it would have, I’d love to see a graph (or preferably an anamoly map) of global temps from oh, say 1890 to 1990, at the onset of this strange procedural change that you’re documenting.

    As you’ve mentioned many times, the minimal rise in temperature would be a powerful blow against the idea of “steadily rising CO2” , lo these 100 years. (Especially for laypeople like myself.)

    With much gratitude,

    Michael

    REPLY: [ Thanks for the compliment, and you are most welcome. Per a global scale dT/dt report: It would not be very hard to do (and I may do it just for you ;-) but it has a ‘flaw’ that makes it a bit less ‘powerful’ a blow than you might expect. That is the lack of any “Gridding” or “boxing”. Due to the way dT/dt works, it’s best for looking at specific sub-sets. If I use it on, oh, France; you can think of it as using a “Grid / box” of France sized proportions (and that’s a reasonable size). Further, as France is down to the dozen or so thermometer range, you are only getting a small set of “splice artifacts”.

    On a global scale, the lack of area scaling means that the extraordinarily large number of thermometers in the USA (and to some extent, in Europe) would mean that you get a graph dramatically dominated by them (and that would look very much like them). Furthermore, you are now taking on the order of 7000+ splices. This means a very high percentage of “splice artifacts”.

    Basically, the “quality of the result” declines as the geography covered approaches the whole globe. It is best quality at the single station level (or the “few” station level – especially if they are long lived stations with few splice artifacts).

    But as long as those limitations are understood, it’s a reasonable result… -E.M.Smith ]

  2. oldtimer says:

    An interesting find.

    I am unsure whether this link to unur`s site will provide what Michael M is looking for. It is an animated anomaly map prepared by Jones in 1990.
    http://www.unur.com/climate/jones-1990-1851-1990.html

    I have not played it all the way through but it appears to change by month by year according to the counter in the bottom RH corner of the screen. I am also unclear what the reference period is against which the anomaly is measured. Could it be 1961-90?

    REPLY: [ The CRUt crew use 1961-90 while the GISS guys use 1951-80. Jones is a CRUt. -E.M.Smith]

  3. Keith Hill says:

    Thanks for that great unur.com link. First cab off the rank for me is to check all Australian Stations.

    EM. If you want a good laugh, next time you look at the site check his Outbox for “glossary of research terms”. It’s better than the IPCC Fourth Assessment Report !!

  4. epicsystems says:

    Dear Sir,

    I have the pleasure to brief on our Data Visualization software
    “Trend Compass”.

    TC is a new concept in viewing statistics and trends in an animated
    way by displaying in one chart 5 axis (X, Y, Time, Bubble size &
    Bubble color) instead of just the traditional X and Y axis. It could
    be used in analysis, research, presentation etc. In the banking
    sector, we have Deutsche Bank New York as our client.

    Link on Chile’s Earthquake (27/02/2010):

    http://www.epicsyst.com/test/v2/EarthQuakeinChile/

    This a link on weather data :

    http://www.epicsyst.com/test/v2/aims/

    This is a bank link to compare Deposits, Withdrawals and numbers of
    Customers for different branches over time ( all in 1 Chart) :

    http://www.epicsyst.com/test/v2/bank-trx/

    Misc Examples :

    http://www.epicsyst.com/test/v2/airline/
    http://www.epicsyst.com/test/v2/stockmarket1/
    http://www.epicsyst.com/test/v2/tax/
    http://www.epicsyst.com/test/v2/football/
    http://www.epicsyst.com/test/v2/swinefludaily/
    http://www.epicsyst.com/test/v2/flu/
    http://www.epicsyst.com/test/v2/babyboomers/
    http://www.epicsyst.com/test/v2/bank-trx/
    http://www.epicsyst.com/test/v2/advertising/

    This is a project we did with Princeton University on US unemployment :
    http://www.epicsyst.com/main3.swf

