The future of Azimuth

John Baez · August 2015

I've been very quiet here since presenting our work at the Neural Information Processing Seminar in December 2014. One reason is that I've been rethinking what I want to do in the Azimuth Project. Another is that I wanted to see what would happen here without me pushing things in particular directions.

My original dream was that the Azimuth Project would take off and fly on its own: that people would use the wiki and forum as a meeting-point to pursue various projects even without me getting involved in all those projects. I'm disappointed at how little this has happened. It's certainly happened to some extent: for example, most recently, Paul Pukite has been using the forum here as a place to talk about his research. And of course when I push, people jump in and help out, accomplishing things I could never accomplish myself: we saw that during the initial setup of the wiki, and the busy phase of the El Niño Project, and various other times.

But I think I miscalculated some things, making mistakes that are hard to fix. One was publicizing the Azimuth Project before accumulating a team of people committed to working on it. This made the project look very "Baez-centric". A small but annoying example is the URL of the Azimuth blog, which contains my name. This was really just a mistake, due to how Wordpress asks you for your name when you start a blog. But it was a bad mistake.

People talk about founder's syndrome, and I'm talking about something like that:

the passion and charisma of the founder or founders, which was such an important reason for the successful establishment of the organization, becomes a limiting and destructive force, rather than the creative and productive one it was in the early stages...

However, I'm not sure the Azimuth Project was ever "successfully established" - certainly not compared to Microsoft or even much smaller ventures - including ones I've been involved in, like the $n$-Category Café and $n$Lab.

Another was failing, largely, to connect the work of Azimuth Project members to the work I'm doing with my academic colleagues. Part of the reason is that the project members tend to like programming, while my colleagues (and myself) tend to like proving theorems. These could be complementary and synergetic activities, but I haven't managed that. If I were a climate scientist, involved in climate models, it might have worked. I tried to become a climate scientist, but I'm quite sure by now that I'll never become a really good one; my heart lies elsewhere.

So those are some of the problems. I also have ideas for a new direction, but maybe I'll stop here and see if anyone has anything to say.

WebHubTel · August 2015

John,
Thanks for your effort.

In the background, I have been pasting links to the Azimuth Project in various climate and energy blogs. I usually preface it by saying that Azimuth is a place where collaborative work can take place. Yet I don't see many takers, which is kind of disappointing.

It also gets frustrating. For example, today, a fellow that runs a relatively popular consensus climate science blog said this:

" Blogs are broadly useless, IMO. What I think people should do is do their best to be as informed as they can be. Normally that would involve interacting with real experts. "

Lots of negativity out there to amateurs with no credentials :( As outsiders to the earth sciences, often times I think we do not own the keys to the castle, and that is a strike against us.

What other approaches are there? I too have noticed that forums that have a focus on solving well-defined puzzle problems are more successful. However, not too many thesis topics come wrapped in a bow-and-tie.

I have a feeling that open-ended problem solving for fun is not everyone's cup of tea.

As always, fishing for ideas.

Paul

nad · August 2015

John wrote:

but maybe I'll stop here and see if anyone has anything to say.

Well I already said something, but I am not sure whether you have noticed what I said here.

I do think that my case is not unique, that is I guess that some of the early Azimuthers left amongst others for the reason that at one point it wasn't really affordable for them anymore to be part of it. Like what happened e.g. to Tim van Beek or Frédéric de Roo or your students? I think I remember that Tim van Beek said something like that he has too much work at a car company, but I don't know about Frédéric and the others. In particular you need to have quite some time surplus to be able to really take part in the discussions here (and partially access to data resources, i.e. in some sense time previously invested).

In my experience people join groups, organisations, companies if with their engagement there comes also something "enriching" with it. And with enriching I don't mean necessarily monetary enrichment (which has a rather high ranking in american culture) but also other enrichments, like for example that you have less bad feelings about the environment, etc. That enrichment is usually balanced and averaged against what you have to invest, in terms of "unpleasant efforts" which are connected with the engagement.

In particular John I don't think, that your "Aura" was the main reason that most people wanted to join Azimuth. That is your contributions are usually interesting (at least for me), so you were certainly "enriching" for some, but I think for most here there were also other motivations playing a role.

People have very different backgrounds and "opinions" about what is "unpleasant" and what an enrichment. I wrote "opinions" and "unpleasant" in quotation marks, since there are of course people starving and it would sound cynical to say that those people have the "opinion" that it is "unpleasant" to join Azimuth, because you don't get something to eat here. I hope it is clear what I mean. That is in some cases one might easily guess what might have been getting too "unpleasant" in balance with a possible "enrichment" in some cases not.

Paul wrote:

Lots of negativity out there to amateurs with no credentials

So one guess is that I think quite some of the more advanced researchers here (like Nathan Urban?) might have felt like this turned too much into a teaching job. In some other cases on the other hand some amateurs work might be in competition with professional work. If competition is strong then people tend to prefer not to support the competitor.

Paul, you seemed for example to have perceived my comments about some of your math as unpleasant or "negativity", since you didn't reply to them or if then only incomplete, but you could have seen them also as an enrichment in the sense that you eventuelly learn something. Finally to a small amount even "negativity" might be enriching in that you learn what you definitely don't want/think is right etc.

I too have noticed that forums that have a focus on solving well-defined puzzle problems are more successful.

I also do think that for quite some people it is important to see immediately what some possible outcomes of their engagement will probably be. But more important I do think a lot of people would tend to observe, what the outcomes of their engagements are on a certain time scale. That is people will have varying demands on knowing what can be achieved with an engagement at Azimuth (at least partially that might also determine perceptions about what would be seen as an enrichment) and they will eventually check back with what had been achieved with their contributions and eventually balance that against how much sacrifices they had to bring in for that contributions.

I do think that there are some things which could eventually be improved by paying more attention not only on content but on the "frame", i.e. organisatorial set-up, improvements in infrastructure etc. but these seem rather costly.

In short: I think at the heart of the problem is that you need a rather big time surplus in order to engage in the Azimuth discussions and that it is currently quite unclear what might be achieved with that time investment on various time scales. Improving the technical "frame" might help to mitigate the time problem but it doesn't really fully adress the core problem.

By the way I think "time is money" holds currently and probably in general only true to a rather limited amount, if at all. On a first guess it seems to be more true if resources are everywhere scarce.

WebHubTel · August 2015

"Paul, you seemed for example to have perceived my comments about some of your math as unpleasant or "negativity", since you didn't reply to them or if then only incomplete, but you could have seen them also as an enrichment in the sense that you eventuelly learn something."

In general, a lack of response is something that goes with the territory. Other than that, without seeing the specific comments that you are referring to, I can't explain why I chose not to respond or provided an "incomplete" response.

I tend not to think that there is a time investment barrier to overcome. I see lots of people whiling their time away doing crossword puzzles or suduko. The equivalent for me is fiddling around with model parameters.

John Baez · August 2015

Nad wrote:

Like what happened e.g. to Tim van Beek or Frédéric de Roo or your students?

Tim switched from programming work to management, and became very busy learning to do well as a manager.

