#### Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Options

# Global warming and thermodynamical quantities

• Options
51.
edited November 2014

I have been saying that the sources of the notebooks I posted were in the same directory, but I had not shared a link to it and it is not accessible from the html notebooks themselves. So here it is.

Comment Source:I have been saying that the sources of the notebooks I posted were in the same [directory](https://www.googledrive.com/host/0B4cyIPgV_VxrX0lxSUxHU2VLN28), but I had not shared a link to it and it is not accessible from the html notebooks themselves. So [here](https://www.googledrive.com/host/0B4cyIPgV_VxrX0lxSUxHU2VLN28) it is.
• Options
52.

Pressure seems to be little better for predict enso3.4 than temperature.

Comment Source:[Pressure](http://www.googledrive.com/host/0B4cyIPgV_VxrbDFJS1dTVWxhV28) seems to be little better for predict enso3.4 than temperature.
• Options
53.
edited November 2014

If you transform the data instead to DELTAt=Xt-Xt-1 and the perform your code on this delta given t-1 is less than the current time, there might be a better fit possible. So the result would be Xt + Forecast(DELTAt). This works best for data that is not overly noisy. Almost all guidance systems work on delta data.

Comment Source:If you transform the data instead to DELTAt=Xt-Xt-1 and the perform your code on this delta given t-1 is less than the current time, there might be a better fit possible. So the result would be Xt + Forecast(DELTAt). This works best for data that is not overly noisy. Almost all guidance systems work on delta data.
• Options
54.

The atmospheric pressure anomaly is a very good predictor of NINO3.4, especially if the Tahiti-Darwin SOI dataset is used. In this case the issue is that the prediction is virtually instantaneous, having no reported lead or lag.

What lag or lead are you seeing Daniel?

Comment Source:The atmospheric pressure anomaly is a very good predictor of NINO3.4, especially if the Tahiti-Darwin SOI dataset is used. In this case the issue is that the prediction is virtually instantaneous, having no reported lead or lag. What lag or lead are you seeing Daniel?
• Options
55.
edited November 2014

Daniel wrote:

I have done some preliminary analyses on the data. The results are here. The same directory also contains the source as both an ipython notebook and a plain python script.

Great, thanks!

The main takeaway is in the scatter plot and the last cross correlation plot. The dependence between the anomaly and the link strength looks weak and the dependence of the crosscorrelation on the lag is erratic.

Ludescher et al claim that the link strength exceeding a certain cutoff predicts an El Niño in the following calendar year, which roughly means the anomaly will exceed a certain cutoff (on average) for some number of months. So they have a rather more tricky - and a skeptic might say, contrived - way of trying to read predictions out of the link strength. It's possible that they are really on to something, and that the effect is real but highly nonlinear: average link strength exceeding a certain value triggers an El Niño. So, to be fair to them, we should do some tests that look for such a nonlinear effect.

I'm in too much of a rush now to see if you were looking only for linear correlations.

Does that mean the link strength data only goes to 1980? Can we get it to go to the present?

It's pretty easy to get it to go to the present. Graham Jones had been avoiding looking at data after 1980, so he could formulate theories and then later test them with that more recent data.

It will take my PC about an hour to compute these link strengths. I'll try to do it later today.

Comment Source:Daniel wrote: > I have done some preliminary analyses on the data. The results are [here](https://googledrive.com/host/0B4cyIPgV_VxraU5pTVFjNkFtYmM). The same directory also contains the source as both an ipython notebook and a plain python script. Great, thanks! > The main takeaway is in the scatter plot and the last cross correlation plot. The dependence between the anomaly and the link strength looks weak and the dependence of the crosscorrelation on the lag is erratic. Ludescher _et al_ claim that the link strength exceeding a certain cutoff predicts an El Ni&ntilde;o in the following calendar year, which roughly means the anomaly will exceed a certain cutoff (on average) for some number of months. So they have a rather more tricky - and a skeptic might say, contrived - way of trying to read predictions out of the link strength. It's possible that they are really on to something, and that the effect is real but highly nonlinear: average link strength exceeding a certain value triggers an El Ni&ntilde;o. So, to be fair to them, we should do some tests that look for such a nonlinear effect. I'm in too much of a rush now to see if you were looking only for linear correlations. > Does that mean the link strength data only goes to 1980? Can we get it to go to the present? It's pretty easy to get it to go to the present. Graham Jones had been avoiding looking at data after 1980, so he could formulate theories and then later test them with that more recent data. It will take my PC about an hour to compute these link strengths. I'll try to do it later today.
• Options
56.
edited November 2014

https://raw.githubusercontent.com/johncarlosbaez/el-nino/master/R/nino3.4-anoms.txt

does not match the link Daniel used, for the end of the file months (please double check in case of I have cache issues in my browser for some reason):

http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/detrend.nino34.ascii.txt

I used John's for SVR publish the results tonight

Dara

Comment Source:This link from John [https://raw.githubusercontent.com/johncarlosbaez/el-nino/master/R/nino3.4-anoms.txt](https://raw.githubusercontent.com/johncarlosbaez/el-nino/master/R/nino3.4-anoms.txt) does not match the link Daniel used, for the end of the file months (please double check in case of I have cache issues in my browser for some reason): [http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/detrend.nino34.ascii.txt](http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/detrend.nino34.ascii.txt) I used John's for SVR publish the results tonight Dara
• Options
57.

https://raw.githubusercontent.com/johncarlosbaez/el-nino/master/R/nino3.4-anoms.txt

does not match the link Daniel used, for the end of the file months (please double check in case of I have cache issues in my browser for some reason):

http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/detrend.nino34.ascii.txt

I used John's for SVR publish the results tonight

The first line of the github file indicates that it is a copy of the NOAA file. This script is just pointing at the original since that is likely to cause less confusion if we share the code outside this group. Also the additional first line in the github file stops it being directly readable by built table-reading functions of various systems.

Comment Source:> This link from John > [https://raw.githubusercontent.com/johncarlosbaez/el-nino/master/R/nino3.4-anoms.txt](https://raw.githubusercontent.com/johncarlosbaez/el-nino/master/R/nino3.4-anoms.txt) > does not match the link Daniel used, for the end of the file months (please double check in case of I have cache issues in my browser for some reason): > [http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/detrend.nino34.ascii.txt](http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/detrend.nino34.ascii.txt) > I used John's for SVR publish the results tonight The first line of the github file indicates that it is a copy of the NOAA file. This script is just pointing at the original since that is likely to cause less confusion if we share the code outside this group. Also the additional first line in the github file stops it being directly readable by built table-reading functions of various systems.
• Options
58.
edited November 2014

The atmospheric pressure anomaly is a very good predictor of NINO3.4, especially if the Tahiti-Darwin SOI dataset is used. In this case the issue is that the prediction is virtually instantaneous, having no reported lead or lag.

What lag or lead are you seeing Daniel?

I am not using the pressure anomaly. I am using the actual pressure measurements, ie I have about 10K input features per time point and I am using those to predict the enso 6 month after that time point. During training the model is given a set of [current pressure map, enso 6 month later] and tries to learn that association. The model is never presented with the enso that goes with the current map.

Comment Source:> The atmospheric pressure anomaly is a very good predictor of NINO3.4, especially if the Tahiti-Darwin SOI dataset is used. In this case the issue is that the prediction is virtually instantaneous, having no reported lead or lag. > What lag or lead are you seeing Daniel? I am not using the pressure anomaly. I am using the actual pressure measurements, ie I have about 10K input features per time point and I am using those to predict the enso 6 month after that time point. During training the model is given a set of [current pressure map, enso 6 month later] and tries to learn that association. The model is never presented with the enso that goes with the current map.
• Options
59.

That's cool Daniel. I have a strong correlation tied into sea-surface height via tidal gauge records. The intriguing connection there is that the inverse barometer effect ties pressure to sea-surface height.

Comment Source:That's cool Daniel. I have a strong correlation tied into sea-surface height via tidal gauge records. The intriguing connection there is that the inverse barometer effect ties pressure to sea-surface height.
• Options
60.

Daniel wrote:

Does that mean the link strength data only goes to 1980? Can we get it to go to the present?

Okay, I computed the link strength from 1948 to 2013 and put it here:

The second column in this file lists the average link strengths S as computed by ludescher.R at 10-day intervals, starting from day 730, and going until day 24090, where day 1 is the first of January 1948. The first column numbers these items from 1 to 2337. For a full explanation see Part 4 of the El Niño Project series.

Comment Source:Daniel wrote: > Does that mean the link strength data only goes to 1980? Can we get it to go to the present? Okay, I computed the link strength from 1948 to 2013 and put it here: * [average-link-strength-1948-2013.txt](https://github.com/johncarlosbaez/el-nino/blob/master/R/average-link-strength-1948-2013.txt). The second column in this file lists the average link strengths S as computed by ludescher.R at 10-day intervals, starting from day 730, and going until day 24090, where day 1 is the first of January 1948. The first column numbers these items from 1 to 2337. For a full explanation see [Part 4](http://johncarlosbaez.wordpress.com/2014/07/08/el-nino-project-part-4/) of the El Niño Project series.
• Options
61.
edited November 2014

I'm afraid I hijacked this thread, which was about "Global warming and thermodynamical quantities". I urge that all further discussion of work mainly related to my NIPS talk - attempts to understand El Niño using machine learning and climate networks - be continued on the thread crunch time, or some other suitable thread.

There's a lot more to say about the interesting things Nad and WebHubTel are doing! I only changed the subject because Dara expressed an interest in helping me. As you all know, until December 10th what I really need help on is that NIPS talk! After that's done I can relax and take stock of all the things we're doing.

Comment Source:I'm afraid I hijacked this thread, which was about "Global warming and thermodynamical quantities". I urge that all further discussion of work mainly related to my NIPS talk - attempts to understand El Ni&ntilde;o using machine learning and climate networks - be continued on the thread [crunch time](http://forum.azimuthproject.org/discussion/1523/crunch-time/), or some other suitable thread. There's a lot more to say about the interesting things Nad and WebHubTel are doing! I only changed the subject because Dara expressed an interest in helping me. As you all know, until December 10th what I really need help on is that NIPS talk! After that's done I can relax and take stock of all the things we're doing.
• Options
62.
edited November 2014

Paul wrote:

John wrote:

2) I’ve been reluctant to embrace Nadja Kutz’s theory that an exact biannual cycle in the Earth’s temperature could arise from a resonance of the magnetic fields of Earth and Sun.


The biannual cycle is a actually pretty straightforward to explain. The sun crosses the equator twice each calendar year. One time exposing the northern hemisphere with more solar energy, and the other time exposing the southern hemisphere more. This leads to a biannual harmonic in the temperature time series.

That's an interesting possibility. It's also a fact that any continuous periodic function has a Fourier series

$$a e^{i \omega t} + b e^{2 i \omega t} + \cdots$$ and quite often $b$ will be fairly large compared to $a$. So, we should expect to see some oscillation occurring twice a year unless there's a good reason for one not to exist.

However, I thought that Nad was talking about a harmonic with period two years, not one with period half a year! That's what she was talking about in her 22 June 2014 blog article.