    A 3 minutes video presentation of above by Professor Alan Krueger
    Bendheim Professor of Economics and Public Affairs at Princeton
    University and currently Chief Economist at the US Treasury using
    Trend Compass :
    http://epicsyst.com/trendcompass/princeton.aspx?home=1

    Latest financial links on the Central Bank of Egypt:

    http://www.epicsyst.com/trendcompass/samples/Aggregate-balance-sheet/
    http://www.epicsyst.com/trendcompass/samples/balance-sheet
    http://www.epicsyst.com/trendcompass/samples/banks-deposits-by-maturity/
    http://www.epicsyst.com/trendcompass/samples/egyptian-banks/
    http://www.epicsyst.com/trendcompass/samples/currency-by-denomination/

    I hope you could evaluate it and give me your comments. So many ideas
    are there.

    You can download a trial version. It has a feature to export
    EXE,PPS,HTML and AVI files. The most impressive is the AVI since you
    can record Audio/Video for the charts you create.

    http://epicsyst.com/trendcompass/FreeVersion/TrendCompassv1.2_DotNet.zip

    All the best.

    Epic Systems
    http://www.epicsyst.com

    REPLY: [ Yeah, it looked like SPAM and is blatant self promotion… but… it’s honest about it’s purpose (to push the product) and it is pertinent to the folks who read this kind of analysis thread. Basically, it’s nice polite and honest SPAM. And I can’t really find fault with someone who is being honest and polite and potentially providing a valuable pointer to a pertinent product. -E.M.Smith ]

  5. pyromancer76 says:

    Thanks for your explanation to Michael M re problems creating a reasonable graph of global temps from 1890-1990.

    My two-cents: I want a long prison term for anyone who permanently tampers with actual “raw data”. Another file of adjusted data with the info on the adjustment method is fine — but change the raw data…..actually, I would like a firing squad.

    REPLY: [ You will probably want to look at the reply per ‘getting a rise out of a step’ on the “QA thread” too. It’s a more detailed look at the issues. Per “raw”: One of the things I’ve come to appreciate out of this whole process is just how corroded the ethos of Science has become. In my high school if you had ANY erasure on a lab book (or worse, a whole page missing) you got an “F” on that assignment. PERIOD. One poor farm town girl just took a blank page out to write a note to a friend (note passing about boys, clothes, and parties being common) and became “a lesson to the rest of us”… The whole point was that ALL data even the MISTAKES was to be preserved “where is as is” FOREVER. If you had a correction, you NOTED IT. It was allowed to put a single THIN strikeout line through something if you were part way through it and realized it was a screwup… but it STAYED.

    Now we have folks fabricating data values, erasing the original recorded values, replacing them with the fabrications, and calling it “QA”. Just Nuts.

    I can only hope that they keep the original “as reported” somewhere. If they don’t, then it’s time for an “F” and a fraud charge. If they do, then it’s only a D and a reprimand…
    -E.M.Smith ]

  6. Keith Hill says:

    EM. Have a look at our old friend Darwin on the unur.com graph and see the big cooling adjustment from 1882 to about the mid-1950’s. Makes the next 50 + years seem warmer to try and give an overall warming trend.
    Which agency was it that held the right thumb on the end of the graph and dragged the early years down?

    REPLY: [ WOW! That graph:
    http://www.unur.com/climate/ghcn-v2/501/94120.html
    is just amazing. “Thumb on the scale”? Heck, it looks like a hand and most of an elbow!

    I can feel a new project coming on… For each of the dT/dt country graphs that found lots of “warming”, find the locations with the bulk of that warming and check their graphs at unur.com for ‘unusual’ adjustments. Could be fun. Now I just need to find the time… Maybe I can restrain myself and just do a few samples ;-) Like, oh, Kenya that looked like a key country in African “warming” and Fiji that was central to the Pacific Islands warming… Probably easier to just write some code to search the delta between the adjusted and the non and rank them… You know, there are times I wish I was a grad student somewhere. There is just SO much material here to work with for a thesis or ten… -E.M.Smith ]