Frederik de Roo (you've made his name French!) has gotten a job working on Transport Processes in the Atmospheric Boundary-Layer at the Karlsruhe Institute for Technology, so he may count as a success story: someone whose interest in Azimuth served as a step toward a career in climate science! On the other hand, it's a bit sad that when he became more professional he no longer wanted to spend time on Azimuth. But that's not surprising, since we don't have many expert climate scientists involved here. So, he's a good example of the problems with the Azimuth Project.

Different students of mine have different stories, but pursuing an academic career makes it hard to spend time on other activities. My student most interested in environmental issues is Blake Pollard. This summer he's been doing great work on network theory and nonequilibrium thermodynamics, and I want to talk more about that on the blog. But precisely because he's busy working on this stuff, he hasn't been hanging out on Azimuth. And that's another example of the problems with the Azimuth Project: I'm an expert on certain things, but those things are mostly not being done "in the Azimuth Project" - though I do blog about them on the Azimuth Blog, and sometimes recruit other people to blog about them.

These are the kind of problems I want to deal with now. I don't want to stop anyone from doing anything that they're doing, but I want to focus my energy on the things I'm really good at. This means I'll do a lot less of certain things on the Azimuth Project, and maybe more of others.

John Baez · August 2015

Nad wrote:

Well I already said something, but I am not sure whether you have noticed what I said here.

Yes, I read that. I can understand you feeling short of time for work on Azimuth. Different people in different situations will have different amounts of time and energy for projects like this, and people will come and go. So, the best we can do with Azimuth is make it a place where people feel able to contribute whatever amount of time they have (a lot, or a little) and accomplish something worthwhile.

Personally I feel that pushing in the direction of El Niño prediction is no longer the best way to make Azimuth into this kind of place. Paul Pukite is reliably interested in this, but it's hard for me to stay excited about it, because this is not something I'm very good at, compared to other things I can do.

I also didn't get a positive response when I hinted to Paul that the Azimuth Project jointly publish a paper on this topic, based on his ideas. Perhaps I was hinting too subtly. Or perhaps he's less interested in publication than I am.

John Baez · August 2015

Once upon a time David Tanzer wrote:

At another point I remember you saying that you felt that the foundations of "green mathematics" (network category theory) were also of deep interest and beauty as mathematics, and had the advantage of significance on a more present timescale than e.g. quantum gravity. Has your perspective on this changed?

I am now very excited about this, because I feel I'm getting to the point of understanding how nonequilibrium thermodynamics connects to biochemical networks! The key new development is my work with Brendan Fong and Blake Pollard on "open detailed balance Markov processes", together with some conversations I had with Matteo Polletini. But this development ties together all the previous stuff I've done on electrical circuits and chemical reaction networks. So, this is what I want to focus on, maybe for the next year.

It took quite a while to reach this point, and I wasn't sure I would!

WebHubTel · August 2015

John, There is a mis-titled climate science blog out there called "And Then There's Physics". When I mentioned the Azimuth Project, someone responded yesterday with:

" (I’ve talked climate in person with John Baez. I like the guy, but he doesn’t really understand climate. Physicists tend to expect more simplicity than the world affords.)" from andthentheresphysics.wordpress.com

That is completely untrue, IMO . You are an expert at all sorts of topics. If you do find some simplifying model when applying network theory to biochemical processes, you will prove this guy wrong. Biochemical dwarfs the complexity of climate! This is really a rallying cry and motivation to work on these topics. The business-as-usual crowd believes that these huge General Circulation Models are the only path forward, which may in fact not be true.

For example, a recent paper [1] demonstrated by using Takens embedding theorem, that the patterns in ENSO may repeat at far shorter periods than previously thought

[1] H. Astudillo, R. Abarca-del-Rio, and F. Borotto, “Long-term non-linear predictability of ENSO events over the 20th century,” arXiv preprint arXiv:1506.04066, 2015.

This is the place for discussing these ideas, as I do not think that the mainstream climate science crowd is equipped to handle this flavor of mathematical physics.

---

As far as taking hints, I admit that I was defensive when you suggested that more thorough information criteria were needed in the ENSO model. I am still struggling with how to do this best. I did a trial-run with writing a white-paper but that did not go over too well.

nad · August 2015

Tim switched from programming work to management, and became very busy learning to do well as a manager.

Germany's economy is quite dependend on selling cars. But for Azimuth his leaving is a pity. I liked his fluid posts. Unfortunately he didn't get to speak about geodesics and the KdV equations, although he intended to do so.

I also didn't get a positive response when I hinted to Paul that the Azimuth Project jointly publish a paper on this topic, based on his ideas. Perhaps I was hinting too subtly. Or perhaps he's less interested in publication than I am.

As I understood Paul did an ENSO-publication, where he published his ENSO model. Did you see that? I have problems with his publication, but I told him that already a couple of times.

I think it is important not to look only at correlations and data fitting but on major mechanism, which lead to ENSO, especially at the QBO and that what I called "chimney effect." Concerning the sofar made Azimuth discussions I think it became clear that still quite some work would need to be done for getting better predictions.

Personally I feel that pushing in the direction of El Niño prediction is no longer the best way to make Azimuth into this kind of place. Paul Pukite is reliably interested in this, but it's hard for me to stay excited about it, because this is not something I'm very good at, compared to other things I can do.

It seems there are rather big commercial interests in better ENSO predictions.

For me discussing the role of methane would be more important in particular I think it is a prerequisite to fully understand the ENSO dynamics.

will have different amounts of time and energy for projects like this

This is a bit off-topic, but above in a playful manner I advertedly left the term "energy" out. That is one crucial observation of quantum mechanics is that higher energy states are sort of inversely proportionate to lifetime. I always wondered how this translates to macroscopic states, in particular in the context that energy is connected with mass and light velocity this may sort of translates into the concept that mass may eventually be seen as some kind of "into space invested/inserted clogged time(s)" sort of in analogy to the fact that the execution time of processes (i.e. "life time") with high job priorities (i.e.high "energy/mass") is smaller with respect to the cpu time than those with lower job priority. If you know about some readable literature which discusses this then I would be interested to hear about.

WebHubTel · August 2015

"I think it is important not to look only at correlations and data fitting but on major mechanism, which lead to ENSO, especially at the QBO and that what I called "chimney effect."

I don't know more what I can do. My ENSO model is as follows: (1) input a waveform as close to QBO as possible as a forcing function to the differential sloshing wave equation, (2) numerically solve the equation w/forcing and adjustments to the initial conditions.

That's it basically.

The QBO data only goes back to 1953 but I extrapolate backward from the data and can accurately model the ENSO data back to 1880. This works because the quasi-periodic nature of QBO is not really that quasi. It is actually very close to a fixed 2.333 year period, with only some jitter around this value. So it extrapolates backward very well.

I am not sure why no one is jumping up and down and getting excited about this model.

"Concerning the sofar made Azimuth discussions I think it became clear that still quite some work would need to be done for getting better predictions."

I can likely make good predictions tomorrow, but that is not what science is about. Science is about characterizing the data that we already have.

