JoNova

A science presenter, writer, speaker & former TV host; author of The Skeptic's Handbook (over 200,000 copies distributed & available in 15 languages).


Handbooks

The Skeptics Handbook

Think it has been debunked? See here.

The Skeptics Handbook II

Climate Money Paper


Advertising

micropace


GoldNerds

The nerds have the numbers on precious metals investments on the ASX



Archives

BIG NEWS IX: The Solar Model!

Dr David Evans, 8 July 2014, David Evans’ Notch-Delay Solar Theory and Model Home

At the introduction to this series of blog posts, we said we’d release the spreadsheet containing all the data, model, and calculations. All in one file for Microsoft Excel. Thanks for your patience.

The model, data, code and calculations are here: Climate.xlsm (20Mb).

Containing 44 datasets, 33 sheets, 90+graphs, and 15,000 lines of code

New Here? See this summary of posts. Evans looked at TSI (total solar irradiance) and Earths temperature, and discovered a mysterious notch filter. That implies some kind of solar effect occurs with an 11 year delay – or one solar cycle after the TSI. He built a model. See the hindcasts, and the prediction of imminent cooling. See the replies to critics.- Jo

(Click to download the Climate.xlsm file. 20Mb)

Why Excel?

I chose to do all the work for this project, right from the beginning, in a single Microsoft Excel spreadsheet for pc. It’s not the fanciest or the fastest, but an Excel spreadsheet is the most ubiquitous, and one of the friendliest programming environments as well. It runs on most computers—any Windows computer with Excel 2007 or later, and possibly on Macs with Excel 2011 or later (in principle it should work, but could someone who tries it let me know if there is anything that definitely does not work on Mac please?)

The models use VBA code, the BASIC programming language that is part of Microsoft Office. There are buttons on the sheets to make models run and so on. You can inspect and run or step through the code; it is all totally open.

The main, long discussion paper is still to come. There is more to this series of blog posts. We don’t want to preempt what is coming, and it’s useful to keep the discussion focused.

Some random screenshots for those who want to oogle without the 20Mb download. (No, it doesn’t begin to capture the sea of data.)

(Click to enlarge)

(Click to enlarge)

(Click to enlarge)

(Click to enlarge)

(Click to enlarge)

(Click to enlarge)

The optimization process by which we found the range of parameter values for the notch-delay solar model, and their most likely values, is complicated by the presence of many local minima. It is lengthy and was guided at hand at some stages, to trim the burgeoning number of possibilities in sensible ways. So at this stage we are also releasing excerpts from the main paper that define and describe the model, the total climate model, and the finding of the solar model parameters.

The spreadsheet was written for doing research—it is not a production version intended for consumers. It assumes the user knows generally what is going on. There are some descriptions and help, mainly in text boxes and comments (the red triangles in the upper right of cells—hold the mouse over the cell and the comment pops up).

People are welcome to make changes, but the only authorized copy of the spreadsheet will be at the download location above. Please send corrections or suggestions for changes to me at david.evans@sciencespeak.com, and I’ll try to incorporate them (no promises about timeliness though, because it has been extremely busy around here since starting the blog posts, with a mountain of comments and so on to read and respond to).

An Open Source Software Project?

If there is sufficient interest, the spreadsheet can be turned into an open source software project. Does anyone know if GitHub is suitable for large Excel files? Software is usually built as many small text files but we have one large non-text file, so it is not clear that version tracking and differencing will work meaningfully. Also, if we go open source there is an administrative overhead for everyone.

Please note that any results you generate using the spreadsheet are not endorsed by me, and if you make graphs other than what is obviously intended in the spreadsheet, please take the “sciencepeak.com” label off them. (Please provide links back to credit the original work, without any endorsement implied, see below.)

Journalists and data?

By the way, this spreadsheet started life as an aid for journalists. The idea was just to have all the main datasets, with instructions on how to download them, and some pretty graphs as examples—to show journalists and news producers how to get the data for themselves.

Soap box time: True authority in science comes from the data. That was the point of the Enlightenment: reason and empiricism triumphed over superstition and abuse by church and state. People learned to trust data ahead of any human authority, and science was born. Empirical data became the highest authority in physics, chemistry, and biology.

However in modern climate science the mainstream media and most politicians go to the government climate scientists as their highest authority, not to the data. Sure the climate scientists show them some data, but only their favorite data—and for a theory to be true it has to agree with all the data. With the Internet it is easy to bypass the authorities and go directly to the data itself, but the old media isn’t doing that yet. Can downloading a file of numbers, reading it into a spreadsheet, and graphing it really be too hard for the media? Come on media people, I’ll show you how.

I was preparing this spreadsheet for journalists in late 2012 when David Stockwell convinced me to look for a low pass filter in the empirical transfer function, assuming the climate was mainly driven by solar radiation (TSI). All the data was there, so I built the Fourier transforms and analysis software right into the same spreadsheet, and got distracted from the journalist project.

By the way, I couldn’t find the low pass filter we were expecting, and I twice gave up on the project because the data analysis was definitely not finding the transfer function of a low pass filter (perhaps the TSI assumption was way wrong?). Then one day, on the point of abandoning it again, I realized there was a notch instead, which was unexpected and interesting.

Sharing and using the model

  • Attribution — Please give appropriate credit to Dr David Evans, provide a link to the Notch-Delay Solar Project Home Page, and indicate if changes were made, with a brief description of the nature of all changes. You may do so in any reasonable manner, but not in any way that suggests David Evans or ScienceSpeak endorses you or your use.
  • ShareAlike — If you remix, transform, or build upon the material, you must distribute your contributions under the same license as the original.
  • Permission -- To use any part of this work in a for-profit project, please email david.evans@sciencespeak.com.
  • Cite as: Evans, David M.W. “The Notch-Delay Solar Theory”, sciencespeak.com/climate-nd-solar.html, 2014.

The world needs more independent science

This large work is offered freely, and has been entirely self-funded and funded through donations to Jo Nova’s blog. There are no government grants, and no UN programs paying for analysis like this. To all those who help make it possible we are more grateful that you can imagine for assistance and contributions of all shapes and sizes.

You can help support more  independent scientific research at the Paypal Tip Jar or by direct deposit or cheque.

……
 
…….
 
……
…….
…….

Click for details
*Due to strange Nanny-state rules, you can’t donate freely to me, but you can buy me quantities of $1 emergency chocolate. (No. I can’t believe it works this way either.)

Jo notes: I dislike group emails, and have not been able to thank every one personally, though I wish I could — I know most people would rather I write and research for public consumption instead. There are some direct depositors who deserve a mention: so thank you to Rodney, David M, Jules, Tom, Fay, Keith, Max, Bartels, Aaron, Phil, Fred D., W.E.B, Peter H, Peter K, Keith, Jim, Lawrie, M.J.B, Black Duck, W.B.C., Reed, I also owe one Peter C a letter in reply. Special thanks to MC, SB, BM, PF, GJ, PM, JD, DE, GB, VM, JP, TL, HC. There are too many good people to name.  We’re grateful to every one of you. I still owe many emails!

Skeptics are winning, against all the odds, but there is still a lot to do, and if you’ve enjoyed the latest revelations, we’d be delighted to get your help so we can do more.

————————————————————

About Dr David Evans:

David Evans, PhD, M.S. (E.E.), M.S. (Stats) [Stanford Uni], B.Eng, M.A., B.Sc. [Syd Uni] worked with Fourier analysis and signal processing, and trained with Professor Ronald Bracewell late of Stanford University. Evans main focus is researching mathematics (Fourier analysis, calculus, the number system, multivariable polynomials, and related topics). He consulted full-time for the Australian Greenhouse Office from 1999 to 2005, and part-time for the Department of Climate Change from 2008 to 2010, and was the lead modeler in developing FullCAM, the carbon accounting model that Australia uses for the Kyoto Protocol. Evans also produces the GoldNerds excel sheets that have become the industry standard for investors in precious metals on the ASX. He is available for contract work.

UPDATE: New version 1.15 posted Wednesday 2pm Perth time, hopefully fixes “clock” compilation problems on 64-bit Excel. Should now run on 32 bit and 64 bit Excel on Windows, and on Mac. Thanks to Mark Gutzwiller, DT Christensen,  and Don Jordan in the comments below for a fix.

VN:F [1.9.22_1171]
Rating: 9.5/10 (110 votes cast)
BIG NEWS IX: The Solar Model!, 9.5 out of 10 based on 110 ratings

Tiny Url for this post: http://tinyurl.com/ksrvrfu

620 comments to BIG NEWS IX: The Solar Model!

  • #

    Willis must be feeling pretty stupid right now….

    500

    • #
      Agnostic

      I can’t understand why he behaved like that. Surely he must have realized that this project was an order of magnitude larger than any of the analyses he has put forward with code on WUWT? Dr Evans even pointed out that some of his analyses and ideas had helped him along the way and acknowledged it. Utterly mystifying.

      I say again as I said to him on “that” thread; it’s perfectly ok to disagree, to be skeptical, to have objections to all manner of things, in fact we count on him to do that because in the response to those objections others can deepen their understanding. But firstly to presume that Dr Evans was some newbie with a wild theory, and then to insist that he was “hiding” his data, when it was explained right at the outset how it was going to be released was so perverse I am actually frowning in consternation as I write this…

      Anyway, enough of that. I watch with great interest as this model is picked over and kicked about. Hopefully a little more respectfully than we’ve seen.

      631

      • #
        Peter C

        Hopefully a little more respectfully than we’ve seen

        Do not bet on it. I am, expecting something like a Shark Feeding Frenzy.

        180

      • #
        TedM

        I think it’s cylcophobia.

        30

      • #
        PhilJourdan

        If this model bears out, it is a paradigm shift in climate modeling. In other words, skeptics and alarmists alike are going to be deflated as the answer goes back to basics – it is the sun stupid (not saying you are stupid, that is just an American colloquialism – kind of like saying DUH).

        Willis has a lot of time and effort invested in not only proving alarmists wrong, but showing a different alternative. As does Leif. But I think they will at least give it an honest shot. If it does not pan out, they will be the loudest critics.

        I suspect most of us are just reading and learning. We do not have a vested interest that can be destroyed by this model. But to a person with a lot of time and effort into other hypotheses, this is ego deflating.

        The good thing is that most non-alarmists will at least study this model and give it an honest shot. The alarmists never will. The former are seeking real answers, the latter are merely protecting their honey pot.

        391

      • #
        Raven

        I can’t understand why he behaved like that. Surely he must have realized that this project was an order of magnitude larger than any of the analyses he has put forward with code on WUWT?

        Reckon you just answered your own question there, Agnostic ;-)

        190

      • #
        Aaron Mead

        “Climate blogging is a bloodsport” TM (c) 2014 Aaron Mead. All rights reserved.

        20

      • #
        Ursus Augustus

        Before we really turn on ‘our own’ I suggest that Willis ( and Leif S) just don’t get what mechanism the notch is modelling so it might seem like some deus ex machina. Not that the alarmists aren’t in very much touch with their rich fantasy lives but what is really missing from the whole discourse, and from th science frankly, has been a robust explanation of the global cliumate mechanism.

        Clearly ( to me at least) the ‘CO2 is evil’ approach is just some carryover from the advent of the atomic bomb (ergo man’s self distruction is manifestly credible, drug addled 60′s through which every thing was evil man and several iterations of the counterculture which has got madder and madder like some Mandelbrot Set of loony cults even without the drugs.

        That said the CO2 concentration is almost trivially simple to evaluate compared to the effective planetary albedo, the small scale mechanisms of the evaporo transpiration system and various other factors that might reasonably be material to the true global circulation mechanism ( as distinct from the various GC models) and so the witch hunt could begin on the flimsiest evidence.

        The evidentiary weakness in the notch model is identifying the underlying mechanism it seems to me. Unless of course I have missed something which would not be the first time. That said, the proposition that the planet has some highly non linear and cumulating-discharging mechanaism associated with oscillating inputs is prima facie prospective and a bloody sight more prospective and rational than just declaring CO2 guity of planetocide.

        60

    • #

      Watts has been allowing Willis to post almost every day for a very long time now. By allowing publication he is in full support?

      141

      • #
        YetAnotherSceptic

        Seriously?? Anthony lets EVERYONE post (except maybe that Connolly clown). That does not imply support. He has been know to ask commentators to tone it down from time to time, and like all sites, moderation occurs.

        31

  • #
    scaper...

    we’d be delighted to get your help so we can do more.

    Time to dig deep, guys. This is important.

    240

  • #
    Another Ian

    Means it won’t work for us Open Office users then?

    50

    • #
      David Evans

      No, it won’t work on OpenOffice because OpenOffice doesn’t have macros (as VBA).

      110

      • #

        They load into LibreOffice (4.1); can be read; won’t run by default (security); but crash out the application when allowed to run. Well, the few I’ve tried all did.

        I’m too busy with paying work to debug that right now.

        Skin-flint skeptics don’t want to send about $1000 to Microsoft just to be able to exercise your spreadsheet as you intend, David. Imagine; it’d otherwise be $1000 that they could send your way instead!

        91

        • #

          Yep with office 97 still doing what i need ,thats me out for a while but would need to relearn basic as well to modify anything. Have not used that much since TI99 days.

          30

        • #

          Bernd, I totally understand about not wanting to feed the MS machine. Unfortunately, it’s really hard to beat Excel, and with 15,000 lines of VB it’s a very big job to get it to go cross platform. Obviously the model needs those macros. But all suggestions welcome.

          210

          • #
            the Griss

            I’m sure some enterprising programming guru will convert the whole thing across to an executable programmed in C++ or something

            …….. NOT !!!! :-)

            I know how long it takes to make 15000 lines of code that actually works.

            A huge achievement is all I can say. :-)

            Hint to those without MS Excel, but using Windows…. If you know a student at a registered uni or college, academic pricing is much lower. .

            70

            • #
              Peter C

              I liked the OLD GRISS! Can you change yourself back?

              90

              • #
                the Griss

                Gees, photo gravatars are only ok for some, hey !!! ;-)

                90

              • #
                Crakar24

                yeah i agree the old Griss had more punch anyway there is a site that may be able to give you excel other than a uni student, just think about pirates and how they used to anchor in bays oh and dont forget the dotcom on the end of course.

                Cheers

                30

              • #
                the Griss

                :-( Ok How do I get the old one back ?

                30

              • #
                TedM

                Yep the smiley Griss was the best one so far. The new Griss definitely retrograde.

                10

              • #
                TedM

                The old Griss was smarter too.

                10

              • #
                Roy Hogue

                Griss,

                If, as I think, that’s a cup of coffee then you’re in no trouble with me. Coffee makes the day.

                10

            • #
              MRW

              It’s $69.99/yr for Microsoft Office online (and downloadable): http://office.microsoft.com/en-us/

              10

            • #
              Roy Hogue

              And also may get you an older version than 2007 because schools may not take advantage of a later version being available.

              Worse than that, students may be eligible for the software for free but it turns out to have been watered down and isn’t fully functional.

              10

            • #
              Duster

              There’s an “old” saying in computerized statistics that “friends don’t let friends use Excel.” Less anti-MS types broadened that to “spreadsheets.” I rather suspect that Dr. Evans real motive for using Excel was to figuratively put a thumb in Phil Jones’ eye.

              Spreadsheets are appallingly easy to break. Worse, even if the creator does things properly and by the book, the routines employed to generate results can be a really problematic. The reason that Excel was singled out in the saying was that MS was extraordinarily bad at listening when users pointed out broken results. I quit using ANY spreadsheets when Excel returned a negative sample variance, which, is a problem.

              Anyone who even momentarily considers using a spreadsheet for this type of modeling should read at least

              http://homepages.ulb.ac.be/~gmelard/rech/gmelard_csda23.pdf

              Melard compares Excel 2003, 2007 and 2010, as well as Calc 3.0 and 3.3 (from Open Office). Melard concludes that 2010 is “much better” than its predecessors and in fact has reached an accuracy comparable to Calc 3.3 and in a couple instances is better. It still has problems with random number generation (potentially a serious problem in modeling), Chi square in some cases, and with fractional degrees of freedom which do occur in some statistical situations. Calc 3.3 has problems of its own.

              However, it is quite important for anyone running the model to know that they can’t expect the same results from Excel 2007 as from Excel 2010. Indeed Melard indicates that Excel 2007 is definitely broken in some places.

              I can view the sheets in Calc 4.1, but the macros are disabled and no pretty pictures are generated. So, OO users, fugeddabaddit. Further, replicating the model in R or any other more appropriate platform, will most definitely have the potential to seriously disagree with Excel output.

              So, happy as I am to see this, are you all really sure that the solar notch isn’t Excel using some pinking shears?

              10

          • #
            Deej

            Perhaps the online version of Excel may be of use to those without access to MS Office,
            or those wishing to collaborate together:

            Excel Online

            Hope this helps…

            Deej

            40

          • #
            freeHat

            Office 365 trial version, good for a month.

            http://office.microsoft.com/en-us/try/

            60

          • #
            Huub Bakker

            In writing software for our experimental robots and machine vision, we use and teach Visual Basic 6, which is now unsupported. It is, however an exceptional fast-prototyping language. None of the modern IDEs (Integrated Development Environments) have the same simplicity and flexibility so that you can forget the language and just tackle the problem at hand. Visual Basic for Applications (VBA) is essentially Visual Basic 6. I totally agree with David’s reasons for using it.

            Huub

            110

            • #
              Mark A

              Yup Huub,
              I’m still maintaining AND writing programmes in VB6 even on Win 8.1 64 bit.
              Apart from the occasional hitch after auto update of the OS it works like a charm, best MS product ever.

              PS, using more modern versions as well of course but some progs. are just too big to convert, easier to keep them in native VB6

              40

              • #
                PhilJourdan

                The irony is that VB is one of the very few programs that Microsoft actually created (well from the start of GW Basic). Most of the rest of their applications were bought, rebranded and then updated – but the underlying code still has some of the original company stuff in it.

                00

          • #
            MikeO

            Jo there is this http://www.microsoftstore.com/store/msaus/en_AU/pdp/Excel-2013-Non-Commercial/productID.260296200?tid=sr1Buq0Ff_dc&cid=5367&pcrid=27911913557&pkw=excel%20program%20download&pmt=b&WT.srch=1&WT.mc_id=pointitsem_Microsoft+AU_google_Excel&WT.term=excel%20program%20download&WT.campaign=Excel&WT.content=r1Buq0Ff&WT.source=google&WT.medium=cpc

            or look on ebay but be cautious Office 2007 there is about $200-$300. Personally I have 2007 and a top end computer (i7 4770 processor). My main display at 24″ is inadequate I wish I had something much bigger, what was it developed on?

            20

        • #
          tom0mason

          I have it running on my vintage ‘screwyouMS’ Linux PC (a Pentium III with ½G of memory !!).
          I’m running the later version of LibreOffice and loaded the .xlsm file into LibreOffice Calc Version: 4.2.5.2 nice as pie.
          Only took 27 minutes to load it and start the macros.
          No error messages so far. :)
          I’ll try it on the big Lenovo later.

          90

          • #
            Lu

            Tomason, I don’t know about your big Lenovo? But I got tired of “LooseDows” 7 & 8 NSA spying and bought a Lenovo B590 laptop that comes without OS except a DOS partition for only about 330 US$ here in Taipei. I installed Linux Mint 15 based on Ubuntu 12, without a hitch. Linux Mint is most user friendly and if I had time I would update to Mint 17 the latest version that uses a stable version of Ubuntu kernel. I have no regrets, except I miss ‘Cintanotes’ & ‘Screenhunter’ for which there are sadly no equivalents in Linux, but which I always need for text & graphic capture & research because of their ‘select-hot-key-instant-capture+save’. Therefore I still have to use XP on my Desktop, inspite of having a dual boot with Mint 15. Just saying, cheap!, plus the latest stable versions of Linux Mint 17 with Libre Office on a normal laptop! By the way not advertising, just helping the low budget people here. Love & Peace,
            Lu

            00

            • #
              Mr Green Genes

              I’m running Mint 16. I’ve tried running the spreadsheet using Libre Office 4.1.3.2 but it throws up syntax errors (frankly, as I expected). I’ve also tried it using Wine and Excel Viewer but it doesn’t run at all.

              I’m going to see if I can get it to do something but with all that code, I’m not hopeful. If someone better than me (not difficult!) can make it work under Linux, I’ll be highly delighted.

              00

        • #
          Dark Distardly

          Well, if you keep your eyes peeled on Ebay, you can pick up unused legit copies of Office 2007 still, for about £10 (I paid that for mine, including postage). I picked one up for Outlook (much cheaper than just buying Outlook), just for a sensible calendar and PIM for my phones (with my memory, which goes blank as soon as I go through a shop door, it works great for even shopping lists). I like 2007 as it is the last version with the sensible interface (sensible to me anyway).

          If you want the newest version legit, computer stalls at markets often do bulk license deals. For example at my local Sunday market I can get Windows 7 Pro, and Office 2013 Pro, for £45 the pair. The stallholder will run off copies of the software (which is legit to do) for a few quid if you don’t have access to them (they are free to download from MS) already.

          For my mobile devices I still prefer to use MS Works 9 (which does all I need while traveling, and there’s no massive install, plus is lightning fast on recent hardware), and I usually get those for £5.

          Personally having tried the alternatives, I don’t like them.

          00

      • #
        tom0mason

        Will cleanly load and the VBA macros in LibreOffice Calc Version: 4.2.5.2.
        But my machine is so slow (see below) I really can’t do much with it, my only indication that it is still running the VBA is the raised processor temperature.
        But it is loaded and it says the macro are loaded.

        (5 programs running processor at 96% memory 412Meg used/264 Megs Swap, and laughing at MS) :)

        70

        • #
          tom0mason

          Not so fast, after waiting the required time, lots of syntax errors :(

          Still can’t expect too much from this 14year old piece of technology.

          70

          • #
            Andrew McRae

            yep, seems like “ReDim Preserve” is VBA that’s not recognised by LibreOffice/OoO.
            The file is all fairly useless to us Linux users.

            As we had quite a bit of fuss over some of the basic theory, like whether a notch filter is even an appropriate abstraction for a predictive model, it might have been better if he’d gotten feedback on the basic theory before writing the 15,000 lines of VBA.

            yes, I know, there’s just no pleasing some people.

            00

      • #
        Rolf

        Open Office have the same extension as Libre office and is running VBA. But there is the same issue with ReDim. I guess a workaround is not that much work ? Takes a long time to load the first time when opening the excel file but once saved as ods it loads in twenty seconds on my notebook, with all code.

        10

        • #
          tom0mason

          ReDim is a problem here. I’m not a VBA programmer, in fact last time I used it was back in the hot days of the 1990s.
          I’m trying to get my head aound what ReDim ( http://msdn.microsoft.com/en-us/library/w8k3cys2.aspx ) actual does and how it fits into Dr.Evans’ code.
          It should act as “Optional. Modifier used to preserve the data in the existing array when you change the size of only the last dimension.”

          Saying that many thanks Dr. Evans for a great project and the free download. If only all modern science was so transparent.

          00

    • #

      Or LibreOffice (4.1)

      It’ll load but the VBA isn’t expected to work as intended by the authors.
      Data are visible and the “Visual Basic” can also be read.

      If you’re aiming to replicate, stritcly speaking you should only need the raw data and the “paper”. If you use the same programs, you’re exercising the same bugs. ;-)

      About a decade ago, I sent an email asking when MS Office would be ported to Linux. I was “surprised” by the lack of a response.

      70

    • #
      sophocles

      Hmm. It should be possible to port to several different languages. OpenOffice can use:
      OO-Basic (OpenOffice Basic, which is not VB nor anything like VB! It’s not a nice Basic to use, IIRC …this includes Libre Office to the best of my knowledge),
      Javascript
      Python
      and a couple of other ones I can’t remember (Java perhaps?) as its macro programming languages. Spoilt for choice! Python would make a good language to port to as there are some powerful analytical tools available in Python eg: numpy. Python is available for most platforms and is free.
      Python is possibly easier to program than VB. There are lots of ‘howto’ books out there. (eg: Learning Python 4th Ed The 5th edition costs. You can find more.
      Learning Python the Hardway is another. Search and ye shall find even more.)

      R is pretty ubiquitous in academia and elsewhere where serious statistical analysis is required, such as data mining. It would be a good port too. It too, is Internet sourced from the CRAN (Comprehensive R Archive Network). It is very powerful and used across the world. Look on mirrors.aarnet.edu.au for it.

      Hmm. wonder if I can suck the VB code out of the sheet … I don’t run any MS software at all so this could be a fun challenge!

      41

      • #
        sophocles

        To think is to Search. Google came back with this vb2py project. Prospects for fun are increasing by the minute!

        And there’s some activity with R as well, which I will have to look into more deeply, sometime.

        At 15000 lines, automated conversion is the way to go.

        40

        • #
          sophocles

          But, the first problem is to get the code out. MS encrypt all the macros and formulae when saving to file. Cute. This makes Excel spreadsheets non-portable, which is why Libreoffice and Openoffice crash after loading the file. I don’t know if there is any way of saving all the code in a non-encrypted text file. I wouldn’t be at all surprised if there isn’t.

          It’s a good reason to go with products which meet ISO standards for their file formats.

          00

          • #
            Deej

            Sophocles, if you disable macros, then hit ALT+F11 you’ll be able to view the code. Scroll down to the ‘Modules’ section in the left-hand pane and double-click on any module to view the code therein. The ‘ComplexArithmetic’ module will keep you occupied for a minute or two :-) )

            Deej

            00

            • #
              Deej

              Ooops, sorry – it seems I misunderstood your post; your requirement is to save the code to file, unencrypted…

              Deej

              00

          • #
            Deej

            Upon further investigation, simply exporting a module produces a module-name.bas file, which is readable as a text file – Notepad has no problem with it, but it does look better using Notepad++ ().

            How are your exported files encrypted?

            Deej

            00

      • #
        Gee Aye

        Yep R.

        20

      • #
      • #
        MikeO

        Have you ever attempted anything like this before? It would be classed as heroic and without an absolute understanding of a correct result always there will be the subtle bugs. If you go ahead I suggest C++ since even if you fail the knowledge will not be a total loss.

        30

        • #
          sophocles

          I hate C++
          I don’t mind C but the Write-Compile-Test cycle is far too slow. C++ is not much better.

          I used to work as a professional programmer, I’ve taught programming from first steps to more for a couple of decades and I still hack for myself.
          So, yes, I have a pretty good concept of the work required. If I attempted it just for myself and by myself it would take me far too long.

          However, David did suggest the possibility of setting up an Open Source Project on GitHub, which is an excellent idea.
          Such a project could:
          – maintain, correct (fix) and extend his VBA code. (Primary concern)
          – enable and direct teams for porting to Python and R
          – establish and create validation requirements and the like

          I suggested Python and R because:
          – they work on everything. VBA doesn’t.
          – they’re free to acquire and use,
          – their licences explicirly permit you to run them on as many machines as you require and to give them away to others
          – they’re open source so they will always be there. VBA is proprietary, the Intellectual Property (MS’s concept) of a corporation.
          – both are well supported for statistical analysis—it’s the raison d’etre for R.
          – both are very fast to develop in —it’s what makes VBA attractive too.
          – both are pretty fast to learn and are supported by a lot of free books.

          Corporations come and corporations go. If MS ever fell over, what then? Nobody expected Digital Equipment Corp (DEC) to die. It did. Nobody expected Sun Corporation to die. It did. In 1991-1993 IBM haemorrhaged 4.5 billion dollars per year. It very nearly died. So it can happen.
          Keeping it proprietary is dangerous for its long term health.

          I consider David’s idea of an Open Source Project to be an excellent one. I’ve laid out my thoughts on it above and my thoughts about why it should be ported to Open Source alternatives. I think I have a pretty good idea of the work involved in what I’m proposing. Interested?

          20

      • #
        Andrew McRae

        I had to consider a language choice a few months ago in a similar situation to this one.
        It also seemed to come down to a battle between R, Matlab (or Octave), and Python.

        I eliminated Matlab on the basis that the language is broken by design; the syntax used for referring to an element of a matrix is actually different depending on the dimensions of the matrix. Sounds like a source of error. As a programmer that just seems bizarre and the rest of the language seemed a bit weird compared to the Java, C#, VB, and javascript that I’m more used to using. If it weren’t for the language quirks, Octave would be a good choice as the free version of Matlab.

        I had misgivings about R because it was intended for statistical analysis of data, whereas a model is trying to produce data from a physical simulation. Maybe that’s a glib reason. More pertinent is that R also has inconsistent results from the same syntax depending on matrix dimensions, as well as other pitfalls and gotchas. Aside from that, R is probably the best choice if conforming to consensus appeals to you. :)

        That basically leaves python. It’s a sensible multi-paradigm language from a programmer’s perspective. Huge numbers of physics and machine-learning libraries for it. Plus it seems to be faster than R with large data sets, though that could vary between algorithms and native C libraries.

        If had to choose I’d pick Python. Mind you that may mean nothing as the R versus Python battle has all the makings of a classic computer industry holy war. The differences are more about style than substance.

        20

        • #

          I don’t know R, but adding matplotlib, scipy and numpy to python provides most/all of the numerical capabilities that one might need, in an optimised form, as well as allowing for the creation of nice graphics.