  7. Keith Hill says:

    To assist anyone wanting to have a look, Darwin’s WMO number is 94120

    REPLY: [ Thanks for the number. Made it easy to find the link (that I stuck in the above comment). FWIW, anyone wanting to put up a similar link can just paste it into a comment. WordPress takes care of all the html tag creation. So just copy the line from your browser bar and paste in the comment. That’s all it takes. ( Oh, and make sure there is white space on each side. So Link:{whatever} won’t work as the “Link:” part confuses wordpress,.. but “Link: {whatever]” will work as the white space after the colon lets wordpress pick out the link… ) -E.M.Smith ]

  8. Ruhroh says:

    Hey, that epic link on the ‘Chilean’ earthquake is hypnotic.

    I snapped out of the trance when my neck got a crink from holding my head rotated by 90 degrees;
    I really prefer to see longitude going sideways and latte tood going up/down. Wish they had a background of the world map.

    But it seems germanium to your porpoises of hi lye ding the related-ness of the pingpong earthquake scenario.
    How did they know?
    Must be math guys, to plot their headline story with the axes reversed…

    It makes me think; how compressible is rock?
    A bit hard for me to to think of it as a (stiff!) spring, but someone must have done the math on this one.

    Or, would it tend to bulge as it is not in a good configuration to just absorb compression without deformation?

    Anyway, now that you’ve ‘gone all graphic onus’, you might have some fun with this.

    BTW, I’d encourage you to forward a copy of the RainboHair graphs to Tufte when you get a round tuit. For his gallery…

    OK, returning to primary obssession of the day job…
    RR

    REPLY: [ Um.. who is “Tufte”? and how does one find his “gallery”? FWIW, IIRC my geology classes, in the short time term and at short distances rock is a stiff spring, in the long term and over long distances it is flowing putty. Somewhere in between it can suffer brittle fracture. The boundaries both in time and space between these transitions of behaviour are unclear and subject to change with temperature, chemical composition, physical form (crystallization / fracture / deformation …), moisture and depth. Helpful, no? So when a quake happens, the rock at the epicenter is having brittle failure, the surface that slips can have plastic deformation and melt artifacts from the friction heat, and both the P and S wave reach you via the rocks acting as a spring and wobbling… and there can be zones of crushing and fracture at any depth. Don’t worry, it’s all settled science and they have computer models for it ;-)
    – E.M.Smith]

  9. GregO says:

    Chiefio,

    Excellent site suggestion – and I tip my hat to the Turkish scientists and researchers for fine work and transparency.

    I was relieved once I discovered for myself that all this CAGW was stuff and nonsense but my relief has since been slowly replaced by what can only be described as a sense of nauseating unease as I see how MSM is continuing to back the AGW theory – just look at the latest edition of Scientific American (Eaarth…come on…quoting the Club of Rome….wow) and Newsweek with its article championing the latest foolishness out of the EPA on CO2. Disturbing and insulting to the intelligence as it is, this is what millions of Americans are reading.

    What is the truth? Look to the thermometers. We’re talking about Global Warming (delta T). We need to measure said warming, right? What does the instrument record tell us?

    Now it’s back to that sense of “nauseating unease” as we see from your work and many others that the very temperature record itself has been manipulated by government sponsored scientists and investigators in order to support CAWG. What gives?

  10. E.M.Smith says:

    For Marble Bar:

    http://www.unur.com/climate/ghcn-v2/501/94315.html

    We can see the “thinning” of the data towards the end and then there is a fairly “dramatic” rise of the bottoms. The tops can be seen fairly flat.

    It’s pretty clear that the “rise” at Marble bar is not from getting hotter, but from the cold bits being “less cold”. It’s that same QA artifact we’ve seen before, IMHO.

    I think I’ll do a “monthly temperatures rainbow graph” for Marble Bar and see what months make the rise (and which ones don’t), though I think it’s pretty clear it will be the winter months rising … The “Anomaly Hair Graph” has only one month with a significant “cold anomaly” after about 1980.