Maybe I'm confused. The word biannual means both "occurring twice each year" and - less often - "occurring every two years". To avoid this ambiguity, I would prefer to say semiannual for "occurring twice each year", and biennial for "occurring every two years".

Comment Source:Paul wrote: > John wrote: > 2) I’ve been reluctant to embrace Nadja Kutz’s theory that an exact biannual cycle in the Earth’s temperature could arise from a resonance of the magnetic fields of Earth and Sun. > The biannual cycle is a actually pretty straightforward to explain. The sun crosses the equator twice each calendar year. One time exposing the northern hemisphere with more solar energy, and the other time exposing the southern hemisphere more. This leads to a biannual harmonic in the temperature time series. That's an interesting possibility. It's also a fact that any continuous periodic function has a Fourier series $$a e^{i \omega t} + b e^{2 i \omega t} + \cdots$$ and quite often $b$ will be fairly large compared to $a$. So, we should expect to see some oscillation occurring twice a year unless there's a good reason for one not to exist. However, I thought that Nad was talking about a harmonic with period _two years_, not one with period _half a year_! That's what she was talking about in [her 22 June 2014 blog article](http://www.randform.org/blog/?p=5572). Maybe I'm confused. The word [biannual](http://www.merriam-webster.com/dictionary/biannual) means both "occurring twice each year" and - less often - "occurring every two years". To avoid this ambiguity, I would prefer to say [semiannual](http://www.merriam-webster.com/dictionary/semiannual) for "occurring twice each year", and [biennial](http://www.merriam-webster.com/dictionary/biennial) for "occurring every two years".
• Options
63.
edited November 2014

The word biannual means both “occurring twice each year” and - less often - “occurring every two years”.

I said this already here in the forum but I used the meaning "occurring every two years", because the words semiannual, annual, biannual, triannual are in a line according to the latin prefix numbering system, where strictly speaking biannual should be duannual. Biennial falls out of this sequence - but who am I to correct the english language, if you say I should use biennial then I probably should - I hope though I don't forget.

However, I thought that Nad was talking about a harmonic with period two years, not one with period half a year! That’s what she was talking about in her 22 June 2014 blog article.

Yes. And I am still looking for a reason for this. I meanwhile implemented a calculation for the station data where I calculated just the simple average anomalies over the station data (after "averaging away" the annual component (since the data is mostly from the northern hemisphere you can see in the average the annual seasons) and I find one can see that biennial oscillation even in the 18th century (where one has to say that the CRUTEM file reflects for this time just the icy Berlin weather (actually so icy that I felt rather sorry for my ancestors)) . So at least it seems the oscillation doesn't seem to be due to some error in the HADCRUT anomaly data calculation. So yes I am still looking for a reason for an exact biennial oscillation.

As I wrote above a different radiation balance of the two hemispheres could play a role (and I think this could eventually be due to the earths orbits ellipticity but maybe it is also linked to Albedo and general thermodynamical considerations, as indicated in the beginning of this thread) and I had sent you a copy of an email where I outline how such a radiation imbalance and in particular the resulting different annual temperature balances could be related to the QBO. The outline was mainly based on the visualization at ugamp, which Paul had linked to I think. So in short the QBO could eventually be the result of an annual and seasonal (semiannual) oscillation and thus it is "sort of exact" biennal. And maybe the biennial temperature oscillations are due to the QBO and the annual and biennial features connected with it.

But this outline is just a working hypothesis and I would need to outline this in more detail especially together with what I wrote in July:

it looks that the heat transfer of temperatures over the northern atlantic to the basin in front of northern south America seem to play a bigger role. That is it looks as if the transfered heat content which had been piled up in the northern atlantic is “bigger” than the heat content over the islands between Australia and China and partially over the indian ocean than this may at least contribute to an El Nino condition (“a chimney over South America”, so to say). For saying wether this “leads” to an El Nino condition I haven’t seen enough data to tell, one would need to look at way more El nino images and probably this aspect of atlantic heat transfer vs Asia is not enough. But I currently think that it probably should be taken into consideration within any model for prediction. But as said I just looked at it for two hours. Temperatures over the northern atlantic and south America are currently piling up, but there is still quite some heat content over Asia. So just by judging with respect to this heat content transfer it currently wouldn’t be so clear wether there would be an El Nino in autumn. This would depend how temperatures develop in the next months. Furthermore that images are also not so good for detecting differences in heat content.

The email outline goes together with this quick description, but it takes still some work to make that more coherent and informed. Like thanks to Jim Stuttard I learned yesterday that this heat content over the indian ocean is called IOD. But also as said above I still look for other reasons too, like the role of the sunwind (and its impact especially on the UV) etc.

Comment Source:> The word biannual means both “occurring twice each year” and - less often - “occurring every two years”. I said this already here in the forum but I used the meaning "occurring every two years", because the words semiannual, annual, biannual, triannual are in a line according to the <a href="http://en.wikipedia.org/wiki/Numeral_prefix">latin prefix numbering system</a>, where strictly speaking biannual should be duannual. Biennial falls out of this sequence - but who am I to correct the english language, if you say I should use biennial then I probably should - I hope though I don't forget. >However, I thought that Nad was talking about a harmonic with period two years, not one with period half a year! That’s what she was talking about in her 22 June 2014 blog article. Yes. And I am still looking for a reason for this. I meanwhile implemented a calculation for the station data where I calculated just the simple average anomalies over the station data (after "averaging away" the annual component (since the data is mostly from the northern hemisphere you can see in the average the annual seasons) and I find one can see that biennial oscillation even in the 18th century (where one has to say that the CRUTEM file reflects for this time just the icy Berlin weather (actually so icy that I felt rather sorry for my ancestors)) . So at least it seems the oscillation doesn't seem to be due to some error in the HADCRUT anomaly data calculation. So yes I am still looking for a reason for an exact biennial oscillation. As I wrote <a href="http://forum.azimuthproject.org/discussion/1485/global-warming-and-thermodynamical-quantities/?Focus=13316#Comment_13316">above</a> a different radiation balance of the two hemispheres could play a role (and I think this could eventually be due to the earths orbits ellipticity but maybe it is also linked to Albedo and general thermodynamical considerations, as indicated in the beginning of this thread) and I had sent you a copy of an email where I outline how such a radiation imbalance and in particular the resulting different annual temperature balances could be related to the QBO. The outline was mainly based on the visualization at ugamp, which Paul had linked to I think. So in short the QBO could eventually be the result of an annual and seasonal (semiannual) oscillation and thus it is "sort of exact" biennal. And maybe the biennial temperature oscillations are due to the QBO and the annual and biennial features connected with it. But this outline is just a working hypothesis and I would need to outline this in more detail especially together with what I wrote in <a href="http://forum.azimuthproject.org/discussion/169/el-nino-southern-oscillation-enso/?Focus=11557#Comment_11557">July: </a> >it looks that the heat transfer of temperatures over the northern atlantic to the basin in front of northern south America seem to play a bigger role. That is it looks as if the transfered heat content which had been piled up in the northern atlantic is “bigger” than the heat content over the islands between Australia and China and partially over the indian ocean than this may at least contribute to an El Nino condition (“a chimney over South America”, so to say). For saying wether this “leads” to an El Nino condition I haven’t seen enough data to tell, one would need to look at way more El nino images and probably this aspect of atlantic heat transfer vs Asia is not enough. But I currently think that it probably should be taken into consideration within any model for prediction. But as said I just looked at it for two hours. Temperatures over the northern atlantic and south America are currently piling up, but there is still quite some heat content over Asia. So just by judging with respect to this heat content transfer it currently wouldn’t be so clear wether there would be an El Nino in autumn. This would depend how temperatures develop in the next months. Furthermore that images are also not so good for detecting differences in heat content. The email outline goes together with this quick description, but it takes still some work to make that more coherent and informed. Like thanks to Jim Stuttard I learned <a href="http://forum.azimuthproject.org/discussion/1529/the-indian-ocean-dipole-iod-and-enso/#Item_8">yesterday</a> that this heat content over the indian ocean is called IOD. But also as said above I still look for other reasons too, like the role of the sunwind (and its impact especially on the UV) etc.
• Options
64.
edited November 2014

I am currently a bit mixed about what to do with all this. That is I think it would be good to get some "results" especially in order to highlight the power of scienitific visualizations - and especially with respect to what I call the "conceptual part of a visualization (see also explanation below), but on the other hand I am not an expert in this matter and the probability to find something new in such an charted research territory -especially as a nonexpert- is about almost zero. I

First -I wrote this earlier - I consider visualizations as a type of art, but I am aware that this is not how it is usually seen in a broader "art" audience, but I think this should be more scrutinized and that's among others what our blog is for. The term "conceptual" is here made with reference to conceptual art.

In a talk in Tokyo, I tried amongst others to illustrate the differences between what I called intra-visualization, teach-visualization and show-visualization and how a "visualization" (where strictly speaking I made use not only of the term "visual" but of the term organoleptic: i.e. using the five senses vision, hearing, taste, sense of smell,sense of touch) may contain components of each of these, depending on audience and purpose. The climate visualizations are rather "intra" and "teach" than "show". I will explain below more what I mean by that below. Unfortunately these "intra" and "teach" aspects are often not so seen as such in a general audience and sometimes even not in a science audience (that's amongst others why I wrote a term paper with some concrete examples...in 1987.....)

Comment Source:I am currently a bit mixed about what to do with all this. That is I think it would be good to get some "results" especially in order to highlight the power of scienitific visualizations - and especially with respect to what I call the "conceptual part of a visualization (see also explanation below), but on the other hand I am not an expert in this matter and the probability to find something new in such an charted research territory -especially as a nonexpert- is about almost zero. I So what about this "conceptual part" of a visualization? First -I wrote this earlier - I consider visualizations as a type of art, but I am aware that this is not how it is usually seen in a broader "art" audience, but I think this should be more scrutinized and that's among others what our blog is for. The term "conceptual" is here made with reference to <a href="http://en.wikipedia.org/wiki/Conceptual_art">conceptual art.</a> In a talk <a href="http://www.f.waseda.jp/martin/conf/arch/g-gv1/gvseminars.html">in Tokyo</a>, I tried amongst others to illustrate the differences between what I called intra-visualization, teach-visualization and show-visualization and how a "visualization" (where strictly speaking I made use not only of the term "visual" but of the term organoleptic: i.e. using the five senses vision, hearing, taste, sense of smell,sense of touch) may contain components of each of these, depending on audience and purpose. The climate visualizations are rather "intra" and "teach" than "show". I will explain below more what I mean by that below. Unfortunately these "intra" and "teach" aspects are often not so seen as such in a general audience and sometimes even not in a science audience (that's amongst others why I wrote a term paper with some concrete examples...in 1987.....)
• Options
65.
edited November 2014

That is "intra" and "teach" are often rather some kind of conceptual art, where with "teach" the overall communicational aspect has more importance (intra is rather intra-communication). For "show" the concept (the science if you want) is not so important. Artists with no science background often use "show" only, that is here the visual (organoleptic) character itself plays a big role, like if people do generative art and play around with code, they are mostly guided by the visuals (or audios) and not by some scientific or mathematical concept. The fact that their work can eventually be related to a concept is here usually rather a side effect, in particular every piece of code is in some sense "a concept", i.e. it is usually related to mathematical structures. Like if I remember what you posted about the roots of polynomials with integer coefficients started out like this. In the art world this "show" aspect is often regarded as art. And it is usually not called scientific visualization, but there exist all sorts of other terms, like computer art, generative art, etc. and it is indeed a good question wether a show-only visualization should still be called a visualization (if it can be related to a scientific concept in retrospective may be eventually yes). Finally a painting of a sunset is in some sense a "visualization of a scientifically describable reality".There are however mixed forms too, especially in art-science collaborations.