The ingredients to the model were in place before the Azimuth prediction project started. What I have been documenting is essentially an on-line engineering notebook on ENSO analysis, both here and on my regular blog http://ContextEarth.com

If you read through it beginning to end, you will see something peculiar in the evolution of my entries, but not totally unexpected.

I started out by assuming the model had to be something more complicated than the obvious solution. After all, people had been working on the ENSO prediction problem for ages and no one had seemed to make any progress, right? Initially, I figured that the answer had to contain significant complexity, otherwise someone else would have come up with a solution. So I spent lots of spare time devising characterization strategies that could manage the complexity.

At first I tried out various automated optimization and fitting approaches, but then after a while, the modeled behavior became simpler and the amount of code and parameters involved began to decrease. That's the nature of first-order physics analysis, as the principal factors start to take hold.

Eventually I removed the automation and worked with the model manually. The more I worked it, the more concise and elegant the model became.

Right now it is in excellent shape, contained in about 20 lines of Mathematica code.

Over at the AnThenTheresPhysics blog, a former GCM coder yesterday decided to deride the model. He said:

"We could profitably do something like what WHT has pointlessly done with ENSO, because unlike his, our models have actual physics in them and actual predictive success."

Does everyone appreciate how frustrating this whole business has become? It has nothing to do with the difficulty in science, but in the difficulty of working in a collaborative environment where the "complementary" and "encouraging" aspects of collaboration are missing.

What I fear most is if Per Strandberg, who is a vocal AGW denier, gets his own model out there before we do. I can't quite tell how close Per is to the basic sloshing model, but suffice to say, it would be very embarrassing if that side makes a breakthrough before the sane among us do.

"For me discussing the role of methane would be more important in particular I think it is a prerequisite to fully understand the ENSO dynamics."

I do not know what to say about that. I will be complementary and encourage you to look into this and see what kinds of correlations that you can find.

Paul

John Baez · August 2015

Nad wrote:

As I understood Paul did an ENSO-publication, where he published his ENSO model. Did you see that? I have problems with his publication, but I told him that already a couple of times.

Sorry, I didn't see it. Where did he publish it?

(I'm probably being old-fashioned, but when I speak of "publishing" a scientific paper I usually mean publication in a respected peer-reviewed journal, and I hope Paul succeeds in publishing his work this way, because anything less will not be convincing to a certain large crowd of people, and I think it's a good exercise to prove that one can convince referees at such a journal that ones work is okay.)

Paul wrote:

As far as taking hints, I admit that I was defensive when you suggested that more thorough information criteria were needed in the ENSO model. I am still struggling with how to do this best. I did a trial-run with writing a white-paper but that did not go over too well.

Is this "white paper" the same thing as what Nad called your "publication"?

I hope you understand: I was hoping I could help you publish your paper in a journal, basically by making it "bullet-proof". This involves looking at the results even more critically than the referees will, coming up with all the objections we can, and answering those objections.

nad · August 2015

in the Context of what i Worte abverlangt:does anybody know about Experiments Measuring the Maß (im sorry im typing on this Thinge with German autocorrection, don't know Howard to Switch that Off) of an Atom beforeandafterexiting?

WebHubTel · August 2015

John, I submitted to Physical Review Letters but it got rejected without any further review. I asked why it didn't warrant further consideration and the editor coughed up the usual baloney.

Nad, looks like you are having lots of problems. :) Don't worry, be happy :)

WebHubTel · August 2015

The information criteria issue is one of having a model of a certain structure and set of parameters and then trying to establish that the agreement of the model to the data is not simply fortuitous. In a past discussion, John correctly pointed out that I was not doing much with the statistical criteria.

There are a couple of general ways of dealing with this situation. One can estimate the probability that a given model can be constructed from a random draw of structures and parameters, or one can use an information criteria such as Akaike and compare the results of that against other models that differ in scope. The latter comparison assumes that one has an alternate model to compare against, which is problematic for me.

A cyclic time series model is tricky because it often doesn't take many sinusoidal components to reproduce an erratic waveform. That is the big hangup for most with experience in fitting.

I posted an entry at Context/Earth yesterday which is on the path to addressing some of these issues for ENSO modeling. What I think is really needed is for others to objectively evaluate this unbelievably simple model in the context of all the other madness going on in the climate science world. I think I know why this model has escaped the attention of everyone that has been involved in the research, but that is not the stuff of journal articles. That is really what these forums really excel at.

WebHubTel · August 2015

One other group is covering similar territory that I am, a group of geophysicists down in Chile:

H. Astudillo, R. Abarca-del-Rio, and F. Borotto, “Long-term non-linear predictability of ENSO events over the 20th century,” arXiv preprint arXiv:1506.04066, 2015.

I sent an email to Dr. Abarca-del-Rio, who I have communicated with before on angular momentum matters to see what his opinion is. He responded with lots of enthusiasm:

Rodrigo Abarca del Rio

hI

Thanks for your interest !! Yes, it's under review. I know it is a breakthrough, but you know how the system is ... Keep waiting.

regards

rodrigo

Their paper is concerning the application of Takens Embedding Theorem to determine the underlying patterns or attractors in the ENSO system, with possibilities outside that realm

http://www.azimuthproject.org/azimuth/show/Attractor%20reconstruction

JimStuttard · August 2015

Thanks for the azimuth wiki page on Takens embedding theorem. I hadn't heard of it until you linked the Astudillo paper a few months ago and tried to get to grips with the wikipedia entry at the time. Maybe this time....

WebHubTel · August 2015

Jim, What I would like to see is the analogous situation of Takens embedding theorem when one is just considering a linear 2nd order differential equation. What is important is not only the natural (i.e. eigenvalue) response, but the forcing conditions that also occur.

So if one assumes that a single sinuosoidal frequency forcing function is driving a 2nd order differential equation, then the result should be a linear combination of the forcing frequency and the natural response frequency. It seems that you want to make sure that the forcing frequency does not get misidentified as an additional eigenvalue response of the system.

But that is really a detail in piecing together what you can find after applying the Takens algorithm. It will work on both the natural response and forcing combination, as long as the forcing is stationary. And that is where the disturbance at 1981 is problematic. Is that a change in forcing, or is it a change in the parameters of the system?

John Baez · August 2015

Paul Pukite wrote:

John, I submitted to Physical Review Letters but it got rejected without any further review. I asked why it didn't warrant further consideration and the editor coughed up the usual baloney.

I believe I could help you get a paper published. I've published about 90 papers, so I've learned some tricks of the trade, and of course I have an academic affiliation, which might help too - though being in a math department might not help so much for climate science, and I expect it to be a tough challenge.

My "price" would be becoming a coauthor, and more importantly, having quite a bit of say over what the paper is like. You may not want to pay this price, and of course I'll understand if you don't.

I think it's important to take the simplest model that does a pretty good fit, and offer extensive evidence that the fit is too good to be due to chance. This means keeping the number of adjustable parameters to a bare minimum, not putting in all the bells and whistles, and doing extensive statistical tests, I hope with help from Steve Wenner and Jan Galkowski. Also, I first heard you were correlating climate to the Chandler wobble, my initial reaction was to dismiss it as absurd. So, I'd want to do whatever it takes to prevent that "instant dismissal" reaction. This will require a bunch of work, and also require not saying everything you believe.

