Options

Is my climate model right? A software engineering perspective

Forum thread for my first blogging attempt, the draft is here: Blog - Climate Model Evaluation. This is just to announce that I maybe will write something in the future, if I actually do, I'll give another heads up here.

Comments

  • 1.

    After some writing I noticed that I should explain some technobabble first, since several people who have written about climate models have confused or misunderstood "stakeholder" and "validation versus verification". So I did that, instead of writing about climateprediction.net, which I did not get running (yet) anyway.

    Anybody interested may have a look at the draft, feedback is welcome.

    Comment Source:After some writing I noticed that I should explain some technobabble first, since several people who have written about climate models have confused or misunderstood "stakeholder" and "validation versus verification". So I did that, instead of writing about climateprediction.net, which I did not get running (yet) anyway. Anybody interested may have a look at the draft, feedback is welcome.
  • 2.

    this is a good start but you should not expect that non-programmers knows about floating-point number and other terms. at least link to wikipedia entries. mabe all technobabble can be like that ? also i think this is more traditional software process than climate model evaluation, so i'd add more on the latter :-)

    Comment Source:this is a good start but you should not expect that non-programmers knows about floating-point number and other terms. at least link to wikipedia entries. mabe all technobabble can be like that ? also i think this is more traditional software process than climate model evaluation, so i'd add more on the latter :-)
  • 3.
    edited June 2011

    Brief comment while I've got net access:

    Nice stuff. One thing: I think it's a good idea to have the first couple of paragraphs of something like this saying in simple declarative statements what you're going to do in this article (series). That helps the audience decide if they're interested in where you're going so they can use it to decide whether to read the whole thing. Something like (possibly slightly longer sentences, but just giving a)

    "Climate change is widely argued to be very important. One important way to analyse climate change is to use computer simulation using physics based models to predict future possibilities. I'm going to talk about running climate models, and in particular a distributed climate model you could run on your own computer. But to sensibly talk about this we'll need to understand what a computer climate model should do, which will involve developing some concepts used in climate science and some used in computer software engineering, as well as understanding who will use the climate model output and how their priorities affect things. Bet you didn't think there was that much involved, did you? Anyway, lets dive in..."

    Comment Source:Brief comment while I've got net access: Nice stuff. One thing: I think it's a good idea to have the first couple of paragraphs of something like this saying in simple declarative statements what you're going to do in this article (series). That helps the audience decide if they're interested in where you're going so they can use it to decide whether to read the whole thing. Something like (possibly slightly longer sentences, but just giving a) "Climate change is widely argued to be very important. One important way to analyse climate change is to use computer simulation using physics based models to predict future possibilities. I'm going to talk about running climate models, and in particular a distributed climate model you could run on your own computer. But to sensibly talk about this we'll need to understand what a computer climate model should do, which will involve developing some concepts used in climate science and some used in computer software engineering, as well as understanding who will use the climate model output and how their priorities affect things. Bet you didn't think there was that much involved, did you? Anyway, lets dive in..."
  • 4.
    edited June 2011

    Quoting Dyson on climate models? Uh, sorry, but ummm... you could as well quote Bjorn Lomborg or Michael Crichton.

    Here is the full quote from Dyson's infamous edge.org essay (emph. mine):

    My first heresy says that all the fuss about global warming is grossly exaggerated. Here I am opposing the holy brotherhood of climate model experts and the crowd of deluded citizens who believe the numbers predicted by the computer models. Of course, they say, I have no degree in meteorology and I am therefore not qualified to speak. But I have studied the climate models and I know what they can do. The models solve the equations of fluid dynamics, and they do a very good job of describing the fluid motions of the atmosphere and the oceans. They do a very poor job of describing the clouds, the dust, the chemistry and the biology of fields and farms and forests. They do not begin to describe the real world that we live in. The real world is muddy and messy and full of things that we do not yet understand. It is much easier for a scientist to sit in an air-conditioned building and run computer models, than to put on winter clothes and measure what is really happening outside in the swamps and the clouds. That is why the climate model experts end up believing their own models.

    (Expert Dyson sure has checked how bad the rather simple model of Hansen 1988 is doing)

    For a detailed exegesis of said essay cf. e.g. Michael Tobis' blog

    P.S.: Having read a bit more of the Dyson climate stuff I want to puke. I guess you physicists admire Freeman Dyson. But meanwhile he became a stupid old clown. A crass example of Dunning-Kruger effect multiplied by theoretical physicists' arrogance.