"Intra" is like "Show" eventually accepted as an art form, if it is rather personal and expressive, like the math doodlings of a mathematician could eventually be considered as some "art" in the art informed world, but usually the strictness of the scientific concepts and the often rather strict symbolism (like temperature blue is cold, red is hot, mathematical symbol conventions etc.) of scientific visualizations which are more in the intra and teach domain are often regarded as having too much impact on the freedom of expression in order to be accepted as "art". Like there exist the term "math art" however that was mostly coined by scientists/mathematicians and a lot of it is -even if it is often thought to be "show"- rather than "intra" or "teach" - perceived by most as "intra" or "teach" because it's visual language is often of a rather particular kind and thus math art is often rather existing as a kind of "niche art". A bit similarily regarded are often rather realistic renderings in computer graphics. Here again the scientific laws, which "rule" the visuals have often to be at least used in an expressive, artsy context in order to be accepted, that is here in this artsy background the concept (i.e. the code) is more seen as a tool, rather than as a concept in itself. A scientific visualization would here be for example the demonstration of the result that changing some parameters has on the crunchiness of a computer generated fabric etc.

So in short "concepts" per se are admitted in art (last but not least see e.g. conceptual art, where one has to say that conceptual art itself has already a somewhat shaky standing in the art world)- however as it seems usually conceptual constraints are usually only accepted if they are more or less "self-imposed" and not "dictated" e.g. by nature/science or historical conventions. In particular since the "dictatorship aspect" (that is how much is "dictated" by nature/conventions) is not so clear for a general audience who does often not know about the involved concepts there seem to be rather fast rejectances of works which adhere to strict scientific representations and there seem to be a rather fast acceptances of works, which are more "playful" etc, but which may eventually completely distort a given scientific reality, i.e. they may rather obfuscate than inform. At least that is the impression I got sofar. This was though not always so. Especially in the Renaissance (where a lot of the science was accessible to a broader public though) there were quite a bit of strong science influences and scientists were more accepted as "artists" and in particular there were hybrids where the distinction was not so clear. Like Leonardo da Vinci was to quite some extend a mechanical engineer. Dürer wrote math books (who where as it seems quite influential in the arts ), both, but especially Leonardo were as it seems rather influenced by Luca Pacioli and Matthias Grünewald was probably also a hydraulic engineer.

So concluding since quite some time I think it is necessary to present examples where the "conceptual part" of a visualization and its power (like with a "result") can be better demonstrated. The climate visualizations appear to me as a quite a good example. Moreover they have quite some components to them which rather clearly touch upon human lives - more than a lot of artworks - I find (even if eventually in another way) and even if their "show" i.e. their visual aspect appears to be neglected.

But as said the "result" aspect is not clear and if there is no half-way "sensational" aspect to it (a la a headline like: "is art helping to predict el nino's?", ceci n'est pas el nino! ) then this example is just one of the examples, which I am -since almost 30 years (....) - on and off trying to communicate. However for being "one of an example" this whole thing is getting now a bit too involved.

Comment Source:That is "intra" and "teach" are often rather some kind of conceptual art, where with "teach" the overall communicational aspect has more importance (intra is rather intra-communication). For "show" the concept (the science if you want) is not so important. Artists with no science background often use "show" only, that is here the visual (organoleptic) character itself plays a big role, like if people do generative art and play around with code, they are mostly guided by the visuals (or audios) and not by some scientific or mathematical concept. The fact that their work can eventually be related to a concept is here usually rather a side effect, in particular every piece of code is in some sense "a concept", i.e. it is usually related to mathematical structures. Like if I remember what you posted about the roots of polynomials with integer coefficients started out like this. In the art world this "show" aspect is often regarded as art. And it is usually not called scientific visualization, but there exist all sorts of other terms, like computer art, generative art, etc. and it is indeed a good question wether a show-only visualization should still be called a visualization (if it can be related to a scientific concept in retrospective may be eventually yes). Finally a painting of a sunset is in some sense a "visualization of a scientifically describable reality".There are however mixed forms too, especially in art-science collaborations. "Intra" is like "Show" eventually accepted as an art form, if it is rather personal and expressive, like the math doodlings of a mathematician could eventually be considered as some "art" in the art informed world, but usually the strictness of the scientific concepts and the often rather strict symbolism (like temperature blue is cold, red is hot, mathematical symbol conventions etc.) of scientific visualizations which are more in the intra and teach domain are often regarded as having too much impact on the freedom of expression in order to be accepted as "art". Like there exist the term "math art" however that was mostly coined by scientists/mathematicians and a lot of it is -even if it is often thought to be "show"- rather than "intra" or "teach" - perceived by most as "intra" or "teach" because it's visual language is often of a rather particular kind and thus math art is often rather existing as a kind of "niche art". A bit similarily regarded are often rather realistic renderings in computer graphics. Here again the scientific laws, which "rule" the visuals have often to be at least used in an expressive, artsy context in order to be accepted, that is here in this artsy background the concept (i.e. the code) is more seen as a tool, rather than as a concept in itself. A scientific visualization would here be for example the demonstration of the result that changing some parameters has on the crunchiness of a computer generated fabric etc. So in short "concepts" per se are admitted in art (last but not least see e.g. <a href="http://en.wikipedia.org/wiki/Conceptual_art">conceptual art</a>, where one has to say that conceptual art itself has already a somewhat shaky standing in the art world)- however as it seems usually conceptual constraints are usually only accepted if they are more or less "self-imposed" and not "dictated" e.g. by nature/science or historical conventions. In particular since the "dictatorship aspect" (that is how much is "dictated" by nature/conventions) is not so clear for a general audience who does often not know about the involved concepts there seem to be rather fast rejectances of works which adhere to strict scientific representations and there seem to be a rather fast acceptances of works, which are more "playful" etc, but which may eventually completely distort a given scientific reality, i.e. they may rather obfuscate than inform. At least that is the impression I got sofar. This was though not always so. Especially in the Renaissance (where a lot of the science was accessible to a broader public though) there were quite a bit of strong science influences and scientists were more accepted as "artists" and in particular there were hybrids where the distinction was not so clear. Like Leonardo da Vinci was to quite some extend a mechanical engineer. Dürer wrote <a href="http://en.wikipedia.org/wiki/Albrecht_D%C3%BCrer#Four_Books_on_Measurement">math books</a> (who where as it seems quite <a href="http://www.dataisnature.com/?p=2048">influential in the arts </a>), both, but especially Leonardo were as it seems rather influenced by <a href="http://en.wikipedia.org/wiki/Luca_Paciolihttp://en.wikipedia.org/wiki/Luca_Pacioli">Luca Pacioli</a> and <a href="http://en.wikipedia.org/wiki/Matthias_Gr%C3%BCnewald">Matthias Grünewald</a> was probably also a hydraulic engineer. So concluding since quite some time I think it is necessary to present examples where the "conceptual part" of a visualization and its power (like with a "result") can be better demonstrated. The climate visualizations appear to me as a quite a good example. Moreover they have quite some components to them which rather clearly touch upon human lives - more than a lot of artworks - I find (even if eventually in another way) and even if their "show" i.e. their visual aspect appears to be neglected. But as said the "result" aspect is not clear and if there is no half-way "sensational" aspect to it (a la a headline like: "is art helping to predict el nino's?", ceci n'est pas el nino! ) then this example is just one of the examples, which I am -since almost 30 years (....) - on and off trying to communicate. However for being "one of an example" this whole thing is getting now a bit too involved.
• Options
66.
edited November 2014

Biennal falls out of this sequence - but who am I to correct the english language, if you say I should use biennal then I probably should - I hope though I don't forget.

It's "biennial", not "biennal". English is unfortunately not a logical language. The word "biannual" is sort of famous for being confusing, since 3/4 of the time it means "two times a year" and 1/4 of the time it means "once every two years". It confused WebHubTel.

But as said the “result” aspect is not clear and if there is no half-way “sensational” aspect to it...

It may not be "sensational", but to me an exact biennial oscillation in the Earth's average temperature would be a shocking and important discovery. Since I'm rather conservative, I won't believe it's real until I see a lot of evidence. And right now, a theory of what causes it seems less interesting to me than a bit more data analysis. Can you do some statistical tests to estimate the probability that the effect you see is just a coincidence? Can you try looking at other climate records besides HadCRUT? It's easy to get global average temperature records from GISS (see the link).

Comment Source:Nad wrote: > Biennal falls out of this sequence - but who am I to correct the english language, if you say I should use biennal then I probably should - I hope though I don't forget. It's "biennial", not "biennal". English is unfortunately not a logical language. The word "biannual" is sort of famous for being confusing, since 3/4 of the time it means "two times a year" and 1/4 of the time it means "once every two years". It confused WebHubTel. > But as said the “result” aspect is not clear and if there is no half-way “sensational” aspect to it... It may not be "sensational", but to me an exact biennial oscillation in the Earth's average temperature would be a shocking and important discovery. Since I'm rather conservative, I won't believe it's real until I see a lot of evidence. And right now, a theory of what causes it seems less interesting to me than a bit more data analysis. Can you do some statistical tests to estimate the probability that the effect you see is just a coincidence? Can you try looking at other climate records besides HadCRUT? It's easy to get global average temperature records from [GISS](http://data.giss.nasa.gov/gistemp/graphs_v3/) (see the [link](http://data.giss.nasa.gov/gistemp/graphs_v3/Fig.A2.txt)).
• Options
67.
edited November 2014

John wrote:

It may not be “sensational”, but to me an exact biennial oscillation in the Earth’s average temperature would be a shocking and important discovery.

Yes. And I am still looking for a reason for this. I meanwhile implemented a calculation for the station data where I calculated just the simple average anomalies over the station data (after “averaging away” the annual component (since the data is mostly from the northern hemisphere you can see in the average the annual seasons) and I find one can see that biennial oscillation even in the 18th century (where one has to say that the CRUTEM file reflects for this time just the icy Berlin weather (actually so icy that I felt rather sorry for my ancestors))

The image with the averaged station data is now here (blue is the averaged temperatures, note 19.11.2014: this is wrong: black is the averaged over a year and their means, and green is the anomalies of that -wrong closed. please see below, no guarantee it is not overly checked back):

I said above that it seems to be possible to find a relation (call it a kind of "explanation") between annual and seasonal oscillations and the QBO. It looks though that the QBO lashes are "starting" in odd years only, with a couple of exeptions, which I described here as "out of sync". Likewise the temperature anomialies oscillation, which might therefore be a result of the QBO, or at least correlated with it. If the QBO would arise from annual and seasonal oscillations only, then one would expect that roughly half of the QBO and temperature oscillations peak in even and half are peaking in odd years. But this doesn't look as if this is the case, especially not in recent years, but eventually this might be wrong.