          Python rocks, and as it is a mulitparadigm language, just like C++, one can very easily incorporate C++ (and C) code as python modules.

          00

  • #

    I’m not surprised that a physical “delay” is more notch-like than of a low-pass nature. It takes time for the heat to flow before a significant temperature gradient exists and for the inertia (a vector quantity) of the moving masses to be overcome.

    Looking at e.g. the Nüsselt number, it has to exceed a certain value (~=1000, IIRC) before natural convection occurs as a result of the driving buoyancy change overcoming the retarding (weight and) viscosity. Without convection, heat transfer is bound by conduction and to a much lesser magnitude within our planet’s climate system; radiation. So stuff has to get warm enough to get moving.

    110

    • #
      Rod Stuart

      If you work for an enterprise that spends a bundle on MS products, many of them host the Microsoft HUP (Home User Program). Basically, if one is an employee eligible for HUP, the entire MS Office suite is available to you essentially for the cost of the postage. (And only for home use and non-commercial reasons of course). Worth investigating. I see no reason that MS would object in such a circumstance to someone using Excel for the worthy purpose of exploring David’s model. That is what HUP is for. (to sell MS Office).

      10

      • #
        MRW

        You don’t need to work for an enterprise. Go to office.microsoft.com. You can buy by the month or the year. The year is $69.99 (was $99 when I bought it six months ago.) Called Office 365 Personal in the US, I think.

        10

  • #
    Rick Bradford

    I deal quite a lot with software, and constantly find that a good but overlooked first step to eliminating trivial errors (typos, copy & paste errors, for example) is to port the code to another language.

    While trying to perform the translation between languages, odd code and especially errors, leap off the page.

    I suppose R would be the obvious choice, but rewriting in a functional language like Haskell could also be valuable for this kind of checking.

    40

  • #
    PeterPetrum

    I’m not even going to try and download it, as I would not have a clue what to do with it! But, golly, will I be watching this site to see what the disparate experts come up with. Thank you David and Jo, can’t wait to see the realisation set in. I’m off to Port Douglass at the weekend for a week of global warming. This Blue Mountains winter sucks!

    320

  • #

    Nir Shaviv on why it is not CO2 and why it is the sun:

    http://www.sciencebits.com/Munich-2012

    140

    • #

      and why it is the sun:

      At 23:15 he also explains why it is not just our sun but other suns!

      70

    • #
      Ron Cook

      See also:

      “The Neglected Sun” by Fritz Vahrenholt I think Plimmer also wrote a book on this subject.

      Ron Cok
      R-COO- k+

      70

      • #
        john karajas

        Actually it’s Ian Plimer and he has been emphasising the influence of the sun on climate for years.

        But if you want to go back further, to the early part of the 1900′s, try Milutin Milankovich. The Milankovich curves have long been known to science but have been conveniently forgotten by the Warmists.

        20

        • #
          Mattb

          Seriously John, and mods, this is rubbish. Milankovich cycles are absolutely NOT ignored by climate science. It’s a 26,000 year cycle though… kinda irrelevent to the timescales being considered whether by warmist cycles or notch filters.

          [I'm kinda busy right now Matt, why don't you type something up as a reply?] ED

          18

          • #
            Mattb

            my reply is already there. My point is that you’d do well to occasionally point out when people post a lot of rubbish on here.

            15

            • #
              the Griss

              when people post a lot of rubbish on here

              But we always point this out to you about your comments, Mattb !

              31

              • #
                Mattb

                indeed. It’s just you guys remain silent on the rubbish as long as you like the sentiment.

                03

              • #

                For a good look at what Mattb calls rubbish.
                Click here
                On this page rates and timings are given. Through out the thing the fact that we are at the to of the warming to cooling cycle peak is laid out. No wonder Mattb wants it rubbished.

                10

  • #

    Nir Confirms the peak at 2 years in David’s Bode plot and also the 11 year delay. It was written in 2007.

    http://motls.blogspot.com/2007/07/nir-shaviv-why-is-lockwood-and-frohlich.html

    First, the response to short term variations in the radiative forcings are damped. This explains why the temperature variations in sync with the 11-year solar cycle are small (but they are present at the level which one expects from the observed cloud cover variations… about 0.1°C). Second, there is a lag between the response and the forcing. Typically, one expects lags which depend on the time scale of the variations. The 11-year solar cycle gives rise to a 2-year lag in the 0.1°C observed temperature variations. Similarly, the response to the 20th century warming should be delayed by typically a decade.

    130

    • #

      Yes. A more complex model could run with multiple lags which are derived from integrations of solar and temperature data done over a number of timescales. I think this would smooth the predicted fall in temperature over a longer period. I think that’s what will happen as the ocean will release some of the excess ocean heat content built up over the higher than average period of solar activity 1934-2003 in the form of El Nino’s. That’s what happened at the end of the C19th as the Sun wound down to a low point in the first decade of the C20th.

      200

      • #

        May be interesting to also try to find two notch frequencies as explained by the distribution at the link below. I wonder how the input should be syncronised with the changing cycle length. This phase error could perhaps be a signal on its own. With a solar magnetic cycle of around 22 years i wonder if months and years should be ignored and instead use daily phase angle at the 22 year position. Like 2prT(days)with T zero reset to odd cycle minimum and 4prT(days) set to the following odd minimum etc. So that a long cycle pair is divided up into longer chunks of time than a short cycle pair.
        http://personal.inet.fi/tiede/tilmari/sunspots.html#tdist

        20

      • #
        gnomish

        possibly of interest to roger:
        it has been found that there is http://phys.org/news/2014-07-hotspot-powerful-cosmic-rays-physicists.html

        and perhaps pertinent to the ND model is http://phys.org/news/2014-07-sun-tsunami-voyager.html

        40

        • #

          gnomish
          Inigo Jones would have been jumping around with excitement about that hot spot find. In his book “My Nephelo Coccygia” he based much of his work on a 1 meter band radio hotspot said to be at the 18th hour of right acension. He mentions many other spots and noted the different types of cosmic radiation.
          From his method he first predicted the solar cycles then from that and old long weather records predicted daily weather for various locations around Australia.
          Here he is in 1950 with a ten year long prediction. I posted this stuff on Jennifer Marohasy’s blog. It looks like BobC has corrected the text for us since then.
          http://trove.nla.gov.au/ndp/del/article/97132477
          I checked to see how a few of the 1954 ones went.

          “heavy rains in January on the Darling downs”
          and also
          “In February the rains are general with heavy falls in the coastal regions north of the tropic and the Gulf and south west and southern parts of the Darling downs.”
          How did that coast north of the tropic bit go?
          http://www.abc.net.au/local/videos/2014/02/21/3949139.htm
          How did that southern Darling downs bit go?
          http://trove.nla.gov.au/ndp/del/article/47574367

          60

    • #
      the Griss

      But gees, doesn’t that vid tie in well with The Notch-Delay model. :-)

      100

      • #
        the Griss

        I should have waited until the end of the vid..

        Nir still predicts some warming. (which would be nice)

        Unfortunately, I suspect he is overestimating, and his 1C per CO2 doubling is still about 1C too high.

        61

        • #

          His 1C is without feedback. He has stated that in his blog post some years back when he expounded on his CO2 model. IIRC his net is on the order of .5C with feedback. Not too different with what others have concluded. He was in the lukewarm crowd (me too – until very recently). I have no idea where he stands today.

          50

    • #
      the Griss

      Does Nir know of this work by David, I assume he does.?

      50

  • #
    Popeye26

    Come on Leif, Willis, John Cook, Mann, Bradley, Hughes et al.

    WHERE ARE YOU ALL??

    Itching for your input and comments – COME ON!!!

    Cheers,

    120

    • #
      Gary

      Willis is at the 9th International Conference on Climate Change http://climateconference.heartland.org/ in Las Vegas. July 9th at 3:00 PM PDT — PANEL 19: Looking Ahead: Future Climates — Dr. Don Easterbrook, Willis Eschenbach, and Dr. Terrence Flower (Moderator: Marlo Lewis)

      51

    • #
    • #
      MRW

      Willis is commenting on WUWT because, as he wrote, the Alexa stats are higher there, so he can spread the word farther, and he can’t get the model to work. I can’t remember the raft of reasons. Further, he says there are too many parameters and quotes Fermi’s answer to the guy from Princeton about parameters: with four you can make an elephant, with five you can get the elephant to wiggle his trunk. He’s verklempt over 11 parameters. So Willis thinks the model as released is an “advertisement” for a model, not science. Something like that.

      30

  • #
    Peter C

    I am with you on that Peter.
    Unfortunately I cannot share the getaway in Port Douglas. Melbourne has been very cold the past fortnight.

    30

    • #
      Peter C

      That is Strange!
      I am sure that I pressed reply to PeterPetrum at #6. My reply somehow ended up here.


      :-( Could be that comment threading bug again? Sorry. – Jo

      20

  • #
    EternalOptimist

    I am a professional software developer. VBA is my tool of choice and I have gone through the code supplied. all I know so far , is that it was not written by a newbie or a cowboy. I will have a good look at it later

    and there is no sign of a ‘Harry Read me’

    250

  • #

    Plenty of people are downloading this right now. my connection is saying it’ll take an hour to get the 18.8meg .xlsx file. And I have to go to work…

    Big thanks to David for making his work freely available – back later.

    260

  • #
    keith from darwin

    For all those that don’t have a copy of Microsoft Office, you do not have to buy a copy to read an Excel file. Microsoft have released a series of “viewers” for “Word”, “Excel” and “PowerPoint”. These programs allow you to open and view Word, Excel or PowerPoint files (a different program for each), although you cannot edit them or create new files. I expect that the Excel viewer will be sufficient for most to at least run Dr. Evan’s model. The Excel viewer can be downloaded from the following link. After downloading, double click on the file to install. Hope this helps.

    80

    • #
      keith from darwin

      P.S. These “viewers” are genuine Microsoft programs and they are free.

      60

      • #
        gary turner

        But do they run on Linux? If not, they’re not worth much to people running a modern (snark alert) OS.

        I admit to wondering why R or Python was not the language of choice, being as they are pretty much the sciency omnium linguarum of programming. (If I’m not overstating the case)

        I would like to see an effort to refactor Dr Evans’s VBA to either of these non-proprietary/Excel linked languages. I wish I were knowledgeable enough in either or in statistal analysis to help with the project.

        gary

        10

  • #
    LightningCamel

    Downloads (30sec), runs, so far so good. I’m looking forward to trying to get my brain around this. I’m also regretting that I feel I can no longer rely on the rigour or objectivity of a couple of people whose opinions on this project I would have given a lot of weight to a couple of weeks ago. Ah well, same story I suppose, if you want it done properly, do it yourself. :)

    Great work David, I hope it survives the next few weeks.

    80

  • #
    handjive

    Those of us who dissent from mainstream thinking about climate change truly are voices in the wilderness, analogous to the Rebel Alliance in the fictional Star Wars’ universe.

    Scattered, underfunded, thin-on-the-ground – that’s us.
    . . .
    Gratz & may the force be with you.

    140

  • #
    the Griss

    Chuckle ;-) ..

    Under “Solar Monthly – Custom data”.

    Cheeky boy, David. ! :-)

    110

  • #
    PhilJourdan

    Just a note of personal thanks to Joanne for sending out the announcement. I have not been active much of late as my time must be devoted to putting bread on the table, so I doubt I will get to read many of the comments or other articles. It helps when she emails a notice of the article. I know it is an imposition on her time, especially with all she has going on with the project. But it sure made my morning!

    Thanks again for the notice! And well done to David Evans and crew! All of you are the epitome of science in action!

    180

    • #
      Mark D.

      Yes Phil, the announcement e-mail is helpful. Since our NH winter was so long, cold and snowy, I’m spending much time getting projects done before summer is over.

      40

  • #
    cedarhill

    It’s just as likely that opponents will simply “go silent and shun” following the trail the BBC is blazing. The thinking seems to be if one even acknowledges an opponent one gives some degree of “ligitimacy” to both the person and what the person’s ideas. One might call this Total Disengagement.

    Even if forced to mention the model, they’ll just say something akin to “Even the deniers think it’s all rubbish. See WUWT and their experts. Next question.”

    It’s a good strategy. It seems to have dawned on the whatever one calls a non-skeptics that they’re simply look ridiculous debating or arguing with the folks hey hate. Plus it follows the rule of “it’s not what’s reported that counts but what’s repeated”.

    150

  • #
    thingadonta

    Good spreadsheet.

    No cloud data?

    A key piece.

    60

  • #

    I have downloaded all 18.5MB. libreOfficeCalc (Ubuntu 14.04) opens with a warning of some scripts. Very nice, I must archive, before I must futz with it! Thank you, again, thank you!

    70

  • #
    thingadonta

    I still think sea level data that is uncertain should be included and also graphed, you just add a proviso etc about where it is uncertain etc, on the graphs.

    (I have a favourite spot on Sydney harbour where I grew up where would mark the highest tide mark every year. It hasn’t changed since the 1970s.)

    100

    • #
      the Griss

      Fort Denison shows 0.65mm/year SLR or something like that.

      But that’s the average SL. I wonder what trend the high tides show?

      50

  • #
    Raven

    Hi Jo & David,
    I can confirm Climate.xlsm opens just fine on a Mac running Excel 2011.

    Of course, that’s a bit like confirming I’ve got you’re Boeing 747 at 33,000 feet.
    I’m sitting in the front seat where the pretty lights are . . . but I can’t actually fly it. :-)

    130

  • #
    ren

    We must see that the ice in the south breaking records not only in winter, but in summer. This means only one thing: the temperature drop atmosphere.
    http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/seaice.recent.antarctic.png
    http://www.cpc.ncep.noaa.gov/products/stratosphere/strat-trop/gif_files/time_pres_TEMP_ANOM_JFM_SH_2014.gif

    60

    • #
      Scrivener

      Are there any measurements regarding sea ice thickness as well as increase in area? Has that been increasing, do you know?

      10

  • #
    Richard Case

    I’ve downloaded it just fine, but cannot get it to open in MS Excel. I get a message saying that there were problems with the file, and did I still want Excel to try to recover as much as possible. I answered yes, but then the program spins awhile until I get a message saying that the file was corrupted. Has anyone else had this issue? I’ll try to download another copy and see if that works.

    20

  • #
    Todd

    BTW,

    You can break up the data feeds into separate files and load them into Excel. That would shrink your Excel file and allow for separate versioning of code & datasets on Github. Just my $.02.

    20

  • #

    “Does anyone know if GitHub is suitable for large Excel files?” Github can host any kind of file. The problem you will have turning it into an open source project is to do with your choice of one large Excel file. Git does merging better than any other source control system, but merging large binary files is difficult.

    Imagine if, instead of using Git to manage the thousands of small text files in the Linux kernel, the Linux team tried to use it to manage the kernel as a binary file. Suppose one contributor changes bytes 4800-5000, and adds bytes 100200-100300, which refer to bytes 4800-5000. Someone else deletes bytes 4800-5000, shifting bytes 5001-5201 into their place. How will you merge the contributions of these two contributors? Source control will be unable to tell if the merge will produce a working spreadsheet, and so will you.

    40

    • #
      David Evans

      If there was sufficient interest one could imagine turning the project into a big bunch of text files, for data and for code in C, and one Excel file to handle the interface or buttons and graphs.

      It could become a general climate investigation workshop. The VBA code is too slow, especially for Monte Carlo or the like, so if we got serious we would run the computation in C code from an add-in.

      110

      • #
        dp

        Binary tracking (it is possible) could break immediately if you updated (different from upgrade) your MS Office. It cannot be saved as a DIF file as it blows out some limits of that format. But you should probably version it immediately and provide a md5 checksum of each version else somebody is going to get their knickers in a knot over legitimate updates to the spreadsheet.

        30

      • #

        Might I suggest that Python be a far better route to go than Excel or C/C++ ?

        http://www.scipy.org/

        Many scientists, engineers, and mathematicians use Python for their studies and modelling.

        Not only would using Python make for a great “porting” project, it’s easy to learn. As you can see from the URL above, there are many science-oriented modules available which could prove extremely useful to a project such as this, and you’d have the added bonus that hosting the project on Github would make updating, and accepting patches an absolute breeze.

        Some people say Python is “slow” – there are ways of compiling Python to speed it up by orders of magnitude; I’ve been writing a GUI application with Python and Qt and have been compiling it with next to no problems.

        For a project like this I’d recommend Python, with PyQt for the GUI front-end, plus whatever math and science modules are listed in the SciPy site above.

        Regards,

        Kevin.

        50

      • #
        Greg Goodman

        It really does go against the stated aim of “accessible to all” if it requires $1000 software to run.

        Also VB/VBA is not a good technical choice for this sort of task.

        Neither would I expect many ( anybody ) to plough through 15000 lines of BASIC code to see how it’s done.

        I think a detailed flow diagram is needed for the method.

        If you want a collaborative, open project, then the code, at least, has to be a separate text file for quick exchange and diffing for version control. Then something like gitHub, cvs, snv for keeping track of it all.

        I would suggest making it modular, not one unwieldy monolithic block.

        Separate code and data, and make code modular ( FFTs, Monte Carlos, fitting, plotting all separate ).

        00

      • #

        David, Hetland, in his book Beginning Python, describes python as pseudocode that works. You start writing “first I want to do this, then that…” except that the text you write runs.

        It will take you a day or two to wrap your head around it to be able to do serious work.

        http://hetland.org/writing/instant-python.html

        As Kevin says, PyQt or the alternative PySide are great for GUIs. matplotlib
        http://matplotlib.org/

        is great for plotting. The documentation worked well for me two years ago for procedural programming but sucked big time for object oriented programming. But others should be able to help you.

        NumPy and SciPy provides all the numerical libraries you might need, coded in c or c++ for speed and available as python modules (i.e. you only ever need to code in python).

        00

  • #

    Impressive!

    Runs on my Home/Student Office version of Excel 2007.
    Home/Student Office version 2013 from ms for $140 US.
    See: http://office.microsoft.com/en-ca/home-and-student/

    Very fast execution (<< 1 second) on my 4 core i5 (<$1000 US).

    It would be an enormous task to translate to another language and another GUI metaphor. Likely not worth the effort.

    The dancing marionettes won't like it. Is too hard/easy/complicated/big/small and doesn't agree with their sacred dogma – he said – she said – they say – it hasn't been published in an authorized journal and reviewed by "recognized climate scientists" – the 97% don't agree – the 0.03325 should have been 0.033248 ….

    Actually, they fear their strings are being cut and they soon won't be able to dance. Dancing is all they have. It defines their existence and purpose in life. They can't do anything else.

    230

    • #

      Minor correction. I believe it should be “prestige journal” and “prestigious scientists”.

      http://youtu.be/b6UAYGxiRwU

      20

      • #
        Lionell Griffith

        They don’t deserve that modifier even in jest. “Authorized” and “recognized” says nothing about quality of performance and has no hint of their having earned their position. It was given to them. “Prestige” and “prestigious” implies a degree of having earned the position because of the quality of their performance over a significant period of time.

        100

        • #

          That little joke came about because Lief was bragging on another thread that he was published in a “prestige journal”. i.e. he was a TOP MAN.

          I wonder if he realizes what a joke he is making of himself.

          I will tolerate a LOT of arrogance if a person has something to contribute. Leif does not qualify for that allowance.

          I have seen it before. Steve Pinker (the brain guy) and I were in a conversation (e-mail as I recall) and I brought up a point (in relation to PTSD and drug use) he airily dismissed. I couldn’t possibly be right. These days that point is integral to his current understanding. TOP MEN.

          90

          • #
            Lionell Griffith

            Dancing marionettes do what dancing marionettes do. They can’t help themselves. After all, their puppet masters are pulling the strings. So they dance and dance some more. Always to the tune of their puppet masters.

            60

          • #
            Robert

            Haven’t been back to that thread in days, but…

            Remember these gems from Dr. Svalgaard?

            So, the valid test would be:
            1) parameterize using ‘my’ TSI
            2) run the model on ‘my’ TSI
            3) compare with the temperature record

            and this:

            The consensus is not about the Earth, but about what the Sun puts out. We have identified and quantified the corrections needed and when they must be applied. A major review of this is in press in Space Science Reviews as we speak. What is remaining is to apply the corrections in their final form, to evaluate error bars, and to publish the series. This will be completed at the fifth workshop early next year [to be held in Boulder Colorado] at which time the series will be submitted to the International Astronomical Union for approval.

            That last bit was, to me, especially ‘informative’ considering past experiences and observations of people. One could speculate that anything that might raise potential questions about “the series” and “‘my’ data” needs to be immediately attacked and discredited lest it interfere with that approval from the International Astronomical Union.

            Nice to see the model released, haven’t done VB in a few years so for me I’m more interested in looking at how the code was put together than what it does. Goes for pretty much any program for me though.

            As others have mentioned, VB can be slow, but you can whip up a working prototype of an application damn fast with it. The other advantage is for general “office work” type applications it works great and is fairly easy for your average manager to learn to do basic macros for presentations etc.

            Pros and cons to everything, but for all the heckling and snide comments I’ve heard from programmers over the years who ALWAYS reply to anything done in VB with a “oh you should have used _________” it works, it isn’t as complicated as whatever their preferred language choice is, and there aren’t a lot of companies out there where it isn’t readily available via VBA on their office desktops.

            Many of us programmers have our favorite language but also understand when other languages are more suited to the task at hand. More often than not we aren’t given the choice but have to use whatever language is handed us due to legacy code or because someone higher up the food chain has already decided for us what the best language to use will be.

            50

            • #

              I like to do bit banging. My favorite is Forth.

              http://spacetimepro.blogspot.com/

              10

              • #

                Forth is an interesting language. It solves the wrong problem magnificently: maximum functionality on a 1K 4 bit CPU with an absolute minimum of code. It is an operating system, programming language, and interpreter embedded in a few hundred bytes of machine code. It had significant utility when it was invented in the 1960s. Unfortunately, when you get beyond a few hundred statements, the program is incomprehensible and unsupportable. However, within extremely limited computing environments (less than your typical digital watch today), there is nothing superior.

                The tool used should match the problem to be solved rather than the tool determining what problems you can solve. The problems I have to solve are many orders of magnitude larger and more complex than a digital watch. Hence, there may be a use for Forth today but what that is escapes me.

                My favorite language is Pascal. Sadly, Pascal is not well supported today except in Borland’s Delphi. I use a strict subset of ANSI C because it is a well supported standard in which one can do procedural, functional, and object oriented programming as needed, all in the same environment. The source is highly portable except for the idiosyncrasies of any given OS. Even those “features” can be encapsulated and limited to a small fraction of one’s code.

                00

              • #

                Lionell Griffith
                July 10, 2014 at 2:06 am ,

                Interesting. None of it is true. None of it. (Well not much) Except possibly that if you want to write programs with a LOT of overhead other languages are better.

                1. Our Forth is 32 bit.
                2. The processor we are using is minimal. The next one will be on a 56K Flash Machine (ARM). We intend to go up from there.
                3. A government inspector (who knew no Forth) once looked at about 20K of code I had written and said it was the best written code he had seen in any language in 3 years. If you write it well it is easy to maintain.
                4. I bit bang. There is no better tool for that.
                5. Portability is over rated. If you need fast, small, cheap, easy to write, self documenting (assuming you are disciplined), easy to incrementally test, there is no better language. You adapt your code to the processor you use.

                As with all languages it depends on what you want to do, how much resource you have to do it, what you like. I like doing low power on minimal resource machines. It is one reason Forth is still used for space applications.

                00

              • #
                gary turner

                @Lionell Griffith

                I, too, learned to love Pascal, in particular Borland’s Delphi neé Turbo Pascal. So much ‘cleaner’ than C/C++. Java, of course simply sucked. The other language for me (because of use in modifying Emacs) is Lisp.

                Unfortunately, I just don’t have a need for much code hacking other than making Emacs bend to my will, and haven’t written a lick of Pascal in over ten years.

                Nice to see someone else appreciates OOPascal.

                cheers,

                gary

                00

              • #
                Lionell Griffith

                I was speaking of Forth as it was in 1967 through to the mid 1980s. It may have advanced since then but if it is Forth, it has the good/bad aspects of Forth. It is a small core, verb oriented, functional language in which its reverse polish notation calculation is one of its better features. Then it used only scaled integers. Perhaps now, it uses some form of floating point. Not a big point for or against.

                If you must use low power minimal resource hardware, Forth might be still valid. Space applications put a premium on low power, low mass, adequate performance for the task, and cost to develop is of no consequence. However, I still don’t believe Forth can be used to produce a product for general distribution with ordinary technical support.

                Knowing about the quality of avionics software, I wouldn’t want to be flying with avionics software written in Forth. I have a true story that gives me white knuckles if I think about flying in a software controlled aircraft. It involves a test pilot who just missed building his memorial crater in the Mojave Desert while going a bit over mach 1 in an F18. It was the direct consequence of a flaw built into the software on purpose to solve an instrumental problem, then forgotten about, and just about destroyed an F18 and killed its pilot in the process (150 feet off the ground and 9G lateral on the pullout).

                There is a premium to be placed upon the quality of that kind of software that makes normal high quality software look sloppy and amateurish by comparison. Even in avionics, I saw much too much sloppy work. Sadly, by Forth’s inherent design, it is far far too easy to be sloppy.

                The nested verb architecture of Forth puts a premium on strict attention on coherence between the name of verb and the content that the verb contains. Otherwise, by the time things get 10 or 20 levels deep, you haven’t a clue as to where a program fault lives and no way to fix it without destroying a hundred other linkages. The verb architecture necessitates such logical depths and massive coupling or all you have is trivial functionality or a program that doesn’t work.

                Fundamentally, quality must be designed into the software at all levels of the design and construction. It cannot be accomplished by inspection or testing after the fact.

                00

              • #
                tom0mason

                Lionell Griffith

                OOPascal is well and alive as aprogamming language. OpenPascal has a free version Delphi of it called Lasarus (available at http://lazarus.freepascal.org/ ). It is multi-platform and can import just about everything Delphi does.

                10

              • #

                Thanks for the tip about OpenPascal and Lazarus. I have downloaded it. I will take a look at it. I left Turbo Pascal when Borland lost the power curve for Windows NT 3.1. I moved to ANSI C and Visual Studio.

                I have a major product of several hundred thousand of executable lines of code in ANSI C and even more lines of text documenting it. Converting it would not yield any advantage and be a very major effort. However, new projects in the Lazarus environment might be worth while.

                10

              • #
                Greg Goodman

                Great as a intellectual exercise but I think the man-hours needed far out-weights the code efficiency.

                I like the irony of the name. They call it Forth because you have think backwards to write it.

                00

              • #

                Greg Goodman
                July 10, 2014 at 9:56 pm

                Great as a intellectual exercise but I think the man-hours needed far out-weights the code efficiency.

                I like the irony of the name. They call it Forth because you have think backwards to write it.

                Actually it takes far fewer man-hours than any other tool I have ever used for bit banging.

                It was named Forth because file names on the first system it was developed on were only 5 characters. Otherwise it would have been Fourth.

                00

              • #

                There is a premium to be placed upon the quality of that kind of software that makes normal high quality software look sloppy and amateurish by comparison. Even in avionics, I saw much too much sloppy work. Sadly, by Forth’s inherent design, it is far far too easy to be sloppy.

                It does make bad programmers worse and good programmers better.

                I once was competing with a team of 30 “C” coders with a team of 3 Forth coders. We got the task done in one month including hardware. The “C” boys weren’t finished after 6 months.

                That made us 60 times as productive. But I knew what I was doing and forced good coding on my boys. That was the job the government inspector looked at and declared, “best code I’ve seen in 3 years”. You basically write a language for the domain you are working in. So you get things like

                Turn-On Motor – as executable code. Well written it reads very much like English.

                00

              • #
                Robert

                Ah, Borland products… Still have the boxed version of Turbo C++ for DOS somewhere around here, floppy disks, manuals and all. If I dug around a bit more I think I have TASM too.

                Whatever language is chosen the biggest problem I seem to run across these days is developers who, if they are even aware of how to do it, pay little if any attention to error handling. Seems to be a mindset of “well if it did that then they weren’t using it the way I intended them to” as if they ever use it the way they were expected to use it.

                Nice little things like a test script trying to apply a load to a power source without first verifying the power source is even running. Makes you feel all warm and fuzzy inside…

                00

              • #
                HAS

                Must say I had a misspent youth in formal logic and got FORTH for my kids C64 (H/T Chuck Moore) and didn’t look back. Fast as machine code, but in a “formal language”, and streets ahead of the old C64 Basic interpreter. Having said that I’d learned to program using Algol on an Elliot 503 so expected languages to be reasonably structured (unlike Fortran that came later, COBOL and other more commercial languages that worked the seedier side of the street).

                I think if I was asked to program instrument control (including a F18) I’d look to FORTH.

                The trouble is all you young uns have got fat and lazy with cheap processors and tonnes of memory. In my day ….

                00

              • #

                Forth used on the Boing 777 avionics.

                http://forth.org/successes.html

                It looks like they use Forth the way Sun did for its workstations – as a test and debug tool. Built in.

                00

  • #
    OregonMiner

    David,

    Fist thing: THANKS. Obviously entailed lots of work.