The more I worked it, the more concise and elegant the model became.

Right now it is in excellent shape, contained in about 20 lines of Mathematica code.

That's very good. Can you tell me in words roughly what these 20 lines do, or point me to something you've written that explains this latest version?

John Baez · August 2015

Paul quoted some guy as saying:

I’ve talked climate in person with John Baez. I like the guy, but he doesn’t really understand climate. Physicists tend to expect more simplicity than the world affords.

The second sentence is true. But I've never claimed to understand climate physics. I haven't put enough work into it to understand it well.

As for "expecting more simplicity than the world affords", that could be true in general for physicists, but it's not true of me. I've never claimed that climate physics was simple - nor that it was complicated; I don't know enough to have a strong opinion, though of course I know people use some very complex models.

Perhaps this guy saw me playing around with some simple models and concluded I thought those models were all one needed to think about. In fact, I was doing it because I need to understand simple things before I can move on to think about more complicated things.

DavidTanzer · August 2015

Hi Guys, this is a thread that is worthy of consideration. In the last few months I've been remote, because life pulled me in some other directions, but I'd like to return to Azimuth, albeit in some modified ways. This is a good time for us to reflect on what has worked well and what has not, and to think of productive ways forward. Will post more later. Best Regards!

WebHubTel · August 2015

John said:

"I think it's important to take the simplest model that does a pretty good fit, and offer extensive evidence that the fit is too good to be due to chance. This means keeping the number of adjustable parameters to a bare minimum, not putting in all the bells and whistles, and doing extensive statistical tests"

John, I am working with the bare minimal model now. All I incorporate is a QBO forcing that matches the available data from 1953 to current. Then I apply it to a 2nd-order differential equation modeled as a characteristic resonant frequency $\omega_0$.

$ f''(t) + \omega_0^2 f(t) = Forcing(t) = qbo(t) $

I adjust the initial conditions and then slide the numerically computed Mathematica output to align with the data.

This is the empirical fit to the QBO, in a Fourier series so I can extrapolate backwards to 1880. QBO

This is where I fit using the first 400 months of ENSO SOI data as a training interval, and validate to the last 800 months (100 years total) first

Fit using the second 400 months of data, and validate forward and backward by 400 months second

Fit using the last 400 months of data, and validate to the first 800 months third

In each case, the overall correlation coefficient is above 0.7, which is as much as can be expected given the CC of the QBO.

BTW, there is a climate shift after 1980 which is well known and that will require another 400 month interval.

"Can you tell me in words roughly what these 20 lines do, or point me to something you've written that explains this latest version?"

It is actually now about half that, if the fitting to the QBO is removed. And the part related to just the DiffEq integration computation is just a couple of lines of Mathematica code. The rest is for handling the data and graphing.

The fit to the QBO can be made as accurate as you want, but I stopped when it got to the correlation coefficent as shown. Then the final result will have a correlation coefficient about that value as well if everything works out, which it appears to have.

The geophysics story is very simple to explain. If I want to go this route I can shorten my current paper from its current 3 pages to perhaps a page and a half.

The statistical test proposed is to compare against an alternate model and then see which one wins with an information criteria such as Akaike. I only have a few adjustable parameters (2 initial conditions, a characteristic frequency, and a time shift from QBO to ENSO), so that this will probably beat any other model proposed. It certainly will beat a GCM, which contains hundreds or thousands of parameters :)

EDITED: eqn notation altered to make sense with the next post

WebHubTel · August 2015

I really should point out an obvious transformation.

$ f''(t) + \omega_0^2 f(t) = qbo(t) $

If we take the Fourier transform of both sides, then:

$ (\omega^2 + \omega_0^2) F(\omega) = QBO(\omega) $

So that the power spectra should contain many of the same components, but scaled at values proportional to the frequency squared. That is why the SOI, i.e. f(t), contains stronger long time period components; while the QBO shows the higher frequency, i.e. 2.33 year period, rather strongly.

DavidTanzer · August 2015

John wrote:

Another was failing, largely, to connect the work of Azimuth Project members to the work I'm doing with my academic colleagues. Part of the reason is that the project members tend to like programming, while my colleagues (and myself) tend to like proving theorems. These could be complementary and synergetic activities, but I haven't managed that.

I see your point, but I wouldn't give up on this aim, even though it poses a real social challenge. The potential rewards from this synergism could be great. So can we keep the problem on the table.

A successful, and well-targeted, blogging effort could be one of the keys to bridging the gap. I started in this direction, but lost focus along the way -- sorry -- so now it's time to refocus. More on this later.

DavidTanzer · August 2015

I think that we would benefit by working to better engage people in discussions about our ideas and projects. I imagine, for example, that a newcomer to the Azimuth forum could be at a loss to see where to even begin to make a contribution.

DavidTanzer · August 2015

On the topic of the foundations of green mathematics, John wrote:

I am now very excited about this [the foundations of green mathematics], because I feel I'm getting to the point of understanding how nonequilibrium thermodynamics connects to biochemical networks!

Great! I'm glad that you have this inspiration, and would like to hear more about it. Suggestion, can you start a thread, with some notes and offhand comments? Consider the challenge of talking to someone, such as myself, who knows very little about thermodynamics. If we can bridge this discourse gap, then we will have widened the horizons of the Azimuth project.

WebHubTel · August 2015

It always boils down to doing something really cool. If we can gain credibility with uncovering some ingenious idea that can go viral, then we can bootstrap the process and attract enough curious and smart people to sustain the project's momentum. IMO, of course.

If it is this nonequilibrium thermo finding that turns out to be a breakthrough, that would be great.

DavidTanzer · August 2015

My vision for the (loose) organization of Azimuth is that it is one broad project, which we've already described, involving broad and general aims for green science, etc., and, within this, we have some finite number of concrete projects to which we dedicate time and commitment. So when people ask, what's going on at Azimuth, we will have specific activities, consonant with our overall aspirations, which we can point to, and invite other people to participate in.

So Azimuth would be a launching ground for projects that are consistent with our goals. Anyone is free to start a project.

But let's face it, projects don't take off on themselves, they need "leadership" -- even if not formalized. Those with the most initiative, experience, and knowledge will naturally take the lead. If they hope to enlist the efforts of others, who don't have the same levels of knowledge, experience or commitment, then a communications bridge will have to be built. This is not easy! But I don't see how it can be avoided.

DavidTanzer · August 2015

Here are some of the projects, whether express or implied, that I perceive at Azimuth:

Green mathematics (John)
El Nino prediction (not active)
ENSO modeling (Paul,...)
Interactive model development (Azimuth Code project), not active
Phylogenetic research (Graham). (I think this is a great subject, and would like to think the Azimuth umbrella includes such topics. It's evolution, after all, which is a cornerstone of our understanding of life!)
Math/science outreach (hibernating) -- blogging for green science, math and programming (Dave)
"Azimuth Code project." (not active) This too thin to qualify as a concrete project. But it could be a launching pad for projects such as interactive model developent, etc.
Machine learning for environmental applications (Dara, ...) , not active

If I've missed anything, or anybody, please chime in, or correct!