    Comment Source:Quoting Dyson on climate models? Uh, sorry, but ummm... you could as well quote Bjorn Lomborg or Michael Crichton. Here is the full quote from [Dyson's infamous edge.org essay](http://www.edge.org/documents/archive/edge219.html#dysonf) (emph. mine): >My first heresy says that all **the fuss about global warming is grossly exaggerated**. Here I am opposing the **holy brotherhood of climate model experts** and the crowd of **deluded citizens** who believe the numbers predicted by the computer models. Of course, they say, I have no degree in meteorology and I am therefore not qualified to speak. But I have studied the climate models and I know what they can do. The models solve the equations of fluid dynamics, and they do a very good job of describing the fluid motions of the atmosphere and the oceans. They do a very poor job of describing the clouds, the dust, the chemistry and the biology of fields and farms and forests. **They do not begin to describe the real world that we live in.** The real world is muddy and messy and full of things that we do not yet understand. It is much easier for a scientist to sit in an air-conditioned building and run computer models, than to put on winter clothes and measure what is really happening outside in the swamps and the clouds. That is why the climate model experts end up believing their own models. (Expert Dyson sure has checked how bad the rather simple model of Hansen 1988 is doing) For a detailed exegesis of said essay cf. e.g. [Michael Tobis' blog](http://initforthegold.blogspot.com/2007/08/dyson-exegesis.html) P.S.: Having read a bit more of the Dyson climate stuff I want to puke. I guess you physicists admire Freeman Dyson. But meanwhile he became a stupid old clown. A crass example of Dunning-Kruger effect multiplied by theoretical physicists' arrogance.
  • 5.
    edited June 2011

    Hi, Tim! Thanks for writing a blog entry! I'll put it on the blog whenever you want.

    I haven't actually read it yet. But I already have two questions:

    1. Is it okay if I read your blog entry and edit it as I do? I have various subroutines for making sentences easier to read, making it easier for the reader to tell what's going on, explaining jargon, and so on. It's almost painful for me to read something and not apply these subroutines, at least in my mind.

    2. Also, Tim: if you want climate scientists worldwide to get angry at you (and therefore, us), quoting Freeman Dyson in an approving way without mentioning any of the rebuttals is a good way to do it. Martin Gisser has illustrated what I mean. Is this really what you want? Maybe you could make your point in your own words, without introducing such a polarizing figure. Or maybe you could surround his quote with some language that explains clearly why you're quoting him. Or maybe you don't mind starting a fight? Or maybe you're trying to get the Koch Foundation to give us money, just like they did with the Berkeley Earth Surface Temperature group?

    Comment Source:Hi, Tim! Thanks for writing a blog entry! I'll put it on the blog whenever you want. I haven't actually read it yet. But I already have two questions: 1. Is it okay if I read your blog entry and edit it as I do? I have various subroutines for making sentences easier to read, making it easier for the reader to tell what's going on, explaining jargon, and so on. It's almost painful for me to read something and not apply these subroutines, at least in my mind. 2. Also, Tim: if you want climate scientists worldwide to get angry at you (and therefore, us), quoting Freeman Dyson in an approving way without mentioning any of the rebuttals is a good way to do it. Martin Gisser has illustrated what I mean. Is this really what you want? Maybe you could make your point in your own words, without introducing such a polarizing figure. Or maybe you could surround his quote with some language that explains clearly why you're quoting him. Or maybe you don't mind starting a fight? Or maybe you're trying to get the Koch Foundation to give us money, just like they did with the [Berkeley Earth Surface Temperature](http://articles.latimes.com/2011/apr/04/local/la-me-climate-berkeley-20110404) group? <img src = "http://math.ucr.edu/home/baez/emoticons/tongue2.gif" alt = ""/>
  • 6.
    edited June 2011

    Tim - I've looked at your blog entry. I think it's very well written! I would only be inclined to make a few small stylistic changes.

    I think you could reduce the number of people who hate you right away by not including the Dyson quote right at the start of the blog entry. Doing so — especially without explanation as you do now — is like waving a red flag in front of a bull.

    It's a bit like starting your blog entry with a quote by Sarah Palin, or for that matter Al Gore... even if it's an interesting quote, it conveys a political message, and that's all most people will see. So many people will stop paying attention to what you're actually saying.

    Comment Source:Tim - I've looked at your blog entry. I think it's very well written! I would only be inclined to make a few small stylistic changes. I think you could reduce the number of people who hate you right away by not including the Dyson quote right at the start of the blog entry. Doing so &mdash; especially without explanation as you do now &mdash; is like waving a red flag in front of a bull. It's a bit like starting your blog entry with a quote by Sarah Palin, or for that matter Al Gore... even if it's an interesting quote, it conveys a political message, and that's all most people will see. So many people will stop paying attention to what you're actually saying.
  • 7.
    edited June 2011

    Thanks for all the comments, I'll have to address them step by step :-)

    Staffan wrote:

    this is a good start but you should not expect that non-programmers knows about floating-point number and other terms. at least link to wikipedia entries. mabe all technobabble can be like that ?

    Actually I don't want all readers to check out what a floating point number is, the mathematicians should just be awed that there is more to programming a mathematical function :-) Maybe I should explain that in the post.

    also i think this is more traditional software process than climate model evaluation, so i'd add more on the latter :-)

    The title does not match the content, because I ended up writing about something else that I had intended, so I'll add a notification box to change the title.