    I wonder if the Australians “tried out” that “QA” process ahead of other folks?….

  11. Sinan Unur says:

    Thank you very much for the kind words. I should note that I did the work in 2007 so the data a snapshot of what things looked like at that point.

    As we all know, (as covered by Steve McIntyre, Anthony Watts and many others), adjustment procedures used by various data sources have a tendency to change the past.

    The terms ‘raw’ and ‘adjusted’ refer to the usage in the documentation provided in the data sources. When plotting the series as they were, I decided not to inject any judgment on whether the term ‘raw’ was appropriate. I am not sure if one can find real raw temperature data in the sense anyone outside of the climate science world would use the term.

    Regardless, I believe a graph like http://www.unur.com/climate/ghcn-v2/223/41020.html really puts into perspective the scarcity of data in various parts of the world.

    You and your readers might also be interested in the animations of various temperature data sets I put together. They are available from http://www.unur.com/climate/

    In addition, I list the top 10 contributors to GISS data at http://www.unur.com/sinan/outbox/070816-us-is-only-two-percent.html Almost half the station identifiers in the GISS are contributed by those 10 countries (about 2% come from Turkey and 30% come from the USA, China and Australia).

    Not only does the data set released by the UK Met Office also display the same tendency to rely on stations in a handful of countries, there are several spelling errors that make me think they were not too careful with the data: http://www.unur.com/climate/cru-station-records/

    Once again, thank you for your kind words and good luck.

    — Sinan

    REPLY: [ Nice to see you noticed! Yes, that ‘few sources’ issue is one that’s been nagging at me. Ought to be highlighted, but not enough time for it yet. Ah well… Per the 2007 time stamp: Not much has changed since then. It’s still a valid snapshot. Per the “raw” usage: A reasonable decision. Though I’m still looking for “really raw” … I’ll be spending a while cruising around your site. I hope you will find things here to entertain your interests as well. Welcome and “hello friend!” – E.M.Smith ]

  12. Michael M says:

    Re: oldtimer @ April 13th

    I saw that link, and I’m not sure that it is what I’m looking for (frankly, I didn’t have the patience to sit through the whole 7 min video!)

    An example of what I’m getting at was posted on WUWT; you may have to scroll down to see where Anthony compares the global anomaly at 1200km and 250km smoothing.

    http://wattsupwiththat.com/2010/03/23/why-joe-bastardi-see-red-a-look-at-sea-ice-and-gistemp-and-starting-choices/#more-17626

    The anomaly goes down, and of course the ‘skeptic’ would say that the 1200km smoothing incorrectly raises the temperature, filling in places it has no business filling in.

    Obviously Chiefio’s work (eg: Kenya ‘spreading the heat around’) would support this theory that GISS is incorrectly filling areas in, and would, I assume lower the temperature as much or more than the 250km chart on Anthony’s above post.

    So, if Chiefio’s right, and the evidence looks compelling, than it would seem plausible to use his data from the dT method and make a (somewhat coarse) anomaly map. Of course I recognize the hurdles that he mentioned in his reply to me; still, one can dream :)

    REPLY[ There is a very important difference between making a “global graph” of all the world in one temperature line (as I’ve made country graphs) and making a “global anomaly MAP” with various boxes filled in with an anomaly from the dT/dt method. Doing the latter has vastly lower issues with “splicing” and other artifacts.

    In fact, there are only 1200 or so currently reporting thermometers globally and GIStemp uses 8000 boxes, so on average there is less than one thermometer per box. Not a lot of splicing artifacts from ONE thermometer! Most of my caveats only apply to make The One Global Average not to making a finer division map…

    So, to that end, it would be modestly trivial to make a variation on dTdt that produced a net anomaly for various lat / long ranges. The Lat and Long are in the data and basically all I’d need to do is make “input” sets for various Lat / Long combinations, then run the dT/dt code AS IS on those sets. (The first step at present is a ‘select for country code/ stationID FOO” and simply changing that to “select for LAT range and LONG range” would produce a “box of thermometers” then run the dT/dt code and “there you go”! I’d probably use fewer boxes than GIStemp just due to the complete silliness of filling 8000 boxes from 1000 thermometers, but it would not be too hard to create the basic data for, oh, 500 boxes. (about 3 x the work to make all the “by country” data). Doing it as ‘equal area’ instead of ‘equal LAT LONG range’ would take a bit more work as I’d have to cook up the LAT and LONG limits to use.