    Second: I am not a computer/code/program person, but I have to ask, (this is going to seem silly); Why are all of the graphs shown ( Sheet AL2302) with straight trend lines? I understand the intent of best fit straight line, especially for the shorter term plots, but it jumps out at me that the longer term graphs should be a curve. A straight trend line shows an elevated temperature rise to the right that simply isn’t there.

    Help,

    Neil

    30

  • #
    Don Jordan

    I’m getting the following error message when I click on any of the buttons:

    “Compile Error: The code in this project must be updated for use on 64-bit systems. Please review and update Declare statements and then mark them with the PtrSafe attribute.”

    I’m running the 64-bit version of Windows 8.1.

    20

    • #
      Don Jordan

      Excel 2010 here, BTW.

      20

    • #

      Interesting. Excel 2007 (32 bit) is running it on my 64 bit Windows 8.1 Pro system without a problem. I did nothing special. It was an install and run operation.

      Is it possible you are running a 64 bit version of Excel? If so, that may be the problem.

      The following MS notice explains why it may not be a wise idea to run a 64 bit version of Excel.

      http://technet.microsoft.com/en-us/library/ee681792%28v=office.15%29.aspx

      You may need to revert back to a 32 bit version.

      20

      • #
        Don Jordan

        Yes, I’m running the 64-bit version of MS Office 2010. There’s nothing wrong with the 64-bit version of Excel. In fact, it allows the creation of spreadsheets that exceed 2GB of data. I added the “PtrSafe” attribute to the offending lines of code and everything seems to be working correctly now:

        Before:

        Private Declare Function QueryPerformanceCounter Lib “kernel32″ (lpPerformanceCount As LARGE_INTEGER) As Long
        Private Declare Function QueryPerformanceFrequency Lib “kernel32″ (lpFrequency As LARGE_INTEGER) As Long

        After:

        Private Declare PtrSafe Function QueryPerformanceCounter Lib "kernel32" (lpPerformanceCount As LARGE_INTEGER) As Long
        Private Declare PtrSafe Function QueryPerformanceFrequency Lib "kernel32" (lpFrequency As LARGE_INTEGER) As Long

        Since I’m a professional C# / C++ developer and not a VB guy, I’m not sure if this change will have ramifications in other section of their code. SO far, so good.

        50

      • #
        Don Jordan

        I’ve been going through page-by-page and running their graphing code and nothing seems to be broken from the code change I added to the project.

        30

      • #
        steven mosher

        64 bit is a problem with the file as downloaded.

        20

  • #
    Don Jordan

    The offending VB code is in your CTimer class and it looks like it’s using 32-bit Windows dependent DLLs.

    Private Declare Function QueryPerformanceCounter Lib “kernel32″ (lpPerformanceCount As LARGE_INTEGER) As Long
    Private Declare Function QueryPerformanceFrequency Lib “kernel32″ (lpFrequency As LARGE_INTEGER) As Long

    30

    • #
      Mark Gutzwiller

      This should work on excel 2007 up:

      #If VBA7 Then
      Private Declare PtrSafe Function QueryPerformanceCounter Lib “kernel32″ (lpPerformanceCount As LARGE_INTEGER) As Long
      Private Declare PtrSafe Function QueryPerformanceFrequency Lib “kernel32″ (lpFrequency As LARGE_INTEGER) As Long
      #Else
      Private Declare Function QueryPerformanceCounter Lib “kernel32″ (lpPerformanceCount As LARGE_INTEGER) As Long
      Private Declare Function QueryPerformanceFrequency Lib “kernel32″ (lpFrequency As LARGE_INTEGER) As Long
      #Endif

      50

      • #
        Don Jordan

        Yeah. I posted the fix earlier. See my comment #29.2.1 above. I’ve been testing it and it doesn’t seem to cause any problems. :)

        20

      • #
        Don Jordan

        But…I forgot to add the preprocessor directives to differentiate between the version of VB.

        20

        • #
          Mark Gutzwiller

          Sorry, I hadn’t seen your fix.
          At least it should work okay, for the the MS systems.

          20

      • #
        DT Christensen

        Depending on whether you have a 32bit/64bit version of Office and the vintage (2007, 2010, 2013, etc.), and 32bit or 64bit version of Windows, I think you would be covered if you used the if statement below…

        #If vba7 and win64 then
        declare ptrsafe function ….
        #Else
        declare function ….
        #End If

        10

  • #
    Brad

    It is going to be a record-breaking post for comments…:)

    Well done, imagine, a scientist that actually stands by their initial proposal and releases both programming and data FREELY. I think we may need to come up with a new name for this wonderful concept. Maybe Evanology??? Open to suggestions…

    150

  • #
    Mikky

    An impressive amount of work, well presented, but …

    Section 6.2 of excerpts.pdf says:

    “There does not seem to be any other possibility here. Given that there is clear evidence of
    notching, and that a notch filter necessarily has a non-causal step response, the only rational
    conclusion is that there must also be a delay. The notch must be accompanied by a delay, because
    the notch on its own is physically impossible.”

    Many people with signal processing experience have said that every part of that paragraph is wrong,
    just look back at the comments on earlier posts.

    Sadly, the slow motion train crash continues, and the board of inquiry has already attributed it to driver error.

    223

    • #
      Ragnaar

      “Many people with signal processing experience have said that every part of that paragraph is wrong, just look back at the comments on earlier posts.”

      So it would be possible for you to state one specific objection you have.

      151

  • #
    Bob Weber

    David, thanx for all the effort and patience. Before looking at your spreadsheet, going into it, my hunch is that you are picking up the solar minimums in your notch model. Temperatures go down when solar flux/TSI drops below a “threshold”, somewhere between 85-100 sfu for F10.7cm, that is experienced primarily near solar minima. My model supports David Stockwell’s solar accumulation model too. In that sense, empirically, the warm-cool threshold looks to be on the low-side of that range, 85-90 sfu.

    70

  • #
    Chester

    Skeptics are winning

    Utterly childish and tribal.

    You actually did what you accused climate modelers of doing – made a model that assumes the sun alone is responsible for global warming. And, of course, there’s the usual rub – please give us money….

    Will you disclose all your sources of income and all the money you raise from this openly? Will you return the money if this is all shot down in flames?

    464

    • #
      bobbyv

      Do you propose climate modelers return billions to tax payers if their models are wrong?

      540

    • #
      Michael Larkin

      Will the climate modellers return all their grant money now that their models have been shown to have no predictive ability? Money that was taken from taxpayers whether or not they wanted?

      That’s tribalism for you: hypocrisy on stilts.

      430

    • #

      Another dancing marionette wildly spinning and prancing. Yet he failed to read and understand the introduction.

      One goal of that “assumption” was to demonstrate that CO2 with a non-physical back radiation positive feedback is the not only thing that can explain the gross behavior of the sun-earth climate system. That goal has been met in spades redoubled. It was worth the effort for that result alone. Does it do anything else. Maybe – maybe not. However, nothing is hidden. You can examine and verify every step of the way.

      The data, the work, and the interaction is fully available for download and testing – FOR FREE! This is quite unlike the work the “team” has done over the past three decades. We the public were forced to pay for their work and don’t get to examine much more than their assertions of “trust us it will be a total catastrophe unless we repent and abandon our evil ways”. Almost interesting is that it does not need multiple multimillion dollar super computers to run it nor an army of programmers to make it happen. A mere mortal can afford a system that will run the models in seconds.

      360

      • #
        the Griss

        “can explain the gross behaviour of the sun-earth climate system”

        LG, I know when you say “gross” in this sentence you mean “total”

        but please that that word for describing people like Chester, instead.

        50

        • #
          Lionell Griffith

          The good/bad thing about language is that words can mean so many different things. Sometimes related and sometimes not. The best one can do is give enough context to support the meaning you wish to have understood. Sadly, a short rapidly written post will sometimes lack sufficient context to make the meaning clear.

          For Chester, I would reserve the phrase “has no couth” is more descriptive of his behavior than “gross”. His communication is not foul or obscene, it is simply socially and intellectually inept.

          I chose “gross” in the sense of high level, general, not detailed. For time series data it implies low frequency behavior. I selected the word specifically because it was terse and did not overwhelm its object “sun-earth climate system”.

          Compare: “gross behavior of the sun-earth climate system.”
          With: “general low frequency behavior of the sun-earth climate system.” They both mean exactly the same thing. However, which way has the stronger punch? I saw it as no contest. The first reduced what I wanted to say to its essence.

          90

          • #
            the Griss

            I actually think the climate system is behaving rather politely at the moment :-)

            The usual minor hiccup, of course, but even a fine lady can get a hiccup.

            History seems to show that the gross stuff happens if things start to cool rapidly. ! :-(

            111

    • #

      Will you disclose all your sources of income

      All the warmists first.

      170

    • #
      Rereke Whakaaro

      You actually did what you accused climate modelers of doing — made a model that assumes the sun alone is responsible for global warming.

      star comment1. We have never, as far as I know, accused the “climate modelers” of anything to do with the sun, apart from accusing them of assuming that it is constant. But at least you are implying that they are not scientists, which is an improvement, I guess.

      2. Dr Evans did not assume that the sun alone was responsible, when he started his research. You have obviously come into this conversation late, and are not aware of all of his earlier work.

      3. The actual philosophical approach to modelling is different in both cases. The previous models all start from the question: “What parameters will be needed, in order to simulate the climate?” That is the normal approach that is taught in most Universities. It is a bottom-up approach.

      4. Dr Evans had the idea of approaching the problem from the top-down, by starting from a model of the observed cyclic changes in the historic data, and then identifying signatures in nature that might explain those cyclic changes. Hence his use of Fourier Analysis to decompose complex signals into its constituent parts. In doing so, he identified an abnormality that acts like a notch filter, and does so with a “regular” time delay, from a known solar cycle.

      5. The modelling leads to the conclusion that the sun is driving climate with a much more defensible process, than the previous CO2 model can sustain. Hence the model is a better approximation of what is actually happening.

      Welcome to the scientific process. It will seem strange to you at first, but only because you are not used to it.

      610

      • #

        Rereke Whakaaro,

        Tour de force. Bravo.

        150

      • #
        Mark D.

        When reading up from the bottom of this thread I kept reading Rereke:”see my post at 34.5″. I also recall an e-mail earlier today where I read a very good post by Rereke.

        Low and behold they are the same post.

        Excellent summary Rereke.

        50

      • #

        “We need a New Climate Model (from now on referred to as NCM) that is created from ‘the top down’ by looking at the climate phenomena that actually occur and using deductive reasoning to decide what mechanisms would be required for those phenomena to occur without offending the basic laws of physics.

        “The general approach is currently to describe the climate system from ‘the bottom up’ by accumulating vast amounts of data, observing how the data has changed over time, attributing a weighting to each piece or class of data and extrapolating forward. When the real world outturn then differs from what was expected then adjustments are made to bring the models back into line with reality.”

        “We have to start with the broad concepts first and use the detailed data as a guide only. If a broad concept matches the reality then the detailed data will fall into place even if the broad concept needs to be refined in the process. If the broad concept does not match the reality then it must be abandoned but by adopting this process we always start with a broad concept that obviously does match the reality so by adopting a step by step process of observation, logic, elimination and refinement a serviceable NCM with some predictive skill should emerge and the more detailed the model that is built up the more predictive skill will be acquired.

        That is exactly what I have been doing step by step since April 2008.”

        from here:

        http://www.newclimatemodel.com/new-climate-model/

        40

        • #

          > The general approach is currently to describe the climate system from ‘the bottom up’

          Yes.

          > by accumulating vast amounts of data, observing how the data has changed over time, attributing a weighting to each piece or class of data and extrapolating forward

          No. That’s not how it works in the GCMs at all.

          211

        • #
          Duster

          Really, such a model is from the “bottom up.” You start with what is observed on the ground so to speak, and then work from that foundation into a structure of theory. “Top down” is implicitly dogmatic more less as the Ptolemaic view of Cosmology operated. Physics especially has turned from deriving theory from observation to seeking observation to match mathematical theory. Some however do get excited when observations such as problems with the size of the proton suggest the universe may not be quite what was calculated.

          10

    • #
      Gary Meyers

      I use to own an ass named Chester! :- )

      110

      • #
        Rereke Whakaaro

        It sounds as though he was perfectly named. :-)

        140

        • #
          Gary Meyers

          Yes he was. We had to sell him because the neighbors complained about his constant braying. Ha!

          130

      • #
        the Griss

        I long time ago we owned a donkey called ‘Otto’ after the Spanish conquistador.

        ?????????

        Donkey Otto, of course !!!
        .
        .
        .
        .
        I’m sorry ! (red thumbs as deserved)

        153

    • #
      Angry

      Very appropriate……..
      Do you remember the Warner Brothers cartoon that had the two dogs in them? One was a big bulldog, and the other was a smaller, lighter dog. The smaller dog, Chester, was always hanging around the bigger dog, Spike, always looking for his favor, giving him compliments, and acting tough when he had Spike to back him up.

      31

      • #

        great… and I suppose you are not forgetting how they ended? With Chester as the tough dog and spike tagging along behind because Spike was actually unable to cope with the situations presented in the story?

        Maybe look before you leap.

        30

        • #
          PhilJourdan

          Yea, I was not going to mention that. Spike always got the panther, and Chester got Sylvester.

          10

    • #
      sophocles

      Proverbs 17:28

      10

  • #
    Chester

    …convinced me to look for a low pass filter in the empirical transfer function, assuming the climate was mainly driven by solar radiation (TSI).

    Imagine the furore on the sceptical blogs if one of the emails contained this phrase:

    …look for a low pass filter in the empirical transfer function, assuming the climate was mainly driven by CO2.

    347

    • #
      Mark D.

      Sure Chester, YOU KNOW that the sun has nothing to do with climate. CO2 is everything right?

      Just another drive by serial troll spamming the thread……

      300

      • #
        Lionell Griffith

        A dancing marionette who’s strings are controlled by his puppet masters. He cannot add to a conversation because he has nothing to offer, all he can do is dance.

        190

    • #

      Chester,

      Evidently you are unaware of the very large low pass filter in the climate system. It is called the oceans. And whether you impute climate change to CO2 or the Sun that low pass filter still has to be accounted for.

      It is amazing the things that the really smart people don’t understand. Makes you wonder what else they are missing.

      290

      • #
        gary turner

        Had to tweet your last paragraph. ( @gtwebdev )

        gary

        10

      • #
        Paul Vaughan

        MSimon (July 9, 2014 at 4:49 am) raises a noteworthy point.

        One of the pillars supporting the IPCC declaration of 1995 was Thomson (1995). Here’s the follow-up paper:

        Thomson, David J. (1997). Dependence of global temperatures on atmospheric CO2 and solar irradiance.
        http://www.pnas.org/content/94/16/8370.full

        From it:

        “[...] the impulse response estimated between Tn(t) and L(t) is noncausal with a width about the length of a solar cycle.” (where T=Temperature & L=soLar)

        “Similar problems are common in transfer function estimation problems ranging from [...] to the design of telephone equalizers.”

        Sound familiar??

        …But generously assuming David Thomson (1995) wasn’t just a mathematically clever political pawn in a grand conspiratorial scheme of vexatious deception aiming for awe-inspiring tax-grabbing world government take-over, we can quickly see from this (if we are honest & competent) that he was at least deeply & darkly ignorant of nature’s beauty.

        Could it be that he’s an entirely decent man who innocently made false assumptions about CO2 & solar-climate relations (that were exploited by others)?

        Well, he’s not here rudely attempting solar-climate thought policing with repeated false (in the black-&-white [not grey] sense) statements echoed by brainless groupies, so based on that it looks plausible that he may be a decent man who made a serious mistake.

        Research Update: I’ve now cataloged 3/4 of the multivariate exploration needed to assemble an extensive report on the 1 page SAM/Sun summary I introduced June 21.

        10

        • #
          David Evans

          Very interesting! Great find Paul.

          David Thomson in 1997 got close, but ignored the clue because of his assumptions. Crazy.

          For last 18 months Joanne and I asked each other: “Why hasn’t this been done before? Isn’t it pretty obvious?”. This paper gives a good clue. He only considers direct, immediate TSI warming, not delayed or indirect TSI effects (i.e. force X).

          Thomson sets out to find the transfer functions of TSI and CO2 to temperature. Great, an obvious approach. He doesn’t graph the transfer function from TSI to temperature, but he obviously came close to what we found, writing: “For example, the impulse response estimated between Tn(t) and L(t) is noncausal with a width about the length of a solar cycle.” This means that the transfer function was non-causal, and that the notch was very roughly near 11 years (why doesn’t he just graph the transfer function? then we could see the notch, and whether it was at 11 years). L(t) is his TSI data, from Lean.

          However he rules out the possibility of a delay by: “Transfer functions implying noncausal response to changes in solar irradiance imply either that the transfer function is unacceptable, or that there is a problem with the assumed irradiance.” So, bad data or does not make sense. Not “it’s telling you to add a delay of 11 years to make it causal, i.e. the TSI is signalling a delayed force X”.

          About the TSI to temperature transfer function being non-causal, he says: “This is an artifact of the near periodicity of the solar cycle as, without other knowledge or assumptions, one cannot assign the response to a periodic process to a given time without ambiguities of N periods”. No, just graph the transfer function in the frequency domain. While he is right that there is ambiguity about the required delay or periodicity, it is not an artifact. The frequency domain graph would give a lower bound on the required delay.

          30

          • #

            It seems like it is an article of faith among some sun watchers that TSI is all that matters. TSI isn’t enough –> it can’t be the sun. But that contradicts the well known cold spells that match reduced solar activity.

            What amazes me – I’m new to this – is if it isn’t the sun what causes the well known long stretches of cool weather? I have yet to see any of the “it isn’t the sun” people come up with an explanation. And it doesn’t seem to trouble them.

            20

          • #

            Strange that David Thomson employs coherence tests for solar variability here in Fig. 11 though.
            http://www.sciencemag.org/content/268/5207/59.full.pdf

            00

          • #
            Bernie Hutchins

            David –

            Are you still insisting, as you did previously, that: “It is well known that notch filters are non-causal.”

            This is totally wrong (even non-physical) of course, and you must know this by now.

            Here is an advanced set of two figures that I have prepared for an upcoming posting of my own:

            http://electronotes.netfirms.com/NotchStepNew.jpg

            I previously posted a hand-drawn version of the circuit and the scope photo (an EXPERIMENT), which you just blew off as:

            “Bernie I’ve no idea and am not going to debug and figure out what you have done exactly.”

            The top of my link is the carefully calculated response by my own Matlab code, computing the step response numerically as the sum of a LP and a HP. Further, it agrees exactly with Matlab’s own built-in step function, and with the neat on-line R-L-C analysis link I provided. And it agrees with the experiment.

            Your step response, Part III Fig. 2 looks the same (you even seem to have chosen critical damping) with two essential exceptions: that it is rotated 180 degrees (non-causal), and is offset by 0.5 (why!). Since you don’t give your transfer function [as in T(s) = ….], I have no way to determine what you got wrong.

            Since the response of a notch IS causal, you don’t need a delay. If you want a delay for your model, just put it in, but don’t blame it on a notch. Further, a delay will not make your step response (your Fig. 2) causal (1) because it is an exponential and would need time reversal, and (2) because it is not really truncated (despite critical damping).

            00

            • #

              If you go back and look at the notch David showed you will find it has zero delay with respect to the step. Yes a real notch is causal. It also has a delay. You can see it in your figure. So a real notch has a delay. That is in fact the point David was making. What David was pointing out was that the notch was not an artifact of the math.

              10

              • #
                Bernie Hutchins

                Of course a “real notch is causal” and “has a delay” – that’s what causal means ;)

                You also said “What David was pointing out was that the notch was not an artifact of the math.” What math in particular are you referring to? Sorry, but I sincerely have no idea what you are suggesting here.

                01

              • #
                Bernie Hutchins

                Oh – another question

                Do you know what the offset of 0.5 is about?

                I’m going to try to reverse engineer it with Matlab’s step.

                00

            • #
              David Evans

              Bernie, the maths and code are all in the spreadsheet.

              Go to the “Analysis” sheet, and see the Solar model section (around cell V54). Turn off the delay and LPF, so the model just becomes the notch filter. Play with the parameters of the notch (hit the “Draw” button underneath, see the step response to the right). Note the non-causality of the notch filter without a delay. Formulae in the excerpts.

              10

              • #
                Bernie Hutchins

                Thanks David -

                I can’t run the spreadsheet, but the pdf helps a lot.

                I interpret Fig. 4 of the pdf to be the step response of two real poles at -1/7.6 and two zeros at 1/11 radius at an angle of (+/-)87 degrees, thus just slightly left of
                (+/-)j(1/11). This is a very ordinary notch with causal step response. Here is Matlab’s step response:

                http://electronotes.netfirms.com/DavidExample.jpg

                What do we disagree on?

                3 AM here so I have to quit for now.

                00

              • #
                David Evans

                Bernie, No access to pc with Excel on it?

                No, the Matlab step response is way off.

                The zeroes are
                2 pi fz (-cos(thetaZ) +- j sin(thetaZ))

                The poles are
                2 pi fp (-cos(thetaP) +- j sin(thetaP))

                The best fit seems to be thetaZ = 87 deg, thetaP = 0 deg, fZ = 1/11, fP = 1/7.6, time in years.

                10

              • #
                Bernie Hutchins

                Thanks David -

                The poles/zeros I used were from your Fig. 4 annotations (see again my plot and the numbers below it) and are identical to the ones you provided at 5:51pm except you multiplied yours by 2*pi. This scaling of the s-plane of course makes no difference regarding causality.

                It remains, fundamentally, quite absurd to suggest a non-causal physical system. Further, you and I agree (essentially) on the filter in an engineering sense (poles/zeros) and every standard method of finding the time response (solving diff eqns, inverse Laplace, numerical integration, Matlab, the online link for R-L-C I gave, experiment, etc.) is causal. You do not say in your “excerpts” how you did the step response. (My Excel loads your spreadsheet but nothing runs, so I can’t use that.) What was your method of inversion?

                Well, I have explained this in the only three ways I know how. The rest is up to you.

                00

              • #
                David Evans

                Bernie – I have provided the full calculations, and they show unambiguously, from scratch, that the notch filter is non-causal. So run it in Excel — you can see each intermediate step, each intermediate time series or spectrum. It is pretty simple.

                I am amazed that you cannot run Excel. Now you have a copy of Excel, good. See the warnings about Microsoft puts up about active contents, internet content, etc? Enable everything, as explained on the “Intro” sheet of the spreadsheet. Most everyone else can run the spreadsheet, but oddly you cannot. It’s about the most widely used computational program on the planet.

                I have no idea of your set up in Matlab. The image you give in Matlab is correct — except it is upside down and (crucially) back to front. The software I have is self-consistent and does all the right things, so I know it is neither upside down nor back to front. You might check the conventions on phase signs and FT use (see the OFT document for the definitions I have used), or simply run the Excel spreadsheet I provided.

                00

              • #
                Bernie Hutchins

                David – thanks for the reply of July 16, 2014 at 3:02 pm

                I found a good classic reference: Franklin F. Kuo, Network Analysis and Synthesis (Wiley 1962)

                Franklin, BLESS HIS HEART!, gave us a closed form. It is in his Problem 6.9, page 179-180

                It is: v(t) = u(t) + 2 e^(-t) u(t) – 2 (t+1) e^(-t) u(t)

                I plotted this and it looks exactly like my previous result. It is of course causal.

                http://electronotes.netfirms.com/Kuo.jpg

                As I noted, your result was rotated 180 degrees (what you say: reversed and upside down) but it also has an offset of -0.5 (was your calculation for a bipolar step?).

                Except for your calculation, I see no support for your claim of a non-causal step response. Me, having essentially no ability with Excel, and you having great ability and having written the code, it makes sense for you to look for your own errors.

                Sincere best wishes.

                Bernie

                00

              • #
                David Evans

                No Bernie. Just open the spreadsheet and play with the solar model. Turn off the delay, turn it back on, play with the notch parameters, and so on. Just do it. No need to look at code, just type in parameters and watch graphs change. Most everyone else can get the spreadsheet working — read the instructions on the “Intro” sheet. You will see that the spreadsheet is fully consistent, with delays, notches, transfer functions, steps, etc all occurring as one would expect. And the notch — when without a delay — is non-causal.

                I have presented the full working code, down to every multiplication and addition. It is easy to inspect, easy to look at intermediate results like arrays, nicely graphed, the works. Everything about it is consistent. It is the same code used to run the model. The onus is on you to check it, since you are claiming it is completely wrong. Yes I have checked it, and I find no errors or inconsistencies.

                You have presented me only with claims that you appropriately fed the parameters into a software package. Have you checked exactly what Matlab does? What its sign conventions are? You are relying on canned software, but are you using it correctly? Why doesn’t your image show the step response before zero? How can your step response start with an infinite value at zero — doesn’t that suggest to you something is just a tiny bit wrong? I can neither check your claims directly nor run Matlab. Just saying I am completely wrong and walking off is not good enough.

                00

              • #
                Bernie Hutchins

                David – PLEASE listen to me! You are wrong.

                I can not get your spreadsheet to run, but I have NO DOUBT that it runs as you say. So I have no incentive to get on the correct learning curve, especially as that would NOT be an independent verification.

                I have my own Matlab calculation (I posted the 11 lines of code which anyone can understand), Matlab’s step function, the online R-L-C calculator, and now Kuo’s closed form, all in addition to the tradition of needing physical causality, and I have done an experiment! Has anyone verified your non-causal step response independently – WITHOUT using YOUR program? I didn’t think so. If so, please indicate!

                The latest I offered was NOT “canned software” but a plot (yes done with Matlab just to plot it) of Kuo’s closed form formula. Let me simplify Kuo’s equation as he manipulated it a bit to be student-friendly in the problem. The closed form is simply:

                v(t) = [ 1 – 2te^(-t) ] u(t)

                (You are familiar with the u(t) notation I feel certain. )

                You asked two questions: “Why doesn’t your image show the step response before zero? How can your step response start with an infinite value at zero?”

                First Question Answer: Because it’s CAUSAL – zero prior to t=0! If you want, I can plot a zero back to -1, or to -1000. All zero – NOT very interesting is it?

                Second Question Answer: Because the value of the step response at t=0 is NOT infinity – it’s 1.

                I guess you haven’t used Laplace transform stuff much recently.

                00

              • #
                David Evans

                So there is a discrepancy between what you do with Matlab and what I do with transforms.

                You are in a position to resolve that discrepancy (you have the spreadsheet, so you can reproduce my results); I am not (not having Matlab or being able to see what you are doing). You won’t even run the spreadsheet.

                If you would run the spreadsheet you would see that it correctly produces step responses for delays and low pass filters and delayed low pass filters, for instance. I’ve looked at the arrays being produced in the code, and it is putting in the transfer function as graphed here (Fig 1) when it produces what I say is the step response. When it puts in the transfer function for a low pass filter instead, it outputs the correct step response for a low pass filter

                Maybe the description of the notch filter here is not being correctly translated into what Matlab needs.

                Yes, sorry, your image shows a value of 1 at t=0, not infinity.

                Bernie, in your image, what is the time of the trough (it looks like about a quarter of the first tick)? If it is like my step response turned 180, it will be about 1.25 years and your time-axis ticks are every 5 years.

                00

              • #
                Bernie Hutchins

                David –

                The easiest answer to your question about the position of the minimum in my causal step responses is to use Kuo’s closed form and differentiate. Thus:

                v(t) = 1 – 2te^(-t)

                had a minimum at t=1 and the minimum is (1-2/e) = 0.2642. This applies to the usual “normalized” case where there are two poles at s=-1 and zeros at s=+- j. This is the nominal case where the frequency “shelves” on either side of the notch are equal. The notch is sharpened as the two real poles split onto a circle and approach the zeros.

                The one plot you mentioned is my least favored, precisely because it uses Matlab’s canned function (step):

                I prefer my own numerical integration:
                http://electronotes.netfirms.com/NotchStepNew.jpg
                and Kuo’s formula above which I plotted here:
                http://electronotes.netfirms.com/Kuo.jpg

                Both of these were done in Matlab but do not use step or any “canning”.

                The numerical integration I wrote uses just the 11 lines:
                VH=zeros(1,5000);
                VB=zeros(1,5000);
                VL=zeros(1,5000);
                dt=0.002
                D=2
                for n=2:5000
                VL(n)=VL(n-1)+VB(n-1)*dt;
                VB(n)=VB(n-1)+VH(n-1)*dt;
                VH(n)=-VL(n-1)-D*VB(n-1)+ 1;
                end
                VN=VH+VL;

                This can be read even if one does not know Matlab – as easy as BASIC. Note that my increment dt=0.002 which is why the time “1” is at 500 on the plot. The second plot does not involve a Matlab computation except for evaluating Kuo’s formula.

                Further, the online calculator:
                http://sim.okawa-denshi.jp/en/RLCbekeisan.htm

                with R=2, C=1, L=1 agrees exactly, as does Matlab’s step (as far as causality is concerned – it indexes time in a way that is not so clear). My experiment agrees. Everyone seems to agree that a physical filter should be causal. So I see my results as having sufficient replication/verification.

                On the other hand, you have not answered as to whether or not anyone else has independently replicated you non-causal response – NOT using your spreadsheet. I assume not.