    Comment Source:Thanks for all the comments, I'll have to address them step by step :-) Staffan wrote: <blockquote> <p> this is a good start but you should not expect that non-programmers knows about floating-point number and other terms. at least link to wikipedia entries. mabe all technobabble can be like that ? </p> </blockquote> Actually I don't want all readers to check out what a floating point number is, the mathematicians should just be awed that there is more to programming a mathematical function :-) Maybe I should explain <i>that</i> in the post. <blockquote> <p> also i think this is more traditional software process than climate model evaluation, so i'd add more on the latter :-) </p> </blockquote> The title does not match the content, because I ended up writing about something else that I had intended, so I'll add a notification box to change the title.
  • 8.

    David wrote:

    I think it's a good idea to have the first couple of paragraphs of something like this saying in simple declarative statements what you're going to do in this article (series). That helps the audience decide if they're interested in where you're going so they can use it to decide whether to read the whole thing.

    That's the hardest part, at least for me! For now, I think I'll stick to a short post that is only about "stakeholders" and "validation versus verification", so I tried to write some introductory sentences for that. I'll get to your draft later, when - and if - I write about the aspects I set out to write about :-)

    Martin wrote:

    Quoting Dyson on climate models? Uh, sorry, but ummm... you could as well quote Bjorn Lomborg or Michael Crichton.

    Dyson is a good example of a physicist getting semi-informed and into thinking he understands it all, I don't know what makes physicists most vulnerable to this kind of behaviour. Anyway, I would cite Crichton or Homer Simpson if the quote

    a) gets people's attention because they have heard of the speaker before,

    b) illustrates my main point,

    which the Dyson quote does. It is a good example of a criticism of climate models that makes implicit assumptions about what the model should do, which are not justified. Others use this kind of criticism to demand a level of certainty of any scientific statement that is impossible to fulfil, the "mearchants of doubts" strategy. I certainly won't discuss that, but an explanation of what climate model validation is and can do is certainly one necessary step against this strategy.

    Comment Source:David wrote: <blockquote> <p> I think it's a good idea to have the first couple of paragraphs of something like this saying in simple declarative statements what you're going to do in this article (series). That helps the audience decide if they're interested in where you're going so they can use it to decide whether to read the whole thing. </p> </blockquote> That's the hardest part, at least for me! For now, I think I'll stick to a short post that is only about "stakeholders" and "validation versus verification", so I tried to write some introductory sentences for that. I'll get to your draft later, when - and if - I write about the aspects I set out to write about :-) Martin wrote: <blockquote> <p> Quoting Dyson on climate models? Uh, sorry, but ummm... you could as well quote Bjorn Lomborg or Michael Crichton. </p> </blockquote> Dyson is a good example of a physicist getting semi-informed and into thinking he understands it all, I don't know what makes physicists most vulnerable to this kind of behaviour. Anyway, I would cite Crichton or Homer Simpson if the quote a) gets people's attention because they have heard of the speaker before, b) illustrates my main point, which the Dyson quote does. It is a good example of a criticism of climate models that makes implicit assumptions about what the model should do, which are not justified. Others use this kind of criticism to demand a level of certainty of any scientific statement that is impossible to fulfil, the "mearchants of doubts" strategy. I certainly won't discuss that, but an explanation of what climate model validation is and can do is certainly one necessary step against this strategy.
  • 9.
    edited June 2011

    Anyway, I would cite Crichton or Homer Simpson if the quote

    a) gets people's attention because they have heard of the speaker before,

    b) illustrates my main point,

    which the Dyson quote does.

    Unfortunately quoting anything without context makes people:

    1. think you agree with it,

    2. think you approve of the person you're quoting, including their activities unrelated to the quote.

    You may complain that this is unreasonable, but nonetheless, it's true. You may think that

    Dyson is a good example of a physicist getting semi-informed and into thinking he understands it all...

    and that sounds about right — but someone starting to read your article won't know you think that!

    Comment Source:> Anyway, I would cite Crichton or Homer Simpson if the quote > a) gets people's attention because they have heard of the speaker before, > b) illustrates my main point, > which the Dyson quote does. Unfortunately quoting anything without context makes people: 1. think you agree with it, 2. think you approve of the person you're quoting, including their activities unrelated to the quote. You may complain that this is unreasonable, but nonetheless, it's true. You may think that > Dyson is a good example of a physicist getting semi-informed and into thinking he understands it all... and that sounds about right &mdash; but someone starting to read your article won't know you think that!
  • 10.

    @John, question 1: Sure, go ahead, I'll look at the history of the page to see what you did and learn :-)

    Question 2: I have no qualms to anger everybody and get into a fight, if it is for the right reasons. Making others think I agree with Dyson certainly isn't one. I also would very much like to get funding from the Kochs, unless they have the right to get their money back if they don't like what I do with it.