Nad, I see that you have made some in-depth contributions to the discussion of the climate data. Is there an implied "project" or goal that you would consider yourself working towards within Azimuth?

Jim, how about you, vis-a-vis the Azimuth code project?

DavidTanzer · August 2015

There may be something to be learned by looking at how projects are organized in software development teams. I will describe the team organization at the company I work for. Consider this only as a point of reference -- I am not proposing that we literally work along these lines.

At any time there are a definite number of well-defined, active projects. There is an overall team of developers, who are each assigned to a project for at least the duration of one "sprint," but can move between projects in the broader coarse of time. Each project has a business analyst (BA), and a tech. lead developer. The BA performs the outreach function of translating between the fuzzy language of user needs and system requirements, and together the BA and tech lead work to write specific "stories" for the general programmers, which describe manageable, empirically testable units of work for a programmer. These language transformations take real work -- it's a job that someone is hired to do -- but it results in a productive collaboration between people with very different kinds of skills and levels of commitment.

Now I will return from this intentional digression back to the topic of projects at Azimuth...

DavidTanzer · August 2015

Before going on, let me emphasize that I am not proposing to institute any procedural mechanisms or hierarchies at Azimuth. Just let's see what we can learn from any partial analogies with the above-mentioned structure, which does produce results.

Only this: anyone can start and declare a project at Azimuth, provided that it is consonant with the overall goals of Azimuth. Here, I think we should take as broad a view of the Azimuth project as is feasible, so as be inclusive of as much math/science/education activity as possible.

If one is happy running their own "one-person" project, that's fine and good. But if one wants to include others, then the outreach issues will arise, and that is where organizational thinking and methods can become useful.

We can also list all of our active projects (which have people and a least a modicum of activity), along with their current participants, on a prominent place on the Azimuth website.

WebHubTel · August 2015

"El Nino prediction (not active) ENSO modeling (Paul,...)"

Umm, these two are the same thing as far as I am concerned. To do prediction, one has to characterize the data that is available. So that backcasting (reverse prediction) is an important part of the modeling process. This is where the machine learning comes in as well.

BTW, most of my forum posts on ENSO are in the Azimuth Code Project category, if people are not aware of that.

This brings up something that is probably too late to do anything about, but I will say it anyways.

If you want the opinion of someone who worked several years on ontologies and the semantic web, the organization of this place leaves a lot to be desired. I run an earth sciences semantic web server, and the organization is not based on these strict categories. In the semantic web universe, what happens is that these projects would be tagged with multiple labels so that a specific project (like what I am working on) could have an ENSO tag, a machine learning tag, a code tag, etc. Then the projects could bleed across categories and a search mechanism would be used to provide a custom view.

My own semantic web server is all open-sourced and I have one version running on Dara's compute server if you want to take a look: http://earthcontext.lossofgenerality.com/

The one I keep up-to-date is here: http://entroplet.com, and I have one running on an Azimuth cloud instance http://23.23.137.157

I anticipate placing the ENSO modeling apps in this environment. I think this is the way of the future, but it takes some discipline and dedication to go this route.

If anyone is interested, I can add you to a GitHub account and you can see the guts of the environment. It's pretty wild stuff.

DavidTanzer · August 2015

Umm, these two are the same thing as far as I am concerned.

Hi Paul, sure.

I was trying to take a rough stab at what might be different between the effort led by John which led to the paper (and which no longer appears to be an active effort), and the ongoing work that you have been posting. I have been observing from afar, and it did seem to me that you were proposing theoretical explanations (sloshing, etc.) which made me think that it was a broader "project" than the machine learning efforts that led to the paper.

But I will stop trying to interpret your work, since I don't have the background needed to follow it in details, and let you describe it in your own terms. As an exercise in communication, can you summarize your quest, and your approach, without using much technical language?

WebHubTel · August 2015

Glad to David.

The equatorial Pacific has a behavior that matches the behavior of a standing-wave dipole -- IOW when the atmospheric pressure increases in one region (say Tahiti) it decreases in another (say Darwin), and vice versa. This is a oscillating (i.e. sloshing) behavior in the underlying thermocline, with both spatial and temporal features. The temperature change is manifested as movement of the thermocline to and away from the surface.

If we can isolate the temporal from the spatial, then a typical solution to a standing-wave dipole is found in the wave equation:

$ f''(t) + \omega_0^2 f(t) = Forcing(t) = qbo(t) $

The unknowns are the natural resonant frequency, $\omega_0$ and the forcing function, Forcing(t). A forcing function is needed to get the standing wave in motion, because it can't be created spontaneously and without an energy source to drive it.

The rather obvious forcing function to apply is that driven by the Quasi-Biennial Oscillation (QBO) in the upper atmosphere winds. This generates a periodic change in the earth's angular momentum, with some frequency and amplitude jitter observed between cycles.

We can use an empirical model fit of the QBO (a primary period of 2.33 years plus Fourier series components to model the jitter) as the forcing in the above equation, and solve the equation using a tool such as Mathematica.

Further, we can do a cross-validation exercise and fit the QBO-forced solution (adjusting initial conditions, scaling, $\omega_0$) in one time interval , and check how effectively it matches the ENSO SOI data across other intervals:

fit

That seems to work well. So the basic physics proposed is explained by a QBO forcing of the wave equation.

The gist is that we know that sloshing must occur in the ocean, just like we know that the moon's orbit has to create tides. The question has always been whether the sloshing was deterministic enough to be modeled and thus capable of being predicted in advance.

You won't find this derivation and calculation anywhere in the research literature, as far as I can tell. Stuff way more complicated than this, yes you will find, plenty of it :)

John Baez · August 2015

Paul quoted some guy as saying:

I’ve talked climate in person with John Baez. I like the guy, but he doesn’t really understand climate. Physicists tend to expect more simplicity than the world affords.

The second sentence is true. I've never claimed to understand climate physics. I haven't put enough work into it to understand it well.

As for "expecting more simplicity than the world affords", that might be true for some physicists, but not me. I've never claimed that climate physics was simple - nor even that it was complicated; I don't know enough to have a strong opinion, though of course I know people use some very complex models.

Perhaps this guy saw me playing around with some simple models and concluded I thought those models were all one needed to think about. In fact, I was doing it because I need to understand simple things before I can move on to think about more complicated things.

John Baez · August 2015

Paul wrote:

All I incorporate is a QBO forcing that matches the available data from 1953 to current. Then I apply it to a 2nd-order differential equation modeled as a characteristic resonant frequency $\omega_0$.

$ f''(t) + \omega_0^2 f(t) = Forcing(t) = qbo(t) $

Your model looks pretty simple, but I don't understand it yet. For starters, what's $f$? You didn't say.

EDIT: Oh, in a later comment you say $f(t)$ is your model's prediction of the Southern Oscillation Index, or SOI. Good.

I think some of the model's parameters are hidden in the method you use to retrodict the QBO. This may become a contentious issue, so it's important to deal with it well... whatever that means.

You haven't said whether you're interested in writing a joint paper or not. What do you think?