That is I find (but thats a possibly biased observation) that there are :

16 even years from 52 possible odd years in the 1720-1820 image, 6-7 ambiguous

18 even years from 52 possible odd years in the 1820-1920 image, 9-10 ambigous

13 even years from 45 possible odd years in the 1920-2020 image, 6-8 ambiguous

So to me it looks as if there is a forcing towards odd years, but of course this "looks as if" is not overly good as a scientific argument and if one counts the ambiguous peaks together with the even ones then this is almost half/half -but still- as said I find the temperature oscillations have a forcing towards odd years and I find this currently rather visible, but if I see other data or other representations I might change my mind. Look for yourself.

I wrote

But as said the “result” aspect is not clear and if there is no half-way “sensational” aspect to it (a la a headline like: “is art helping to predict el nino’s?”, ceci n’est pas el nino! )

With "sensational" I was rather meaning "sensational" in the respective climate science circles and rather joking. This would -if it is true- at most a side notice in some science media. And yes it might be shocking to you but who knows, whats known in the climate science community about that.

Can you do some statistical tests to estimate the probability that the effect you see is just a coincidence? Can you try looking at other climate records besides HadCRUT? It’s easy to get global average temperature records from GISS (see the link).

I am not sure wether statistical test are leading much further here. Like I asked that already somewhere here on the forum: could you see a forcing towards odd years with a fourier analysis?? Eventually yes -but this sounds not so easy. In principle I could look at other temperature data, but first I spend already WAY TOO MUCH TIME on this (especially given my pension money) and secondly it seems there is not so much other quality checked temperature data available. So it seems that statistical tests may only be somewhat useful given the data set. I know that "qualitative" descriptions are often of less value than quantitative descriptions in the science literature (often for a good reason) and I am aware of the fact that one could write a science article on that claim only with more statistical evidence, but as said I spend already too much time and nerve on this and then it still might be that all this is already well-known but just hidden in some literature so that finally we would get the nice headline: "House wive of Marzahn finally learned about 100 year old scientific knowledge"(i might wear a Marzahn traditional pink jump suit for the photo next to that). I mean the temperature data is now almost 300 years old, how likely is it that something like that would go undetected?

By the way, why is the blog post about the sparse temperature data not published? Did you forget it or is there a problem, like that I "oversaw" (didn't know about) a major temperature data set?

Similar to the case of bioinformatics, the study of climate change provides a data-rich scientific domain in which cutting-edge tools from machine learning can make a major impact.

that is: where is that data-rich domain with regard to temperatures?

Comment Source:John wrote: >It may not be “sensational”, but to me an exact biennial oscillation in the Earth’s average temperature would be a shocking and important discovery. I <a href="http://forum.azimuthproject.org/discussion/1485/global-warming-and-thermodynamical-quantities/?Focus=13419#Comment_13419">wrote:</a> >Yes. And I am still looking for a reason for this. I meanwhile implemented a calculation for the station data where I calculated just the simple average anomalies over the station data (after “averaging away” the annual component (since the data is mostly from the northern hemisphere you can see in the average the annual seasons) and I find one can see that biennial oscillation even in the 18th century (where one has to say that the CRUTEM file reflects for this time just the icy Berlin weather (actually so icy that I felt rather sorry for my ancestors)) The image with the averaged station data is now here (blue is the averaged temperatures, note 19.11.2014: this is wrong: *black is the averaged over a year and their means, and green is the anomalies of that* -wrong closed. please see <a href="http://forum.azimuthproject.org/discussion/1485/global-warming-and-thermodynamical-quantities/?Focus=13573#Comment_13573">below</a>, no guarantee it is not overly checked back): ![average tempertures 1720-1820](http://www.randform.org/blog/wp-content/2014/11/1720-1820.jpg) ![average tempertures 1820-1920](http://www.randform.org/blog/wp-content/2014/11/1820-1920.jpg) ![average tempertures 1920-2020](http://www.randform.org/blog/wp-content/2014/11/1920-2020.jpg) I said above that it seems to be possible to find a relation (call it a kind of "explanation") between annual and seasonal oscillations and the QBO. It looks though that the QBO lashes are "starting" in odd years only, with a couple of exeptions, which I described <a href="http://forum.azimuthproject.org/discussion/1498/is-there-an-exact-biannual-global-temperature-oscillation/#Item_26">here</a> as "out of sync". Likewise the temperature anomialies oscillation, which might therefore be a result of the QBO, or at least correlated with it. If the QBO would arise from annual and seasonal oscillations only, then one would expect that roughly half of the QBO and temperature oscillations peak in even and half are peaking in odd years. But this doesn't look as if this is the case, especially not in recent years, but eventually this might be wrong. That is I find (but thats a possibly biased observation) that there are : 16 even years from 52 possible odd years in the 1720-1820 image, 6-7 ambiguous 18 even years from 52 possible odd years in the 1820-1920 image, 9-10 ambigous 13 even years from 45 possible odd years in the 1920-2020 image, 6-8 ambiguous So to me it looks as if there is a forcing towards odd years, but of course this "looks as if" is not overly good as a scientific argument and if one counts the ambiguous peaks together with the even ones then this is almost half/half -but still- as said I find the temperature oscillations have a forcing towards odd years and I find this currently rather visible, but if I see other data or other representations I might change my mind. Look for yourself. I wrote >But as said the “result” aspect is not clear and if there is no half-way “sensational” aspect to it (a la a headline like: “is art helping to predict el nino’s?”, ceci n’est pas el nino! ) With "sensational" I was rather meaning "sensational" in the respective climate science circles and rather joking. This would -if it is true- at most a side notice in some science media. And yes it might be shocking to you but who knows, whats known in the climate science community about that. >Can you do some statistical tests to estimate the probability that the effect you see is just a coincidence? Can you try looking at other climate records besides HadCRUT? It’s easy to get global average temperature records from GISS (see the link). I am not sure wether statistical test are leading much further here. Like I asked that already somewhere here on the forum: could you see a forcing towards odd years with a fourier analysis?? Eventually yes -but this sounds not so easy. In principle I could look at other temperature data, but first I spend already WAY TOO MUCH TIME on this (especially given my pension money) and secondly <a href="http://forum.azimuthproject.org/post/13056/">it seems there is not so much other quality checked temperature data available.</a> So it seems that statistical tests may only be somewhat useful given the data set. I know that "qualitative" descriptions are often of less value than quantitative descriptions in the science literature (often for a good reason) and I am aware of the fact that one could write a science article on that claim only with more statistical evidence, but as said I spend already too much time and nerve on this and then it still might be that all this is already well-known but just hidden in some literature so that finally we would get the nice headline: "House wive of Marzahn finally learned about 100 year old scientific knowledge"(i might wear a Marzahn traditional pink jump suit for the photo next to that). I mean the temperature data is now almost 300 years old, how likely is it that something like that would go undetected? By the way, why is the blog post about the sparse temperature data not published? Did you forget it or is there a problem, like that I "oversaw" (didn't know about) a major temperature data set? ....or <a href="http://forum.azimuthproject.org/discussion/1523/crunch-time/?Focus=13298#Comment_13298">are you eventually afraid to repel the machine fraction</a> who <a href="https://nips.cc/Conferences/2014/Program/event.php?ID=4267">writes amongst others</a>: >Similar to the case of bioinformatics, the study of climate change provides a data-rich scientific domain in which cutting-edge tools from machine learning can make a major impact. that is: where is that data-rich domain with regard to temperatures?
• Options
68.

I got the GISS data John mentioned in #67. There are 134 years. I made 67 adjacent pairs of years, and found out how often an odd year was hotter than the preceding even year. It was 41 out of 67. For a two-sided test, its just statistically significant at the 0.05 level.

fpath <- "C:/Users/Work/Desktop/giss.txt"
odd <- which((1:length(x)) %% 2 == 0)
even <- which((1:length(x)) %% 2 != 0)
# even and odd swapped cos first year is 1880 which is x[1]
y <- x[odd]-x[even]
# y[1] = T(1881)-T(1880), etc
n <- length(which(y>0))
n
pbinom(n, length(y), 0.5)

Comment Source:I got the GISS data John mentioned in #67. There are 134 years. I made 67 adjacent pairs of years, and found out how often an odd year was hotter than the preceding even year. It was 41 out of 67. For a two-sided test, its just statistically significant at the 0.05 level. ~~~~ fpath <- "C:/Users/Work/Desktop/giss.txt" x <- read.table(fpath)[,2] odd <- which((1:length(x)) %% 2 == 0) even <- which((1:length(x)) %% 2 != 0) # even and odd swapped cos first year is 1880 which is x[1] y <- x[odd]-x[even] # y[1] = T(1881)-T(1880), etc n <- length(which(y>0)) n pbinom(n, length(y), 0.5) ~~~~
• Options
69.

It seems to me that if there was a warming trend in the GISS time-series, then it would be better than even odds that "an odd year was hotter than the preceding even year"

That's global warming for you! It plays havoc with everything, even a simple statistical experiment ! :)

Comment Source:Graham, was this GISS data detrended? It seems to me that if there was a warming trend in the GISS time-series, then it would be better than even odds that *"an odd year was hotter than the preceding even year"* That's global warming for you! It plays havoc with everything, even a simple statistical experiment ! :)
• Options
70.

When I was looking at previous research on biennial cycles, most of the papers that I ran across were references to detection of two-year cycles in fish abundance based on commercial harvest numbers. The even versus odd year abundance flips every once in a while.

[1] Ishida, Yukimasa, et al. "Interannual variability in stock abundance and body size of Pacific salmon in the central Bering Sea." Progress in oceanography 55.1 (2002): 223-234.

Comment Source:When I was looking at previous research on biennial cycles, most of the papers that I ran across were references to detection of two-year cycles in fish abundance based on commercial harvest numbers. The even versus odd year abundance flips every once in a while. [1] Ishida, Yukimasa, et al. "Interannual variability in stock abundance and body size of Pacific salmon in the central Bering Sea." Progress in oceanography 55.1 (2002): 223-234.
• Options
71.

The even versus odd year abundance flips every once in a while.

Does it actually flip or does it gradually phase shift bacause the period is not exacly 2 years?

Comment Source:> The even versus odd year abundance flips every once in a while. Does it actually flip or does it gradually phase shift bacause the period is not exacly 2 years?
• Options
72.

The salmon catches are actually locked in to an odd year for quite a span [1]. In figure below, the red squares are odd years and the blue is even.

Do the binomial statistics on North America and it is hard to refute. But we must remember that biology is strange.

[1] Irvine, J. R., et al. "Increasing Dominance of Odd-Year Returning Pink Salmon." Transactions of the American Fisheries Society 143.4 (2014): 939-956. http://www.tandfonline.com/doi/full/10.1080/00028487.2014.889747

Comment Source:The salmon catches are actually locked in to an odd year for quite a span [1]. In figure below, the red squares are odd years and the blue is even. ![salmon](http://www.tandfonline.com/na101/home/literatum/publisher/tandf/journals/content/utaf20/2014/utaf20.v143.i04/00028487.2014.889747/20140711/images/medium/utaf_a_889747_f0001_oc.jpg) Do the binomial statistics on North America and it is hard to refute. But we must remember that biology is strange. [1] Irvine, J. R., et al. "Increasing Dominance of Odd-Year Returning Pink Salmon." Transactions of the American Fisheries Society 143.4 (2014): 939-956. <http://www.tandfonline.com/doi/full/10.1080/00028487.2014.889747>
• Options
73.