    Anyway, I replaced the Dyson quote at the beginning with another reason why I started to write about "validation versus verification" which is probably less controverse, the blog post of Steve Easterbrook.

    I think you could reduce the number of people who hate you right away by not including the Dyson quote right at the start of the blog entry. Doing so — especially without explanation as you do now — is like waving a red flag in front of a bull. It's a bit like starting your blog entry with a quote by Sarah Palin, or for that matter Al Gore... even if it's an interesting quote, it conveys a political message, and that's all most people will see. So many people will stop paying attention to what you're actually saying.

    Geez!?

    The part of Dyson's statement that I did quote is a matter-of-fact criticism of climate models, it should be possible to discuss this calmly! BTW: I don't agree with the positive part either, I don't think that it is evident that climate models handle the fluid dynamics parts well. In fact, they don't, they don't reproduce tropical storm formation the way that climate scientists would like them to, for example. Maybe it is not possible to discuss this out in the open without getting instrumentalized by political interest groups, but then we should try to create some kinf of save haven where this is possible.

    Comment Source:@John, question 1: Sure, go ahead, I'll look at the history of the page to see what you did and learn :-) Question 2: I have no qualms to anger everybody and get into a fight, if it is for the right reasons. Making others think I agree with Dyson certainly isn't one. I also would very much like to get funding from the Kochs, unless they have the right to get their money back if they don't like what I do with it. Anyway, I replaced the Dyson quote at the beginning with another reason why I started to write about "validation versus verification" which is probably less controverse, the blog post of Steve Easterbrook. <blockquote> <p> I think you could reduce the number of people who hate you right away by not including the Dyson quote right at the start of the blog entry. Doing so — especially without explanation as you do now — is like waving a red flag in front of a bull. It's a bit like starting your blog entry with a quote by Sarah Palin, or for that matter Al Gore... even if it's an interesting quote, it conveys a political message, and that's all most people will see. So many people will stop paying attention to what you're actually saying. </p> </blockquote> Geez!? The part of Dyson's statement that I did quote is a matter-of-fact criticism of climate models, it should be possible to discuss this calmly! BTW: I don't agree with the positive part either, I don't think that it is evident that climate models handle the fluid dynamics parts well. In fact, they don't, they don't reproduce tropical storm formation the way that climate scientists would like them to, for example. Maybe it is not possible to discuss this out in the open without getting instrumentalized by political interest groups, but then we should try to create some kinf of save haven where this is possible.
  • 11.
    edited June 2011

    Hi again, Tim.

    I have no qualms to anger everybody and get into a fight, if it is for the right reasons. Making others think I agree with Dyson certainly isn't one.

    Good. I agree that annoying people is fine if it's done for good reasons. One wants to pick the right battles.

    So I guess you can ignore my last comment... except for the deep wisdom it contains, which is true quite generally.

    The part of Dyson's statement that I did quote is a matter-of-fact criticism of climate models, it should be possible to discuss this calmly!

    I agree — but when explaining things to an audience, it's not very relevant what they "should" be able to do. People can discuss this calmly if they know you and know you're a calm and reasonable person. But many people reading your first blog post won't know that.

    There is a war going on, a "climate war", and most people have already taken sides. If you're in a war, and you wave a flag around, it's no use saying "I was just waving this around as an example of a flag." Before you can say that, people will start shooting.

    [...] I don't think that it is evident that climate models handle the fluid dynamics parts well. In fact, they don't, they don't reproduce tropical storm formation the way that climate scientists would like them to, for example.

    I think it's possible to talk about these very important scientific issues in a calm way. I would never want you to avoid them! It just requires a little bit of care.

    Comment Source:Hi again, Tim. > I have no qualms to anger everybody and get into a fight, if it is for the right reasons. Making others think I agree with Dyson certainly isn't one. Good. I agree that annoying people is fine if it's done for good reasons. One wants to pick the right battles. So I guess you can ignore my last comment... except for the deep wisdom it contains, which is true quite generally. <img src = "http://math.ucr.edu/home/baez/emoticons/tongue2.gif" alt = ""/> > The part of Dyson's statement that I did quote is a matter-of-fact criticism of climate models, it should be possible to discuss this calmly! I agree &mdash; but when explaining things to an audience, it's not very relevant what they "should" be able to do. People can discuss this calmly if they know you and know you're a calm and reasonable person. But many people reading your first blog post won't know that. There is a war going on, a "climate war", and most people have already taken sides. If you're in a war, and you wave a flag around, it's no use saying "I was just waving this around as an example of a flag." Before you can say that, people will start shooting. > [...] I don't think that it is evident that climate models handle the fluid dynamics parts well. In fact, they don't, they don't reproduce tropical storm formation the way that climate scientists would like them to, for example. I think it's possible to talk about these very important scientific issues in a calm way. I would never want you to avoid them! It just requires a little bit of care.
  • 12.