(I've found it pays, in collaborations, to be really clear on who is going to be an author. It avoids people getting wrong ideas and, later, bruised feelings.

I'm not pushing for a commitment: it's perfectly fine if you say "I don't want you to be a coauthor until you actually do something." But how much work I put into this project will depend on whether I'm a coauthor who gets some say in the details of this paper, or whether I'm just talking to you about your paper here on the Forum. If I'm a coauthor, I can promise to do my damnedest to actually get something published.)

John Baez · August 2015

David wrote, about my work on thermodynamics of networked systems:

Suggestion, can you start a thread, with some notes and offhand comments? Consider the challenge of talking to someone, such as myself, who knows very little about thermodynamics. If we can bridge this discourse gap, then we will have widened the horizons of the Azimuth project.

Sure! Glad to see you back.

John Baez · August 2015

David wrote:

I think that we would benefit by working to better engage people in discussions about our ideas and projects. I imagine, for example, that a newcomer to the Azimuth forum could be at a loss to see where to even begin to make a contribution.

At present that's true, unless they want to help Paul Pukite or Nad - because I've been holed up working on papers with grad students, not talking much here. But at other times, like when the Azimuth Wiki was in its first growth spurt (mainly before you showed up), or when we were hard at work on the El Niño project, it was fairly easy for people to join in, and they did join in.

WebHubTel · August 2015

John, If I have you as a co-author, I will have to add a bunch of other people out of fairness. I can send you this list in private so you can get a feel of who I have been collaborating with.

WebHubTel · August 2015

I have been thinking about what David said about the nature of the original project and my spin on the ENSO model. The original project was about finding or evaluating linked climate indices and making sense out of those.

But consider that QBO is actually just another climate index, and completely spatially separated from the ENSO index, yet it has a significant linkage to ENSO. Thus it falls under that category of linked climate indices. This is actually good because it fits the spirit of what we were trying to find in the first place -- some sort of linkages that would eventually allow us to make better predictions!

There could be a common mechanism which drives ENSO and QBO in tandem. The ENSO only appears erratic because of the huge inertial resonant lag of the ocean's volume, while the QBO is much more periodic as it responds to the mechanism with very little inertial lag. So if we find the basis for QBO and how it jitters slightly and we me have figured out the whole ball of wax.

nad · August 2015

I wrote:

in the Context of what i Worte abverlangt:does anybody know about Experiments Measuring the Maß (im sorry im typing on this Thinge with German autocorrection, don't know Howard to Switch that Off) of an Atom beforeandafterexiting?

Paul wrote:

Nad, looks like you are having lots of problems. Don't worry, be happy

As I wrote I had to struggle with that german autocorrection, since I used someone elses phone I didn't know how to swich that off. I wanted to write: In the context of what I wrote above:does anybody know about Experiments measuring the mass of an atom before and after exiting?

Pursuing this discussion in this thread any further would be quite off-topic. So I just mentioned this in case someone (JOhn?) would know some experiments connected with this.

Paul wrote:

In general, a lack of response is something that goes with the territory. Other than that, without seeing the specific comments that you are referring to, I can't explain why I chose not to respond or provided an "incomplete" response.

I meant amongst others this comment: https://forum.azimuthproject.org/index.php?p=discussion/1471/qbo-and-enso/&Focus=12470#Comment_12470

Anyways I don't have anymore the time to discuss the things in the comment any further. Sorry.

Paul wrote:

Does everyone appreciate how frustrating this whole business has become? It has nothing to do with the difficulty in science, but in the difficulty of working in a collaborative environment where the "complementary" and "encouraging" aspects of collaboration are missing.

What I fear most is if Per Strandberg, who is a vocal AGW denier, gets his own model out there before we do. I can't quite tell how close Per is to the basic sloshing model, but suffice to say, it would be very embarrassing if that side makes a breakthrough before the sane among us do.

As I said it seems there are commercial interests in getting better climate predictions. So I wouldn't wonder if there are apart from the academic/academic volunteer context also private companies, like assurances etc. having their own models and predictions, But they might be interested to keep their findings for themselves.

David wrote:

Nad, I see that you have made some in-depth contributions to the discussion of the climate data. Is there an implied "project" or goal that you would consider yourself working towards within Azimuth?

John wrote:

At present that's true, unless they want to help Paul Pukite or Nad - because I've been holed up working on papers with grad students,

I described the current situation already here.

That is I have to reduce my time for unpaid voluntary commitments. Moreover there are currently happenings in my direct environment which seem to be rather urgent. In addition I currently have an editing job, where I have to sit for hours in front of the screen. Sitting too long in front of the screen affects my health. So currently sitting much longer at the screen, as it is necessary for the Azimuth projects thus aquires currently a rather existential dimension. Given that I probably don't need to mention that climate science is not really what I like to do all my free time. It is something I do in order to eventually help to prevent further catastrophies. So in short again: I have to dramatically reduce my engagement with Azimuth.

I did outline again the major tasks and problems in the comments starting 11.07.2015 to Johns Google+ post at https://plus.google.com/117663015413546257905/posts/BnZCchyzSoR. David for your convenience I summarized them below after the lines. I have to cut them and paste them in seperate comment spaces, because I got the warning: "Body is 16095 characters too long." I hope didn't mess up with the pasting.

If you read them you understand that "my contributions" indicate that the current CO2 doctrine may need to be eventually rather urgently "revised". Given the problems connected with this (like trustworthiness of climate science etc.) this thus needs to be much further investigated with way more "in-depth" before making any assertions. Moreover it is clear that the possibly involved "corrections" and "revisions" of the official CO2 view of climate science may eventually go way less smoothly and unnoticed as for the case of the temperature data. I don't have the means for that more "in-depth" investigations. And apart from the fact -as explained above- that I don't have the (screen) time, I also don't have access to all ressources. Maybe I should also say at this place that depending on how big the impact of this "methane thing" will turn out, this might well be something which could quite affect myself in my lifetime, but even given that personal component - it currently makes no sense to me to work here any much further.

So concluding: Even if my findings about the quality of climate science give me a lot of doubt - I can only hope that some of the climate scientists became aware of the problems and start to work on it, like it was the case for the temperature data. John eventually goes to climate science congresses he might talk to people.

nad · August 2015

REPASTE FROM JOHNS GOOGLE PLUS ACCOUNT https://plus.google.com/117663015413546257905/posts/BnZCchyzSoR

Nadja Kutz 11.7.2015 OK John you know that since quite some years I find that "methane thing" a bit more alarming (mildly speaking) than what has been expressed here and that in particular I question the role of methane in the global warming as has been propagated by the IPCC.

I just wrote a post on our blog about the new corrections to the temperature data, which had been previously used by the IPCC.

If the methane data is correct then there was a pause in the rise of methane emissions from 1999 to 2007 and thus one indication of that role of methane would be to compare the exact shape of the new temperature curve to that of the methane curve (that is by looking at the NOAA image: http://www.noaanews.noaa.gov/stories2015/images/no slow down in global warming.jpg

one can roughly see a plateau but as said thats roughly. I don't have access to the data, otherwise I would have eventually already done this comparision as I had done previously. I really think this is important. The other aspect of methane (namely eventually a possible reduction in oxygen) is another important fact.