I ran a spectral analysis on the the HadCRUT4 data. The most prominent spike is for a period of 6 months. This probably is due to earths's axis tilt. However most of the spectral power is in the low frequency range reflecting the long range trends. Here is the notebook, source and script.

Comment Source:I ran a spectral analysis on the the HadCRUT4 data. The most prominent spike is for a period of 6 months. This probably is due to earths's axis tilt. However most of the spectral power is in the low frequency range reflecting the long range trends. Here is the [notebook](https://5619417f7fb3a489ed01c7f329cbd1e9b70a10d6-www.googledrive.com/host/0B4cyIPgV_VxrX0lxSUxHU2VLN28/biannual.html), [source](https://5619417f7fb3a489ed01c7f329cbd1e9b70a10d6-www.googledrive.com/host/0B4cyIPgV_VxrX0lxSUxHU2VLN28/biannual.ipynb) and [script](https://5619417f7fb3a489ed01c7f329cbd1e9b70a10d6-www.googledrive.com/host/0B4cyIPgV_VxrX0lxSUxHU2VLN28/biannual.py).
• Options
74.

I got the GISS data John mentioned in #67. There are 134 years...

That’s global warming for you! It plays havoc with everything, even a simple statistical experiment ! :)

Thanks Graham and Paul for discussing this. I think Paul's argument has to be taken into consideration, so a question could be how often the difference of an odd year minus the consecutive even year is bigger, if there would still be a trend - even with warming- then this would be something. But then the GISS data is the mean over a whole year (at least in this file, which as I understood you took) and I am not so sure wether comparing the annual averages/mean would give much of a difference anyways. Finally a sharp peak and a broad valley might give the same annual mean.

And while thinking about all that and especially after seeing that you took actually an interest in that what I wrote I detected that I was actually wrong when I wrote:

blue is the averaged temperatures, black is the averaged over a year and their means, and green is the anomalies of that

somehow I obviously didn't remember correctly what I had drawn there, when writing the comment :0 - I am sorry. I am now looking into the code, while writing this comment :) :

so let me say what the curves are exactly: blue are the monthly temperatures as given from the stations, averaged over all stations, i.e. sum over all valid station values for that month divided by the number of stations, who provide that valid value. Call that array of monthly averages (or means) "monthlyaverage". So that was correct. As you can see at the annual oscillation the northern hemisphere clearly dominates the average.

the jagged "black" (its actually dark grey) curve is however already an anomaly curve that is I summed first the monthlyaverage for a given month over all years, divided by the number of years and then I deduced that mean from the respective monthlyaverage that array of monthly anomalies is called monthlyanomall. The "black curve" through it is the mean of that.

for(m=0;m<12;m++){

dummy=0;


monthlyanomall[m]=0;

for(y=startyear;y<endyear-yearoffset;y++){

monthlyanomall[m]=monthlyanomall[m]+monthlyaverage[y][m];

dummy=dummy+1;

}

monthlyanom[m]= (1/dummy)*(monthlyanomall[m]);

}


/calculation of monthly/

var monthly = new Array();

var idx = 0;

for(k=startyear;k<endyear;k++){

for(m=0;m<12;m++){

monthly[idx]=monthlyaverage[k][m]-monthlyanom[m];

idx++;

}

}


/calculation of monthly mean/

var meanmonthlyaverage = new Array();

for(j=5;j<monthly.length-7;j++){

meanmonthlyaverage[j-5]=0;

for(i=-5;i<7;i++){

meanmonthlyaverage[j-5]=meanmonthlyaverage[j-5]+1/12.*monthly[j+i];

}

}


The green curve uses the function diff12 (I don't know why it is called diff12, but apparently climate scientists use this term), it is the annual change of the monthly value:

/calculation of diff12/

function diff12(array){

var datafilterdiff12 = new Array();

for(j=12;j<array.length;j++){

    datafilterdiff12[j-12]=array[j]-array[j-12];

}


return datafilterdiff12;

}

so the green curve is diff12(meanmonthlyaverage) and my "per hand counting" of peaks was done with repect to the green curve, but one could do this also for the anomaly curve. There it is eventually not as good as visible.

Daniel wrote:

I ran a spectral analysis on the the HadCRUT4 data. The most prominent spike is for a period of 6 months. This probably is due to earths’s axis tilt. However most of the spectral power is in the low frequency range reflecting the long range trends. Here is the notebook, source and script.

Does it actually flip or does it gradually phase shift bacause the period is not exacly 2 years?

As far as I understood that the gradual phase shift is what is usually assumed to happen at least for the QBO it seems usually said that the average period is about 27 months, like here it is written: > Die Periode dieser Schwingung schwankt zwischen 22 und 34 Monaten. Im Mittel beträgt sie 27 Monate. (The period of this oscillation is between 22 and 34 months. On average it is 27 months)

To me it looks though as if the temperature oscillation is "forced back" to peak at odd years (and at the moment it looks to me that this holds also for the QBO) and as said I might be wrong. It looked to me so visible that I have already been starting to look for possible reasons, while still investigating wether this force back is true at all.

So this is shaky grounds. Moreover I actually do not know how to investigate such a "force back" via fourier transforms, wavelets or similar. I think it should be detectable in some kind of resolution analysis but I am not an expert for that. I am not even sure wether you could eventually depict it via looking for special forms of the fourier spectrum.... So unfortunately I can't tell wether one could see such a feature in your spectral analysis. Do you know? But as said this is shaky grounds -it might not be worth the effort. I don't know how much surplus time you have for such endeavors.

Pauls remarks:

The salmon catches are actually locked in to an odd year for quite a span [1]. In figure below, the red squares are odd years and the blue is even.

are though interesting in this context.

Another problem is that I was actually more interested in this methane problem and that biannual temperature oscillations appeared so to say as a byproduct of those investigations and I have not really much of an idea yet how important this forcing is in this context. I think if there is a forcing then it would be important to understand the reasons for it but thats still quite a gut feeling.

Comment Source:>I got the GISS data John mentioned in #67. There are 134 years... >That’s global warming for you! It plays havoc with everything, even a simple statistical experiment ! :) Thanks Graham and Paul for discussing this. I think Paul's argument has to be taken into consideration, so a question could be how often the difference of an odd year minus the consecutive even year is bigger, if there would still be a trend - even with warming- then this would be something. But then the GISS data is the mean over a whole year (at least in this <a href="http://data.giss.nasa.gov/gistemp/graphs_v3/Fig.A2.txt">file</a>, which as I understood you took) and I am not so sure wether comparing the annual averages/mean would give much of a difference anyways. Finally a sharp peak and a broad valley might give the same annual mean. And while thinking about all that and especially after seeing that you took actually an interest in that what I wrote I detected that I was actually wrong when I <a href="http://forum.azimuthproject.org/discussion/1485/global-warming-and-thermodynamical-quantities/?Focus=13564#Comment_13564">wrote: </a> >blue is the averaged temperatures, black is the averaged over a year and their means, and green is the anomalies of that somehow I obviously didn't remember correctly what I had drawn there, when writing the comment :0 - I am sorry. I am now looking into the code, while writing this comment :) : so let me say what the curves are exactly: blue are the monthly temperatures as given from the stations, averaged over all stations, i.e. sum over all valid station values for that month divided by the number of stations, who provide that valid value. Call that array of monthly averages (or means) "monthlyaverage". So that was correct. As you can see at the annual oscillation the northern hemisphere clearly dominates the average. the jagged "black" (its actually dark grey) curve is however already an anomaly curve that is I summed first the monthlyaverage for a given month over all years, divided by the number of years and then I deduced that mean from the respective monthlyaverage that array of monthly anomalies is called monthlyanomall. The "black curve" through it is the mean of that. for(m=0;m<12;m++){ dummy=0; monthlyanomall[m]=0; for(y=startyear;y<endyear-yearoffset;y++){ monthlyanomall[m]=monthlyanomall[m]+monthlyaverage[y][m]; dummy=dummy+1; } monthlyanom[m]= (1/dummy)*(monthlyanomall[m]); } /*calculation of monthly*/ var monthly = new Array(); var idx = 0; for(k=startyear;k<endyear;k++){ for(m=0;m<12;m++){ monthly[idx]=monthlyaverage[k][m]-monthlyanom[m]; idx++; } } /*calculation of monthly mean*/ var meanmonthlyaverage = new Array(); for(j=5;j<monthly.length-7;j++){ meanmonthlyaverage[j-5]=0; for(i=-5;i<7;i++){ meanmonthlyaverage[j-5]=meanmonthlyaverage[j-5]+1/12.*monthly[j+i]; } } The green curve uses the function diff12 (I don't know why it is called diff12, but apparently <a href="http://www.azimuthproject.org/azimuth/show/Does+global+warming+lag+or+lead+a+rise+in+greenhouse+gas+concentration%3F">climate scientists use this term</a>), it is the annual change of the monthly value: /*calculation of diff12*/ function diff12(array){ var datafilterdiff12 = new Array(); for(j=12;j<array.length;j++){ datafilterdiff12[j-12]=array[j]-array[j-12]; } return datafilterdiff12; } so the green curve is diff12(meanmonthlyaverage) and my "per hand counting" of peaks was done with repect to the green curve, but one could do this also for the anomaly curve. There it is eventually not as good as visible. Daniel wrote: >I ran a spectral analysis on the the HadCRUT4 data. The most prominent spike is for a period of 6 months. This probably is due to earths’s axis tilt. However most of the spectral power is in the low frequency range reflecting the long range trends. Here is the notebook, source and script. >Does it actually flip or does it gradually phase shift bacause the period is not exacly 2 years? As far as I understood that the gradual phase shift is what is usually assumed to happen at least for the QBO it seems usually said that the average period is about 27 months, like <a href="http://www.geo.fu-berlin.de/met/ag/strat/produkte/qbo">here</a> it is written: > Die Periode dieser Schwingung schwankt zwischen 22 und 34 Monaten. Im Mittel beträgt sie 27 Monate. (The period of this oscillation is between 22 and 34 months. On average it is 27 months) To me it looks though as if the temperature oscillation is "forced back" to peak at odd years (and at the moment it looks to me that this holds also for the QBO) and as said I might be wrong. It looked to me so visible that I have already been starting to look for possible reasons, while still investigating wether this force back is true at all. So this is shaky grounds. Moreover I actually do not know how to investigate such a "force back" via fourier transforms, wavelets or similar. I think it should be detectable in some kind of resolution analysis but I am not an expert for that. I am not even sure wether you could eventually depict it via looking for special forms of the fourier spectrum.... So unfortunately I can't tell wether one could see such a feature in your spectral analysis. Do you know? But as said this is shaky grounds -it might not be worth the effort. I don't know how much surplus time you have for such endeavors. Pauls remarks: >The salmon catches are actually locked in to an odd year for quite a span [1]. In figure below, the red squares are odd years and the blue is even. are though interesting in this context. Another problem is that I was actually more interested in this <a href="http://www.azimuthproject.org/azimuth/show/Does+global+warming+lag+or+lead+a+rise+in+greenhouse+gas+concentration%3F">methane problem</a> and that biannual temperature oscillations appeared so to say as a byproduct of those investigations and I have not really much of an idea yet how important this forcing is in this context. I think if there is a forcing then it would be important to understand the reasons for it but thats still quite a gut feeling.
• Options
75.