    Just to illustrate what I mean, let's type "Dyson climate change" into the all-knowing Google and see what comes out on top. First Wikipedia, that's fine. Next, this blog:

    Freeman Dyson is on the cover of yesterday's New York Times Magazine. Inside a baseball writer (a very good baseball writer, but still) gives the man an opportunity to explain why he doesn't believe climate change is something to worry about.

    Others have lamented the attention devoted by the nation's leading newspaper to the thoughts of someone who has no expertise in the field. I share Chris Mooney's reservations about the writer's understanding of the way scientific skepticism is supposed to work. But Dyson cannot be easily dismissed if for no other reason than he has proven himself to be a brilliant mind, one that enjoys widespread respect in the community of science. Fortunately, I'm going to have the chance to see him talk about the subject this week.

    That's not so bad... Then the New York Times:

    Lately, however, since coming “out of the closet as far as global warming is concerned,” as Dyson sometimes puts it, there has been noise all around him. Chat rooms, Web threads, editors’ letter boxes and Dyson’s own e-mail queue resonate with a thermal current of invective in which Dyson has discovered himself variously described as “a pompous twit,” “a blowhard,” “a cesspool of misinformation,” “an old coot riding into the sunset” and, perhaps inevitably, “a mad scientist.”

    And then the New York Times Magazine:

    What is up with baseball and climate change denial?

    First we had George Will: Baseball lover, climate change denier.

    And now we’ve got a writer named Nicholas Dawidoff, whose Wiki bio suggests writing about baseball to be his chief area of expertise, but who has just waded into a minefield with a sympathetic profile of climate change skeptic Freeman Dyson in the New York Times Magazine. (Joe Romm detonates the ground beneath Dawidoff here.)

    And further down I'm sure you'll get a bunch of 'climate skeptics' who support Freeman Dyson and complain about the article I just quoted.

    My point is that this is not a good battlefield to enter unless there's a good reason.

    Comment Source:Just to illustrate what I mean, let's type "Dyson climate change" into the all-knowing Google and see what comes out on top. First Wikipedia, that's fine. Next, [this blog](http://scienceblogs.com/islandofdoubt/2009/03/freeman_dyson_climate_change_s.php): > Freeman Dyson is on the cover of yesterday's New York Times Magazine. Inside a baseball writer (a very good baseball writer, but still) gives the man an opportunity to explain why he doesn't believe climate change is something to worry about. > Others have lamented the attention devoted by the nation's leading newspaper to the thoughts of someone who has no expertise in the field. I share Chris Mooney's reservations about the writer's understanding of the way scientific skepticism is supposed to work. But Dyson cannot be easily dismissed if for no other reason than he has proven himself to be a brilliant mind, one that enjoys widespread respect in the community of science. Fortunately, I'm going to have the chance to see him talk about the subject this week. That's not so bad... Then the [New York Times](http://www.nytimes.com/2009/03/29/magazine/29Dyson-t.html): > Lately, however, since coming “out of the closet as far as global warming is concerned,” as Dyson sometimes puts it, there has been noise all around him. Chat rooms, Web threads, editors’ letter boxes and Dyson’s own e-mail queue resonate with a thermal current of invective in which Dyson has discovered himself variously described as “a pompous twit,” “a blowhard,” “a cesspool of misinformation,” “an old coot riding into the sunset” and, perhaps inevitably, “a mad scientist.” And then the [New York Times Magazine](http://blogs.discovermagazine.com/intersection/2009/03/26/new-york-times-magazine-on-freeman-dyson-climate-change-skeptic/): > What is up with baseball and climate change denial? > First we had George Will: Baseball lover, climate change denier. > And now we’ve got a writer named Nicholas Dawidoff, whose Wiki bio suggests writing about baseball to be his chief area of expertise, but who has just waded into a minefield with a sympathetic profile of climate change skeptic Freeman Dyson in the New York Times Magazine. (Joe Romm detonates the ground beneath Dawidoff [here](http://thinkprogress.org/romm/2009/03/25/203866/new-york-times-magazine-profile-global-warming-crackpot-freeman-dyson-slander-james-hansen/).) And further down I'm sure you'll get a bunch of 'climate skeptics' who support Freeman Dyson and complain about the article I just quoted. My point is that this is not a good battlefield to enter unless there's a good reason.
  • 13.

    Great. But is it okay then to place the quote in context at the end as it is now?

    Comment Source:Great. But is it okay then to place the quote in context at the end as it is now?
  • 14.
    edited June 2011

    Yes, that's fine.

    I've edited your post. It's a nice article!!! I've really been wanting to get get more blog articles from other people, and this is a great example of what I'd like: clear, tackling an important issue, fun to read, and so on.

    I added a few questions in boldface - things for you to fix, or decide not to fix. Please get rid of those questions in either case.

    I like to be quite systematic in using boldface for defined terms on the blog. So various terms are now in boldface.

    I made a bunch of small formatting changes that bring the article closer to the point where I can stick it on the blog and have it work. Wordpress doesn't use the same markup language as the wiki.