                The second issue was whether I had correctly interpreted the poles/zeros from your Part III, Fig. 2. I had, and you also gave these to me separately. Your poles are not the nominal case as the poles are at -1/7.6 and the zeros at 1/11 at an angle 3 degrees to the left of the j-omega axis. Allowing for frequency scaling (de-normalization), this is quite similar to the nominal case. Certainly there is no difference that would rotate and offset the step response.

                I hope there is an “ominous ring of truth!” coming through here.

                I suspect that with considerable effort (and time!) on my part, I could get Excel running (my 2007 version is “non-commercial”) and learn to run your spreadsheet. I have no doubt I would get your result. But it would still be wrong, and you, rather than I, would be in a far better position to fix it, were you to become so inclined.

                I appreciate your time here which I hope will be rewarded in the end result.

                Bernie

                00

              • #
                Bernie Hutchins

                David – some progress as I have a better idea where we agree!
                Please see:

                http://electronotes.netfirms.com/Compare.jpg

                The upper left has the pole/zero plot that corresponds to the poles/zeros you give. I have normalized to 1/11 years = 1. So the zeros at just slightly left of +- j and the two poles at about -1.51. This gives a very agreeable match of your frequency response and mine.

                The step responses are still quite different, although they have certain features in common. If we were to take the regions of my graph and yours between vertical 0 and 0.44, they are 180 degree rotations. What differs is the direction of time of course – the causality.

                Note that since there is no zero at s=0, there is a DC response amounting to 0.44 that show up at zero frequency and as well, correctly, as the limit of my step response. So we understand we don’t need to converge on a step level.
                I think we agree on the delay at the bottom of the trough. You get about 1.75 years. I get a normalized 1, which is 11 years, but if we divide 11/(2*pi) that is 1.75.

                In as much as adding a delay shifts your non-causal response to the right, dagger and all, it certainly does not make it causal in the sense that mine occurs naturally.

                00

              • #
                David Evans

                Bernie,

                Is Kuo using one sided or two sided Laplace transforms? Not sure using the one-sided Laplace transform could get to grips with any non-causality, because it ignores times before zero. (Is Matlab doing that too, in some way?) In EE we usually use one-sided transforms. Yes it’s a while since I have used them, but I recall that they were one-sided and we used u(t) to enforce the one-sided-ness, sort of like a built in assumption of causality. The purpose of the one-sided-ness is to deal with functions that increase without limit as t goes to positive infinity.

                In climate work none of the time series (functions) become infinite as t goes to +- infinity, so there is no requirement to use Laplace transforms. We can just use Fourier transforms, and stay strictly on the frequency (j-omega) axis on the complex frequency plane. I suspect this is at the heart of our discrepancy. I wonder if we are tripping over some assumption built into the usual EE approach that we’ve all forgotten about. Usually in EE the behavior of interest repeats over and over, so we’d never notice a small delay that makes a circuit realizable.

                I used Fourier transforms, found the spectrum of a step function, modified it in accordance with the notch filter as shown in my transfer function graph (noting both amplitude and phase), and added the resulting sinusoids (ie inverse-transformed). Conceptually very simple, and no (real) exponentials involved. And the step response comes up non-causal every time (no matter what parameter values I used in the notch filter).

                Yes I parametrized my notch filter and designed it in the complex frequency plane, which is why it is easy to think I know what its poles and zeroes are, but it is only its behavior on the frequency (j-omega) axis that matters for the conclusions I am drawing.

                The trough on your step response: It is at t=1 in the units shown, but what is it in terms of frequencies expressed in cycles per year? I want to be sure it is at around 1/11 cycles per year, after taking normalization into account.

                Can you display the phase of your filter? You have the amplitude matching mine, presuming the trough is in the right place.

                The circuit you built with the op-amps is realizable and therefore causal, but does it have the filter response I had in the transfer function, or a delayed version? Adding the delay to the notch as I did in the model just fiddles with the phases. Similarly any circuit we build is realizable, and so is causal.

                00

              • #
                Bernie Hutchins

                David – a couple of neat results.

                First, your questions. (1) Yes Kuo uses 1-sided Laplace transforms. (I don’t think anyone even imagined non-causal filtering until z-transforms and digital FIR filters a decade or two later.) (2) And the dip in the frequency response is at 1/11 Years as I marked below the dip. All my frequencies are normalized to 1/11 Years. (3) The experimental circuit did not use your poles/zeros but poles/zeros on the same radii. Probably 2/3 the analog music synthesizers in the world offer this state-variable notch. Nothing special.

                The phase you asked about is neat.

                http://electronotes.netfirms.com/DavidPhase.jpg

                I was kind of expecting a sharp whip-around by 180 degrees typical of notch filters, but here the zeros are not on the j-omega axis. It seems to agree very well with your result.

                Another neat thing is that the difference in your step response and mine is a rotation as we have both observed. I also claimed an offset of about 1/2. Please see:

                http://electronotes.netfirms.com/Flip.jpg

                Because your pole/zero choice puts the poles and zeros at different radii (nothing wrong with that), the frequency response “shelves” are unequal, with the low-frequency shelf (DC response) about 0.44 instead of 1.00 (yours too). Hence my step response heads for 0.44 for long time. If we simply time reversed mine to make yours, we would have 0.44 for long negative time, and it could not be made causal with ANY delay. So we can understand that SOMETHING forced that left side end condition to 0 (approximately critically damping to 0). (Mine is zero there by causality.) In consequence, the right side of yours (the flat region) is matched to about 0.44, as is mine. So both ends match (dark red ovals)! This sure sounds like one of those differential equation problems where you get two solutions matching the boundary conditions, but rule one out on physical grounds.

                I am looking at the Fourier transform ideas you suggested.

                Bernie

                00

              • #
                Bernie Hutchins

                David –

                I have looked at your Fourier transform comments some. You said:

                “I used Fourier transforms, found the spectrum of a step function, modified it in accordance with the notch filter as shown in my transfer function graph (noting both amplitude and phase), and added the resulting sinusoids (ie inverse-transformed). Conceptually very simple, and no (real) exponentials involved. And the step response comes up non-causal every time (no matter what parameter values I used in the notch filter).”

                Well, if you took the Fourier transform of a step (Heaviside) I hope you got (1/2)delta(omega) – j/(2*pi*omega) which is what everyone has gotten before (including Bacewell!). I hate the singularities here. You can then also write the transfer function from the poles/zeros as a function of omega and multiply. Taking the inverse FT of the product would seem challenging!

                You speak of “adding the resulting sinusoids” (adding suggest to me a finite number) which suggested to me you might have meant the DFT and not the FT. The problem with the DFT is that there is no step in the DFT world, since everything is periodic. You can very easily notch with the DFT (remove some DFT frequency bins) but the result is “ringing” (Gibbs phenomenon) on BOTH sides of the jump, while yours is perfectly flat on the positive side. So I don’t think this is what you did. In fact I don’t see in any case how you summed sinusoids and got a flat result (in fact with an angular transition).

                You comment on “sinusoids” suggests to me you may be using your OFT, which I have only glanced at.

                All and all, I am pretty comfortable with Laplace transform here.

                Bernie

                00

              • #

                Bernie,

                I certainly don’t follow the math. But I do follow what was done. The reason for the OFT is to get the frequencies with more exactness and a minimum of sidebands that an DFT would generate where the frequency spectrum is not harmonically related to the sampling spectrum.

                00

              • #
                David Evans

                Bernie, Have replied but elevated it to a blog post.

                00

            • #
              Paul Vaughan

              Bernie,
              -
              1. How are you representing climate’s spatial dimensions and spatiotemporal sampling & aggregation issues in your time-only circuit analogies?
              -
              2. Are you familiar with this article?

              Thomson, D.J.; Lanzerotti, L.J.; Vernon, F.L.; Lessard, M.R.; & Smith, L.T.P. (2007). Solar modal structure of the engineering environment. Proceedings of the IEEE 95(5), 1085-1132.
              http://solarmuri.ssl.berkeley.edu/~schuck/public/manuscripts/HOLD/04266886.pdf

              10

              • #
                Bernie Hutchins

                Thanks Paul -

                I don’t understand your first question. Notch filters are analyzed both in frequency and in time using Laplace transform, mostly in frequency of course.

                As for the paper, no I had not seen that, although it was your first reference to the Thompson PNAS that got my attention here. I noted Thompson said:

                “Transfer functions implying noncausal response to changes in solar irradiance imply either that the transfer function is unacceptable, or that there is a problem with the assumed irradiance.”

                Quite so.

                But here all I am debating about is EE – notch filters in the sense of a few components on the workbench. Engineering. Nothing about climate. None of these filters puts out a signal before some input arrives.

                David needs to put together a self-contained ordinary scientific paper, say five pages and five illustrations max, and post it. This is not easy I know.

                But it is no surprise that so many EEs and physicists are apparently just running away.

                02

              • #

                But it is no surprise that so many EEs and physicists are apparently just running away.

                Well I’m not running away (EE). But that is because I have followed the series from the beginning and understand what David is trying to accomplish.

                From what I can tell you have parachuted in and then say, “This makes no sense.” Start at the beginning. Go through all 5K or 10K comments (I have) and then get back with the parts that you still don’t understand.

                Well it may be a bit much to read all the comments. But at least read all of David’s work. Or just give up and go away. It will make no difference to the process. Which is as yet incomplete.

                10

              • #
                Bernie Hutchins

                MSimon said to me in part July 16, 2014 at 3:19 am:
                “From what I can tell you have parachuted in and then say, “This makes no sense.” Start at the beginning. Go through all 5K or 10K comments (I have) and then get back with the parts that you still don’t understand.”

                If you have read all the comments, you know that I have been here a fair amount going back to one of the earliest parts but kind of gave up when David replied to me: “Bernie I’ve no idea and am not going to debug and figure out what you have done exactly.”

                But no problem at the moment as David has given us the poles/zeros:

                “The zeroes are
                2 pi fz (-cos(thetaZ) +- j sin(thetaZ))

                The poles are
                2 pi fp (-cos(thetaP) +- j sin(thetaP))

                The best fit seems to be thetaZ = 87 deg, thetaP = 0 deg, fZ = 1/11, fP = 1/7.6, time in years.s given us the two poles and two zeros of his step response”

                and you are an EE, so the next step is obvious. Why don’t YOU calculate the step response and report back. Take your time. Oh – please show your work ;)

                01

              • #

                Bernie Hutchins
                July 16, 2014 at 2:35 am

                If you have read all the comments, you know that I have been here a fair amount going back to one of the earliest parts but kind of gave up…

                That explains why I don’t remember you.

                If you have been following along you will know that it is possible to get non-causal filters from analysis. David has shown that to make the filter from his analysis work it needs a delay of 10 to 20 years.

                The next step is to try and figure out what signal from the sun (or the rest of the system) has a delay in that range.

                If that explanation does not clear up your difficulties may I suggest you give up permanently.

                BTW David has said from quite early on that he may be headed into a blind alley. So that in fact may be your difficulty. In that case may I suggest you give up permanently.

                I am not at this time convinced that it is a blind alley. I may come to that eventually. Or not.

                10

              • #
                Bernie Hutchins

                MSimon – thanks for your reply of July 16, 2014 at 10:20 am

                I noticed that you apparently evaded my “trap” of suggesting you do the step response yourself ;) . Or are you working on it? Most EEs have not completely forgotten how to do this, but remember that it was really assigned by instructors mainly as a form of torture!

                You do remember how to form an s-domain transfer function from the poles and zeros. Then you multiply by 1/s and do the inverse LT, usually by partial fractions. Pitfalls all over the place. Numerical integration is far less tedious – that’s what I did. And it agreed with experiment.

                David gave us the poles/zeros, and his is accordingly a perfectly ORDINARY notch (two stable poles, and two zeros near or on the j-omega axis), and the step response is of course causal. All and all – pretty simple to get right.

                01

              • #

                Bernie Hutchins
                July 16, 2014 at 11:11 am

                If I needed to do that kind of analysis I’d plug it into a tool. My area of expertise lies elsewhere.

                Since I trust David I didn’t check his work. I’m interested in the bigger picture – what is the cause for the delay.

                If I actually had a need to do that sort of work I’d be able to do it after a week or two of study. I never got a degree in engineering. Or anything else. Worked my way up to aerospace engineer from bench technician.

                With the help of a few friends I taught myself Polywell Fusion.

                I’m generally two weeks away from anything I need to know. Polywell was rougher. It took me on the order of two or three months. But I learned it well enough to get a thank you from Dr. Bussard (the inventor) just before he died.

                20

              • #
                Bernie Hutchins

                MSimon – good for you. Credentials have a value going as e^-t where t is in years. Engineering needs more like you.

                Believe me, we have had many “excellent students” (honors ribbon at graduation) who would not have received my recommendation. On the other hand we had a student who was near the bottom of the class who could not even get an on-campus interview. We put him on the bus for Boston with a note (to a friend) that said don’t worry about this guys grades. He came back with the job. HE got the job himself – we just got him in the door. Later the company president told me “Bernie – you know that hiring Mike saved us a year and a half of engineering man hours”.

                21

              • #
                Paul Vaughan

                Bernie,
                -
                A) It has been clear for some time that you want David to say “causal with lag”. Protracted quibbling over semantics isn’t advancing the discussion.
                -
                B) I asked you the following:

                “How are you representing climate’s spatial dimensions and spatiotemporal sampling & aggregation issues in your time-only circuit analogies?”

                Your (lack of) response confirmed (as expected) the suspicions that led me to inquire.

                You haven’t yet established sufficient foundations in climate circulation, sampling, & aggregation to competently apply your time series skills to advanced solar-climate exploration.

                Towards adequate foundations:

                • concise overview of heat engines
                = p.433 [pdf p.10] here:

                Sidorenkov, N.S. (2005). Physics of the Earth’s rotation instabilities. Astronomical and Astrophysical Transactions 24(5), 425-439.

                • elaboration on heat engines
                = section 8.7 (begins on p.175 [pdf p.189]) here:

                Sidorenkov, N.S. (2009). The Interaction Between Earth’s Rotation and Geophysical Processes. Wiley.

                00

              • #
                Bernie Hutchins

                Paul Vaughan said July 16, 2014 at 1:18 pm

                “Bernie,
                A) It has been clear for some time that you want David to say “causal with lag”. Protracted quibbling over semantics isn’t advancing the discussion.”

                Nope – you are wrong. It is not even a matter of semantics. I want him to STOP saying that all notch filters are non-causal, because that’s exactly WRONG. In fact, all notch filters are causal. It does not “advancing the discussion” to misunderstand and manipulate terms – it just gives the impression, prima facie, that you don’t know what you are talking about. Causal MEANS there is a lag (phase shifts) of course. If anyone wants to add a delay to any model, that’s fine – just do it. Delay with superposition has spectral implications (such as comb filtering – periodic notching in fact). But don’t add a delay under pretense that it “fixes” a non-causal situation. A delay would not in fact fix a step response that is time reversed, and exponential.

                Paul also said:
                “B) I asked you the following: “How are you representing climate’s spatial dimensions and spatiotemporal sampling & aggregation issues in your time-only circuit analogies?”

                I DID respond to this at July 16, 2014 at 2:35 am above. If I was not clear, I specifically objected to your suggestion of “time-only circuit analogies”. My analysis dealt with time-domain step-response because David was talking about step response. (Most interest in notch filtering involves, in fact, the frequency domain primarily.) And while I am talking about what are classically called “Analog Circuits” these are actual electrical circuits in their own right, and not analogies FOR climate or for anything. (The term “analog” comes from the SIGNAL, not the SYSTEM. Is this new to you?) The electrical notch filter is decades (many decades) old. And I clearly said in my 2:35 response that I was ONLY discussing EE matters. Further to this point…..

                Paul also said:
                “Your (lack of) response confirmed (as expected) the suspicions that led me to inquire. You haven’t yet established sufficient foundations in climate circulation, sampling, & aggregation to competently apply your time series skills to advanced solar-climate exploration……….”

                Where did I, in the slightest, suggest that my comments extended even a millimeter beyond electrical engineering filter analysis and design? Where? Strawman! Obfuscation! Knock it off.

                We are talking about David’s misuse, in the engineering and physics sense, of causality.

                01

              • #

                Bernie Hutchins
                July 16, 2014 at 2:49 pm

                It is obvious that analysis can give you a non-causal filter. David did that analysis. Evidently he got a non-causal filter. The next question is – what would make the filter causal? It turns out that a delay on the order of one sunspot cycle gives a causal filter. So the next question –> what other output of the sun is delayed one cycle from TSI and also can affect TSI?

                I don’t understand your quibble. Yes a real filter MUST be causal. If you START with that filter. That is not where David started.

                The above is why I asked if you had been following along from the beginning.

                David also admits that this all may lead to a dead end. I want to see it all play out before judging. The work – in so far as it has been presented on the ‘net is not done.

                10

              • #
                Paul Vaughan

                Bernie,

                Something to consider:

                Hinnov, Linda A. (2013). Cyclostratigraphy and its revolutionizing applications in the earth and planetary sciences. Geological Society of America Bulletin 125(11-12), 1703-1734.
                -
                Rather than connect the dots to realize Thomson’s serious error and then proceed to solve the solar-climate puzzle the way you think it should be done, you instead opt to do nothing but repeat yourself.
                -
                The time I have this morning is better spent doing EOF analyses of ICOADS wind fields.
                -
                It’s clear you’re not here to advance the discussion.

                10

              • #
                Bernie Hutchins

                MSimon said in part July 16, 2014 at 3:46 pm:
                “It is obvious that analysis can give you a non-causal filter. David did that analysis. Evidently he got a non-causal filter. ……I don’t understand your quibble. Yes a real filter MUST be causal. If you START with that filter. That is not where David started.”

                The “start” was apparently Fig. 2 of Part III of David’s presentation. There he shows a step response that is INCONSISTENT with the poles/zeros annotated on the SAME graph, and subsequently he posted these poles/zeros to me. As I have said, the Fig. 2 is similar to the correct step response, but rotated 180 degrees (not just time reversed) and offset by -0.5. In as much as the step response must be causal for the poles/zeros he supplied, I conclude he made some simple math errors, and MY result is consistent with Matlab’s step function, with the online R-L-C calculator, and with experiment, as well as a century or more of understanding linear circuits.

                It may be useful as well for me to post the Matlab code I wrote myself and ran to produce the figure I posted, JUST to indicate how short and simple it is. Don’t blink!

                VH=zeros(1,5000);
                VB=zeros(1,5000);
                VL=zeros(1,5000);
                dt=0.002
                D=2
                for n=2:5000
                VL(n)=VL(n-1)+VB(n-1)*dt;
                VB(n)=VB(n-1)+VH(n-1)*dt;
                VH(n)=-VL(n-1)-D*VB(n-1)+ 1;
                end
                VN=VH+VL;

                This mistaken non-causality is hardly a “quibble” as he uses it to justify a delay added to his model.

                11

              • #
                Bernie Hutchins

                Paul –

                Once again you confuse me with someone who has made claims regarding climate models while I have not done anything beyond EE circuit theory. Again your strawman.

                I’m not sure how one “advances a discussion” by repeatedly ignoring the error in step 1.

                11

          • #
            Paul Vaughan

            David,
            `
            Look at how Thomson’s (1995) work is framed here:
            http://www.heatisonline.org/contentserver/objecthandlers/index.cfm?id=3458&method=full

            The trail doesn’t end with the 1995 & 1997 articles.

            Something dreadfully simple was overlooked in a 2007 report. What I find most telling about developments since 2009 is what was omitted from a 2013 report.

            If people are inclined to start exploring the trail independently, I’ve provided several links from comments here:
            http://wattsupwiththat.com/2014/06/27/mapping-the-hottest-day-of-the-year-in-the-usa/

            Paid work demands my attention now, but I look forward to an evolving conversation as time & other obligations permit.

            Best Regards

            10

    • #
      Rereke Whakaaro

      You are cherry picking your quotations Chester, see my comment #34.5

      120

  • #
    Michael Larkin

    The download appears okay in my copy of Excel 2007, but I had to enable macros for the buttons to work (not as if I have much of a clue what to do with the spreadsheet, mind :-) ).

    I tested the free Excel Viewer, but the buttons don’t work, full stop. So you have to scroll to see graphs, etc. But it might well be of some use if you don’t want to buy excel or play with the data.

    The viewer I tested can be downloaded here:

    http://www.microsoft.com/en-us/download/details.aspx?id=10

    30

  • #
    Travis Casey

    @Chester – Of course that is what the GCMs do is assume that the main driver of Earth’s climate is GHG. Dr. Evans has been clear from the beginning that he was making a “solar assumption”

    110

    • #
      Chester

      Of course that is what the GCMs do is assume that the main driver of Earth’s climate is GHG.

      Give me just one example and explanation that demonstrates the validity of this claim.

      130

      • #

        Kyoto. Copenhagen.

        That is two. There are more.

        220

      • #
        gnomish

        the ipcc made the basis of their co2 driver argument: ‘we can’t explain it any other way’
        if you weren’t sleeping through that, maybe you should catch up now – on your sleep, i mean.

        190

      • #
        Rereke Whakaaro

        See my comment #34.5

        70

      • #
        tom0mason

        Chester
        “Of course that is what the GCMs do is assume that the main driver of Earth’s climate is GHG.”

        “Give me just one example and explanation that demonstrates the validity of this claim.”

        The example is UN-IPCC.
        The explaination is UN-IPCC.
        Look-up why the UN-IPCC came into being and their mandate. Who ultimately directs the work on the GCMs – the UN-IPCC.

        210

      • #
        Duster

        The sun is treated as a constant because of the presumed near invariance in output. The climate however is not and never was even close to invariant, so logically whatever “drives” it is not the sun. The sun powers the climate, the drivers “steer” it. The GHG debate is about whether CO2′s feet can reach the pedals while holding the steering wheel. Geological evidence says that CO2 is too short. I suspect that what’s needed is the equivalent of a climate Reynolds number. That postulates that climate variability (change to the unwashed) is turbulence and is not “driven” by anything in particular.

        10

    • #
      Rereke Whakaaro

      Actually he made no such assumption. He suspected the sun may have an influence, as we all did, but he did not assume. Hence the use of Fourier Analysis to isolate the various frequency signals.

      230

  • #
    Brad

    Jo and David, can you put up a counter so we can watch how many times it is downloaded?
    Just curious…

    140

  • #

    David / Jo – Have you folks taken a look at Google Documents? It is widely available, stores in the cloud, and has a spreadsheet capability. Would require you to operate out of Google Drive. Note that Google Sheets does not run VBA. However, there are some extensions called Google Apps Scripts that fill the same function. A couple links follow. Cheers -

    http://www.google.ca/sheets/about/

    https://developers.google.com/apps-script/overview

    22

  • #
    dp

    I’m using MS Office 2011 for Mac. It loads find and the several tests I ran all completed. The solar model reproduced the notch. Clearly a lot of effort has gone into the spreadsheet. Time will tell if it can use data that passes the stringent Svalgaard sniff test.

    50

    • #
      CC Squid

      Svalgaard has already posted that it is not designed the way he wanted it to be. It does not allow him to enter the data that he would enter. Whine!!!

      150

      • #
        the Griss

        That’s the beauty of the spreadsheet.

        If he is clever enough he can redesign it so it does allow him to.

        He just has to do a bit of work, instead of verbal regurge.

        171

  • #
    Eliza

    Due to the fact that most surface temp data may be “not true” would this model run against any temp (imaginary or true) curve? Perhaps the CET raw data is the best?

    60

  • #
    Don Jordan

    Ha! I purchased a panoramic monitor (21:9 aspect ratio) for my system. This monitor was made for a big spreadsheet like this! :D

    LG 34UM95-P monitor review

    30

  • #
    Brad

    Drive-bys from LS and SP over on a post at wuwt…. Very sad.
    And why does Anthony say the timing couldn’t be worse? That is very negative and implies potential bias on his part?

    130

  • #
    Don Jordan

    A bit off-topic, but…

    I just purchased a new panoramic monitor (34″ size and 21:9 aspect ratio) for my computer. This monitor was made for a big spreadsheet like this since I don’t have to scroll horizontally 99% of the time to see everything. :D

    LG 34UM95-P monitor review

    20

  • #
    Philip

    FYI: Mac Word 2011 — loads and appears to work fine.

    Download was a few seconds, loading into Excel took maybe 10 seconds.

    80

    • #
      Rud Istvan

      Ditto. Code shows and can be modified. Cells show underlying arguments. Can be cross checked and modified per standard excel. Runs and graphs. I have not had time today to wring it out beyond the first superficial hour, but looks good to go.
      Simply assembling all the data and references in one place is an enormous contribution.
      Well done.

      100

  • #
    Matt Thompson

    Great stuff David.

    80

  • #
    Pat K

    … from my viewpoint the timing of this could not be worse, given that a number of people including myself are in the middle of the ICCC9 conference in Las Vegas.

    Anthony could have phrased that better. Surely WUWT is not implying that the release should have been preceded by a comprehensive investigation into various people’s travel arrangements.

    151

    • #
      Ross

      Yes Pat K. absolutely pathetic comments from Watts.

      Equally pathetic from Willis E –the poor guy cannot get it to run so it’s not science and it’s just an advertisement according to the self proclaimed oracle. He starts his comment with a sort of congratulations and ends with his advertisement comment –so the start wasn’t very sincere.

      As Brad comments above LS is up to his usual drivel.

      I hope David and Jo just ignore them from now on.

      I think the big issue here is David and Jo have made such a radical departure from the norm, in their approach to the issue ( both the science/engineering approach as simply outlined above by RW and their approach to the delivery of the information ) that even the so called well qualified guys like WE, LS and AW cannot get their heads around the basics so their view is ‘coloured” from the start.

      230

      • #
        Brad

        Agreed, some posters over at WUWT have apparently decided to have a turf war over this.
        They can’t seem to realize the world’s economy and the lives of millions of people are at stake. Very sad…

        140

      • #

        In a thread over at WUWT Willis dismissed what he referred to as (approximately) “EE understanding”. Leif on several threads here has made similar remarks. As if EE was a useless and reviled occupation and had no useful tools for analysis of difficult problems.

        I’d call it unprofessional jealousy.

        210

      • #
        the Griss

        perhaps Willis should contact Phil Jones for some help. !?

        130

    • #
      Mike Jowsey

      Yes, Pat – on the one hand Willis is “begging” David to release the data and condemning him for not doing so, then when David does release it he is condemned by Anthony for poor timing. Wuwt??? And by the way, Anthony, you have known about this project for a long time, so sorry that you weren’t consulted as to the perfect timing in your world, but you have had plenty of opportunity to investigate the claims and projections before now.

      Many comments over there (e.g Pamela Gray) indicate that the comments are first-impression knee-jerks rather than considered opinions having researched the whole series leading up to the Excel release. I tell you, WUWT has slipped way down the ladder in my estimation. Really sad to see.

      140

  • #
    Richard

    Trying to download it to my Xbox 360. Nothing doing.

    60

  • #
    Brad

    OMG…
    Posted at WUWT by Willis:

    Willis Eschenbach commented on Solar Notch-Delay Model Released.
    in response to Anthony Watts:
    Readers may recall the contentious discussions that occurred on this thread a couple of weeks back. Both Willis Eschenbach and Dr. Leif Svalgaard were quite combative over the fact that the model data had not been released. But that aside, there is good news. David Archibald writes in to tell us that the model has […]
    Brad says:
    July 8, 2014 at 2:19 pm
    Can anyone explain why all this discussion is occurring here in WUWT and not at the publishers website? I would call that snubbing (dissing, shunning, etc), which is typical of drive-by snipers.
    Thanks, Brad. First, we discuss a host of things here rather than on the “publisher’s website” in order to expose the ideas to a much wider and larger variety of readers. Alexa ranks their site as number 82,590 in the world, while WUWT is number 9,370 … so we have a much, much larger audience. You should be thanking WUWT for bringing their ideas out to the wider scientific world, not busting us.
    Second, for me, I got frustrated over there because they hadn’t published either the model or the results of the out-of-sample tests which they said were already done. By doing so, they gained a host of adherents among the credulati. These folks had already become true believers in the model despite the lack of either the model or any testing, and those are the worst kind of believers. As a result of their belief being grounded in … well … nothing, they are most unreceptive to even the slightest criticism of the model.
    So I gave up, and told David and Jo I’d return when they published their model and their out-of-sample test results.
    At present, they’ve published only part of their model. They have not published the part which actually fits the arbitrary parameters. In addition, they have not published their out-of-sample test results. So we are prevented from doing the very simplest of tests on their model, the out-of-sample tests … and they have not published their results from the out-of-sample tests despite the fact that they have already been done.
    When they do publish the tests and the rest of their model, I’ll return and discuss it there. Until then … I’m here.
    w.

    80

    • #

      they are most unreceptive to even the slightest criticism of the model.

      IMO Willis is confusing some of the commenters here with David and Jo. Both D and J have been most gracious in their response to real problems or even some imagined ones. The are in fact looking for useful criticism in order to improve their work. Good engineering practice.

      In any case most of the imagined problems have been taken care of in the comments. Sometimes ungraciously. But none the less addressed by the army of EEs now flocking to this site. I have expressed more than once how much I enjoy being in the company of so many EEs.