I think Paul at #70 is right. A quick check using odd years before even years makes the bias go away. I guess there's still a slight bias overall, but not an interesting one. Yes, Nad, I used http://data.giss.nasa.gov/gistemp/graphs_v3/Fig.A2.txt.

Comment Source:I think Paul at #70 is right. A quick check using odd years before even years makes the bias go away. I guess there's still a slight bias overall, but not an interesting one. Yes, Nad, I used <a href="http://data.giss.nasa.gov/gistemp/graphs_v3/Fig.A2.txt">http://data.giss.nasa.gov/gistemp/graphs_v3/Fig.A2.txt</a>.
• Options
76.
edited November 2014

Graham wrote at 76:

I guess there’s still a slight bias overall, but not an interesting one.

OK Upon popular request I did a plot like the salmon plot blue are the odd years starting in 1711 and red are the even years starting in 1712 in order to avoid the global warming trend, below are the differences:

I find until around 1800 (I estimate this, since I don't feel like drawing the year numbers, this is all handmade in javascript...) the difference line stays clearly below the null line, i.e. the blue line on top of the red line, then things are not so clear anymore and it looks almost as if the opposite is true and then around the 1950s (where the steep jump is)** the difference line is again rather clearly below the null line. One could sum all up ("integrate") but for me thats more than enough already.

this is not 1920s but then the numbers are only estimates so just in case I would like to keep for reference this link here

Comment Source:Graham wrote at 76: >I guess there’s still a slight bias overall, but not an interesting one. OK Upon popular request I did a plot like the salmon plot blue are the odd years starting in 1711 and red are the even years starting in 1712 in order to avoid the global warming trend, below are the differences: ![differences](http://www.randform.org/blog/wp-content/2014/11/difference1.jpg) ![differences](http://www.randform.org/blog/wp-content/2014/11/difference2.jpg) I find until around 1800 (I estimate this, since I don't feel like drawing the year numbers, this is all handmade in javascript...) the difference line stays clearly below the null line, i.e. the blue line on top of the red line, then things are not so clear anymore and it looks almost as if the opposite is true and then around the 1950s (where the steep jump is)** the difference line is again rather clearly below the null line. One could sum all up ("integrate") but for me thats more than enough already. this is not 1920s but then the numbers are only estimates so just in case I would like to keep for reference this link <a href="http://forum.azimuthproject.org/discussion/1504/symbolic-regression-machine-learning-and-enso-time-series/?Focus=13141#Comment_13141">here</a>
• Options
77.

This is what I am thinking is happening in terms of a causal trail from QBO to ENSO to GISS.

1. The QBO has a clear 2.333 year period over the last 60+ years of stratospheric radiosonde measurements. The 2.333 year period leads to a 7 year beat where the QBO can align with an annual signal. i.e. 2.333*3 = 7 years.

2. I showed this a few days ago in another thread, but one can see the 7 year reinforcement in the ENSO signal, i.e. the following SOI autocorrelation and cross-correlation. The upper arrows are on 7-year correlation boundaries. This may have some relevance in terms of the appearance of El Nino events at the same relative time of a calendar year. Since it is on a 7-year period, then the seasonal alignment may propagate that way. This is a new way to think about the synchronization.

3. And then of course, the El Nino propagates to GISS as El Nino is a significant contributor to the global temperature variability.

This is all reinforced by the sloshing model that I recently finished up which applied QBO as a major forcing for ENSO: http://contextearth.com/2014/11/18/paper-on-sloshing-model-for-enso/. I should probably do an autocorrelation for the model but I have no doubt that the 7-year alignment would show up there as well. Take a look at the wavelet scalogram.

Comment Source:This is what I am thinking is happening in terms of a causal trail from QBO to ENSO to GISS. 1. The QBO has a clear 2.333 year period over the last 60+ years of stratospheric radiosonde measurements. The 2.333 year period leads to a 7 year beat where the QBO can align with an annual signal. i.e. 2.333*3 = 7 years. 2. I showed this a few days ago in [another thread](http://forum.azimuthproject.org/discussion/1523/crunch-time/?Focus=13530#Comment_13530), but one can see the 7 year reinforcement in the ENSO signal, i.e. the following SOI autocorrelation and cross-correlation. The upper arrows are on 7-year correlation boundaries. This may have some relevance in terms of the appearance of El Nino events at the same relative time of a calendar year. Since it is on a 7-year period, then the seasonal alignment may propagate that way. This is a new way to think about the synchronization. 3. And then of course, the El Nino propagates to GISS as El Nino is a significant contributor to the global temperature variability. ![xcorr](http://imagizer.imageshack.us/a/img537/4074/jZCphA.gif) This is all reinforced by the sloshing model that I recently finished up which applied QBO as a major forcing for ENSO: <http://contextearth.com/2014/11/18/paper-on-sloshing-model-for-enso/>. I should probably do an autocorrelation for the model but I have no doubt that the 7-year alignment would show up there as well. Take a look at the wavelet scalogram.
• Options
78.

This ENSO deal is getting ridiculously easy. The folks at NOAA have another ENSO index called BEST, see http://www.esrl.noaa.gov/psd/people/cathy.smith/best/enso.ts.1mn.txt This is an interesting index because it combines SOI and SST, getting the benefits of SOI via a long history and then getting the noise reduction of the SST data.

After doing machine learning on it with the sloshing model, we can pick out disturbances such as Pinatubo very easily. See the large spike at 1992, 122 years after 1870 on the following residual chart:

The analysis is fully documented here : http://contextearth.com/2014/11/18/paper-on-sloshing-model-for-enso/

The sloshing model is working so well that it is now able to pick out isolated disturbances that aren't part of ENSO.

Comment Source:This ENSO deal is getting ridiculously easy. The folks at NOAA have another ENSO index called BEST, see <http://www.esrl.noaa.gov/psd/people/cathy.smith/best/enso.ts.1mn.txt> This is an interesting index because it combines SOI and SST, getting the benefits of SOI via a long history and then getting the noise reduction of the SST data. After doing machine learning on it with the sloshing model, we can pick out disturbances such as Pinatubo very easily. See the large spike at 1992, 122 years after 1870 on the following residual chart: ![BEST](http://imageshack.com/a/img673/9030/lKWH1V.gif) The analysis is fully documented here : <http://contextearth.com/2014/11/18/paper-on-sloshing-model-for-enso/> The sloshing model is working so well that it is now able to pick out isolated disturbances that aren't part of ENSO.
• Options
79.
edited November 2014

Ok here for my 15 min Azimuthforum fame, I made the integration and put the curves now behind number bars:

So as you can see the integral stays clearly below zero (it is actually even scaled by 0.01), but indeed between 1840s or 1860s and 1920s the temperatures seem to flip on average even and odd years at least for that station data and for this case of directly comparing the months in cosecutive years. That is the integral curve ascends. This could thus be correlated with the alledged phase reverse of the Chandler wobble, however I don't see anything in 2005. But this might eventually be due to the rottening of the temp file. No guarantee for correctness I just hacked that in.

The analysis is fully documented here : http://contextearth.com/2014/11/18/paper-on-sloshing-model-for-enso/

Paul just as for the CSALT model I am impressed about how the curves fit, moreover I could imagine that the Chandler wobble, total solar insolation and QBO play a role in ENSO model (in fact the above shows that e.g. the Chandler wobble might eventually be even rather visible in the temperatures), I could even imagine that some kind of Floquet theory type equation might be used in this context, but I don't see how this all goes together in your sloshing model. You don't give much details in your article (like strength of the various influences etc.) and motivations for your choices (at least not sufficient for me). Moreover maybe even the (Mathematica?) code is somewhere and thus someone (i.e. someone with that program, i.e. in particular not me) could e.g. test how the equation change with a small disturbation. But even if all those tests would be there I would still wonder a lot if your sloshing model would predict the ENSO also in the upcoming years.

Comment Source:Ok here for my 15 min Azimuthforum fame, I made the integration and put the curves now behind number bars: ![integrateddifference](http://www.randform.org/blog/wp-content/2014/11/differenceInt1.jpg) ![integrateddifference](http://www.randform.org/blog/wp-content/2014/11/differenceInt2.jpg) So as you can see the integral stays clearly below zero (it is actually even scaled by 0.01), but indeed between 1840s or 1860s and 1920s the temperatures seem to flip on average even and odd years at least for that station data and for this case of directly comparing the months in cosecutive years. That is the integral curve ascends. This could thus be correlated with the <a href="http://forum.azimuthproject.org/discussion/1504/symbolic-regression-machine-learning-and-enso-time-series/?Focus=13141#Comment_13141">alledged phase reverse</a> of the <a href="http://en.wikipedia.org/wiki/Chandler_wobble">Chandler wobble</a>, however I don't see anything in 2005. But this might eventually be due to the rottening of the temp file. No guarantee for correctness I just hacked that in. >The analysis is fully documented here : http://contextearth.com/2014/11/18/paper-on-sloshing-model-for-enso/ Paul just as for the CSALT model I am impressed about how the curves fit, moreover I could imagine that the Chandler wobble, total solar insolation and QBO play a role in ENSO model (in fact the above shows that e.g. the Chandler wobble might eventually be even rather visible in the temperatures), I could even imagine that some kind of Floquet theory type equation might be used in this context, but I don't see how this all goes together in your sloshing model. You don't give much details in your article (like strength of the various influences etc.) and motivations for your choices (at least not sufficient for me). Moreover maybe even the (Mathematica?) code is somewhere and thus someone (i.e. someone with that program, i.e. in particular not me) could e.g. test how the equation change with a small disturbation. But even if all those tests would be there I would still wonder a lot if your sloshing model would predict the ENSO also in the upcoming years.
• Options
80.

Where do you think the two major volcanic eruptions are in this BEST ENSO time series?

My machine learning buddy, Eureqa, can find them. Can you?

The point is that ENSO is largely a stationary oscillation, and the only problem is in decoding it. Glitches such as volcanic eruptions are apparently the only events that will disturb the underlying pattern.

Comment Source:Where do you think the two major volcanic eruptions are in this BEST ENSO time series? ![BEST](http://imageshack.com/a/img674/4940/xEHRI8.gif) My machine learning buddy, Eureqa, can find them. Can you? The point is that ENSO is largely a stationary oscillation, and the only problem is in decoding it. Glitches such as volcanic eruptions are apparently the only events that will disturb the underlying pattern.
• Options
81.

Here another link to studies on plant growth and CO2: Dirty Pool: Soil's large carbon stores could be freed by increased CO2, plant growth

As trees and other vegetation flourish in a carbon dioxide-rich future, their roots could stimulate microbial activity in soil that in turn accelerates the decomposition of soil carbon and its release into the atmosphere as carbon dioxide, the researchers found.