    By the way, your main grammar mistake is using commas to combine two sentences into one, where English demands semicolons. But in most such cases, I find it's more peppy to leave them as two separate shorter sentences.

    You may want to change the smiley to a different style from those available here:

    http://math.ucr.edu/home/baez/emoticons/

    or even add a new one to my collection!

    Also, please pick the title you want for the blog article. (Short is good.)

    Comment Source:Yes, that's fine. I've edited your post. It's a nice article!!! I've really been wanting to get get more blog articles from other people, and this is a great example of what I'd like: clear, tackling an important issue, fun to read, and so on. I added a few questions in boldface - things for you to fix, or decide not to fix. Please get rid of those questions in either case. I like to be quite systematic in using boldface for defined terms on the blog. So various terms are now in boldface. I made a bunch of small formatting changes that bring the article closer to the point where I can stick it on the blog and have it work. Wordpress doesn't use the same markup language as the wiki. By the way, your main grammar mistake is using commas to combine two sentences into one, where English demands semicolons. But in most such cases, I find it's more peppy to leave them as two separate shorter sentences. You may want to change the smiley to a different style from those available here: [http://math.ucr.edu/home/baez/emoticons/](http://math.ucr.edu/home/baez/emoticons/) or even add a new one to my collection! Also, please pick the title you want for the blog article. (Short is good.)
  • 15.

    Ok, I tried to address all your points, I also changed the title. I'd like to include the scary graphic from GCM, what do you think?

    Only to show that GCMs can be quite complicated :-)

    I still have to look up an example where climate model output lead to the discovery of a systematic error in some observational equipment, I think I read about this in "A Vast Machine". I'll look it up when I'm back home.

    Comment Source:Ok, I tried to address all your points, I also changed the title. I'd like to include the scary graphic from [[GCM]], what do you think? Only to show that GCMs can be quite complicated :-) I still have to look up an example where climate model output lead to the discovery of a systematic error in some observational equipment, I think I read about this in "A Vast Machine". I'll look it up when I'm back home.
  • 16.
    edited June 2011

    The only problem with that scary graphic is that images on the blog can be at most 450 pixels wide, so it will look like this:

    Of course I can do the "click to enlarge" thing I always do... it's up to you.

    When you say your article is ready, I'll put it on the blog. It would be nice to do that soon!

    Comment Source:The only problem with that scary graphic is that images on the blog can be at most 450 pixels wide, so it will look like this: <img width = "450" src = "http://www.azimuthproject.org/azimuth/files/bretherton.gif" alt = ""/> Of course I can do the "click to enlarge" thing I always do... it's up to you. When you say your article is ready, I'll put it on the blog. It would be nice to do that soon!
  • 17.

    Hi Tim,

    nice post. May I add a few little remarks?

    In between the text (Short review of...) there's still the comment:

    (A MILLION of WHAT?)

    And there are a few others still.

    Less noticeable, when you quote Dyson his quote is torn into two parts, one as a quote, another part as a normal paragraph, i.e.

    They do a very poor job of describing the clouds [...]

    Is Dyson's harsh critic:

    But there is no reason to believe the same fudge factors would give the right behaviour in a world with different chemistry, for example in a world with increased CO2.

    actually correct? Can't fudge factors change smoothly with smooth changes in the chemistry? Btw, is fudge factor the same as a calibrated parameter (I'll look it up myself)

    Comment Source:Hi Tim, nice post. May I add a few little remarks? In between the text (Short review of...) there's still the comment: > (A MILLION of WHAT?) And there are a few others still. Less noticeable, when you quote Dyson his quote is torn into two parts, one as a quote, another part as a normal paragraph, i.e. > They do a very poor job of describing the clouds [...] Is Dyson's harsh critic: > But there is no reason to believe the same fudge factors would give the right behaviour in a world with different chemistry, for example in a world with increased CO2. actually correct? Can't fudge factors change smoothly with smooth changes in the chemistry? Btw, is fudge factor the same as a calibrated parameter (I'll look it up myself)
  • 18.

    @Frederik: Yes, there was a version with John's remarks in it, and a missing quotation mark, which I added. It should be Ok now.

    Is Dyson's harsh critic...correct? Can't fudge factors change smoothly with smooth changes in the chemistry?

    The latter question should be addressed when we talk about ensemble runs, the ones which test how the model output varies with the different parameterizations. Acutally the "fudge factors" are both parameterizations and corrections of errors of numerical approximations, that I would also like to talk about.

    If a system has some preserved quantities, the discretized system will often violate this. So people add "fudge factors" to counteract these effects, so that the discretized system still respects mass and energy conservation, for example. This is a serious problem and a very interesting topic! But more of that later...