      250

    • #
      Spotted Reptile

      Reading between the lines here from Willis:

      I have a sneaking suspicion that David and Jo might be on to something big, but I cannot admit this as it would severely undermine my strides towards greatness and credibility despite some rather incorrect conclusions I jumped to a few years ago and have been trying desperately to cover with a blanket of oblivion ever since, so I must trumpet loudly about something irrelevant in order to a) distract everyone from the actual data and its implications and b) make myself appear above the hype and with such a high standard of critical thinking that I cannot be ‘seduced’ by someone else’s theory. I must also throw about some rude and boorish epithets such as ‘credulati’ and make out that everyone who praises David and Jo is either stupid or in on a vast conspiracy against me and others like me.
      Therefore I will latch on to something that has very little significance and hype it up so that it becomes more important than the theory, so I can say that the theory has huge holes in it when really all I am doing is distracting from the fact that I actually believe deep down that the theory has credence but to admit so would make me wrong so that is not going to happen and I hope everyone still reveres me because that is everything to me.
      Oh, and site traffic-measuring is really relevant too. Just like the other measuring thing.

      350

    • #
      dp

      Willis and the Whiners, the whinge and dance act of WE, LS, SM, and others at WUWT are failing miserably the reasonable customer test. One does not ask for today what the vendor has not said will be available today. One does not ask on Monday what is advertised for Tuesday. One does not demand today what has been promised for another day. Three ways to say it – you get it when the vendor makes it available.

      Possibly the most pathetic and self-damning thing Willy [snip] has said so far is a direct comparison of DE and JN with Michael Mann regards releasing code and data. MM has refused and leaves no doubt he will never comply and I am not alone believing he is incapable of complying. DE and JN have been honest about how the information will become available and are living up to their claims. For his effort Willis wins the coveted Imperious Ass award for best display of boorish manners in a public forum.

      The only possible way to remove that stain from his record is an honest open apology. Not a cute mea culpa followed by additional criticism as he has done in the past – a stand-alone apology that carries no other message.

      I don’t expect it and regards the stain – it’s lost among the other stains that tarnish his reputation.

      110

  • #
    the Griss

    Only just starting to look at it. On Excel 2007.

    I seem to be missing some drop down menus where the “How to run solar model” says there should be some, (have asked Jo/David about this via email)

    But it draws the graphs etc when I press compute all, so something is working.

    30

  • #
    Mikky

    Can I suggest a separation of the “observational” part of the work from the “model” part.

    The observational part is pure science, the claim of finding a notch, i.e. (for example) a measured response
    of global temperatures to 6-year variations in TSI. Nobody has ever found that before AFAIK.
    Will be a tough sell though, as there is not much of a 6-year signal in the input.

    15

    • #

      Of course it hasn’t been found before. To my knowledge no one has ever done a Bode plot of the climate system before.

      ======================

      Sometimes when you look at things a different way you see things others have not previously seen.

      EEs are well enough educated in David’s methods to “get” what David is doing. It seems that most everybody outside EE is at a loss. That knowledge is, to me, even more amazing than the model. Most people in developed countries depend on the work of EEs and yet even some very bright people are not just light in the understanding (that is reasonable) but they don’t even have a clue. And worse they seem uninterested in getting one.

      And you don’t have to pay much for a clue. The information is out there at any level you want to approach it from. Just do a www search. Here are some hints:

      Bode Plot
      Lag
      Delay
      Integrator
      Filter
      Low Pass Filter
      Frequency Selective Filter

      That should do for a start. I’m not even going to go into instrument design which Leif – at the level of understanding he would like to claim – needs a host of graduate courses in. And Leif. Just in case you are dropping in here – it was a trick question. Did you find the trick?

      180

      • #
        Rud Istvan

        Well done post. We use Bode plots extensively for my energy storage materials business. and all that delay lag stuff we end up calling power density, also using Nyquist plots which are an inversion of the imaginary part of the complex number describing the electrical response (really only phase shift, except we work in a pulse mode DC world where the RC time constant ‘tau’ is a big deal.
        Even better done was your trick. Bet you good old American dollars to donuts LS will not respond. He seems to get quite nasty when swimming out of his depth.

        100

        • #
          the Griss

          “when swimming out of his depth”

          The first step is realising and accepting that he is out of depth.

          Once he does that, he can learn to dog-paddle.

          80

  • #

    At present, they’ve published only part of their model. They have not published the part which actually fits the arbitrary parameters. In addition, they have not published their out-of-sample test results. So we are prevented from doing the very simplest of tests on their model, the out-of-sample tests … and they have not published their results from the out-of-sample tests despite the fact that they have already been done.

    There is so much wrong with that it is difficult to know where to start.

    They have not published the part which actually fits the arbitrary parameters.

    The tests may be ad hoc and just slapped together in a way that requires a LOT of hand holding. It may not in fact be a deconstructor. It could be just plugging things in and seeing what works. I have done that often enough. Sometimes it is the easiest way to do things. Or it gives you insights. For a problem you intend to solve only once it is not a bad way to go.

    So what could the WU Company do to resolve the issue? Well they could add to the work by making their own parameter extractor.

    So we are prevented from doing the very simplest of tests on their model, the out-of-sample tests … and they have not published their results from the out-of-sample tests despite the fact that they have already been done.

    In fact no one is prevented from doing any tests on the model. Or changing the model. Or using different input data as a check.

    In any case there will be enough people doing testing. And then there are the predictions. If they hold up there may be something to the model.

    =======================

    And that bit about Rank. It is in fact rank. For this kind of work at this stage it is not the numbers in the audience but the quality of the audience. And from my observations here the audience is of a higher quality than you find at WUWT. At least in relation to the problem being discussed. I remember discussing this work in its early stage of release over at WU with Willis. There was another EE on the thread. We both gave up in disgust.

    Willis doesn’t get “EE thinking”. His loss.

    290

    • #
      Rereke Whakaaro

      The only test of a prediction that matters is the future.

      The artifacts that Willis drones on about, are just that, artifacts. They are only necessary if a) you are desperate to get your views accepted before the fact, or b) you are searching for irrelevant measures that you can uses to debunk what you see as a competitor.

      David has made some predictions about global warming some two to four years from now. As far as I am aware, Willis has studiously avoided taking the step of making any verifiable predictions at all.

      I am betting my farm, that David is correct.

      90

  • #
    Alan Poirier

    Great spreadsheet, David, Jo. Easy to work with.

    130

  • #
    Scute

    I just downloaded this to my iPod Touch 5 (gasp!). However, it swallowed it whole and is now finding it hard to chew on. I’m getting data tables but they seize up sometimes when scrolling. Not sure if I could run the model.

    I went straight for the ocean heat content tables (and TSI but that was the one that seized up). Having all that Argo data in one place is a godsend. I was looking for the 2011-14 Argo data yesterday and it was hard to find. Also, the most-cited papers based on Argo seem to be from 2011 and before. Now we can look at the more recent patterns.

    David, I couldn’t see a column for implied radiative imbalance based on change in OHC as all the papers I’ve read do. Is that possible? I have done rough calculations before but it’s laborious and I’m probably missing some parameter. I just would have thought that OHC will be the first to show up the radiative imbalance swinging negative. It would be nice to know what the wattage is as it approaches zero.

    20

  • #
    Mike Singleton

    I chuckle at the comments about number of parameters and wiggling elephant trunks. Do those making that statement not understand the difference between a designed laboratory experiment, where one has full control of all variables, versus the real world where there is no control and the number of parameters are pre-determined by nature.

    It’s quite laughable.

    Kudos to David and Jo.

    260

    • #
      Rud Istvan

      Agree. If you research the actual history of his quip, he was talking about the well known problem of degrees of freedom in statistics. A 10 variable equation fit to ten sets of data will have an r2 of 1 (perfect) and also be meaningless. I have actually encountered such in real life! But a 10 variable equation fit to, say 200 sets of data (here years of all the necessary stuff) has a very high degree of freedom and is probably statistically reliable. (We overlook here confounding problems like autocorrelation and heteroskedasticity).
      Superficial whiningbynthose who have bothered to yet read the accompanying excepts from the long not DE has promised is coming, which contains some delightful surprises for some of his harshest critics, specifically WE. Well done table turn, which WE over at WUWT is as yer clueless about. This might be fun to watch.

      80

      • #
        Rud Istvan

        His=John VonNeumann, of course. Said IIRC in 1951. Father of game theory, linear programming, and the heirarchical CPU computer architecture.

        10

    • #
      Raven

      I chuckle at the comments about number of parameters and wiggling elephant trunks.

      It’s going to be even funnier if it turns out that there’s no actual elephant. ;-)

      40

  • #
    mesoman

    It’s clear that Leif and Willis simply do not want this to succeed because it will upstage them (Willis, after confidently proclaiming no 11 year solar effect, and Leif, who fancies himself the world’s gatekeeper to solar science). They immediately launched personal attacks when the first bit of info was released (big red flag), Leif repeatedly ignored corrections to his assertions and set up about 150 strawmen to knock down, and now that the model has been released they move on to attacking its complexity, layout, lack of this, lack of that, etc.

    They’ve been nothing but bitter about this project from the beginning and that is a huge indicator of where they’re coming from–and it’s not really about science, despite their claims.

    290

    • #

      Nothing?
      http://daedalearth.wordpress.com/2014/06/27/an-11-year-solar-signal-in-the-atmosphere/

      Of course an effect is in many datasets it’s just that it is very minor.

      I don’t at present agree with David but I watch to see where things go.

      60

      • #
        mesoman

        I was referring to Willis’s fairly recent WUWT post where he essentially blew off any 11 year solar effect and brashly challenged anyone to prove him wrong…then along came Dr. Evans. It was coincidental timing, but suddenly Willis had someone coming to the opposite conclusion, as he dared them to do.

        150

        • #
          mesoman

          By the way, I have no idea if the solar notch model is correct, so I can’t defend it (I’m an interested observer and I’m eager to see what others find). I’m only analyzing the behavior of two people that really surprised me after reading their material and comments for years.

          160

        • #

          I have a short bit (part of a longer comment) about my run in with Willis on that WUWT thread here:

          http://joannenova.com.au/2014/07/big-news-ix-the-model/#comment-1505381

          I had read the Willis post and thought he would be excited about the news of David’s discovery which in fact corroborated Willis’ findings. And in fact David gave Willis props for the post as it confirmed his effort.

          Instead of being happy about his contribution to David’s work he got (as my English friends like to say) stroppy. (I said to Willis something to the effect – “have you seen this” – he had – a discussion ensued)

          60

      • #
        David Evans

        Hi Tim! Yes, here are 11 year signals in all sorts of climate variables, including temperature records — but not, as it happens, in the global mean surface air temps.

        As mentioned in comments on earlier posts, presumably force X and TSI affect different variables by different amounts, but come pretty close to cancelling for global surface temps. We’ll be doing a post soon on crowd sourcing force X, and one of the clues has to be what has an 11 year signal and what doesn’t.

        Btw, thanks for the earlier discussions on transforms. For those who don’t know, Tim, who writes often on Tallbloke’s Talkshop, has some very advanced software for analyzing time series into sinusoid, i.e. Fourier analysis. Tim has been focusing on finding the main few cycles in datasets.

        150

        • #
          the Griss

          The fact that the 11 year cycle appears in quite a few local or regional and other climate variable shows that cyclic solar effects DO occur at these levels.

          To think that they don’t occur also at the global level, as some people seem to think, is quite bizarre.

          This points to the only possibility, that being that the cyclic solar effects are “well smoothed” at the global level.

          This implies delays, storage and release etc etc and quite probably many different delays/storages etc at different time spans that culminate in a reasonably uniform smoothing at the global level.

          I doubt we will ever know all of these delay mechanisms, and certainly it will take many years to quantify and time them inside any climate model.

          That is what makes David’s approach rather unique.

          The many different mechanisms aren’t individually all that important, its their overall end effect that is.

          Working outside->in, rather than inside -> out

          And David’s method might just have shown a method to figure this out.

          Time will tell.

          120

        • #
          Richard C (NZ)

          David #56.1.2

          >”Yes, here are 11 year signals in all sorts of climate variables, including temperature records — but not, as it happens, in the global mean surface air temps.”

          Not? Here’s an analysis of GISTEMP:

          ‘On the relationship between global, hemispheric and latitudinal averaged air surface temperature (GISS time series) and solar activity’

          Souza Echer et al (2011)

          http://www.researchgate.net/publication/233853522_On_the_relationship_between_global_hemispheric_and_latitudinal_averaged_air_surface_temperature_%28GISS_time_series%29_and_solar_activity/file/9fcfd50c209b0f2370.pdf

          Click “View” or “Download”, see Table 2, page 4. The 11-yr solar signal is clearly evident in globally averaged mean surface temperature, GISTEMP in this case.

          Similarly, the tropospheric NCEP analyses of Coughlan and Tung 2004 (see Tim Channon’s post above) and subsequently by Zhou and Tung 2012, includes near-sfc temperature (1000 hPa):

          ‘Observed Tropospheric Temperature Response to 11-yr Solar Cycle and What It Reveals about Mechanisms’

          Zhou and Tung (2012)

          ABSTRACT
          Using 54 yr of NCEP reanalysis global data from 1000 to 10 hPa, this study establishes the existence and the statistical significance of the zonal-mean temperature response to the 11-yr solar cycle throughout the troposphere and parts of the lower stratosphere

          http://depts.washington.edu/amath/old_website/research/articles/Tung/journals/Zhou_and_Tung_2013_solar.pdf

          These papers have been posted numerous times (well before Tim) but just get a brushoff. You can’t ignore the evidence for ever David as long as these papers exist in the literature. Neither can you dismiss NCEP because it doesn’t meet your criteria of globally averaged series only. You are simply ignoring the evidence when and where it occurs around the globe at varying lengths of cycle..

          00

          • #
            David Evans

            We checked GISTEMP. No, a strong 11 year signal was not found — it was no higher than surrounding frequencies.

            First paper (Souza), table 2 on page 4 as recommended, global data, lists the periods in years of stronger signals found as: “2–2.8;3.7–6.6;7.7;8.3;9.1;10.4;11.5;20.6;26.3;29.6 and 65″ 11 years isn’t exactly standing out, like it does in the TSI data. Seems compatible with Figure 4 here.

            Zhou and Tung are looking at zonal temperatures, not global.

            30

            • #
              Richard C (NZ)

              David,

              >”11 years isn’t exactly standing out,” [in GISTEMP]

              Of course it isn’t and it wont. The “11-yr” cycle is NOT an exact 11-yr cycle. The range is 9 – 13.7 yrs. Souza Echer et al found low correlation 1880 to 1950 but higher correlation after that These are the SC lengths in regard to Souza Echera et al:

              SC, Duration
              1, 11.3
              2, 9.0
              3, 9.3
              4, 13.7
              5, 12.6
              6, 12.4
              7, 10.5
              8, 9.8
              9, 12.4
              10, 11.3
              11, 11.8

              Souza Echer study begins 1880.

              12, 11.3 (1878 start)
              13, 11.9
              14, 11.5
              15, 10.0
              16, 10.1
              17, 10.4
              18, 10.2
              19, 10.5 (1954 start)
              20, 11.7
              21, 10.3
              22, 9.7
              23, 11.7

              Obviously the durations are either side of the 11.1 yr mean 1755 – 2008 so no surprise that Souza Echer et al found “11-yr” SC periodicity at 10.4 yrs and 11.5 years either side of the 10.78 mean 1878 – 2008.

              I repeat, the “11-yr” cycle is NOT an 11 year cycle.

              >”Zhou and Tung are looking at zonal temperatures, not global.”

              There’s the brushoff again but not so fast, they explicitly state in 3. Results on page 2 (my emphasis):

              Troposphere

              The existence of the solar cycle signal is established
              for the entire troposphere for the first time at the 95%
              confidence level
              in the annual mean

              The “entire troposphere” is GLOBAL David.

              And see page 3:

              FIG. 1. (left) CMD patterns as a function of latitude of the annual-mean zonal-mean NCEP temperature, and (middle) the time series obtained by projecting the temperature data onto this pattern with the TSI index superimposed. The correlation coefficient r between these two time series is indicated in (middle). (right) The statistical confidence level using the Monte Carlo– enerated synthetic data and their correlation coefficients with the TSI (the null distribution) are shown at each pressure level.

              http://depts.washington.edu/amath/old_website/research/articles/Tung/journals/Zhou_and_Tung_2013_solar.pdf

              The “zones” are latitude and pressure. 1000 hPa corresponds to near-surface and you can see the 1958 – 2011 time series for the surface “zone” in the bottom panel Fig 1.

              00

              • #
                David Evans

                Are not Souza Echer et al consistent with Figure 4 here?

                We are looking at global surface air temperatures. Whole of troposphere temperatures show the 11 year response because the temperatures off the surface show the 11 year signal, as both of us have pointed out before.

                30

              • #
                Richard C (NZ)

                David #56.1.2.2.1

                >”Are not Souza Echer et al consistent with Figure 4 here?”

                Hard to tell GISTEMP and Moberg (2005) apart in your graph due to the colour coding identical.

                But “consistent”? I don’t think so. They state specific periodicity at 10.4 yrs and 11.5 yrs either side of the 10.78 mean. What specific periodicity about 11 yrs does your analysis of GISTEMP return to be “consistent” with Souza Echer et al i.e. what are your datapoints for GISTEMP in Figure 4?

                >”We are looking at global surface air temperatures.”

                Yes, exactly as I pointed out above in regard to Zhou and Tung (2012). Their 1000 hPa “zone” is nearest to GLOBAL SURFACE AIR TEMPERATURE., not mid-troposphere or tropopause – but at near-surface.

                Standard temperature and pressure:

                1. *Standard atmospheric pressure of 1013.25 mb absolute at sea level
                2. *Standard temperature of 288.15 K (15°C)

                1000 hPa = 1000 mb

                >”Whole of troposphere temperatures show the 11 year response….”

                Well yes it does but that is not the issue. The issue is that 3 papers find the “11-yr” solar signal in global surface air temperatures which you apparently don’t accept because you didn’t.

                00

        • #
          Duster

          Dr. Evans, are the signals lagged differently between different the different variables?

          00

      • #
        Paul Vaughan

        well-known, interesting paper based on reanalysis data

        30

      • #
        Richard C (NZ)

        Tim Channon #56.1

        >”Nothing? [link to Coughlan and Tung (2004) post]”

        This is the paper (in full at the link below):

        ‘Eleven-year solar cycle signal throughout the lower atmosphere’

        K. Coughlin and K. K. Tung (2004)

        Abstract

        [1] A statistically significant atmospheric signal, which represents the influence of solar radiation changes on our climate, is found in global data (1958–2003). Using a nonlinear, nonstationary time series analysis, called empirical mode decomposition, it is shown that atmospheric temperatures and geopotential heights are composed of five global oscillations and a trend. The fourth mode is synchronized with the 11-year solar flux almost everywhere in the lower atmosphere. Statistical tests show that this signal is different from noise, indicating that there is enhanced warming in the troposphere during times of increased solar radiation.

        1. Introduction

        [4] We have decided against using subjective methods such as Fourier analysis for detecting the solar cycle. Fourier analysis is subjective because it assumes a priori that the signal should have a constant period and a constant amplitude throughout the length of the time series. Neither of these assumptions is applicable to the solar cycle index or to the atmospheric signal correlated with it. This has the potential effect of reducing the amplitude of the extracted signal. Given that our length of record is already short, the statistical significance of the signal cannot be demonstrated at all pressure levels. Figure 1 shows the Fourier spectra of the 600-hPa geopotential height (January 1948 to February 2004) data averaged from 20° to 90°N. The statistics are naively calculated, with the mean (magenta line) fitted to the spectra of the data without the annual cycle and the 95% confidence interval calculated with a student t test. In Figure 1, there is a Fourier spectral peak near the 11-year period. Its amplitude is reduced from that of the “true” signal because of its entanglement with other atmospheric signals. This Fourier signal is also difficult to see everywhere over the globe. At 700 hPa, there is less noticeable power at 11 years, and when only the years after 1958, where the data are considered more reliable, are used, the power at this period also diminishes. The statistics calculated here do not take into consideration the choice of height or the choice of years included in the spectrum. This would also diminish the significance of the signal. A more objective method should be one which can disentangle the atmospheric signals from one another

        http://onlinelibrary.wiley.com/doi/10.1029/2004JD004873/full

        10

  • #
    ColA

    Congratulations Dr Evans,
    I have had a preliminary look at the spread sheet and having done a bit of Excel work for others to use I can really appreciate the time and effort you have made in the presentation and functionality of the programme. The whole thing has truly been a mammoth task and I can see why it was 2 years in development – must have been more then a few ‘Ah stuff it moments’.
    I do hope everyone takes the time and makes the effort to work through the programme and understand what is happening – not shooting off from the hip with their brain half cocked! And I hope questions and criticisms are positive, helpful and building.

    240

  • #
    Hasbeen

    God I love watching & listening to people who really know what they are talking about, & who know a lot more than me.

    Even if something is not quite right, watching the process is satisfying just in itself.

    210

    • #

      Hasbeen,

      I’m with you on that.

      I just love reading this stuff. Six years ago, it would have been over my head. It still is now, but at least it’s a helluva lot closer, and I can pick up bits and pieces, and upon further thought, other things slide into place, that years ago would have still been out there.

      However, I get this impending feeling that I need to keep watching and reading, because I know that this is something significant.

      I also love the way others are coming in and leaving ….. constructive ….. comment, other than the flamers, who (my perception only) seem worried that something like this could shoot down their ….. religion. They need to snipe away with their one liners in the vain belief that we will come around to their fractured way of thinking, and if you consider it, then something like this must have them worried.

      What comforts me is that even though it is that little bit over my head, it’s Space Shuttle distance over the heads of those flamers, and getting further out.

      Tony.

      270

      • #
        Mike Jowsey

        then something like this must have them worried.

        After 10-plus years of following blogs and research on CAGW, this is as significant to me as Climategate 1 was. Here is a list of differentiators that make this so significant:
        1. A falsifiable hypothesis about the climate
        2. From a hitherto un-utilised profession
        3. From a highly-qualified scientist with an impecible CV
        4. Showing a completely new way of looking at the factors affecting global climate
        5. Openly publishing the thesis, data and programming and inviting comments
        6. Completely self-funded
        7. A short timeframe for falsifiability
        8. A polite, humble and gracious countenance in the face of unwarranted vitriol

        … must have them worried!

        390

        • #

          Worried? They are terrified. Their self built pedestal is crumbling as we speak and the high horse they are trying to ride is about to dump them.

          The source of their horror is that someone is doing something without their permission, using a method they don’t understand, getting results that blows their stuff out of the water, using (Gasp!) Excel and Basic on mostly obsolete PCs, and doing it in the open for all to see and use. Then the final blow: all they can do is quibble over the ragged edges, dibs, dabs, and drips but about nothing of real substance.

          My bottom line is that David’s top down methods are correct. It will eventually converge onto some significant truths. What those truths will be is part of the discovery process. The evidence drives the process. Reality sets the rules. Pet theories need not apply for employment.

          290

      • #
        PeterK

        Amen Tony!

        60

      • #
        Hasbeen

        Yes Tony, it is just a bit over my head, & trying to get up there can give me a headache.

        However I have enough still there to see where it’s going & love stretching myself to try to keep up.

        It really is wonderful to be given the opportunity to understand, rather than be talked down to by self styled experts, who believe it is beyond all us peasants.

        150

        • #

          Hasbeen, you say here:

          It really is wonderful to be given the opportunity to understand, rather than be talked down to by self styled experts, who believe it is beyond all us peasants.

          What is really good here is how this is in fact complex Science, well complex for the average (huge emphasis on that word average) person anyway, which is after all, 97% of the general populace.

          Now, those on their side of the argument say that what we have here from Dr. Evans is just so much charlieromeoalphapapa.

          Yet extrapolating it out, it’s no different really to their brand of complex Science with their brand of complex models, and this gives the impression that those willing believers on that side totally, utterly, without question, implicitly, (you get the idea) ….. understand ….. completely their brand of complex Science and complex models, if you can see the point I’m attempting to make here.

          Charlatans the lot of them who believe utterly their Science and not some from the other side of the debate, without the exact understanding of either. (and here note Chester who thinks it is obviously being sponsored by, dare I say it even ….. Big Oil, Big Coal etc, and wants not an explanation, but a list of non existent people paying for it, according to him)

          Tony.

          90

          • #

            > Charlatans the lot of them who believe utterly their Science and not some from the other side of the debate, without the exact understanding of either

            That perfectly describes you, and many another commentator here. You’ve already admitted that you don’t understand DE’s stuff – but you love it anyway despite that – and its clear that you understand none of the IPCC’s stuff either.

            > it’s no different really to their brand of complex Science

            I and others have repeatedly pointed out that DE’s model is non-physical, unlike say the GCMs. Its pretty clear that no-one here even knows enough to understand the distinction. And yet you keep puffing your Science-with-a-capital-S.

            231

            • #
              Rereke Whakaaro

              Oh look, everybody, it’s Will-I-am Con-knolly.

              He is under starters orders, and he is off, to a great start, with a quotation taken out of context, and cherry-picked for maximum effect. So we can expect …

              Yes, it has happened, he has switched to the drive-by ad hominem of the day. And yes folks, it is down to his usual low standard, with his rambling sentences that have no subject, and go nowhere.

              But wait, he has suddenly reverted to a second quotation taken out of context, which might be a risky strategy at this stage …

              But no, he has managed to recover by expressing a personal opinion in a form that he can attribute to unnamed others, but then fails by making statement that, that is totally meaningless and also contains a spelling error.

              What a pity, He started so well, but the judges are going to have to mark him down on that performance, due to his poor finish … And yes, they can only award him the ranking of moron. What a shame. It was such a valiant effort, at the start.

              281

            • #

              So black box analysis is now out of bounds? Please give your reasoning. Also consider editing the Wiki on that subject. I’m sure it needs “improvement”.

              120

              • #

                You’re evading the point. Or perhaps you didn’t understand it. I said nothing that can be interpreted as “black box analysis is out of bounds”. I said that DE’s model is non-physical, and that none (actually, I think there were a couple on earlier threads, but they’ve pretty well given up now) of you appear to even understand that point. You appear determined to prove me correct.

                323

              • #
                the Griss

                It seems very evident that it is YOU that doesn’t understand what this is all about.. Its way beyond you.

                You have made that VERY obvious since your first post on this whole topic.

                All you keep doing is making continued stupid vacuous comments, zero contribution, a waste of space.

                Because that’s all you can do.

                Its all you are.

                131

              • #

                Black box analysis is unphysical. You can look it up. Try the wiki.

                111

              • #
                James Bradley

                William,

                It’s just that the CO2 models don’t refelct reality.

                So there are other factors that have more influence on the atmosphere than CO2.

                One may be the little space heater at the centre of the solar system – it’s there for a reason.

                80

            • #
              Raven

              […] – and its clear that you understand none of the IPCC’s stuff either.

              The IPCC is a political lobby group.
              What’s to understand?

              130

            • #

              A Dancing Marionette

              He cannot resist the motion.
              His masters pull the strings.
              Without will or devotion.
              The dancing is all he brings.

              He dances into the night.
              No will to do but prancing.
              Having no thought or delight.
              A dancing marionette is dancing.

              70

            • #
              Rereke Whakaaro

              William,

              It is you that misses the point, I fear.

              Please see my comment at #34.5, then we might have a serious debate about science, and the scientific method.

              90

  • #
    Ragnaar

    Looking at the process is interesting in itself. Even if this doesn’t work out, one could review the evolution and the steps of the process and learn something. The feedback they are getting may have value. As far as I know they’ve tried something new, and in the future others will follow them in ways that are similar.

    I am reminded of a quote, “I know you’ve taken it in the teeth out there, but the first guy through the wall. It always gets bloody, always. It’s the threat of not just the way of doing business, but in their minds it’s threatening the game.” – Moneyball

    I post something similar to the above on WUWT’s current solar model thread.

    100

    • #
      Hat Rack

      “Even if this doesn’t work out, one could review the evolution and the steps of the process and learn something.”

      Quite right Ragnaar. When confronted by a problem, one of the many things I learnt as a (retired) tradesman was to investigate all avenues, even those that seemed “out there”. It was amazing how many times such ideas formed part of the solution.

      My tip is that, even if the Solar Model is not the answer, at least some parts of it will be useful as scientists achieve a better understanding of climate.

      Congratulations to David and Jo for their ability to think outside the square.

      90

  • #
    Brad

    Just got a post snipped by Anthony:
    ***************************************
    Willis Eschenbach says:
    July 8, 2014 at 6:07 pm
    Brad says:
    July 8, 2014 at 5:51 pm

    Anthony,
    Willis did not investigate the problem or post a comment on Jo’s website. How is that “reporting”?

    I investigated the problem and found exactly where it was coming from. I reported it here. I am extremely reluctant to post anything at Jo’s website, because at this point it is infested with people like yourself who will never be satisfied with what I do. In any case, the same bug was indeed reported there by others.

    w.
    ****************************************************************************************************

    I responded to this comment, stating that Willis’ own actions were responsible for the attitude, and got the following from Anthony:

    Brad says:
    July 8, 2014 at 6:21 pm
    [ok - we are done with the food fight Brad, you don't like what Willis has to say, we get it...move on to something relevant like posting what you've learned from the model - Anthony]
    *****************************************************************************************************
    You get it Anthony but the other readers didn’t, and won’t because of your apparent bias. You have effectively shut down any free speech that does not benefit your supporters.