    So if it’s a “by box map” that’s much more reasonable to make… (though someone else would need to do the “turn anomalies into map” step. I’m not ready for that level of graphics work (yet!)… -E.M.Smith ]

  13. dougie says:

    Hi E.M.

    probably you are aware, but in case not

    http://climateaudit.org/2010/04/12/the-wisdom-of-solomon/

    your efforts are being recognised. power to your quest.

  14. oldtimer says:

    re Tufte above

    Edward R Tufte originally made his name with a book called The Visual Display of Quantitative Information dating from 1983. Envisioning Information Narratives of Time and Space appeared 1990. Fascinating stuff with many examples from many different environments from economics to railway timetables to science.

    You can find his books via Amazon. It would be better that you did not get a copy as it would distract you from your task.

    REPLY: [ You seem to have discovered EXACTLY how to get me running after something really fast… “It’s very interesting and fascinating stuff… but you ought not to look a it.” Rather like “Yes Mr. Investigator, you don’t need to look in the crypt behind the garage.” It’s a guaranteed trip to the crypt… ;-)

    FWIW one of my favorite example of “visual display” is a side by side chart of “Sales This Year”. One showing sales per month, declining. The other showing rising sales; as “cumulative sales to date” by month… Same data, different story 8-}
    -E.M.Smith ]

  15. Ruhroh says:

    Thermometers per box as a function of box size?

    Let’s ignore the equal-area issue briefly.

    What system of boxes doesn’t have primarily
    empty and multi-thermometer boxes to deal with?

    Anyway, as I pop in briefly to check out Cheifio’s outbox,
    I find myself wishing for that map (or mappable list) of the influential thermometers that have the most disproportionate impact on OGA as practiced by the mainstream folks.

    Oops, oork calling, must go…
    RR

    REPLY: [ The globe is divided into thermometer poor regions (largely the oceans, Antarctica, and ‘third world’ countries – though with odd exceptions) and thermometer rich regions ( USA, Europe, Japan, Turkey ) so by definition we are mostly measuring that “thermometer rich” group. The “boxing” system will always have to deal with the fact that there simply isn’t any data for substantially all of The Southern Ocean, for example, and with the fact that Canada is reported almost entirely from the US border, and with the fact that Africa is reported from The British Empire, then often ends with independence…

    So, IMHO, it’s not possible to “fix the data” via grids and boxes. You can help the presentation a bit, but not a lot. And no system of grids or boxing can get around that. The best I could dream up was “by country” as a form of “area proportional” fix. Though even there I could see, for example, grouping the smaller countries into regional groups and breaking the land giants (Russia, Canada, USA, China, Australia, Brazil) into smaller regions. I just decided not to do so in the first pass. (What? You thought I was done? ;-) At present the “political boundary boxes” of “by country” can act as a baseline test for any form of geographic boxing. Does the alternative hide more than it reveals? Or reveal more than it hides?