John Baez 11.07.2015+

+James Salsman - yes, I'm glad Colin reminded me about it because I'd forgotten. The Society for Industrial and Applied Mathematics, SIAM, has a new Activity Group on Mathematics of Planet Earth. Unfortunately it's only open to SIAM members, and I'm more connected to the AMS, the more 'pure-leaning' American mathematical society. But maybe I should pay the dough to join SIAM and this group.

Dear Colleague,

We are pleased to announce that the SIAM Council and Board of Trustees have endorsed the proposal to establish a

SIAM Activity Group on Mathematics of Planet Earth (SIAG/MPE).

This activity group will focus on planet Earth, its life-supporting capacity, and the impact of human activities.

With the establishment of SIAG/MPE, SIAM recognizes the responsibility of the mathematics and computational science community to address the issues of global change, loss of biodiversity, and sustainable development. The new Activity Group will provide a forum in SIAM to discuss the mathematical and computational issues of climate, sustainability, ecology, natural resources, food systems, the environment, socio-economic systems, and related topics. SIAG/MPE will join forces with other disciplines to further interdisciplinary research in these new application areas.

Activities will include a biennial SIAM Conference on Mathematics of Planet Earth, minisymposia at SIAM Annual Meetings, workshops, and other conferences. The inaugural meeting of SIAG/MPE is scheduled for the fall of 2016.

Membership is open to SIAM members, who can join the Activity Group at https://my-helper.siam.org/forms/join_siag.htm or by calling SIAM Customer Service (http://www.siam.org/contact/). SIAM membership is free for many students and includes membership in two SIAGs, so we encourage students to join SIAM and add SIAG/MPE to their subscription. When you join SIAG/MPE, your name will be added to our e–mailing list, which is used to announce conferences and opportunities related to Mathematics of Planet Earth. A SIAG/MPE web portal is in preparation.

We sincerely hope that you will join us in this new endeavor and are looking forward to your active participation.

The Officers of the SIAM Activity Group on MPE,

Hans Kaper, Georgetown University, Chair

Mary Lou Zeeman, Bowdoin College, Vice-Chair

Antonios Zagaris, Twente University, Secretary

Hans Engler, Georgetown University, Program Director John Baez 11.07.2015

+Nadja Kutz - You say "Methane may however play eventually also a role in a way more dramatic environmental context. In my point of view that context should also be investigated URGENTLY."

I agree that it should be urgently investigated. Luckily people are investigating it. I particularly point you to the Arctic Methane Emergency Group:

http://ameg.me/

When climate scientists are saying methane is not currently the main problem, that doesn't imply it's not a problem or that it doesn't deserve investigation. I hope you've read this:

http://www.realclimate.org/index.php/archives/2013/11/arctic-and-american-methane-in-context/

(By the way, you put html in your comments here, but that doesn't work. The best thing is to simply include URLs.)

nad · August 2015

Nadja Kutz 11.7.2015: Thanks for pointing out ameg.me - I looked very briefly into it. I got though the impression that first ameg is not a big "group", but rather a person called John Nissen and maybe a few others, secondly that the citations are not done very carefully and that the group seems rather not an expert group, a fact which need not be bad apriori. Finally I am also not an expert. But it's clear not being an expert challenges credibility, because you are more likely to say something wrong than an expert. It would be good if there would be a better informed methane task force.

I didn't say that that the IPCC claims that methane is not a problem or that it doesn't deserve investigation - I wrote: "I question the role of methane in the global warming as has been propagated by the IPCC."

In particular I wrote quite a while back that by (too briefly though) looking into the papers of the IPCC reports that I got the suspicion that the global warming potential of methane might have been eventually misestimated and that in particular the contributions from a UV absorption line may have been forgotten. But this is a highly speculative suspicion, in particular I have no means to investigate this directly. I can only look at secondary things like temperature correlations etc. But even if this speculative - frankly the way how the temperature data question was handled by the IPCC has rather hardened my suspicion that such an omittance really could have taken place. In particular I had the feeling that not too many independent groups did elaborate on that really important question.

The point is that IF there would be a big mistake in the estimations then this has rather dramatic consequences. For everybody.

Concerning the realclimate post. Thanks for the link. I don't fully agree with sentences like that: "Is this bad news for global warming? Not really, because the one real hard fact that we know about atmospheric methane is that it’s concentration isn’t rising very quickly." Of course its debateably what you mean with quickly, but if there would be a clathrate gun event then concentrations could rise VERY quickly. I redid some calculations of David Archer https://johncarlosbaez.wordpress.com/2015/03/04/melting-permafrost-part-4/#comment-64687 which I more or less confirmed but I think if there is so much at stake then first those type of calculations are by far not sufficient and secondly to me a sentence like "The Siberian Arctic, and the Americans, each emit a few percent of global emissions. Significant, but not bombs, more like large firecrackers" might be largely misleading, because as said that what we can observe now MAY change rather rapidly.

But all in all I have at the moment not the means to do much research myself, I currently have thefeeling that I had already sacked more time and energy into all those climate discussions than I can actually afford. So I hope that my comments may spur the development of a professional methane task force.

John Baez 11.07.2015

+Nadja Kutz - maybe I should have cited the Permafrost Carbon Network, which is a different group, not only focused on methane, but probably worth paying attention to:

http://www.permafrostcarbon.org/

nad · August 2015

REPASTE FROM JOHNS GOOGLE PLUS ACCOUNT https://plus.google.com/117663015413546257905/posts/BnZCchyzSoR

Nadja Kutz 12.07.2015

Thanks for the link. On a first glance this link looks a bit more professional. Unfortunately the information there is not very accessible. Even their survey article: "It draws together information from all the Permafrost Carbon Network working groups in order to outline the current state of knowledge about the impact of thawing permafrost carbon on climate in a future warmer world." is behind a paywall, depite the fact that I saw for example among the researchers people from the german Max-Planck-Institute which is as far as I know fully payed by the german tax payer.

+John Baez wrote: "which is a different group, not only focused on methane, but probably worth paying attention to"

Sure the other carbons are also important, but at the moment it just seems to me that the whereabouts of methane are especially important to understand. In particular it is of course very important to know the amount of greenhouse gases in permafrost, but what about the seepages from deep water hydrates, like mentioned in this article: http://www.sciencedaily.com/releases/2015/02/150205101921.htm

As you know I had expressed my concerns about a possible misestimation of the stability of the hydrates in deep water, which might be based on the fact that the usual phase diagram of methane may look differently in bubbly water and eventually also from the fact that the gas which comes (supposingly) from below due to tectonical happenings may eventually (?) be quite hot. Moreover this tectonical aspect needs to be investigated more. You know that I suspect that it could be that the eurasian plate is breaking because the Indo-asia-australian plate is putting it under pressure. So this which is called a "passive margin" in the article might eventually be more active than previously thought. The earthquake in Nepal may be eventually related to this. And if there are more tectonics to be going on then this might eventually DRAMATICALLY influence the methane seepage. What I wrote above is also contained in a post on our blog, which contains also an image with the suspected tectonical breaks:

http://www.randform.org/blog/?p=5768#more-5768

The methane from e.g. wetlands and the antarctic is of course also important. Moreover as said already above I am not so sure about the global warming potential -yes CO2 is about 200 times more abundant (if we believe wikipedia:https://en.wikipedia.org/wiki/Atmosphere_of_Earth) but methane is 20 times more potent (to be more exact: ca. 25/100ys) thats only a factor of ten, and on a short term scale that factor is even smaller- is an error in that order of size completely unthinkable? What is the role of methane in cloud formation?