    What I have learned from the model deals with psychology, not physical sciences. Deviation from your mainstream agenda apparently won’t be tolerated…. simple as that.
    ************************************
    It is in moderation… will let you know if it gets posted but I doubt it.

    180

    • #
      Brad

      My reply, shown above, was also snipped:

      Brad says:
      July 8, 2014 at 7:59 pm
      [snip - covered earlier, let's not continue the food fight between you and Willis....or anyone else for that matter.

      If you have something useful to discuss about the model, that doesn't deal with personalities, please post it. - Anthony]

      So Willis can “report” a problem he has with his own software on another website, and Anthony considers it acceptable, not even up for discussion.

      Amazing.

      151

    • #
      Mike Jowsey

      http://wattsupwiththat.com/2014/07/08/solar-notch-delay-model-released/#comment-1680602

      As skeptics we have to be doubly hard on each other, pal review has no place here. That said, I’ve asked for more courtesy.

      Further, Jo and David and I are on quite good terms. I’ve sent them some tools to help, and offered some advice. All is well between me and them.

      If you think you know what is going on between Jo, David, and I and want to lecture me further on “respect” …think again.

      Anthony Watts

      112

      • #
        Brad

        Wow, Missed that one at WUWT. He apparently has no control over WE and LS yet refuses to suspend/ban them. SM and PG are also in the mix.

        Anthony, if you read this, know that I do respect what you do and have referenced your website countless times. It is just this latest barrage of egotistical attacks from your regulars is really over the top.

        What is the reference to 1.3 million comments? Over how long a time period and how many posts? How many moderators? Some unanswered questions…

        71

        • #
          Paul Vaughan

          positive ID of 4 most darkly negative AW defenders (where AW = anthropogenic warming)

          60

        • #
          the Griss

          lol.. my posts all go into moderation over there, so I don’t bother anymore.

          Maybe because I kept referring to Mosh as a used car salesman ;-)

          110

          • #
            Crakar24

            no its that silly smiley avatar of yours :-)

            30

            • #
              the Griss

              Hope this one is more to your liking. !

              30

              • #
                Rereke Whakaaro

                You mean the bad-hair-day and the hobnail boots one?

                Very professional.

                The remaining question: what profession?

                40

              • #
                the Griss

                Those are not hobnail boots.

                They are purple fluffy slippers (with a mouth, nose and eyes). It is winter, afterall. !!

                30

          • #

            Have you tried “junk car salesman” ?

            70

            • #

              LOL. I love it.

              Re going into moderation over at WUWT. I always do. I thought at first it was because my email address is not my name (but looks like a name – it’s actually the name of our house), but finally came to the conclusion that it’s just a timing thing on the automated posting. My posts pop up usually within ten minutes, if not sooner.

              Of course, I could be wrong. :)

              10

        • #
          Paul Vaughan

          Mike Jowsey (July 9, 2014 at 1:48 pm) linked to an exchange that includes 2 interesting highlights:

          • John Hewitt (July 8, 2014 at 3:12 pm) wrote:
          “Your failure to control the total negativity of both Leif and Willis diminishes your credibility further.”

          • Anthony Watts replied:
          “Comments like yours are the ones that make me think I should just shut off the blog and walk away some days.”

          Brad (July 9, 2014 at 2:00 pm) correctly IDs 4 dark roots for AW.

          Sensible solutions address the dark agency:
          • easy solution: simple blog policy (one sentence)
          • if that fails: banishment’s even more efficient

          In summary:
          Solar-climate thought-policing had it’s day, but maybe things change…

          70

    • #
      Mark D.

      Brad, from the bits I’ve read, Anthony is trying his best to keep the discussion from devolving (to what happened some days ago).

      I think Anthony deserves some credit for trying to prevent an escalation in poor will. (yes there might be a pun there)

      182

      • #
        Rud Istvan

        It is an object lesson on why the ‘WildWest’ Internet needs sheriffs like Wyatt Earp (sorry if the American analogy does not correctly internationalize). Watts is paid as much as Jo. Norhing. He has a bigger target painted on his back by warmunists (a term coined and explained in the next book). So I Agree with Mark D.
        On the other hand he let stuff get out of control that even Judith Curry ( even more forgiving of BS) has reigned in. You publish a blog, you own it, and it’s policing. I prefer Judith or Jo’ s model any day of the week. But nothing is going to be perfect given all the nutters and ‘drunk driving’ blog equivalents out there.
        Hey, it is a new medium. it will sort itself out, as this kerfuffle is showing.

        151

      • #
        Brad

        I have given him credit for his overall efforts but it is his website and a simple matter to correct.
        No cross-website sniping. PERIOD. (Yes, that is an OBAMA pun…:)~ ).

        Be ethical or be banned.

        40

    • #

      Brad, thanks for sticking up for us on WUWT. I do appreciate it (and did notice!). Yes, Willis has my email, and free access to commenting here. No he has not reported anything to us.

      I had a good conversation with Anthony the other day which I was grateful for. I sympathize that it’s very hard to manage 800 comment threads. And especially so if there are commenters who have little interest in being accurate. Leif and Willis are making his work harder.

      260

      • #
        Brad

        Jo,
        I am really sticking up for open science, which you and David currently embody.
        I appreciate Anthony’s situation but it is one he got himself into. It is not mine or your problem.
        He can change it all with a simple lockout. It is his call, not ours. He does himself no favors by allowing false charges on his website. WE and LS, and anyone else, should be required to respond to you directly instead of being given “safe-haven” to take potshots.
        He is damaging his hard-earned reputation in the world, which hurts all of us.
        I understand his being preoccupied with the conference but that does not prevent him from calling a hiatus for a few days. Shut it down and reassess the situation after the conference.
        The fact that he is stifling dissent by blocking comments should be reason enough.

        Again, I am not a climate expert but I am pretty good at analyzing logic. So far, the naysayers have no logic other than you are impacting their salary/egos. I have dealt with this hundreds of times, and the complaints are always the same:
        Authority-How dare you question me!!
        Who are you, what are your credentials?
        Change-the-subject switcheroo
        Ignore it all together and create a strawman instead to attack.
        And then there is my all time favorite-just lie through your teeth and scream loudly. ( My kids favorite) from Calin and Hobbs….
        You and David have a great conce

        121

      • #
        Brad

        Jo,
        I am really sticking up for open science, which you and David currently embody.
        I appreciate Anthony’s situation but it is one he got himself into. It is not mine or your problem.
        He can change it all with a simple lockout. It is his call, not ours. He does himself no favors by allowing false charges on his website. WE and LS, and anyone else, should be required to respond to you directly instead of being given “safe-haven” to take potshots.
        He is damaging his hard-earned reputation in the world, which hurts all of us.
        I understand his being preoccupied with the conference but that does not prevent him from calling a hiatus for a few days. Shut it down and reassess the situation after the conference.
        The fact that he is stifling dissent by blocking comments should be reason enough.

        Again, I am not a climate expert but I am pretty good at analyzing logic. So far, the naysayers have no logic other than you are impacting their salary/egos. I have dealt with this hundreds of times, and the complaints are always the same:
        Authority-How dare you question me!!
        Who are you, what are your credentials?
        Change-the-subject switcheroo
        Ignore it all together and create a strawman instead to attack.
        And then there is my all time favorite-just lie through your teeth and scream loudly. ( My kids favorite) from Calin and Hobbs….
        You and David have a great concept to share with everyone. The divergence with Co2 is already upon us. It needs to be broadcast to every corner of the world to shut down current policies.
        I live in Washington state, USA. One of the worst climate screaming places in the world. I continually deal with those who make a living off it, selling bad projects in its name.
        IT NEEDS to STOP!

        140

        • #
          Mattb

          It’s worth noting that if jo herself only ever raised issues with other folks’ work directly then this blog wouldn’t even exist…

          021

          • #
            Winston

            The difference, Matt, is that David and Jo are entirely open, transparent and, if wrong, are prepared to acknowledge it.

            If mainstream climate science policed itself in such an honest and self-effacing fashion, been prepared to acknowledge their failures and limitations, and had not committed serial crimes against scientific method, then blogs like Jo’s wouldn’t be necessary. Any private communication to such people would be given a “talk to the hand” level of disdain.

            It’s simple cause and effect, their lack of honesty makes it necessary for such accountability to be forcibly and forcefully applied.

            270

            • #
              Mattb

              I’m not criticising Jo’s website. I’m criticising Brad’s random and inconsistent standards he’d like upheld.

              07

              • #
                Brad

                Mattb,
                Please explain…

                10

              • #
                Mattb

                “He does himself no favors by allowing false charges on his website.” is he?

                “WE and LS, and anyone else, should be required to respond to you directly” I mean how ridiculous this is the blogosphere.

                “So far, the naysayers have no logic other than you are impacting their salary/egos.” rubbish comment I’m not sure how Jo is impacting Mosher or Willis’ salary or ego.

                “Again, I am not a climate expert” at least you are honest here. But look who cares if someone is rude if they are right. I think BoM is rude/arrogant/basically a prat… I also think he is wrong and that’s the main issue.

                06

              • #
                Brad

                Mattb
                Thanks for responding here.
                Nothing you listed is random or inconsistent, as you had claimed.
                Your comments do not warrant any specific responses.

                50

              • #
                Mattb

                so you cannot answer how Steve M’s salary is being impacted?

                05

              • #
                Mattb

                which appears to be a “false charge”.

                05

  • #
    Ragnaar

    Top-down solar modulation of climate: evidence for centennial-scale change
    http://iopscience.iop.org/1748-9326/5/3/034008/fulltext/
    “This implies that `top-down’ solar modulation could be a larger factor in long-term tropospheric change than previously believed, many climate models allowing only for the `bottom-up’ effect of the less-variable visible and infrared solar emissions. We present evidence for long-term drift in solar UV irradiance, which is not found in its commonly used proxies.”

    100

    • #
      MRW

      Dr. Willie Soon was discussing this in his first panel in Vegas. It’s available for screening, I think.

      40

  • #
    Paul Vaughan

    Piers Corbyn 7 July 2014:
    “There is no observational evidence in the real world of real data in the real atmosphere and real ocean of their deluded claim that CO2 changes drive temperature changes.”

    Agree.

    200

  • #
    Ron Van Wegen

    “reason and empiricism triumphed over superstition and abuse by church and state.”

    Boy, that’s one heckuva simplistic analysis!

    Dog whistle… Flat Earth, Dark Ages, Galileo etc.

    Tiring.

    115

    • #
      ianl8888


      Tiring

      Still better than being burned at the stake on a sunny Saturday afternoon auto de fe … authority’s metaphorical answer to informed dissent

      Doesn’t mean I am at all convinced of the utility of the Evan’s model. My observations suggest coupled, non-linear, dynamic chaos of the various forces, rendering prediction pointless. But we will see

      30

  • #
    Rud Istvan

    Bravo for the fast fix to what was ultimately just another Microsoft problem. Double Bravo to the code experts who knew what the problem was from the gitgo and posted the VBA fixes.

    This whole experience shows the incredible power of the Internet to create a ‘supermind’.
    It has enormous potential, used wisely. That is our collective responsibility. Or downfall.

    60

  • #

    Excel isn’t suitable for Github. The issue is branching and merging of changes, I think this would get kind of ugly with an Excel file.

    20

  • #
    sophocles

    This is WOT (Way Off Topic) and may be a repeat.

    Nigel Calder, ex-editor of New Scientist, author of Einstein’s Universe, The Magic Universe and The Chilling Stars with co-author Henrik Svensmark, died on 25th June. See his blog for more. There’s a brief notice in Google and his Wikipedia entry is up to date. He was a fierce espouser of the Scientific Method—somewhat similar to our hosts.

    Requiescat in pace, Nigel.

    [Added links:] ED
    “The Independent, 27 June 2014″ http://www.independent.co.uk/news/obituaries/nigel-calder-prolific-journalist-and-author-who-did-much-to-educate-the-public-in-the-understanding-of-science-9569398.html
    “The Global Warming Policy Foundation 27 June 2014″ http://www.thegwpf.org/nigel-calder-1931-2014/
    “Tallbloke’s talkshop 27 June 2014″ http://tallbloke.wordpress.com/2014/06/27/sad-news-nigel-calder-1931-2014/
    “Bishop Hill 28 June 2014″ http://bishophill.squarespace.com/blog/2014/6/28/nigel-calder.html
    “The Guardian, 1 June 2014″ http://www.theguardian.com/media/2014/jul/01/nigel-calder

    70

  • #

    The team at WUWT are very skillful at containing the argument within their own parameters, where they know the outcome. They continually refer to TSI having such a small variance that cannot be a driver of climate, and UV radiation only makes up a small part of the spectrum and also only makes up a very small fraction of the ENERGY involved.

    With respect to UV which can vary by as much as 100% over a solar cycle at the extreme end, the important function is CHEMICAL. There are many processes linked to EUV and FUV that control the ozone balance of the atmosphere, little of this is totally understood but there has been data collected showing variances over the solar cycle at different levels of the atmosphere.

    There is a growing area of research suggesting the current ozone concentrations at varying altitudes are influencing the position of the jet stream, this itself is a huge impact on climate with the possibility of cloud cover changes also being plausible.

    I am jumping the gun somewhat but this could be Factor X, it is possible that a cooling force influenced by ozone redistribution is balancing against the TSI peak at cycle max….it would take much research to falsify this theory but it could be one of the areas worth investigating.

    So next time you hear the energy argument over at team wassup, remember the chemical component that they will shy away from.

    100

    • #
      the Griss

      Geoff, I prefer to think of it as one of many parts that might make up Factor X.

      There appear to be many possible parts of this force with different timing and amounts, giving a large smearing effect to the solar inputs.

      It is highly unlikely we even know but a few of them, certainly not to any great detail, which is why the normal climate model is probably pretty much doomed to failure (as we have seen)

      It is also why a whole-istic cause and effect type model as David has devise, probably stands more chance of success.

      21

      • #

        Whatever force x is it must be capable of causing the climate zones and jetstream tracks to shift latitudinally in response to solar variability.

        That must involve changes in the gradient of tropopause heights between equator and poles.

        Those heights are determined by ozone concentrations.

        Force x must alter the ozone creation / destruction balance differently at different heights and latitudes.

        The UV and EUV wavelengths affect ozone amounts but additionally other chemical by products of solar interaction with the atmosphere can do so.

        The way forward is to carefully monitor the changing ozone balance in three dimensions above the tropopause.

        60

        • #
          the Griss

          I just think that Force X is highly likely to be a multi-faceted entity, rather than a single big force.

          That’s what makes this model so interesting.. It doesn’t really matter what Force X is, exactly.

          Its the final effect that matters., and trying to pin it down to a single over-riding factor is probably counter-productive. :-

          61

        • #

          Hi Stephen, the end result I think is the planet has more low pressure systems that the jet stream is forced to go around that pushes the jet stream more meridional. Ozone/tropopause gradient heights or whatever seem to correlate with the increase in low pressure systems.

          The data should be available to check ozone concentrations at different latitudes and heights over the past few solar cycles, this would be a project worth funding.

          20

          • #

            Thanks Geoff.

            I think that both high and low pressure systems become larger, more intense and slower moving when the sun is quiet.

            When the sun is active they become smaller, less intense and faster moving.

            What goes up must come down so you can’t have more rising air in low pressure cells without more falling air in high pressure cells.

            20

            • #

              Yes agree, the blocking high pressure cells also seem to be related. We had a large permanent high pressure cell over the most of Australia for almost the entire month of May.

              30

  • #

    Brad
    July 9, 2014 at 4:55 pm · Reply

    Jo,
    I am really sticking up for open science, which you and David currently embody.
    I appreciate Anthony’s situation but it is one he got himself into. It is not mine or your problem.
    He can change it all with a simple lockout. It is his call, not ours. He does himself no favors by allowing false charges on his website. WE and LS, and anyone else, should be required to respond to you directly instead of being given “safe-haven” to take potshots. …

    I would have love posting this reply in the proper place, #60.4.1, but I don’t see how to do that now. No “reply” buttons are seen today. Oh well.

    My comment is that Mr. Watts is not the “friend” of skeptics that many of us think he is. His snipping and moderation of anything that he can’t answer is something I don’t see here or elsewhere on the skeptic side of the debate. Even his “replies” inside a comment where the poster can not answer is off-putting to me. He allows Willis and Lief to say any damn thing they care to say and those who disagree should say nothing or limit their response to a minimum. I find the 3 of them to be very, very thin skinned and childish at times. (often even)

    At Steve Goddard’s site (aka Tony) there is almost no moderation. You can post any damn fool thing you want to post — but the crowd will jump on you if it is stupid. I also have never had a post here at Ms. Nova’s site run into trouble.

    Final thought. There are those at WUWT who can “dish it out” but they can not “take it”. Sad to say this, because the site is a force for good. It could be better with a more mature host.

    100

    • #

      I now see the “reply” button. I think I may be losing it in my old age. :-(

      10

      • #
        Rereke Whakaaro

        You are not “loosing it”. The “reply” nesting is limited to a certain number of levels. It is a “Feechur” of WordPress.

        A few of us lobbied Jo, a while back, to increase the number of nesting levels. The results were less than ideal, since we now loose the last number in the comment identifier, or it goes screwy. On other occasions, nesting just seems to give up, puts on its coat, and goes off to the pub.

        60

    • #

      Mark, thanks for your comments at WUWT. I did notice and appreciate it.

      Anthony writes inline (and I do that too sometimes) because if he arrives late to a thread posting a reply 100 comments down is a hard way to do a conversation. I appreciate reading his inline replies, but I can also appreciate if you disagree and one is placed in your comment, it would be frustrating.

      Having open science policy on comments makes skeptics stronger. I have learnt so much from commenters. (Thank you).

      I use the inline replies when a commenter is particularly dominant, repetitive and refuses to acknowledge a point. Some commenters appear to be here for advertising rather than a conversation, and inline replies neutralize that advantage. I always try to add a comment at the bottom of the thread to list those inline additions as a courtesy, so the commenter is aware of it.

      Nesting is both good and bad, I wish I had a button at the top of the thread to re-sort the comments into chronological order (so people could see which were the latest), perhaps a red dot on the last 20 comments would be good instead). No plugin seems to be available for either yet :-( Please let me know if anyone finds one.

      90

  • #
    bit chilly

    in relation to tsi ,this is worth a watch . appears very different to lief,s proposals .http://climateconference.heartland.org/breakout-1-streaming/

    00

    • #

      Re: http://climateconference.heartland.org/breakout-1-streaming/

      You need to give the actual session number. Even better give the running time. Otherwise it is hard to tell.

      10

      • #
        bit chilly

        thanks for that m.simon,i made the mistake of assuming clicking the link auto played the correct session.sorry about that.

        10

    • #

      In the 58:28 video at 17:16 the speaker (a solar guy) shows flat temps to 2100.

      At 19:00 the speaker (who is difficult to understand) has a slide that says 2014 is the beginning of a little ice age. The slide of 24:21 says the delay is 20 +/- 8 years. 31:49 – warming on Mars slide.

      At 34:29 Willie Soon. IPCC = Gangster Science. I’m down with that.

      30

      • #

        58:28 – 36:44 TSI – The instruments are poor because the aperture holes are small. So why make the holes small? It reduces the required instrument power. The Boulder instrument (discussed on the LS thread with LS providing a maximum of ignorance) has an aperture that lets in 68mw of sun power. What is .1% of that? 6.8 uW. And the presumed accuracy is supposedly 1/2 of that. 3.4uW. Willie says that is not credible. I’m inclined to agree. And the precision? It is .68uW. And the resolution is .34uW.

        Anyway I suggest you have a look at the video. esp the instrument designers among us.

        I’d like to know how much power that the 68 mw represents in comparison to the total power required to keep the instrument at 31C.

        I can see all kinds of other confounding problems. Like radiation loss variations of the instrument to the spacecraft.

        10

        • #

          The instrument does measure something but is it TSI? Perhaps what it does measure can be demonstrated on the ground and even in a vacuum chamber. Can it be done after several years, in a satellite, in space, and exposed to the full space environment? That is still to be proved. I suspect a small and unknown fraction of what it measures is only a part of the actual TSI.

          What is the offset, variance, and scale of an unknown? It can be anything, everything, and nothing all at the same time”. Hence, you can process the apparent data and get any answer you want. How convenient? In the words of the SNL Church Lady: “Now that is ‘Special’!”.

          20

        • #
      • #
        Richard C (NZ)

        MSimon #69.2

        >”At 19:00 the speaker (who is difficult to understand) has a slide that says 2014 is the beginning of a little ice age. The slide of 24:21 says the delay is 20 +/- 8 years”

        Habibullo Abdussamatov.

        In his 2012 paper he calculated planetary thermal inertia at 14 +/- 6 years for an ocean depth of its active layer about 200-500 м, (page 2 – page 4):

        http://icecap.us/images/uploads/abduss_APR.pdf

        Equation (1) where:

        S is the TSI,
        ΔS – increment of TSI,
        А – Earth global albedo (Bond albedo), A – Bond albedo increment,
        ε – emissivity of the system Earth- tmosphere,
         – Stefan-Boltzmann constant;
        Тр – thermodynamic temperature of the planet,
        Е – specific power of the enthalpy change for the active layer of the atmosphere and the Ocean (W/m2),
        С – specific surface heat capacity of the active layer of the atmosphere and the Ocean, with regard to the total surface of the planet (J/м2К),
        t – time

        He observes (page 4)

        “Due to quite large heat capacity of the Ocean the thermodynamic temperature of the planet (with account for
        emissivity) changes rather slowly”

        But he puts no figure on oceanic delay specifically in the 2012 paper. 20 +/- 8 years comes in his 2013 synopsis (SPM) page 2:

        “Quasi-bicentennial variation of the solar radiation absorbed by the Earth remains uncompensated by the energy emission to space over the interval of time that is determined by the thermal inertia of the World ocean (20±8 year).”

        http://scienceandpublicpolicy.org/images/stories/papers/originals/grand_minimum.pdf

        I don’t know how he arrived at 20±8 years though.

        30

        • #
          Richard C (NZ)

          Should be:

          “ε – emissivity of the system Earth- atmosphere,”

          00

        • #
          Wayne Job

          Not sure about the delay Richard, but he is in charge of measuring every thing about the sun from the international space station, he mentions that it varies a lot and causes cold periods and hot periods. He also mentions all the harmonic cycles that cause ice ages much like what Roger Tallbloke talks about. Very interesting.

          20

        • #
  • #
    • #

      My response:

      jmorpuss says:
      July 9, 2014 at 5:34 am

      I have had nothing to do with that research. I do think there is something to it. Lightning is a frequently noticed side effect of earthquakes.

      I have a design for a very low cost instrument for measuring such things (earthquakes vs magnetic field changes) and would like to get it crowd funded so it could be widely deployed esp in active earthquake zones.

      Any one who would like to help with that can contact me at http://spacetimepro.blogspot.com/ my e-mail is on the sidebar

      20

      • #

        I do not know when Jo will have the What can force X be, post?
        Perhaps time for many many SWAGs, of both kinds. To be carefully indexed into groups.
        I frown on direct electromagnetic radiation or TSI, David has eliminated most all receiver, time constants on EMR. The 11 year apparent delay is most problematic at these distances. Note the 22 yr magnetic field cycle corresponds to earthling cycle time.
        Electrostatic discharge? I like that one, and want to be in charge, Lightning bolt at whatever may annoy, to be patiently delivered in 11 years! How many bolts I get?
        Magnetic field interaction delayed by rotational inertia? Many others, none to be produced by any greasy statistical mechanic!

        10

  • #
  • #

    http://wattsupwiththat.com/2014/07/08/solar-notch-delay-model-released/

    They are still going on like a bunch of scared rabbits.

    Typical “we can’t test it unless we know all the details of how it was made” – roughly.

    I dunno. Typically when I want to test something I turn it on and see how it runs.

    ==================

    A sceptic model that makes predictions. Further improvements (or scrap) possible. You would think they would be excited. Well they are. But not in a good way.

    And of course we have – “900 days of fake data” – still. As if that was an input to the model. It is just a way to estimate a transition point for estimation purposes (when the cold spell will start). If that is the wrong number the estimation will be off. Why don’t they wait to crow about that – when it happens?

    Oh yeah. We are all groupies here. No independent thinkers.

    Now I’m not deep in the math (Haven’t done any significant calculus in 50 years – when I need to do stats I pull up an application that is almost trivially simple, When a company needs heavy math for a project I’m working on they add a mathematician to my team) but the model derivation makes sense to me. I can’t wait to see the test reports.

    It is as if they were being willfully stupid. It will catch up to them.

    130

    • #
      Rereke Whakaaro

      Typical “we can’t test it unless we know all the details of how it was made” – roughly.

      A clear indication that the whole approach to climate science has been bottom up, and based on the perceived influences. It is a Computer Science approach.

      Typically when I want to test something I turn it on and see how it runs.

      … and then explore any interesting factors … which is the more generally accepted approach to real scientific research, in the physical sciences.

      Those differences are at the heart of this debate.

      I would doubt that any of “the names” in climate science have any knowledge of Fourier Analysis, other that asking, “… is there an app for that?”

      110

    • #
      steven mosher

      “They are still going on like a bunch of scared rabbits.

      Typical “we can’t test it unless we know all the details of how it was made” – roughly.

      Actually no.
      Although I can assure your that test pilots want to know how a plane was made.

      The model is not the science. a spreadsheet that runs the model is not the science. it is a tool.

      What Willis has asked for and What I have asked for is EXACTLY the same sort of stuff that folks asked Mann for.

      1. The scientific claim is that this model was derived from data and certain analytic steps. it was not pulled out of thin air.
      2. To assess THAT claim we need

      A) the data that was sued and the code that was used to derive the model.
      B) the out of sample tests that were done.

      Absent that there isn’t much interesting in an 11 parameter tuned function. Heck you can get a better hindcast with one or two.

      There are two science claims at issue here

      1. the claim that the model was derived from certain data using certain steps. Show that work
      2. the claim that this model has predictive power

      two claims. Both claims need to be addressed.

      For example, if you look at #1 and find some odd bug ( as was found in #2) the whole house could come tumbling down

      Imagine.. 5 years from now and the prediction fails. and then somebdoy looks at how the model was constructed and finds
      a problem in THAT bit of science.

      two claims. two bits of science

      The first is the model construction. we can look at that now if david releases
      the second bit, the prediction, of course we will wait to see it prove wrong.. in the mean time.. the full bits and nothing but the bits

      214

      • #
        Mattb

        Seems bloody reasonable to me Steve.

        19

      • #

        Let me do the homogenising and it will fit the data, the 99.7% confidence interval if not the 70%.

        70

      • #
        Ross

        Have you read the first 8 information releases Steven ?

        60

      • #

        Mosher, If you really want to know, start by reading the lead up posts and see the 34 page extension file.

        The summary page is here: http://sciencespeak.com/climate-nd-solar.html

        110

        • #
          dp

          We both know he doesn’t really want to know. He wants his cookie and he wants it now. Cookies are on a later menu but he wants them right damn now and your schedule doesn’t matter. The good news is he has nothing useful to offer as he, Willis, and Leif have made up their minds. You’ve harshed their paradigm by introducing things they hadn’t thought of.

          120

          • #
            Backslider

            We both know he doesn’t really want to know.

            That would be a waste of time. He is only a squarker and would not comprehend anything given to him.

            00

        • #
          GabrielHBay

          Must confess: did not notice the extension file before. Just had a look. Seems comprehensive and beyond my maths. I can guess, however, that it would take even someone much more competent than me weeks to fully digest and be crystal clear about what is all going on here. That realization confirms my previous suspicion: the crowd at Wussup could not possibly have digested enough of the whole thing yet to justify their snidy negativity. So: it is all just bluster and an attempt to mask their feeling of emasculation at being confronted by something so out of their usual box. I would ignore most of what the prime triad and their groupies there assert and wait for comment from some-one who has actually studied and who really understands what is being presented. For what it is worth, from my pretty wide systems experience, I do think I understand the gist of the underlying philosophy and the approach taken, and see real potential here. Will be watching with great interest. Thanks a mil Jo and David.

          50

      • #

        Steve,

        The model is a predictor. You test a predictor by seeing if the predictions come true.

        You can test it with various noise sets to look for biases (a la Mann hockey).

        You can change the tuning parameters to see how sensitive the model is to those.

        I could probably think of other tests that don’t require any of the “missing” information.

        You could ask David/Jo for the bits you are missing. I have been corresponding with Jo frequently of late. I have found her most gracious and accepting of my input.

        The consensus at WUWT is that the model is untestable. All that tells me is that you lot are incompetent to run even the most cursory tests until the information you claim you need is available. If I was doing a reverse engineering project I could test it. For something like that I’d probably need a math person to help. My area of expertise lies in a different direction. But I could certainly direct such an operation. In fact I have. The company I was working for assigned me a mathematician. And I found errors in her work despite my math skills not being up to the general task. And we ultimately got a model that was a good representation of the servo system I was working on. It made some predictions and on the basis of those I was able to greatly improve the operation of the servo. The customer was happy. So why did I have to reverse engineer a product made by the company? Well the engineer who did the design left the company before committing all the details to “paper”.

        My take from all the whining at WUWT is that you guys are not near as good as you think you are.

        110

        • #
          Mattb

          “Steve,

          The model is a predictor. You test a predictor by seeing if the predictions come true.”