    Comment Source:@Frederik: Yes, there was a version with John's remarks in it, and a missing quotation mark, which I added. It should be Ok now. <blockquote> <p> Is Dyson's harsh critic...correct? Can't fudge factors change smoothly with smooth changes in the chemistry? </p> </blockquote> The latter question should be addressed when we talk about ensemble runs, the ones which test how the model output varies with the different parameterizations. Acutally the "fudge factors" are both parameterizations and corrections of errors of numerical approximations, that I would also like to talk about. If a system has some preserved quantities, the discretized system will often violate this. So people add "fudge factors" to counteract these effects, so that the discretized system still respects mass and energy conservation, for example. This is a serious problem and a very interesting topic! But more of that later...
  • 19.

    If a system has some preserved quantities, the discretized system will often violate this. So people add "fudge factors" to counteract these effects, so that the discretized system still respects mass and energy conservation, for example.

    I have a friend who works on heart models and once said that some older models enforce charge conservation by hand by setting the remainder to zero again after each loop...

    It would be very interesting indeed if you could point out similar constructions in climate models.

    Comment Source:> If a system has some preserved quantities, the discretized system will often violate this. So people add "fudge factors" to counteract these effects, so that the discretized system still respects mass and energy conservation, for example. I have a friend who works on heart models and once said that some older models enforce charge conservation by hand by setting the remainder to zero again after each loop... It would be very interesting indeed if you could point out similar constructions in climate models.
  • 20.

    Frederik wrote:

    It would be very interesting indeed if you could point out similar constructions in climate models.

    The buzzword is "flow corrections". I cannot point to a climate model code snippet, but I think I can and should explain what this is about, in principle. This is an interesting topic of its own, and also a topic of active research, both in climate models, and in numerical mathematics.

    When you say your article is ready, I'll put it on the blog. It would be nice to do that soon!

    Okay, shoot.

    If this does not turn out to become a traumatic experience, I may write another one.

    Comment Source:Frederik wrote: <blockquote> <p> It would be very interesting indeed if you could point out similar constructions in climate models. </p> </blockquote> The buzzword is "flow corrections". I cannot point to a climate model code snippet, but I think I can and should explain what this is about, in principle. This is an interesting topic of its own, and also a topic of active research, both in climate models, and in numerical mathematics. <blockquote> <p> When you say your article is ready, I'll put it on the blog. It would be nice to do that soon! </p> </blockquote> Okay, shoot. If this does not turn out to become a traumatic experience, I may write another one.
  • 21.

    Maybe you mean "flux corrections"? If so, as I understand them they aren't traditionally used to fix numerical non-conservation. They were more commonly applied, e.g., when you spin up (equilibrate) subcomponents of a model independently (e.g., atmosphere and ocean) and then couple them together for a transient forced simulation. The atmospheric and oceanic states so obtained aren't necessarily consistent with each other when the coupling is turned on, and so "flux corrections" at the boundary between components may be applied to reconcile them and maintain a climatologically plausible atmospheric-ocean state. In the last 10 years or so the coupled models have improved to the point that most of them no longer require such flux corrections. I'm not sure if this term applies to artificially adjusting quantities to satisfy conservation laws, or whether climate models ever do that (although I believe I've heard of such things).

    Comment Source:Maybe you mean "flux corrections"? If so, as I understand them they aren't traditionally used to fix numerical non-conservation. They were more commonly applied, e.g., when you spin up (equilibrate) subcomponents of a model independently (e.g., atmosphere and ocean) and then couple them together for a transient forced simulation. The atmospheric and oceanic states so obtained aren't necessarily consistent with each other when the coupling is turned on, and so "flux corrections" at the boundary between components may be applied to reconcile them and maintain a climatologically plausible atmospheric-ocean state. In the last 10 years or so the coupled models have improved to the point that most of them no longer require such flux corrections. I'm not sure if this term applies to artificially adjusting quantities to satisfy conservation laws, or whether climate models ever do that (although I believe I've heard of such things).
  • 22.
    edited June 2011

    Personal opinion time: I think it's dangerous to use phrases like "If you are interested in both climate science and computer science" because it invites people to think, say, "actually, I'm not interested in computer science, so I'll skip this article". In general, there's a problem writing about thing's you've learned: although learning tends to occur backwards ("I want to understand climate, so I discover I need to understand computer models of climate and their issues, so I discover I need to understand what makes a computer model good, which ...") the tendency is to do the write-up forwards. This is fine, except it helps to let your audience know where your eventual destination is (and if you're thinking of writing any blog articles of which this would be a prerequisite, I'd hint about those plans here).

    Again, personally I'd tempted to make the introduction more contextual and assertive, e.g.: "If you want to understand climate change in the near term (rather than watching as whatever happens happens), you need to understand computer models, And computer models are tricky things, where some specific concepts have been developed to help analyse and discuss them. Although it looks like innocuous English, the precise meanings assigned to terms make it important to be one the club who know the patter."

    As I read the current introduction, I (imagine if I didn't know) the broader significance of what you say you're going to talk about isn't obvious.