This effect counters current key projections regarding Earth's future carbon cycle, particularly that greater plant growth could offset carbon dioxide emissions as flora take up more of the gas, said first author Benjamin Sulman, who conducted the modeling work as a postdoctoral researcher at the Princeton Environmental Institute.

....the article doesn't say anything on methanosarcina and concrete experiments.

Comment Source:Here another link to studies on plant growth and CO2: <a href="http://phys.org/news/2014-12-dirty-pool-soil-large-carbon.html">Dirty Pool: Soil's large carbon stores could be freed by increased CO2, plant growth</a> >As trees and other vegetation flourish in a carbon dioxide-rich future, their roots could stimulate microbial activity in soil that in turn accelerates the decomposition of soil carbon and its release into the atmosphere as carbon dioxide, the researchers found. >This effect counters current key projections regarding Earth's future carbon cycle, particularly that greater plant growth could offset carbon dioxide emissions as flora take up more of the gas, said first author Benjamin Sulman, who conducted the modeling work as a postdoctoral researcher at the Princeton Environmental Institute. ....the article doesn't say anything on <a href="http://www.azimuthproject.org/azimuth/show/Closed+artificial+ecosystem">methanosarcina and concrete experiments</a>.
• Options
82.

I tell you why, check these out:

How Plants Survived Chernobyl

Discovery of the secrets that enable plants near Chernobyl to shrug off radiation

Botanical molecular structures are quite dynamic and via turning DNA switches on and off and mutations on and off, software-IF-THEN-ELSE-like. They do that to deal with arid climate changes or even radiation. I am sure there are such gene switches for CO2 rich soil and air. Specially since in the deep past of botanical life on this planet, the atmosphere was toxic gas/soil rich, and planets dealt with that.

So papers as such appear drawing child-like diagrams from 1960s books and completely ignoring the fantastic molecular genetics of the botanical molecular structures...

I could not emphasize how dynamic the botanical molecular structures are, our imaginations are crude given the results from the flourishing of Chernobyl plant and animal life.

….the article doesn’t say anything on methanosarcina and concrete experiments.

But an insider revealed a woman had been excluded from the running because her presence could cause awkward sexual tension which could ruin the experiment.

These non-awkward non-sexual men run our science and educational institutions! heh heh heh

Comment Source:Hello Nad this article **dirty pool..** is a good candidate for crackpot science. I tell you why, check these out: [How Plants Survived Chernobyl](http://news.sciencemag.org/2009/05/how-plants-survived-chernobyl) [Discovery of the secrets that enable plants near Chernobyl to shrug off radiation](http://www.acs.org/content/acs/en/pressroom/presspacs/2010/acs-presspac-december-8-2010/discovery-of-the-secrets-that-enable-plants-near-chernobyl-to-shrug-off-radiatio.html) Botanical molecular structures are quite dynamic and via turning DNA switches on and off and mutations on and off, software-IF-THEN-ELSE-like. They do that to deal with arid climate changes or even radiation. I am sure there are such gene switches for CO2 rich soil and air. Specially since in the deep past of botanical life on this planet, the atmosphere was toxic gas/soil rich, and planets dealt with that. So papers as such appear drawing child-like diagrams from 1960s books and completely ignoring the fantastic molecular genetics of the botanical molecular structures... I could not emphasize how dynamic the botanical molecular structures are, our imaginations are crude given the results from the flourishing of Chernobyl plant and animal life. >….the article doesn’t say anything on methanosarcina and concrete experiments. >But an insider revealed a woman had been excluded from the running because her presence could cause awkward sexual tension which could ruin the experiment. These **non-awkward non-sexual men** run our science and educational institutions! heh heh heh
• Options
83.
edited December 2014

who knows - its behind a pay wall. three authors are though from princeton, which is supposed to be a good university.

I am sure there are such gene switches for CO2 rich soil and air. Specially since in the deep past of botanical life on this planet, the atmosphere was toxic gas/soil rich, and planets dealt with that.

i am sure that there are some species who can deal with increased CO2 levels, but not all, sudden big changes usually let to mass extinctions.

I tell you why, check these out:

How Plants Survived Chernobyl

Discovery of the secrets that enable plants near Chernobyl to shrug off radiation

Funny - the chemical society omitted to mention some "minor aspects" like the decreased size of the beans, but write:

Scientists are reporting discovery of the biological secrets that enable plants growing near the Chernobyl Nuclear Power Plant to adapt and flourish in highly radioactive soil — legacy of the 1986 nuclear disaster in the Ukraine.

Yoohoooo - everything is flourishing nicely!!!! From the other article:

"The radiation zone beans looked odd even before the protein analysis. They weighed half as much and took up water more slowly than their low-radiation counterparts. And on a molecular level, the beans were even stranger, the researchers report in the June Journal of Proteome Research. When compared with normal plants, beans from the high-radiation area had three times more cysteine synthase, a protein known to protect plants by binding heavy metals. "

You actually don't need to got to Chernobyl for deformations, like in Switzerland the artist Cornelia Honegger documented deformed insects in the vicinity of nuclear power plants

These non-awkward non-sexual men run our science and educational institutions! heh heh heh

well -similar to the smaller beans- males being submitted to higher radioactivity levels may eventually grow smaller penisses - problem solved. :)

but bad jokes aside - tensions can arise from all sorts of things and often they arise from competitions and sometimes competitions are for a mate but not always. In particular in this particular case it seems they didn't check for tensions which may arise from homosexual partners, but maybe that part was omitted by the media. I was once invited to an all male gay party - amongst others also for reducing the sexual tensions, as I was told.

Comment Source:>Hello Nad this article dirty pool.. is a good candidate for crackpot science. who knows - its behind a pay wall. three authors are though from princeton, which is supposed to be a good university. >I am sure there are such gene switches for CO2 rich soil and air. Specially since in the deep past of botanical life on this planet, the atmosphere was toxic gas/soil rich, and planets dealt with that. i am sure that there are some species who can deal with increased CO2 levels, but not all, sudden big changes usually let to mass extinctions. >I tell you why, check these out: >How Plants Survived Chernobyl >Discovery of the secrets that enable plants near Chernobyl to shrug off radiation Funny - the chemical society omitted to mention some "minor aspects" like the decreased size of the beans, but write: >Scientists are reporting discovery of the biological secrets that enable plants growing near the Chernobyl Nuclear Power Plant to adapt and flourish in highly radioactive soil — legacy of the 1986 nuclear disaster in the Ukraine. Yoohoooo - everything is flourishing nicely!!!! From the other article: >"The radiation zone beans looked odd even before the protein analysis. They weighed half as much and took up water more slowly than their low-radiation counterparts. And on a molecular level, the beans were even stranger, the researchers report in the June Journal of Proteome Research. When compared with normal plants, beans from the high-radiation area had three times more cysteine synthase, a protein known to protect plants by binding heavy metals. " You actually don't need to got to Chernobyl for deformations, like in Switzerland the artist <a href="http://www.randform.org/blog/?p=5706">Cornelia Honegger documented deformed insects in the vicinity of nuclear power plants</a> >These non-awkward non-sexual men run our science and educational institutions! heh heh heh well -similar to the smaller beans- males being submitted to higher radioactivity levels may eventually grow smaller penisses - problem solved. :) but bad jokes aside - tensions can arise from all sorts of things and often they arise from competitions and sometimes competitions are for a mate but not always. In particular in this particular case it seems they didn't check for tensions which may arise from homosexual partners, but maybe that part was omitted by the media. I was once invited to an all male gay party - amongst others also for reducing the sexual tensions, as I was told.
• Options
84.

i am sure that there are some species who can deal with increased CO2 levels, but not all,

How do you know that not all could keep up with CO2 increases?

sudden big changes usually let to mass extinctions.

How do you know that? Please provide the research, this is again under the consensual science heading in my book. By research I mean an actual way of showing or proving the hypothesis as opposed to conjectures of a few men of certain backgrounds.

I was once invited to an all male gay party - amongst others also for reducing the sexual tensions, as I was told.

I had a similar experience when I was invite to the local Bee Keepers Association!

three authors are though from princeton, which is supposed to be a good university.

The next Galileo will be prosecuted by guys like these authors.

Comment Source:>i am sure that there are some species who can deal with increased CO2 levels, but not all, How do you know that not all could keep up with CO2 increases? >sudden big changes usually let to mass extinctions. How do you know that? Please provide the research, this is again under the **consensual science** heading in my book. By research I mean an actual way of showing or proving the hypothesis as opposed to conjectures of a few men of certain backgrounds. >I was once invited to an all male gay party - amongst others also for reducing the sexual tensions, as I was told. I had a similar experience when I was invite to the local Bee Keepers Association! >three authors are though from princeton, which is supposed to be a good university. The next Galileo will be prosecuted by guys like these authors.
• Options
85.

I had a similar experience when I was invite to the local Bee Keepers Association!

Are all irish bee keepers on average lesbian or just the one in Dungaravon (sorry forgot the correct spelling of your town)?

Comment Source:>I had a similar experience when I was invite to the local Bee Keepers Association! Are all irish bee keepers on average lesbian or just the one in Dungaravon (sorry forgot the correct spelling of your town)?
• Options
86.

Are all irish bee keepers on average lesbian or just the one in Dungaravon (sorry forgot the correct spelling of your town)?

heh heh heh no. BUT It was a funny story I tell you later on for laughs.

Comment Source:>Are all irish bee keepers on average lesbian or just the one in Dungaravon (sorry forgot the correct spelling of your town)? heh heh heh no. BUT It was a funny story I tell you later on for laughs.
• Options
87.
Comment Source:Nad check this out [Good News on Forests and Carbon Dioxide ](http://science.nasa.gov/science-news/science-at-nasa/2014/31dec_forests/)
• Options
88.
edited January 2015

For about 25 years, most computer climate models have been showing that mid-latitude forests in the Northern Hemisphere absorb more carbon than tropical forests. That result was initially based on the then-current understanding of global air flows and limited data suggesting that deforestation was causing tropical forests to release more carbon dioxide than they were absorbing.

In the mid-2000s, Stephens used measurements of carbon dioxide made from aircraft to show that many climate models were not correctly representing flows of carbon above ground level. Models that matched the aircraft measurements better showed more carbon absorption in the tropical forests. However, there were still not enough global data sets to validate the idea of a large tropical-forest absorption.However, there were still not enough global data sets to validate the idea of a large tropical-forest absorption. Schimel said that their new study took advantage of a great deal of work other scientists have done since Stephens' paper to pull together national and regional data of various kinds into robust, global data sets.

Schimel noted that their paper reconciles results at every scale from the pores of a single leaf, where photosynthesis takes place, to the whole Earth, as air moves carbon dioxide around the globe.

Thanks Dara for pointing out the link and yes the article doesn't say anything about the root problem, which were mentioned in the previous article you cited. On a first glance it looks as if Boulder forgot about the root problem, but maybe it is just not mentioned in this science news report. If this isn't a reporting problem then if Boulder didn't take the roots into account they would/should at least have said something like that their result has to be taken with a grain of salt or that the root problem is not important because ...etc. and because they didnt it seems they havent noticed that there might be a problem.