          Their point is if the full information is shared, you could run the model on the 1st half of the data and see if it predicted the 2nd and already known half. You could run it on the 2nd half and see if it back-casted the 1st half. There is precious little reason to wait 5-10 years when you could find out next week if it is worth waiting.

          They’ve outlined exactly what they need, it is information Jo/David claim to have. It is up to Jo/David to choose when to release it, and until such a time the WUWT folks are, quite rightly, not going to join the masses in lapping it up.

          010

          • #
            Rereke Whakaaro

            No Matt, that is a fallacy.

            As I understand it, we are not talking about a theory defining the model, as Willis and Mosher are doing, we are talking about the data informing the model.

            The data has certain characteristics, that can be modelled with a degree of accuracy. The model holds, and can be used for predictive purposes for as long as the data supports the model. If the data changes in ways that the model does not support, it is falsified, and may require refinement.

            But backcasting is not valid for this type of model, because it is circular reasoning.

            70

            • #
              Mattb

              “Oh I do love the smell of real science in the morning …” is the perfect sign off to cover up dodgy reasoning.

              08

              • #
                Rereke Whakaaro

                OK Matt, I will humour you.

                In what way is that dodgy reasoning?

                30

              • #
                Mattb

                The dodgy reasoning is insisting that we have to sit back 8 years to see if it is robust. It is dodgy because the whole #72 thread is denying Steve and Willis the information that they want, and that Jo has, in order for them to do what they would like to do to test the model.

                I’m also not convinced by your “But backcasting is not valid for this type of model, because it is circular reasoning.” because what the model does is estimate the temperature based upon a number of parameters. Whether a point of time where each of those parameters is a particular value occurs in the past or in the future the temp would be at least similar.

                I have no problems with the fact no one has a clue what so called Force X is.

                Steve’s post hs not been dealt with at all as far as I can see though, so best you worry yourself with that and not on me though.

                07

              • #
                Backslider

                Steve’s post hs not been dealt with at all as far as I can see though

                But, Steve is just a troll…. so?

                50

              • #
                the Griss

                Steve’s post has not been dealt with at all as far as I can see though

                Mosh is a used car salesman hired by BEST as a frontman..

                He can wait until David and Jo are ready. or go suck his dummy !!

                91

              • #
                Rereke Whakaaro

                Matt,

                If you take a whole bunch of data, and analyse that whole bunch of data, and identify patterns in that data, and then construct a theory around that data, and build a model that implements the theory, based on that data, you cannot then use that data to “prove” the veracity of your model. It is a circular argument. Of course your data will “prove” your model, because the model was derived from that data.

                You need different data in order to prove the model, and that data cannot have been manipulated or preprocessed, by any other system. Where do you get that? The future is where you get that.

                100

              • #

                Mattb July 10, 2014 at 4:43 pm

                “The dodgy reasoning is insisting that we have to sit back 8 years to see if it is robust. It is dodgy because the whole #72 thread is denying Steve and Willis the information that they want, and that Jo has, in order for them to do what they would like to do to test the model.”

                “We”, are you pregnant, or have mice in your pockets? “Robust” wtf?,
                You cannot sit back 8 years, you must do now! Suggestion: Collect your troll friends and collectively jump into a volcano “for the sake of the children”!

                51

              • #
                Mattb

                RW but if I used the 1st half of the data “and analyse that whole bunch of data, and identify patterns in that data, and then construct a theory around that data, and build a model that implements the theory, based on that data, ” I could see how it turned out in predicting the 2nd half of the data.

                like wow it’s like I just invented time travel!

                06

              • #
                Mattb

                WillJ ““We”, are you pregnant, or have mice in your pockets?”

                that’s a pretty odd statement. Unless everyone else has a tardis they are not telling me about. yes WE!

                17

              • #

                Mattb July 10, 2014 at 10:27 pm

                WillJ ““We”, are you pregnant, or have mice in your pockets?”

                “that’s a pretty odd statement. Unless everyone else has a tardis they are not telling me about. yes”

                That was not a statement but a question!
                You seem a BOT, infected with multiple personalities. Who is “we” and why should anyone care of the BOTs bliovating?

                50

              • #

                Mattb
                July 10, 2014 at 10:25 pm

                RW but if I used the 1st half of the data “and analyse that whole bunch of data, and identify patterns in that data, and then construct a theory around that data, and build a model that implements the theory, based on that data, ” I could see how it turned out in predicting the 2nd half of the data.

                The data IS the theory for this stage of the exercise.

                The model is not physics – at this point. That comes at a later stage. Yeah. It is hard to get your head around it because that is not the way it has been done in climate science – up ’til now.

                60

              • #

                > The model is not physics – at this point

                Remember to tell DE that. http://joannenova.com.au/2014/06/big-news-part-i-historic-development-new-solar-climate-model-coming/ says: “We get a physical model with physical interpretations (that is, not just curve fitting)”.

                Alternatively: just admit that you haven’t understood what DE is doing.

                010

              • #
                Mattb

                it looked a lot more like a statement with a question mark.

                06

              • #

                William Connolley
                July 10, 2014 at 11:52 pm

                Brilliant W.C.

                Would you care to point out where the oceans reside in the current model?

                As I understand it so far there is only a black box model. Is there a white box model I’m unaware of?

                40

              • #

                > Would you care to point out where the oceans reside in the current model?

                These conversations are long, slow, painful and rambling because you don’t bother read what people write.

                All I did was point out that you and DE disagree on whether the model is physical or not. If you think its a non-physical, block-box model, then your disagreement is with DE, not with me. If you want to know “where the oceans reside in the current model” then ask DE, not me.

                17

              • #
                Richard Case

                MattB writes:
                The dodgy reasoning is insisting that we have to sit back 8 years to see if it is robust. It is dodgy because the whole #72 thread is denying Steve and Willis the information that they want, and that Jo has, in order for them to do what they would like to do to test the model.

                Ok, let me see if I have this straight… Two guys with no formal science education whatsoever (not even a BS in science between them from even a community college) are hugely critical of another guy with something like 6 science/engineering degrees from world-class universities for not following their high standards and the proper rules of science ?

                What is wrong with this picture?

                70

              • #

                > What is wrong with this picture?

                You, pushing credentialism. Make up your minds: either (a) credentials don’t matter, in which case your comment has no virtue. Or (b) credentials do matter, in which case you “skeptics” all lose to the IPCC, the least of whom have credentials far exceeding the best of yours. I think you’re going to go for (a).

                18

              • #
                Mark D.

                Will quotes ONE person then says “make up your minds”

                Can anyone not see what a simple failure in reasoning this demonstrates?

                Take my word for it Will, at least 97% will see the failure in your reasoning.

                40

          • #
            Rereke Whakaaro

            Once the model is robust, then we can discuss its characteristics and come up with a number of theories about why what we observe happens.

            Oh I do love the smell of real science in the morning …

            130

            • #
              Mattb

              “Oh I do love the smell of real science in the morning”, while risking being accused of pursuing a logical fallacy… are you even a real hard physical scientist? I thought you were all on linguistics/social stuff? I mean when did you even smell it in the morning?

              112

              • #
                Mattb

                Scenario in a nutshell though. “I’m a skeptic. I’ve released an idea. 4 skepitcs think I’m jumping the gun (Lubos, Mosher, Eschenbach, Svalgaarg) and ask for some more info, info that I have and could easily provide.”

                Now even if I think they don’t need that info, and that they are barking up the wrong tree, well I’d provide it.

                010

              • #
                the Griss

                4 skepitcs think I’m jumping the gun

                roflmao.. Mosher is never a skeptic. He’s hire by BEST, among the warmest of warmists.. !!!

                80

              • #
                the Griss

                And Willis and Leif seem to have their own baskets of apples, which they are panicking might get tipped over.

                Ludos… no idea where he is coming from.. I don’t think he does either.

                60

              • #
                Ross

                Mattb

                I think you have joined the four you mention in that you don’t understand
                (or don’t want to understand) the different approach being taken by David & Jo to the development of the model, as Rekere has clearly explained in #34.5 , for us laymen.
                Also it seems really strange to me that many on this thread who know how these works are saying it is great , runs well and understand what is going on, but the vocal few who have been negative from day one continue on the same path (one of them has even said he cannot get the model to work !!)

                70

              • #
                Rereke Whakaaro

                In response to MattB @ 72.2.5.1.1

                … are you even a real hard physical scientist?

                Interesting question: Am I hard? Perhaps, I am special forces trained, if that makes a person hard. Am I a physical scientist? I hold a BSc in Military Science, which contains papers in Maths, Physics and Chemistry, which gives a good all round appreciation of the physical sciences (including blowing things up). I also hold a BA in Geopolitics and World History. As for the linguistics and social stuff? Well, it is hard to work in Intelligence for forty odd years, especially at Government level, without picking up a few of those skills as well.

                But my comment, “I love the smell of real science in the morning”, was really derived from, “I love the smell of napalm in the morning”, from the movie, “Apocalypse Now”. I thought it was appropriate because the Global Warming/Climate Change scare, is as much of a lost cause, as the Vietnam War turned out to be.

                150

              • #

                Mattb July 10, 2014 at 4:45 pm ·

                “Oh I do love the smell of real science in the morning”, while risking being accused of pursuing a logical fallacy… are you even a real hard physical scientist?”

                Can you recite a recent date when any “real hard physical scientist” did anything usefull? 1905 perhaps or 1926? Anything more recent?
                What is “your” verified CV?

                30

              • #
                Mattb

                Will I didn’t make the comment that I loved the smell. I actually do have a genuine certifiable degree in physics (mjr Atrophysics)… but I hated the smell in the morning to be honest and got out of there.

                RW from comments here I always figured you worked in marketing/psychology of communications sort of stuff. I doubt that most folks who quote Apocalypse now would do anything other than sh*t their pants if they smelled Napalm in the morning:)

                07

              • #
                Mattb

                “Can you recite a recent date when any “real hard physical scientist” did anything usefull?”

                “What have the Romans ever done for us?”

                06

              • #
                Rereke Whakaaro

                MattB #72.5.1.1

                As I explained at 34.5, the actual philosophical approach to modelling, that Lubos, Mosher, Eschenbach, and Svalgaarg are used to, is a bottom-up approach, that starts with the question: “What parameters will be needed, in order to simulate the climate?”. They are then normally skeptical about various details in that specific paradigm.

                David Evans is approaching the problem from the top-down, by starting from a model of the observed cyclic changes in the historic data, and then identifying signatures in nature that might explain those cyclic changes. Hence his use of Fourier Analysis to decompose complex signals into its constituent parts. It is a new paradigm.

                The parameters requested by “The Four”, are artifacts of the bottom up approach, and are therefore essentially meaningless, in a top-down approach.

                So there is nothing to provide. If they want to go away and sulk over the matter, then that is their concern.

                150

              • #
                Mattb

                But the model still ascribes weightings etc to those signatures. E.g. the discussed weighting of nuclear testing effects. People want to know have that for example was derived.

                06

              • #
                Rereke Whakaaro

                You should never assume. Marketing/psychology and communications are all in the tool bag, with or without formal qualifications.

                As Harold Wilson once remarked, “You do not need a Political Science degree to be a politician. In fact, having a Political Science degree can actually be an impediment for a politician.”

                70

              • #
                Mattb

                Also the made up last 5 years of data for the 11 year smoothed graph is a sticking point. it should not be there.

                09

              • #
                Mattb

                RW your talk of bottom up and top down… the claimed reality is that Jo has said they have done the testing that some of these 4 would like to do. She’s not said it is irrelevant because it is a top-down model. Why would she/DE have done it if it was not useful or necessary?

                07

              • #

                Mattb
                July 10, 2014 at 7:16 pm

                Also the made up last 5 years of data for the 11 year smoothed graph is a sticking point. it should not be there.

                Do you understand the purpose of that graph? It has NOTHING to do with the analysis. NOTHING.

                The purpose is strictly an aid in estimating the timing. And for that purpose adding some data at the end to extend the TIMING estimation is only a sticking point for those who have no clue about what is going on here.

                You might as well say you can’t use a magnifying glass to examine things. It distorts the image. True. But irrelevant.

                Adding data points at the end definitely screws up things. But as an aid to determining transitions it is probably OK. Surely OK? Nothing is absolutely sure. Time will tell if it is useful. In any case it has nothing to do with the analysis because it is just used to get a fix on a timing marker.

                60

              • #

                the Griss July 10, 2014 at 5:14 pm

                “And Willis and Leif seem to have their own baskets of apples, which they are panicking might get tipped over.”

                I like the slippers!

                “Ludos… no idea where he is coming from.. I don’t think he does either.”

                That is Lubos Motil, a really sharp guy on string theory, on opening a can of beans, not so much!
                I think it should be “Force Lubos”, then you can explain it in language ununderstandable by everyone, who nod up and down!

                30

              • #

                Mattb July 10, 2014 at 6:39 pm

                “(“Can you recite a recent date when any “real hard physical scientist” did anything usefull?”)”

                “What have the Romans ever done for us?”

                I take it that that you can recite no such date!

                Jif peanut butter debuted in 1958.

                10

            • #
              Rereke Whakaaro

              I know it will come as a bit of a shock, but I cannot read Jo’s mind, nor do I know exactly what David is doing.

              What I presume is happening, is that David is identifying the key climate influences, in terms of frequency and magnitude, and then comparing each influence against the relevant parameter (if one exists) in the various climate models. Whether such an exercise would be useful or not, remains to be seen.

              But the exercise is not irrelevant, at least in the short term, because it acts as a bridge, or verification, between the Climate models, and Davids’ analytical model. If you take two paths and get to the same point, that is great. If you take two paths and get to different points, then you have identified a potential problem with one path or the other.

              Either way, you end up knowing more than you did when you started.

              110

              • #

                > I cannot read Jo’s mind, nor do I know exactly what David is doing.

                Neither can anyone else. The point is, we shouldn’t need to.

                > What I presume is happening, is…

                Either DE has explained what he is doing in sufficient detail, in which case anyone competent knows what is going on, and anyone who is “presuming” is really only admitting they haven’t understood what he has said (in which case, no-one cares about your guesses-which-you-call-presumption); or DE hasn’t, in which case no-one’s guesses are of any real interest, and you should be pressing DE to reveal more so that you don’t have to guess.

                Either way, filling up the thread with pointless guesses is pointless.

                216

              • #
                Rereke Whakaaro

                As are pointless comments about pointless guesses – and so ad infinitum …

                100

              • #
                the Griss

                RW, all the WC has ever had is pointless meaningless comments.

                Its what he does..

                Its what he IS. !!

                71

              • #

                I got your pointed comment about pointless guesses and the point I wish to make is that there is no point to this point. And I intend to make it sharply or if you will pointedly.

                60

              • #
                PhilJourdan

                Speak for yourself Billy. Your ignorance is not universal.

                70

              • #
                the Griss

                “in which case anyone competent knows what is going on”

                There ya go !! .. I told y’all he hasn’t got a clue what is going on, now he admits it.

                30

          • #

            Mattb,

            That would work if only 1/2 the data was used to train the model. If you use all the data – for better predictive power – you have to wait.

            51

          • #

            William Connolley
            July 11, 2014 at 12:45 am

            These conversations are long, slow, painful…

            Honored to be of service.

            30

            • #
              the Griss

              “These conversations are long, slow, painful… “

              And the one constant factor is the WC. !!

              He is slow and painful.

              20

        • #
          Steven Mosher

          You also test it by seeing how it was built.

          Simple.

          Why wait years when we can see the mistake
          Today

          00

      • #

        Mosh says:

        Absent that there isn’t much interesting in an 11 parameter tuned function. Heck you can get a better hindcast with one or two.

        But that isn’t the point of the model. The model is just a first step. After the first step you go looking for things in nature that explain the model behavior.

        Honestly Mosh. You are just doing a drive by without understanding the project.

        We have a pretty good idea about the lags. Oceans, land, even atmosphere, depending on the time scale. The delay is the big question. And David is fairly certain that the delay is solar. That is a big deal.

        If solar variations (not just TSI) drive the climate that changes everything we think we know about climate (warmists and most sceptics alike). David thinks he has found a signature of that. We shall see. The work is far from done and it may take months to years to work out something that is robust and backed with data.

        100

    • #

      “Typically when I want to test something I turn it on and see how it runs.”
      Best to have someone else turn it on, then come back if no smoke!

      80

      • #
        Rereke Whakaaro

        That is the most sensible advise I have seen on this thread in the last half hour. :-)

        70

        • #

          Thanks for being concerned about my health and well being. But I greatly prefer to take my own risks. I like being there for the excitement.

          50

  • #

    Explosive volcanic eruptions triggered by cosmic rays
    “An anti-correlation of silica-rich magma eruptions and solar magnetic activity is found: Eight out of eleven eruptions took place in the period when the sun is not active. This anti-correlation suggests that silica-rich magma eruptions are triggered by bubble nucleation in magma, induced by cosmic-ray muons.”

    http://www.researchgate.net/publication/234022172_Explosive_volcanic_eruptions_triggered_by_cosmic_rays_Volcano_as_a_bubble_chamber

    or here

    http://www.sciencedirect.com/science/article/pii/S1342937X10001966

    Found out about it here.
    https://www.youtube.com/watch?v=FNQj0v1rgSQ

    20

  • #
  • #
    Marc Major

    I am a first time contributor and have no expertise to offer other than as a person convinced that the AGW model is not sufficient to explain climate. However, as a computer systems analyst, I confess much empathy for Dr. Evans’ approach to climatology and completely understand his “black box” approach. In systems analysis, your desire is to model “what is and what should be” and thus, seek to implement the result of that understanding.

    As a consequence of the “black box” approach, you may not necessarily understand all causes, but your hope is that the final result of the system you deliver is that it achieves real world expectations.

    To that end, I understand that Dr. Evans model, does not seek to seek to explain physical causation but rather, model it and hope that by modeling it, understanding of the physical causations will later follow. To this end, I am constantly reminded of Einstein’s “thought” experiments; understand the principle and then seek the causation.

    In my limited understanding, I believe that, simply, this is what Dr. Evans is providing, an environment in which we can conduct many “thought” experiments by which we may come to understand the principle.

    Thank you all for a wonderful and entertaining blog.

    210

    • #
      Rereke Whakaaro

      I am constantly reminded of Einstein’s “thought” experiments; understand the principle and then seek the causation.

      Thank you for reminding us of that.

      Unfortunately, it is often overlooked today, especially by those invested in the idea that exhaling should be taxed.

      30

      • #

        It is better to collect the fare, when such gets “on” the bus! Goes way back before the physical sciences, err, perhaps not before “those” physical sciences!

        10

  • #
    Marc Major

    And by entertaining, I mean thoughtful and provocative.

    60

  • #
    Ross

    I thought this comment from the WUWT thread was well worth sharing :

    farmerbraun says:

    July 9, 2014 at 12:09 am

    Willis Eschenbach says:
    July 8, 2014 at 10:48 pm

    Willis , if you start to get a bit tired, I’ve got a spare 20 ton excavator that I could lend to you :-)

    60

    • #
      the Griss

      Australia is somewhere near the opposite side of the globe from the USA,

      ……. perhaps he is planning a sneak attack by digging through . :-)

      60

      • #
        Roy Hogue

        A roughly 8,000 mile long hole, half of it straight up? I wish him luck!

        10

        • #
          the Griss

          But if he gets enough momentum coming down, (and he certainly seems to be trying to)

          … he may just reach the other side. !

          30

          • #

            And perhaps learn something of the internal energy of this Earth!

            10

          • #

            Are there any mathematicians willing to compute the maximum velocity of a neutrino at the center of earth in an eliptical orbit with zero minor diameter. Will that give an 11 year delay?

            10

          • #
            Roy Hogue

            But if he gets enough momentum coming down, (and he certainly seems to be trying to)

            … he may just reach the other side. !

            Only if you can get a pendulum without any input except gravity to go as high on the first swing as the height from which you released it. There’s always that real world to take into account, even by the fanatic denier of the laws of physics. What a problem, nothing cooperates.

            Now of course, if he has some rocket engines to give him a boost…

            And speaking of climate change, he’ll get a good dose of what really hot means on his way through. I’d recommend asbestos coveralls. ;-)

            20

      • #

        I still covet those sloppers!

        00

  • #
    Roy Hogue

    I figured getting through this would take some time. But seeing the detail is daunting.

    Wish me luck.

    50

    • #
      Rereke Whakaaro

      Good luck Roy, we will be thinking of you … :-)

      30

      • #
        Roy Hogue

        Thank you, Rereke. :-)

        I will be working on it a little each day as I can and If something becomes a sticking point I won’t be afraid to ask for a little help. But the first problem is to simply look at what’s there and try things. You learn your way around that way and I’ve already squirrelled away a copy of the downloaded file so I can replace my working copy if I manage to screw something up. This is a good idea for anyone in my position, by the way — cheap, easy recovery with no additional download required (a lesson from years of modifying software).

        30

        • #

          I can replace my working copy if I manage to screw something up. This is a good idea for anyone in my position, by the way — cheap, easy recovery…

          I do the same when writing code. Years of hard experience have taught me that lesson.

          20

          • #
            Roy Hogue

            I once had to type a whole program into the machine again. Fortunately I had printed the whole thing the day before. It took an entire day. But I counted myself lucky because there were several weeks of work in that lost code file.

            10

  • #
    Gregorio

    I must agree with most here about the responses over at WUWT vs Joanne Nova’s site. I’ve read Climate Audit and WUWT for years, and only recently have been reading here more. I don’t comment often, hopefully, occasionally to add something of value.
    I was extremely disheartened to read the WUWT thread last night. I don’t blame Anthony Watts; I think he tries hard to keep it together, but some of the louder egotistical voices seem to get the better of him. I think this site can be a model of civility in problem solving, and maybe some of it will rub off on others in the blogosphere.
    My work in various fields over the years has put me in contact with very egotistical people, always very nasty to deal with. I don’t think I’ll quit reading WUWT blog, but I must say, I will probably put it down for a while.

    170

    • #
      rogueelement451

      Well said Gregorio , I have just dumped one on that thread ,couldn’t help myself ,the response from Willis (picnic)Eschenbach was quite sickening in its sanctimoniousness. I love it when he says people who know me will tell you i,m the first to admit it when i,m wrong , and then proceeds to add insult to injury ,[snipped]

      70

  • #

    As an electrical engineer with a good understanding of control theory I welcome this new open mind thinking.

    100

  • #
    Wayne Job

    Just watched the solar scientists giving their presentation in Las Vegas, very interesting, diametrically opposite to Leif and Willis, as Willis is in attendance I do hope he actually listened.
    A new day is slowly dawning on climate science as more people are becoming aware of the political nature of the CO2 scare. The internet is our best last chance of real open science that
    seeks truth, I have waited most of my life for science to release its shackles from the moribund consensuses of standard models, like all other branches of science before them the AGW crowd have settled on their standard model. Dr Evans, sir, you are upsetting the apple cart, this is most irregular, the amount of flack you receive will depend on how close you are to the target, be undaunted.
    Those whom seek you downfall will be those with most to loose, they will be the biggest of the charlatans and their minions, or have some called them useful idiots.

    Your project is quite worthy and a mammoth effort, many in science will support you, the faithful followers pushing the political side of AGW will be those most dangerous to both your health and wealth. Being a rebel right or wrong is fraught with danger in the science world as history shows, whilst the internet remains unmolested, you and others have a chance to make a difference and shine a little light on real science. Thank you sir.

    210

  • #
    Alex

    TSI? Like a broken clock that is right twice in 24 hours, except in this case it is right twice a year at the equinox. January- 1412 W/m2. July – 1321 W/m2. 1361 W/m2 at the equinox- 1AU. Some people have said that this doesn’t make any difference and that 1361 is the annual average. I beg to differ. The northern hemisphere is land heavy and receives snow on it in winter and the southern hemisphere is mainly water and doesn’t get much snow resting on the surface. To me it means there is a great imbalance of energy from the sun between the north and south ie a cycle. If one chooses to they can average this cycle out. Unlike Willis I don’t live on a chequerboard planet.

    60

  • #

    Interesting.

    The list of dancing marionettes grows. They continue to dance to the old tunes believing they know the rules. David is using a set of rules they have never encountered or at least not considered relevant. David approaches the task as an engineer who intends to make something that really does work. To do that, he must reverse engineer the system he is intending to replicate. He is using a set of rules that every engineer knows explicitly or implicitly. It was once called the scientific method. The dancing marionettes haven’t a clue and just keep dancing.

    Reverse engineering cannot be done by guessing the bits and pieces of a thing to be replicated (aka the internal physics). This is because the thing to be replicated is a natural system of unknown components with their individual behaviors and interconnections also unknown. Further, it cannot be disassembled to test each part by itself because you also don’t know where to slice and dice it. You can only guess. Some guesses will be right by accident but most of them will be wrong. You will have no way to know which is which.

    Even if you did slice and dice it, the major portion of the system behavior, that which arises out of the interconnections, will be destroyed. Hence, an atomized bottom up reconstruction will lead to something but not the system to be replicated. At least not one you can verify and validate compared to the real system, who’s behavior, you have not attempted to understand.

    Imagine trying to discover a sodium atom by starting at the level of an arbitrary set of sub-atomic particles and attempting to construct all possible atoms from the parts. Then select only the sodium atom.

    You might be able to construct something but can you construct a sodium atom. If you think you have, can you verify that you did without first discovering the properties of sodium metal and from that the properties of the sodium atom? You can’t. In fact you can’t even have the concept of a sodium atom or sub atomic particles using a bottom up approach.

    The only way man discovers anything is starting at his level of perception, thinking about what he sees, and doing experiments to find and verify the next layer of knowledge just underneath the surface. Then just underneath that layer. Eventually working down to the discovery and verification of the fundamental forces, particles, and interactions to be found in his environment.

    In other words, a top down approach IS science in the only way it can be done by a being who must learn everything rather than just knowing it. Once the science is done, then you can assemble the pieces to make whatever you want. That is, if the pieces you know and understand will permit it. This is engineering.

    It is a contradiction to expect to be able to make something that works if you don’t know what it is nor how it behaves nor all the pieces that must be or can be used to build it. Yet that is exactly what the bottom up approach to wannabe science attempts to do.

    That is worse than alchemy or astrology. They are merely ineffective approaches to discovering reliable knowledge. The bottom up approach with incomplete knowledge combined with arbitrary guesses simply won’t work except by accident.

    Even if by some miracle it does work, you won’t know that it works.
    The most you can do is assert “the science is settled.” Even that assertion is bogus because nothing can be settled by such an approach. It eventually devolves to “he said”, “she said”, then onto “I will bash your head in if you don’t agree with me.” Sadly, we are coming all too close to the bashing of heads stage.

    Yet, the dancing marionettes continue to dance as that is all they can do.

    130

    • #

      The bottom up approach with incomplete knowledge combined with arbitrary guesses simply won’t work except by accident.

      It is implicit in the IPCC approach. And near everyone, warmist and sceptic alike, has fallen for it.

      If you go to the Monckton video http://www.ustream.tv/recorded/49805723 at the Heartland conference you will find that when he does the audience query they ALL are lukewarmists. They believe – on no sufficient evidence – that CO2 affects climate. Which means they are reduced to picking holes in the IPCC models. David’s approach has the possibility of blowing their models out of the water.

      The funny thing is that the “sceptics” (not all of them) are defenders of the IPCC approach.

      The Emperor ha NO clothes.

      90

  • #
    Brad

    NOAA just issued a polar vortex alert for the US Midwest? 10 to 30 degrees below average.

    40

  • #
    mesoman

    I hope Steve McIntyre will take a look at the model, try to reproduce the results, and offer constructive criticisms. It would be good to see because he’s meticulous and does not have an ego the size of a blimp, unlike a couple of other guys I know.

    70

    • #
      steven mosher

      doubtful.
      The first things he will ask for is
      1. out of sample tests, something he regularly requires.
      2. code and data used to construct the model.

      its pretty simple.

      Then he would consult Leif about which TSI dataset is the preferred data. Leif used to be the resident
      Solar expert at CA before he went to WUWT.

      You see Willis and I are not asking for anything more than we were trained to ask for by spending years on ClimateAudit

      010

      • #
        Backslider

        Willis and I are not asking for anything more than we were trained to ask for

        Ruff ruff!

        How about asking for something you might understand? The code and data are there.

        70

        • #
          steven mosher

          yes the code for excercising the model and the data are there. Downloaded and examined as promised.

          The missing bits

          1. The out of sample testing that david conducted ( well he says he did)
          2. The code used to construct the model.

          I remember the first model I was given to run. A model for an ESA radar. It was an equation and some data tables.
          Running it told me nothing other than the model ran.

          Some guys suggested that we run a big test of the model against real world results.

          I asked the guy for his code that he used to build the model.

          he supplied it.

          he forget a sin() term in a transformation matrix. The model was bogus. No need to test the garbage out when it was evident that a mistake had been
          made in model construction. And he used some outdated data. Once those were fixed the model became testable. There is not too much use in testing a model
          you haven’t audited.

          Its simple.

          23

          • #
            Backslider

            1. The out of sample testing that david conducted ( well he says he did)

            Why don’t you do your own?

            2. The code used to construct the model.

            Perhaps you need to learn how to code? It’s there.

            The fact is that you have it all, to do as you wish with it. I think you listen too much to Svaldork’s whining.