    Perhaps it’s that it is early morning and I’ve been awakened by the family (against my will…) and have barely smelled the coffee pot… but what is an “OGA” ? Old Global Anomaly? Oddly Growing Anomaly? Ohmy God Anthropogenicity? Enquiring minds want more coffee…

    And when you are done communing with Oork, tell ’em Mork and Mindy are still waiting for a visit ;-)

    But yeah, now that we have all the “by country” graphs showing who is hockey sticked and who is not (flat, truncated, or dropping) a logical next step is to go “inside those boxes” and find out what makes each one tick. What individual instruments “do the deed” and is it done by “bottom thinning”, by “splice artifact”, by “Cherry Flavored Thermometers”, or simply by “Jet Age Jetsons” thermometers on a rocket ride… Much as Boballab did for Costa Rica a few posts back. But that is going to be at least 10 x the work of the original “by country” graphs (and that took me a month… pushing…) so we’re looking at a “staff-year” to “Get ‘er Done!” and if it’s just me, that’s “an issue”…

    So I’m going to “sample” instead of “exhausting”, er, make that “exhausted”, er, “COMPLETE”…

    I was thinking of starting with France (since they have been making anti-AGW noises lately, the least I can do is offer some ammo to them… and point them at King George, or is it Prince Jones these days? ;-) (Hey, what’s a couple of hundred years between friends?) WIth only a few surviving thermometers it’s not that hard to make exhaustive sets of graphs for things like Rainbow Temps and “by instrument and Duplicate Number Anomalies” and show who gets spliced to whom and which DupNums get their bottoms feathered…

    But we’ll see…. That Unur chart of Darwin has me thinking Australia would be a rich field to plough too… (But then there was that Quake in China and I ought to put up something about it too, as it continues the clockwise rotation of quakes around the Pacific Plate and that means next on the list is North America… and I’m there, er, here…)

    Oh, bag it all.

    I’m going to go make more coffee first… and wake up some more…

    -E.M.Smith ]

  16. oldtimer says:

    Your story about the sales trend reminds me of the occasion when a colleague, sliding his ruler down the page to read the data below the chart, solemnly intoned “1964, 1965, 1966, 1967! Aha, a rising trend”

    It was, as you can see, was a long time.

  17. Sinan Unur says:

    @E.M. Smith: It was hard to miss the spike in traffic ;-) You are very popular (and deservedly so).

    A couple of years ago, in my quest to find raw data, I was able to locate scanned PDFs of hand-written temperature measurements from Istanbul, Turkey. I never downloaded them (I should have) but I did spot check the values for some months from 1930s and they matched the GHCN data files. How I wish I had at least bookmarked them if not saved all the files. *Sigh*

    There are even older hand-written records in the Meteorological Musem in Turkey. See, for example, the last page of http://www.dmi.gov.tr/files/kurumsal/ekitap/4mevsim2/s0102.pdf Unfortunately, I cannot read Arabic script so I can’t tell what’s on this page.

  18. vjones says:

    What a flurry of interesting comments. You reminded me I had bookmarked Sinan’s site back in January, but forgotten about it (very useful Sinan – thank you) – even when I was writing the post on Turkey.

    I’ll also be interested to look up the Tufte books (at the risk of getting immersed in them) as visualising data is very important.

    You all might also be interested in the Climate Data Visualiser at http://www.appinsys.com/globalwarming/climate.aspx which will draw the annual mean type graphs (and the waaayyy tooo muuch info at: http://www.appinsys.com/GlobalWarming/)

  19. E.M.Smith says:

    @Sinan: Your work is what’s driving the traffic. All I can do is recognize it…

    @Vjones: Yeah, now that all those graphs are out of the way ;-)

  20. vjones says:

    @Ruroh (last comment) re thermometers in boxes.

    E.M., it strikes me that there are two types of empty boxes: those that do not have “their own” thermometers and those that are completely empty because they are so far away from any station (i.e >250km and >1200km depending on which level of gridding is used).

    Let’s ignore the completely empty ones for the moment…..

    The boxes with a one or a few thermometers in are interesting. What does a box temperature look like with just the stations in it vs what does it look like when it is homogenised with additional stations surrounding the box?

    This sounds tough no?

    Has anyone here been following RomanM’s R-Code for combining stations and producing anomalies:
    http://statpad.wordpress.com/2010/03/13/faster-version-for-combining-series/
    http://statpad.wordpress.com/2010/03/18/anomaly-regression-%E2%80%93-do-it-right/
    And what Jeff Id has done with it?
    http://noconsensus.wordpress.com/2010/03/24/thermal-hammer

    This does not have to use the whole data set – it would work at box level, except distance weightings would be required. Can anyone else toss in a few ideas?