    Comment Source:Personal opinion time: I think it's dangerous to use phrases like "If you are interested in both climate science and computer science" because it invites people to think, say, "actually, I'm not interested in computer science, so I'll skip this article". In general, there's a problem writing about thing's you've learned: although learning tends to occur backwards ("I want to understand climate, so I discover I need to understand computer models of climate and their issues, so I discover I need to understand what makes a computer model good, which ...") the tendency is to do the write-up forwards. This is fine, except it helps to let your audience know where your eventual destination is (and if you're thinking of writing any blog articles of which this would be a prerequisite, I'd hint about those plans here). Again, personally I'd tempted to make the introduction more contextual and assertive, e.g.: "If you want to understand climate change in the near term (rather than watching as whatever happens happens), you need to understand computer models, And computer models are tricky things, where some specific concepts have been developed to help analyse and discuss them. Although it looks like innocuous English, the precise meanings assigned to terms make it important to be one the club who know the patter." As I read the current introduction, I (imagine if I didn't know) the broader significance of what you say you're going to talk about isn't obvious.
  • 23.

    Nathan wrote:

    Maybe you mean "flux corrections"?

    Yes.

    If so, as I understand them they aren't traditionally used to fix numerical non-conservation. They were more commonly applied, e.g., when you spin up (equilibrate) subcomponents of a model independently (e.g., atmosphere and ocean) and then couple them together for a transient forced simulation.

    I think I read about both aspects, so it would be a good idea to invent a name for both, like "numerical fudge factor" and "coupling fudge factor".

    (Okay okay okay, "fudge factor" will be used immediatly against climate models. How about "numerical balancer" and "coupling balancer".)

    David wrote:

    ...personally I'd tempted to make the introduction more contextual and assertive...

    Maybe I should do that, the problem is that I don't know my audience. I often get feedback like "why do you clog your text with banalities" and "thank god that finally someone explained this" at the same time, when I write a document for 5 people to read that I do know. Like, for example, a system specification :-)

    In this case I'd expect that most readers of John's blog are interested in climate science and have read about climate models. If that were not true, I think I should be allowed to add the Dyson quote at the beginning of the post without causing an uproar :-)

    Maybe John comes up with a better introduction.

    Comment Source:Nathan wrote: <blockquote> <p> Maybe you mean "flux corrections"? </p> </blockquote> Yes. <blockquote> <p> If so, as I understand them they aren't traditionally used to fix numerical non-conservation. They were more commonly applied, e.g., when you spin up (equilibrate) subcomponents of a model independently (e.g., atmosphere and ocean) and then couple them together for a transient forced simulation. </p> </blockquote> I think I read about both aspects, so it would be a good idea to invent a name for both, like "numerical fudge factor" and "coupling fudge factor". (Okay okay okay, "fudge factor" will be used immediatly against climate models. How about "numerical balancer" and "coupling balancer".) David wrote: <blockquote> <p> ...personally I'd tempted to make the introduction more contextual and assertive... </p> </blockquote> Maybe I should do that, the problem is that I don't know my audience. I often get feedback like "why do you clog your text with banalities" and "thank god that finally someone explained this" at the same time, when I write a document for 5 people to read that I do know. Like, for example, a system specification :-) In this case I'd expect that most readers of John's blog are interested in climate science and have read about climate models. If that were not true, I think I should be allowed to add the Dyson quote at the beginning of the post without causing an uproar :-) Maybe John comes up with a better introduction.
  • 24.

    @John: If you like to, post it :-)

    Comment Source:@John: If you like to, post it :-)
  • 25.
    edited June 2011

    Okay, Tim, I'll try to post your blog entry tomorrow morning (Singapore time). Unfortunately, when I click "edit" to see the source code at

    Blog - Climate Model Evaluation

    I am thrown back to the homepage of the wiki.

    Ah, I see the problem! Somehow you managed to change the page's title without creating a redirect. This should happen automatically when you click "Change page name" while editing the page. The redirect is necessary to keep people from getting confused like I just did.

    The new page name is

    Blog - Your model is verified, but it is not valid! Huh?

    (It seems that wiki pages ending in question marks are hard to link to... I had to use a trick.)

    Comment Source:Okay, Tim, I'll try to post your blog entry tomorrow morning (Singapore time). Unfortunately, when I click "edit" to see the source code at [[Blog - Climate Model Evaluation]] I am thrown back to the homepage of the wiki. Ah, I see the problem! Somehow you managed to change the page's title without creating a redirect. This should happen automatically when you click "Change page name" while editing the page. The redirect is necessary to keep people from getting confused like I just did. The new page name is [Blog - Your model is verified, but it is not valid! Huh?](http://www.azimuthproject.org/azimuth/show/Blog+-+Your+model+is+verified%2C+but+it+is+not+valid!+Huh%3F) (It seems that wiki pages ending in question marks are hard to link to... I had to use a trick.)
  • 26.

    Sorry, I'll be more careful in the future :-)

    Comment Source:Sorry, I'll be more careful in the future :-)
Sign In or Register to comment.