            50

      • #

        Mosh,

        Perhaps Willie Soon might have something to say about the correct solar data set.

        He discusses it in this video http://www.ustream.tv/recorded/49735731 starting at 34 minutes in.

        Leif is a charming fellow but I’m not convinced he understands the radiometers used to gather the data. Wille – from his appearance in the video – seems better grounded.

        ==================

        BTW it would be useful to get the bits you want, I’m sure. But you don’t need them to get started.

        You could test the model against white noise, red noise, or any other data set you like to check for biases – to start. You know. What was done for M. Mann.

        40

        • #
          steven mosher

          Actually I do need them to get started.

          I have a principle. It is simple. Follow the data to the source first.
          Unpact the models to the underlying models first.

          Why should I treat David E different than I treated Mann?
          Why should I treat his work with less skepticism than I treat GCMs

          Second. I would not rely on Soon. I don’t need to. The data, the actual hand written data is there to check.
          You get Leifs work. You understand his method. You see the data. You take that data and re work from scratch.
          Why wait for Soon? and if Soon comes out with something he gets the same fricking treatment. hey Willie show your work

          It is brutally simple. And consistent. david makes a claim.

          A) this data was processed with methods to produce a model.

          really?

          cool. show that work.

          But yes, the spreadsheet works. it outputs numbers. that’s not interesting. I’m interested in the claim about how it was built, not whether or not a spreadsheet works

          02

          • #

            Well. OK then. You can just hang on and see if DE’s predictions turn out to have some merit. If we go into a little ice age as predicted by both David and Habibullo Abdussamatov (both predictions on solar grounds) then there may be something to their models.

            I’m sure you have read about the summer polar vortex coming to the US Midwest. Did you know that Russia is also having a similar experience?

            http://notrickszone.com/2014/07/14/polar-vortices-everywhere-central-russia-hit-by-mid-summer-freak-abnormal-snowstorm-snowdrifts/

            Of course it is just weather for now, If it continues or gets worse in the coming years it may mean the climate is shifting. Look for period doubling. i.e. twice as many such events (2 or 4 or 8 or chaos). The rate that that happens (if that happens) will give an indication of how fast climate is changing.

            ===================

            The point about Willie Soon IS the data. Have a look at the video and see what he is complaining about.

            ===================

            My area of expertise is not in thrashing data. I’m an instrument designer. That keeps me occupied well enough. So even looking at the data wouldn’t do me much good. The best I can do in that area is understand the general principles.

            In my work if I need a mathematician I hire one (or one is hired for me).

            ===================

            You might get some amusement out of this:

            http://chiefio.wordpress.com/2014/07/14/a-rather-useful-archive-of-individual-station-modification-graphs/

            The graph at the bottom right shows that the past was cooled, and the present warmed. Clearly and obviously.

            Now, to me, it isn’t so much the warming present and cooling past, as that pretty much every graph has more change from adjustments than it does from actual trend. Those that are not changed generally are so short of data that there isn’t much point. (Though there are graphs that are unchanged).

            What’s the net-net of it? Hard to say, but I’d say mostly a “Global Warming” signal that comes out of the adjustments, not out of the data.

            =========

            Well that is always the danger of adjusting the record. It gets corrupted rather than improved. If you knew how the system worked making adjustments can have at least some rational grounds. But adjusting to find out how the system works is a fools errand. IMO.

            00

          • #
            • #
              steven mosher

              neither are the least bit relevant.

              it is simple.
              link to data.
              link to code.
              papers are just advertisements for the science. the science is in the application of the code to the data.

              First question when reading a paper. Are the claims, the tables, the graphs, ACTUALLY produced by the methods and data referred to in the text.
              First check.

              without the data as used and the code as run, you cant do the first check.

              it is simple. just treat every paper and every claim to the same test and procedure

              OR you could believe the conclusions first and then defend it. Like folks who believe in Mann’s work

              10

              • #

                I’m not a believer. But more than a few people say we are headed towards a cooling period despite CO2. And there is a fair amount of evidence that CO2 has no effect on climate – other than making plants grow faster.

                So what have I thought about the last 15 or so years of no temperature rise? The flat top of an approximately sine wave before the decline in said wave.

                David’s analysis seems to confirm that. Habibullo Abdussamatov predicts it.

                http://hockeyschtick.blogspot.co.uk/2014/07/analysis-solar-activity-ocean-cycles.html

                Analysis: Solar activity & ocean cycles are the 2 primary drivers of climate, not CO2

                Dan Pangburn has updated his analysis identifying the two primary drivers of global temperature:

                1) the integral of solar activity
                2) ocean oscillations [which are in-turn driven by solar activity and perhaps lunar-tidal forcing].

                The correlation of the integral of solar activity and ocean cycles to global temperature is 90.49%, and with the addition of CO2 the correlation only improves very slightly to 90.61%, demonstrating CO2 change has no significant effect on climate.

                00

              • #
                Steven Mosher

                [SNIP. Sorry, we can't discuss the Holocaust, or Holocaust denial on this site. - Jo]

                01

              • #

                Steven Mosher
                July 16, 2014 at 11:08 am

                Given the adjustments to the data some people are claiming that we are already well in to a cooling period.

                And then Habibullo Abdussamatov says the cooling will be noticed in 2014 and accelerate in a downward direction from there despite rising CO2.

                My take? Ocean cycles were confused with CO2 action. If that is the case actual CO2 effects could be as low as zero. As this analysis indicates:

                http://hockeyschtick.blogspot.co.uk/2014/07/analysis-solar-activity-ocean-cycles.html

                If it is all Solar+Ocean Cycles – the people standing up for the CO2 hypothesis are going to be in a world of hurt.

                00

      • #

        Don Jordan
        July 11, 2014 at 6:53 am

        Now, my dabbling into this subject is certainly not as deep as others on this blog. but as a software developer, this is a fascinating spread sheet to play around with. Kudos to Dr. Evans for releasing this work in such a fashion: it doesn’t get more transparent than this.

        It would be nice if the other side were so forthcoming about allowing the public to examine their work. After all, since they want to tax us to death, the least they can do is let their work be subject to the same scrutiny as Dr. Evans.

        ===============

        And this model is just the first major step along the way. The next is looking for inputs whose amplitudes and frequencies match that model (Bode Plot) in order to turn the work into physical reality.

        You might also find this of interest:
        http://wattsupwiththat.com/2014/07/10/unusual-weather-were-having-aint-it/#comment-1682599

        I’ve heard of Piers Corbyn saying we’re right now in the transition to the meridional jet stream based Little Ice Age pattern that we may be stuck with for the foreseeable future.

        =================

        That (Little Ice Age pattern) is also confirmed by prediction by Habibullo Abdussamatov in the above linked video. You can start at about 33 minutes in for the punch line.

        ==================

        I’m also finding that volcanic activity correlates with low solar activity – which would indicate that solar activity (unspecified for now) controls volcanoes.

        http://iceagenow.info/2014/05/ice-ages-correlate-increases-volcanic-activity/

        =============

        You might also find a series of comments that discusses volcanoes with respect to the sun of interest. Starting here:

        http://joannenova.com.au/2014/07/big-news-ix-the-model/#comment-1506121

        ==========

        You might also want to search Google for “volcanoes and solar activity” especially have a look at the NASA Goddard pages.

        40

        • #
          the Griss

          “I’m also finding that volcanic activity correlates with low solar activity ”

          Might be interesting to try a correlation against the Sun’s MAGNETIC activity.

          50

      • #
        CC Squid

        Have you heard about the solar vortex hitting the US next week? Temps will be 20-30 f lower than normal. Colorado has been cloudy and cool and the temps are the coldest I have experienced in 30 years… It has been the same in Georgia.

        40

  • #
    Don Jordan

    I’ve been following Joanne since she was a guest on John Loeffler’s “Steel on Steel” radio show / podcast. Now, my dabbling into this subject is certainly not as deep as others on this blog. but as a software developer, this is a fascinating spread sheet to play around with. Kudos to Dr. Evans for releasing this work in such a fashion: it doesn’t get more transparent than this.

    It would be nice if the other side were so forthcoming about allowing the public to examine their work. After all, since they want to tax us to death, the least they can do is let their work be subject to the same scrutiny as Dr. Evans.

    60

  • #
    larrygeary

    I just wanted to add a thought about the “delay” component of the hypothesis. If “force X” is related to cosmic rays, then there may be an additional delay due to propagation. I’m thinking in terms of the time required for the particles emitted in a Coronal Mass Ejection (CME) to reach the heliopause – just over one year. If the sun’s magnetic field weakens, of course that influence travels outward at the speed of light, but incoming cosmic rays might take much longer to travel from the heliopause to earth. Cosmic rays come in a large range of energies, and thus speeds, from near light-speed to relatively slow – and which ones are important is unknown – but my point is that there may be an additional delay of up to a year or more before any increased or decreased cosmic ray flux reaches earth.

    Larry, thanks, I’ve sent you an email. – Cheers – Jo

    50

    • #
      Scute

      I agree, larrygeary. Also, both Voyager 1 and 2 have recorded known CME’s on the time scale you mention. For example, Voyager 2 detected the October 2003 Halloween Solar Storm in April 2004. It was still a long way from the heliopause at that point, very roughly two thirds of the way. So that’s definitive proof for any alarmist doubters who may question this notion on the basis that the CME’s slow down. They do slow but by nothing like enough to wane away to nothing. The Voyagers also detected the Bastille Day storm of 2000 on a similar time frame. Paper here:

      http://onlinelibrary.wiley.com/doi/10.1029/2002GL014729/full

      Also, confirmation of 410-day Propagation time to heliopause here (paywalled):

      Gurnett, D. A., W. S. Kurth, S. C. Allendorf, and R. L. Poynter, “Radio emission from the heliopause triggered by an interplanetary shock”, Science, 262, 199, 1993.
      CrossRef,CAS,Web of Science® Times Cited: 149,ADS

      00

      • #

        Please explain what good a 1 year delay does if you are looking for a 3 to 15 year delay? In addition the Sun is about 8 minutes from Earth at light speed. At 1/100th of light speed that is 800 minutes. That is 13 hours. 1/1000th of light speed is about 5 days. The slowest CMEs take about 75 days to reach the Earth. Typical is on the order of 4 days. The time scales don’t match.

        00

        • #

          MSimon
          Just as lightning causes the Schumann resonance to circulate in the earth ionosphere cavity, could this cause the whole heliopause/sun cavity to resonate at a far far longer period.

          or

          As i have said in previous comments if the trip from the heliopause on to the next stars 4 and a bit light years away is magnetic then an SWR type reflection would take a good round trip time.

          00

          • #

            SWR? Hadn’t seen that in a non EE forum in a long time.

            =============

            The timing may be right at 8 light years but the attenuation is a killer.

            The ringing of the heliopause/sun cavity may be more promising. If some one is looking for it.

            I’m still in the solar magnetic direct camp on this one. Not that my opinion is worth anything against evidence.

            00

        • #

          Good thinking. A faulty lever to where, would produce an 11 yeay delay?
          Do we need to repair the whole damned Universe

          10

  • #
  • #
    jim2

    Dr. Evans. Not sure if this would be useful to you, but here is a technique to make any two arbitrary frequencies orthogonal. Being an electrical engineer, you probably already know, but on the off chance you don’t, here is is:

    http://electronics.stackexchange.com/questions/58221/orthogonal-frequency-division-multiplexing-calculate-what-frequencies-are-orth

    10

    • #

      Funny. They have an OSH Park PCB mfg ad at the site. They have been my favorite for prototypes for 3 years. Very good work. Their motto? “We’re fab.” http://oshpark.com/

      As to to OFDM. That helps if you are generating frequencies. If you are trying to decompose them David’s version of the Fourier transform is what you want. Especially if real time is not a requirement.

      20

  • #
    Andrew

    While reading some of these comments, ozone is regularly mentioned.

    This paper ( 3 papers ) came to mind http://globalwarmingsolved.com/2013/11/summary-the-physics-of-the-earths-atmosphere-papers-1-3/
    You can jump the first as it is an introduction

    The second is interesting, may well be relavent. ” multimerization of atmospheric gases above the troposphere ”
    It proposes an new mechanism for the formation of jet-streams, hi/lo pressure systems, with associated albedo effects

    I think Stephen Wilde has read and commented on it a the Talkshop some months ago.

    If force X is linked to the spectral change and or Ozone then this paper may be relavent.

    10

    • #

      In the end I decided that it introduced unnecessary complexities that might not be sound.

      I prefer to stand by the simplicity of my New Climate Model which I hope will be found to be consistent with David and Jo’s work.

      43

  • #

    http://www.theregister.co.uk/2014/06/16/brit_boffin_to_obama_science_advisor_you_are_wrong_on_climate_change/

    16 Jun 2014

    Climate change is unlikely to lead to more days of extreme cold, similar to those that gripped the USA in a deep freeze last winter … [Recent changes in the Arctic climate have] actually reduced the risk of cold extremes across large swathes of the Northern Hemisphere.

    =============

    Mother Nature certainly has a sense of humor:

    http://www.abc2news.com/weather/weather-blogs/polar-vortex-to-take-a-vacation-south-next-week

    10

  • #
    Paul Vaughan

    Due to the darkly ignorant &/or deceptive solar-climate thought-policing, I stopped reading the comments in WUWT solar-climate threads a few years ago.

    To see who’s still there, I did a quick-scroll through the names of commentators on the wuwt thread running parallel to this JoNova thread.

    I was pleased to see the names of so many sensible commentators completely absent.

    There has been an out-migration (from wuwt) of sensible solar-climate commentators.

    Perhaps soon the only people left commenting on wuwt solar-climate threads (aside from a few brainless groupies) will be the excessively rude, darkly ignorant &/or deceptive wuwt solar-climate thought police.

    WUWT may have other uses we can all agree on (e.g. recent climate news, insightful Bill Illis comments, Bob Tisdale SST graphs, etc.), but its utility for productive solar-climate discussion is 100% dead.

    40

  • #
  • #
  • #
  • #

    In the past there was a 10 to 12 delay in cooling when there was a change from a very active solar magnetic cycle to a Maunder minimum. There is a physical reason for the delay in cooling.

    http://wattsupwiththat.com/2013/02/20/its-the-sun-stupid-the-minor-significance-of-co2/#comment-1229333

    00

  • #

    http://www.jpl.nasa.gov/news/news.php?feature=2420

    Study confirms theories that Earth’s liquid outer core is slowly “stirred” in a series of regularly occurring waves of motion that last for decades.

    Measurements of Earth’s magnetic field from observatory stations on land and ships at sea were combined with satellite data to determine common patterns of movement within Earth’s core.

    The findings give scientists new insights into Earth’s internal structure, the mechanisms that generate its magnetic field, and its geology.

    Earth’s magnetic field shields us from harmful solar radiation and has many practical applications, ranging from navigation to archaeology.

    20

    • #

      Study confirms theories that Earth’s liquid outer core is slowly “stirred” in a series of regularly occurring waves of motion that last for decades.

      10

    • #
      the Griss

      All these many things affecting the Earth’s climate, at many different frequencies….

      ….. no wonder there is little or no TSI frequency signal in the global average temperature.

      Just a whole mess of different delays and smudging at a whole heap of different frequencies.

      No-one can ever hope to model every single effect in its correct proportion ….

      ……but maybe, perhaps, someone has found a way to identify the overall effect.

      20

      • #

        Thank you for the sloppers, nice, lovely, or perhaps enchanting!!!

        Can anyone tell me what Global average temperature may mean?
        Is that something like the Global average flavor of spices?
        Do even stastical trends in such, have any meaning?
        Spend lots to correct the adverse trend in flavor of weeds, like Oregano?
        ————Think of the Children!!!
        How much has already been spent? Stupid earthlings….

        10

    • #
      Andrew

      Regarding the magnetic field.

      http://notalotofpeopleknowthat.wordpress.com/2014/07/11/earths-magnetic-field-weakening/

      Surprised this hasn’t gathered more attention

      20

  • #
  • #

    Nevertheless, the beautiful thing is that within the errors in the data sets (and estimate for the systematics), all three sets give consistently the same answer, that a large heat flux periodically enters and leaves the oceans with the solar cycle, and this heat flux is about 6 to 8 times larger than can be expected from changes in the solar irradiance only. This implies that an amplification mechanism necessarily exists. Interestingly, the size is consistent with what would be expected from the observed low altitude cloud cover variations. – Nir Shaviv

    http://www.sciencebits.com/calorimeter

    http://wattsupwiththat.com/2013/02/20/its-the-sun-stupid-the-minor-significance-of-co2/#comment-1229591

    10

  • #

    Dr Norman Page says:
    February 21, 2013 at 9:55 am

    http://wattsupwiththat.com/2013/02/20/its-the-sun-stupid-the-minor-significance-of-co2/#comment-1229738

    To me there is little doubt that Shaviv is more or less right. The Iris effect seems to have the largest effect on climate. Look at Fig 3 and 4 above. As mentioned in my post there is a good general 20th century match allowing for the lag time. The 20th century peak on your red curve is at the Cycle 22 max and is the highest point on the entire chart.Look at the NGRIP data in the Berggren paper you referred to. The correlation of the high Be flux with the Maunder and Dalton minima is incontrovertible as is the Be low with the 18th century warmth from about 1730 on and the generally falling trend and low flux values in the 20th century. I dont know what the TSI was doing for much of time frame – but the percentage variation in TSI is much less than the variation in incoming GCRs which is why the latter is the controlling influence on climate at decadal,centennial and millenial time scales.

    As you rightly point out the Ap number during the 23 /24 minimum in late 2009 was down to 1900 levels which is why I will make a testable prediction that by 2019 -20 NH temperatures could well be surprisingly cool perhaps even 0.8 below current levels – though that seems a shocking amount even to me.

    10

  • #
  • #

    Good bit with lots of links:

    http://wattsupwiththat.com/2013/02/20/its-the-sun-stupid-the-minor-significance-of-co2/#comment-1229941

    And the good stuff (referring to Leif):

    Then there are all the other studies that support Shaviv’s position and not yours.

    Here is an interesting one from 25 years ago that predicted the current quiet sun.

    00

  • #
    Paul Vaughan

    From a 1997 article written by an author named David:
    “transfer function [...] telephone equalizers”
    “noncausal with a width about the length of a solar cycle”

    -
    Here’s the dead simple reason why it’s important for readers here to familiarize themselves with that author’s work:
    The sunnot CO2 — governs the first multivariate mode of terrestrial climate.

    40

    • #
      David Evans

      Different David! See this reply.

      10

    • #

      Putting aside my anti-CAGW prejudices the preceding David J. Thomson paper ‘The Seasons, Global Temperature and Precession’ equally gripping….
      http://www.sciencemag.org/content/268/5207/59.full.pdf

      Warning – slowness/refusal to load the free pdf possible due to the double dots (i.e. may not pass through MailScanner on some Unix servers) but should get there eventually.

      10

      • #

        It is not a free pdf for me. I’m not a member of AAAS.

        00

        • #
          • #

            Nor the ACS.

            But this is a useful gem too:

            http://people.earth.yale.edu/sites/default/files/files/Park/Park_2011_CO2coherence.pdf

            ABSTRACT.
            Large-scale carbon-cycle feedbacks within Earth’s climate system can
            be inferred from the statistical correlation of atmospheric CO2 and other climate observations. These statistical relationships can serve as validation targets for global carbon-cycle models. Fourier-transform coherence between atmospheric CO2 measured at Mauna Loa, Hawaii, and Hadley Centre global-average temperatures changed in the late 20th century at interannual frequencies, from a 6-month time lag to a 90° phase lag that scaled CO2 fluctuations to a time-integral of the
            global-average temperature anomaly. Wavelet coherence estimates argue that this change occurred with a recognized ocean-circulation climate transition during the late 1970s. General features of these CO2-temperature correlations are confirmed using global-average temperature from other sources and atmospheric CO2 measured at other locations, though only the Mauna Loa CO2 record is long enough to resolve well the coherence properties before the 1970s transition.
            The CO2-coherence phase for the global- average surface-air temperature time series from NASA-GISS and the lower- troposphere temperature series from the MSU satellite is more complex than for the Hadley-Centre dataset, the only estimate that incorporates sea-surface temperature
            (SST) observations. Near f = 0.25 cyc/year, 4-year oscillation period, the CO2-coherence is particularly strong for the Hadley-Centre gridpoint temperature-anomaly time series from low-latitude oceans. This suggests that sea-surface temperature is a primary driver of the correlation, at least for the 0.2 < f < 0.5 cyc/yr bandpass where the El-Nino/Southern-Oscillation (ENSO) climate process dominates. Outside the ENSO bandpass coherence is significant
            between 14 long-running GLOBALVIEW CO2-observing sites and the sea-level-pressure-based Southern Oscillation Index (SOI) and North Atlantic Oscillation (NAO) time series, consistent with wind stress and mixed-layer-thickness influences on ocean-atmosphere CO2 flux, independent of temperature fluctuations. Evidence for terrestrial biosphere influence is strongest in the leading
            principal component of GLOBALVIEW CO2-variability at f = 0.25 cpy, where a larger amplitude and a 4-month phase shift distinguish the mid- and high- latitude Northern Hemisphere CO2 fluctuations from those of the tropics and the Southern Hemisphere. The terrestrial signal we infer, however, coheres more strongly with oceanic-gridpoint temperatures than to continental-gridpoint temperatures.

            Perish that 'heretical' thought that 'Force X' may be CO2 cycle-driven….(;-) or indeed that we at the mercy of an 'Intelligent Design' albeit one evolved by a bunch of dumbass photosynthetic Unicellular organisms whose bandwidth just happens to cover the 400 – 700 nm window wherein TSI is most variable.

            Shock horror – one might be tempted to call that divine justice!

            00

        • #
  • #

    We are in a Little Ice Age already:

    Thursday, May 31st 2012, 8:01 AM EDT

    Piers Corbyn to make changes to his Solar Lunar Action Technique(SLAT) to fall in line with the NEXT “Little Ice Age”

    Piers Corbyn, Astrophysicist of WeatherAction.com said:

    “After studying weather developments in the last two months we have introduced some modifications to our Solar Lunar Action Technique. SLAT7a – which was about improving selection of detailed look-backs using lunar factors more strongly – did superbly at predicting the exceptionally cold weather in the first three weeks of May but does not well see the VERY RAPID (“Little Ice Age” type) changes experienced and specifically missed the sudden warming of around 22nd.

    “Although our ‘at least 6 out of 8 weather periods basically correct’ in a month was met in May we want to improve on that and better handle the “approach to LIA weather types” REGIME the world is now in.

    http://climaterealists.com/?id=9720

    10

  • #

    http://climaterealists.com/index.php?tid=717

    A Graph to Debunk AGW: Solar Geomagnetic Activity is highly correlated to Global Temperature changes between 1856-2000

    A paper published in the journal of the Italian Astronomical Society finds that solar geomagnetic activity was highly correlated to global temperature changes over the period from 1856-2000. The authors “show that the index commonly used for quantifying long-term changes in solar activity, the sunspot number, accounts for only one part of solar activity and using this index leads to the underestimation of the role of solar activity in the global warming in the recent decades. A more suitable index is the geomagnetic activity which reflects all solar activity, and it is highly correlated to global temperature variations in the whole period for which we have data.”

    20

  • #
    No Body Important

    EXCEL #0: Those complaining about use of Excel not being ‘open-source’ (i.e. free). If you can’t get a basically ‘free’ copy of Excel then you are not trying hard enough, … student version? employer Home User Program? eval version from MS? etc etc etc

    EXCEL #1: Bravo Zulu for putting up the entire XLSM with the VBA code viewable by whoever downloads it; although the inevitable s/w language flame war is to be expected (Excel vs R vs Matlab vs Mathematica vs XYZ). At least it will not get as nasty as the EMACS vs VIM text editor battles.

    EXCEL #2: Princess suggestion with tab naming convention. In addition to color coding, suggest data tabs start with “DATA – “, model output tabs start with “OUTPUT – “, etc

    EXCEL #3: Version control, regardless of method (gitHub, other), suggest _YOU_ do a chksum/hash of the version _YOU_ release once a week as the ‘official’ version

    FUN PROPOSAL: Time for an “Evans-Svalgaard wager” where Anthony puts a count-down clock in upper right corner of WUWT, ditto JoanneNova, ditto ClimateAudit, ditto … counting down to date at which David Evans’ model says XYZ will happen. Call it a beer wager akin to the “Simon–Ehrlich wager”, except call it the “Evans-Svalgaard wager”, maybe rig up a PayPal thing to let the audience participate with monies going to favorite charity of winner. Imagine there would have to be some negotiation of the terms: how measured, by whom, plus-minus predicted measurement, plus-minus time, who measures, exceptions like asteroid strike exceeding certain size, volcano eruption exceeding certain size, nuclear weapons exchange, Canary Islands mega-tsunami, man-made aerosol discharge exceeding certain size by wealthy loony person delusional about saving world, war, etc etc etc)

    20

  • #
    Andrew

    This just popped up

    Solar & ocean cycles explain earths climate. No co2 required http://hockeyschtick.blogspot.co.uk/2014/07/analysis-solar-activity-ocean-cycles.html

    00

  • #
    Truthseeker

    David and Jo,

    Willis Eschenbach has made an apology and has outlined his objections in a “Canadian” way … I think that means being nice instead of nasty …

    00

    • #
      Deej

      T,
      the link provided doesn’t work…

      00

    • #
      • #
        Rod Stuart

        Willis has been publishing a delightful little travel diary since his visit to Vegas. His reference to travelling with his “ex-fiancée” would suggest that he just got married.
        Presumably, the change of heart may be related to a release of frustrations (or something).

        01

        • #

          He has been married a long time. Ex-fiancée is sort of a joke. He uses it instead of the term “wife”. I use “first mate”. Being a Navy man from a Navy family. The last of the line so far. None of my kids are into the sea.

          00

        • #
          Paul Vaughan

          Bill Illis is the only sensible commentator on that thread.

          10

    • #
      gnomish

      the title of that post is ‘mending fences’.
      it starts with a kind of hillary on benghazi ‘what does it matter’ kind of dismissal of any particulars and seems to develop into ‘i’m sorry you’re so stupid’.
      he then proceeds to wave the ‘anonymous coward’ label at anybody who questions whether it was an actual apology…
      i’m a fan of his good stuff – and maybe that’s why it’s been a violation of my expectations to see how, over the months, he’s become so image conscious and defensive. vast, cool and indifferent he’s not. also, when he brags about his humility, i just have to laugh. he doesn’t even get it? he needs a real friend to give him a clue. his imaginary ones only agree with him and cherish his bad aggrandized self.

      10

      • #
      • #

        i’m a fan of his good stuff – and maybe that’s why it’s been a violation of my expectations to see how, over the months, he’s become so image conscious and defensive. vast, cool and indifferent he’s not. also, when he brags about his humility, i just have to laugh. he doesn’t even get it? he needs a real friend to give him a clue. his imaginary ones only agree with him and cherish his bad aggrandized self.

        Yeah. It particularly hurts me because we have some “friends” in common from the 60s. And then he goes off on me with the epithet “EE thinking” (roughly).

        00

        • #
          gnomish

          i think it’s ‘guru syndrome’, where a person insulates himself from feedback of reality with a layer of sycophants.
          i don’t care for the slander, tho. no excuse for it.
          so ‘mending fences’ was definitely not what that post was about.
          oh well.

          20

          • #

            Evidently he is now doing slander without anger. HUGE improvement. /sarc

            00

            • #
              gnomish

              lol- i was tempted to call that point out over there about using wuwt as a platform for that – but since you noted it, i reckon it must have been quite plain.
              poor the mods… they try to be fair but the biggest troll, now… ah… n/m
              i just shot 286 beautiful parts in 150 minutes with my injection molder in the mold i finished cutting tonight and i’m stoked. can’t dampen my spurts!
              sad to lose a hero or a friend. life ain’t always according to disney, tho.
              but in the end- we all go out the same door, eh?

              10

              • #

                Do you do IM for a living or a hobby?

                If you live in the States I have some ideas. If you live elsewhere shipping costs are too high.

                Contact me by e-mail. It is on the sidebar.

                00

            • #
              gnomish

              whoa- i just looked at your site!
              ima raid your smorgasbord. very cool.

              00

              • #

                If you want something at my site let me know. I’ll work something out with you if you are not a builder. If you are a builder I can “hold your hand” through the process.

                I have some things not yet announced. And I will do custom for a very reasonable price. Thanks for the TI link. I do use those principles for interfacing between logic voltages.

                00

            • #
              gnomish

              check this out:
              http://www.ti.com/lit/an/scla011/scla011.pdf
              it’s especially about using hct for interfacing different logic levels.
              maybe useful for you.

              00

  • #

    This is fabulous! So many studies and models are hidden behind paywalls.

    If I ever find the resources, I would like to support this any way I can. I am a software developer and would be willing to convert this to a web based application for: 1) Greater accessibility, 2) Expanded education on how it works, 3) Continuous updates – automatically sourcing observations so that on any given day, one can view the conclusion.

    For now it’s all just promises, I hope we can amplify this and expand to broader study.

    00

  • #

    [...] cycle does not show up in temperature or precipitation data. Most recently, Dr. David Evans has introduced his “notch filter” answer to the problem. I think the answer is much [...]

    00