  21. kenskingdom says:

    Gday Chiefio

    Good stuff! I have set myself the (ambitious) goal of comparing Australia’s “high quality” (i.e. adjusted) data sites with the apparently raw data from the BOM website.. then later comparing with GHCN, GISS etc. I had started on GISS adjustments because some of them are pretty spectacular but recent climate statements from BOM and CSIRO are just begging to be checked. One site I will be checking more closely is Roma (043091).

    Cheers
    Ken S

  22. Geoff Sherrington says:

    on April 13, 2010 at 7:11 am Keith Hill wrote about Darwin airport and Sinan Unir re http://www.unur.com/climate/ghcn-v2/501/94120.html

    There is much more than meets the eye with this site. Various people have written a lot about it on WUWT, CA and tAV (apologies if I have missed any). Modestly, I’m one of the writers. For some background to help context, try searching Geoff Sherrington Darwin on these sites and that will lead you to others. This is not self-promotion, it’s just a quick way to better understanding and links. (For a start, the airport station was not used before WWII).

    It is very difficult to get truly raw temperature data about Australia. The graph that is mentioned probably has been adjusted before GISS or whomever saw the data.

    Indeed, the same is becoming evident for most of the world.

    This is in no way meant to criticise those who contributed to the post. The code will be equally useful when the true raw data become public and there are inexorable moves in this direction, thanks to blogs.

  23. Keith Hill says:

    Hi Geoff Sherrington. Welcome to EM’s informative site (if you haven’t been here before). In my 77th year I think I’ve learnt more here and at other helpful sites, than I ever did about science and other matters in years of schooling, but I’m still very much a novice. I have read and enjoyed many of your posts round the blogosphere and there certainly is “much more than meets the eye” to Darwin.

    To give other posters an idea of the history, the following is an excerpt from the site of the late John L Daly. The first part was a contribution by Darwin resident Ken Parish and John Daly makes a very pertinent comment at the end.
    (I love the bit about the postmaster either forgetting or being too busy to move the unscreened instrument out of the sun !) quote:-

    “The main temperature station moved to the radar station at the newly built Darwin airport in January 1941. The temperature station had previously been at the Darwin Post Office in the middle of the CBD, on the cliff above the port. Thus, there is a likely factor of removal of a slight urban heat island effect from 1941 onwards. However, the main factor appears to be a change in screening. The new station located at Darwin airport from January 1941 used a standard Stevenson screen. However, the previous station at Darwin PO did not have a Stevenson screen. Instead, the instrument was mounted on a horizontal enclosure without a back or sides. The postmaster had to move it during the day so that the direct tropical sun didn’t strike it!
    Obviously, if he forgot or was too busy, the temperature readings were a hell of a lot hotter than it really was! I am sure that this factor accounts for almost the whole of the observed sudden cooling in 1939-41.
    The record after 1941 is accurate, but the record before then has a significant warming bias.” (Ken Parish)

    “The whole point of this investigation of just one station is that although Darwin shows an overall cooling due to that site change, this faulty record is the one used by GISS and CRU in their compilation of global mean temperature. More importantly, most of the stations used by them will have similar local faults and anomalies, rendering any averaging of them problematical at best. The best statistical number-crunching cannot eliminate these errors.” (John Daly)

    It seems to me nothing much has changed in the last ten years !
    Link: http://www.john-daly.com/darwin.htm

  24. Keith Hill says:

    EM. Thanks for the “how to link” tip It worked and I’m still learning !

    REPLY: [ You are quite welcome. I spent a couple of weeks (about 3 years ago) trying different fancy things to “tag” links right for wordpress… then someone pointed out to just paste it in… Sometimes the simple things are the hardest to accept. ;-) -E.M.Smith ]

Comments are closed.