Another thing is the above mentioned oxygen problem. Frankly I had thought myself that this might eventually be a bit overdramaticing things or as you say in german "aus der Mücke einen Elefanten machen" but I am not so sure of that anymore. There was this strange mentioning of a "lack of oxygen" in Novy Urengoy a city next to the russian gas fields in that post I linked above. What is the amount of natural gas which would lead to breathing problems, are there other health problems for higher doses, toxicity etc.? This article here: http://journals.lww.com/epidem/Fulltext/2011/01001/Methane_and_Natural_Gas_Exposure_Limits.771.aspx speaks about no toxiticity of methane and an asphyxiant work exposure limit of 5300 ppm which seems (if I got the factors right) off by roughly a factor of 26 from current average methane concentrations of 179 ppm. (the other components of natural gas taken aside for a moment). This elder book http://www.nap.edu/openbook.php?record_id=689&page=95 writes though: "Thus, the toxic effect of methane is much greater than that of nitrogen when available oxygen is low, but methane has little effect when oxygen is readily available. It seems that the toxicity of methane should bediscussed not alone, but rather with respect to the partial pressure of oxygen in the atmosphere in question."

I do think there should be a professional methane/natural gas task force.

nad · August 2015

REPASTE FROM JOHNS GOOGLE PLUS ACCOUNT https://plus.google.com/117663015413546257905/posts/BnZCchyzSoR

Nadja Kutz 12.7.

I wrote above: "- is an error in that order of size completely unthinkable?"

The Global Warming Potential (GWP) is computed from the Radiative Forcing (RF), here are the definitions I found in chapter 2 of AR 4 (lead authors:Piers Forster (UK), Venkatachalam Ramaswamy (USA)): http://www.azimuthproject.org/azimuth/show/Global+warming+potential

In particular there the RF is described in chapter 2 as: "The definition of RF from the TAR and earlier IPCC assessment reports is retained. Ramaswamy et al. (2001) define it as “the change in net (down minus up) irradiance (solar plus longwave; in W m^2) at the tropopause after allowing for stratospheric temperatures to readjust to radiative equilibrium, but with surface and tropospheric temperatures and state held fixed at the unperturbed values”.

In the supplement to AR5 https://www.ipcc.ch/pdf/assessment-report/ar5/wg1/supplementary/WG1AR5_Ch08SM_FINAL.pdf page 7 sectin 8.SM.3 it is written: "The formulae used to calculate the radiative forcings (RFs) from carbon dioxide (CO2), CH4 and nitrous oxide (N2O) are taken from Myhre et al. (1998) Table 3 as in Third Assessment Report (TAR) and Fourth Assessment Report (AR4). They are listed here for convenience."

That formula for CH4 is listed in table 8.SM.1. It depends onvalues M, M_0, N, N_0 measured in ppm. Where its not said what these quantities are, but luckily there is a copy of the paper Myhre, G., E. J. Highwood, K. P. Shine, and F. Stordal, 1998: New estimates of radiative forcing due to well mixed greenhouse gases. Geophys. Res. Lett., 25, 2715–2718 on Gunnar Myhre's (probably old) postdoc homepage: (I can't find a recent page by Gunnar Myrhe) http://folk.uio.no/gunnarmy/ That is the article is at the URL:

http://folk.uio.no/gunnarmy/paper/myhre_grl98.pdf

There it is explained that the variables are concentrations which are used together with certain models, called NBM, LBL and BBM. On page 2715 it is written that: "The LBL model is used to calculate optical depth (Edwards1992) and radiative fluxes are calculated as in the work of Myrhe and Stordal (1997). The NBM is the 10 cm^(-1) narrow band radiative transfer scheme of Shine (1991)....The LBL model and the BBM use spectrocopic data for the halocarbons from the HITRAN96 database and for the other WMGG as in Myrhe and Stordal (1997).

So in order to asses the quality of the assessments at least of the RF it seems one would need to understand those models.

The coauthor Frode Stordal seems to be still reachable: http://www.mn.uio.no/geo/english/people/aca/metos/frodes/

As a side remark: There is a typo in the formula under Notes (I hope its only a typo and this is not what was used....) that is in the formula for f(M,N) it should be (MN)^(0.75) and not as written there: (MN)0.75

DavidTanzer · August 2015

...continuing the line of thought from above...

In software development the "stakeholders" include end-users and programmers. The business analyst functions as an interpreter between two very different languages. Key to this effort is highly targeted literature -- specifications or "stories," that explain specific needs and can be translated into specific technical requirements.

DavidTanzer · August 2015

Now let's consider the "stakeholders" at Azimuth:

Researchers who take the lead in specific projects.
Students and colleagues of these researchers, involved in Azimuth-related projects.
The overall, heterogeneous Azimuth community at the Azimuth Forum, who have an interest in understanding this research -- this is a generality, of course, degree of interest will vary widely across individuals.
A subgroup of Azimuth consisting of software developers.
Potentially, if we develop: That wider section of the public, who are interested in science and the environment, and would be interested to know, at the appropriate level of detail, what is going on at Azimuth.
Readership of the blog. This group may cut across all of the above groupings.

DavidTanzer · August 2015

Certainly there are many different "languages" spoken here, and some kind of translations will need to be performed in order for these groups to effectively collaborate.

The appropriate forms of translation will depend on what the specific objectives are.

DavidTanzer · August 2015

Suppose, for example, that we had researchers who were performing in-depth studies of network dynamics, and wanted to run complex experiments involving simulation software -- and perhaps even to create new types of simulations. Then they would have an active interest in communicating these ideas to the programmers, in order to start a productive collaboration.

Although this appears not to be the case at present, let's consider, as a thought experiment, what would be the communication issues here, and they could be effectively addressed.

WebHubTel · August 2015

John said:

"You haven't said whether you're interested in writing a joint paper or not. What do you think?"

As a reminder, I sent you a private Forum message to tell you who the other collaborators would be. I know it is easy to not notice the notification indicator in the display.

This topic has got me pumped, and some others too, but in the wrong way -- enough that it has gotten under the skin of a lot of the blogosphere climate science "gatekeepers". Everyone immediately thinks that the ENSO model is a case of over-fitting and that I am being too pushy.

With 4 known factors, the model can correctly identify 40 of the 43 inflection points in the 100 year ENSO record running from 1880 to 1980, and those three missing points are arguably at the noise level.

This is from another thread that Graham started recently

/discussion/1640/predictability-of-the-quasi-biennial-oscillation

Azimuth Forum

Howdy, Stranger!

Categories

The future of Azimuth

Comments

---