I don't trust those computer models - especially after I had seen how some climate researchers dealt with temperature data.

They should do more concrete measurements and things like the (quasi-) closed ecosystem experiments in my point of view.

That is one can try to do models on shaky grounds, but it should be made clearer that they are based on very shaky grounds and with very shaky I mean more than a sentence like:

"However, there were still not enough global data sets to validate the idea of a large tropical-forest absorption."

Or in other words:

A new NASA-led study shows that tropical forests may be absorbing far more carbon dioxide than many scientists thought, in response to rising atmospheric levels of the greenhouse gas.

By looking at that root and data problem it may be that that "may be" is already quite an exageration and especially the headline "Good News on Forests and Carbon Dioxide" definitely appears to be exagerated, but I am aware that a lot of people prefer to read "good news" (only), so the headline and the hookline may have been chosen for this reason. I just fear that it this what sticks in peoples heads.

Comment Source:From the <a href="http://science.nasa.gov/science-news/science-at-nasa/2014/31dec_forests/">Colorado Boulder fraction</a>: > For about 25 years, most computer climate models have been showing that mid-latitude forests in the Northern Hemisphere absorb more carbon than tropical forests. That result was initially based on the then-current understanding of global air flows and limited data suggesting that deforestation was causing tropical forests to release more carbon dioxide than they were absorbing. >In the mid-2000s, Stephens used measurements of carbon dioxide made from aircraft to show that many climate models were not correctly representing flows of carbon above ground level. Models that matched the aircraft measurements better showed more carbon absorption in the tropical forests. However, there were still not enough global data sets to validate the idea of a large tropical-forest absorption.However, there were still not enough global data sets to validate the idea of a large tropical-forest absorption. Schimel said that their new study took advantage of a great deal of work other scientists have done since Stephens' paper to pull together national and regional data of various kinds into robust, global data sets. >Schimel noted that their paper reconciles results at every scale from the pores of a single leaf, where photosynthesis takes place, to the whole Earth, as air moves carbon dioxide around the globe. Thanks Dara for pointing out the link and yes the article doesn't say anything about the <a href="http://forum.azimuthproject.org/discussion/1485/global-warming-and-thermodynamical-quantities/?Focus=13921#Comment_13921">root problem</a>, which were mentioned in the previous article you cited. On a first glance it looks as if Boulder forgot about the root problem, but maybe it is just not mentioned in this science news report. If this isn't a reporting problem then if Boulder didn't take the roots into account they would/should at least have said something like that their result has to be taken with a grain of salt or that the root problem is not important because ...etc. and because they didnt it seems they havent noticed that there might be a problem. I don't trust those computer models - especially after I had seen how some climate researchers dealt with temperature data. They should do more concrete measurements and things like the (quasi-) closed ecosystem experiments in my point of view. That is one can try to do models on shaky grounds, but it should be made clearer that they are based on very shaky grounds and with very shaky I mean more than a sentence like: >"However, there were still not enough global data sets to validate the idea of a large tropical-forest absorption." Or in other words: >A new NASA-led study shows that tropical forests may be absorbing far more carbon dioxide than many scientists thought, in response to rising atmospheric levels of the greenhouse gas. By looking at that root and data problem it may be that that "may be" is already quite an exageration and especially the headline "Good News on Forests and Carbon Dioxide" definitely appears to be exagerated, but I am aware that a lot of people prefer to read "good news" (only), so the headline and the hookline may have been chosen for this reason. I just fear that it this what sticks in peoples heads.
• Options
89.
edited January 2015

I don’t trust those computer models - especially after I had seen how some climate researchers dealt with temperature data.

We have no other tool at this point of time other than computer modelling to deal with planetary climate data. And at the same time the climate researcher are not computer scientists, from what I have read from them I consider them as astrologists at best. Their dealings with computer models are questionable and results show.

I prefer NASA media and papers, they are on taxpayers funding and if I email them today they get back to me tomorrow, the other elite scientists in those papers we all discussed here, will not reply to any critic of their works.

My bet is on NASA.

I get daily emails from their scientists and even when they goof and err they email us and let us know.

Comment Source:>I don’t trust those computer models - especially after I had seen how some climate researchers dealt with temperature data. We have no other tool at this point of time other than computer modelling to deal with planetary climate data. And at the same time the climate researcher are not computer scientists, from what I have read from them I consider them as astrologists at best. Their dealings with computer models are questionable and results show. I prefer NASA media and papers, they are on taxpayers funding and if I email them today they get back to me tomorrow, the other elite scientists in those papers we all discussed here, will not reply to any critic of their works. My bet is on NASA. I get daily emails from their scientists and even when they goof and err they email us and let us know.
• Options
90.

I email them today they get back to me tomorrow

Did you email them and let them know about the roots problem?

Comment Source:>I email them today they get back to me tomorrow Did you email them and let them know about the roots problem?
• Options
91.

Did you email them and let them know about the roots problem?

No but you could the scientists email is at the bottom of the report.

Comment Source:>Did you email them and let them know about the roots problem? No but you could the scientists email is at the bottom of the report. Also send him your comments, and challenge them.
• Options
92.
edited January 2015

No but you could the scientists email is at the bottom of the report.

I think its not the scientists themself, but I send an email to the adress given there with the kind request to transmit the information.

I dont want to challenge them - I just want to point out that there might be a problem!!!

Comment Source:>No but you could the scientists email is at the bottom of the report. I think its not the scientists themself, but I send an email to the adress given there with the kind request to transmit the information. I dont want to challenge them - I just want to point out that there might be a problem!!!
• Options
93.

I dont want to challenge them

Challenge them, and hold them responsible, if a medical doctor makes a mistake in an intricate surgery that might cost 1 life we hold him responsible in courts of law, but climate scientists making mistakes 1000s die and millions of lives ruined and we do nothing to hold them responsible

Comment Source:>I dont want to challenge them Challenge them, and hold them responsible, if a medical doctor makes a mistake in an intricate surgery that might cost 1 life we hold him responsible in courts of law, but climate scientists making mistakes 1000s die and millions of lives ruined and we do nothing to hold them responsible
• Options
94.
edited January 2015

when they built the southern campus of the munich technical university my father was attending a public presentation of the architectural plans for the campus. the architect was also there. my father studied the plans and then asked the architect: And where is the subway going to be located? (Because he couldnt find it n the plan). The architect replied: which subway?

...turned out they had forgotten to tell the architect about the planned subway. so in short my father "challenged" the involved participants and saved the state of Bavaria probably thousands if not millions of euros with his simple "critical" question.

Comment Source:when they built the southern campus of the munich technical university my father was attending a public presentation of the architectural plans for the campus. the architect was also there. my father studied the plans and then asked the architect: And where is the subway going to be located? (Because he couldnt find it n the plan). The architect replied: which subway? ...turned out they had forgotten to tell the architect about the planned subway. so in short my father "challenged" the involved participants and saved the state of Bavaria probably thousands if not millions of euros with his simple "critical" question.
• Options
95.

if a medical doctor makes a mistake in an intricate surgery that might cost 1 life we hold him responsible in courts of law, but climate scientists making mistakes 1000s die and millions of lives ruined and we do nothing to hold them responsible

in this Boulder study the scientists made a "maybe" suggestion. they didnt present a very reliable model and made this clear - so I dont know what type of liability you are asking for here in this particular climate study.

Comment Source:it seems the comments overlapped >if a medical doctor makes a mistake in an intricate surgery that might cost 1 life we hold him responsible in courts of law, but climate scientists making mistakes 1000s die and millions of lives ruined and we do nothing to hold them responsible in this Boulder study the scientists made a "maybe" suggestion. they didnt present a very reliable model and made this clear - so I dont know what type of liability you are asking for here in this particular climate study.
• Options
96.

My point in general is clear and this report is no better or worse than any other research papers posted here.

Comment Source:My point in general is clear and this report is no better or worse than any other research papers posted here.
• Options
97.

I also tend to pay more attention to NASA scientists. They seem to understand that climate science is really a combination of geophysics and atmospheric/oceanic sciences. The evidence and clues are scattered all over the place and a cross-disciplinary approach is necessary.

Comment Source:I also tend to pay more attention to NASA scientists. They seem to understand that climate science is really a combination of geophysics and atmospheric/oceanic sciences. The evidence and clues are scattered all over the place and a cross-disciplinary approach is necessary.
• Options
98.

Paul my only choices are these NASA fellows, they are giving me gigantic amounts of data, at least communicate and suggest ideas and report bugs and errors from satellites, I do not have to kowtow to any of them to get any of that.

Comment Source:Paul my only choices are these NASA fellows, they are giving me gigantic amounts of data, at least communicate and suggest ideas and report bugs and errors from satellites, I do not have to kowtow to any of them to get any of that.
• Options
99.
edited January 2015

Dara wrote:

Your statement is casting a strong innuendo against the authors of the paper. You've gone out on a limb, so please substantiate your case.

You said:

I tell you why, check these out:

How Plants Survived Chernobyl

Discovery of the secrets that enable plants near Chernobyl to shrug off radiation

Botanical molecular structures are quite dynamic and via turning DNA switches on and off and mutations on and off, software-IF-THEN-ELSE-like. They do that to deal with arid climate changes or even radiation. I am sure there are such gene switches for CO2 rich soil and air. Specially since in the deep past of botanical life on this planet, the atmosphere was toxic gas/soil rich, and planets dealt with that.

But at most that can show that they made a mistake, and failed to consider some factor.

How does that make them crackpots?

Comment Source:Dara wrote: > Hello Nad this article dirty pool.. is a good candidate for crackpot science. (The article is <a href="http://phys.org/news/2014-12-dirty-pool-soil-large-carbon.html">Dirty Pool: Soil's large carbon stores could be freed by increased CO2, plant growth</a>.) Did you read the full article? Your statement is casting a strong innuendo against the authors of the paper. You've gone out on a limb, so please substantiate your case. You said: > I tell you why, check these out: > > [How Plants Survived Chernobyl](http://news.sciencemag.org/2009/05/how-plants-survived-chernobyl) > > [Discovery of the secrets that enable plants near Chernobyl to shrug off radiation](http://www.acs.org/content/acs/en/pressroom/presspacs/2010/acs-presspac-december-8-2010/discovery-of-the-secrets-that-enable-plants-near-chernobyl-to-shrug-off-radiatio.html) > > Botanical molecular structures are quite dynamic and via turning DNA switches on and off and mutations on and off, software-IF-THEN-ELSE-like. They do that to deal with arid climate changes or even radiation. I am sure there are such gene switches for CO2 rich soil and air. Specially since in the deep past of botanical life on this planet, the atmosphere was toxic gas/soil rich, and planets dealt with that. But at most that can show that they made a mistake, and failed to consider some factor. How does that make them crackpots?
• Options
100.
edited January 2015

This is a separate point, but I also don't understand your chain of reasoning.

The paper is claiming that increased plant growth could trigger an increase in microbial growth in the soil, which could lead to the release of large amounts of carbon dioxide stored in the soil.

Comment Source:This is a separate point, but I also don't understand your chain of reasoning. The paper is claiming that increased plant growth could trigger an increase in microbial growth in the soil, which could lead to the release of large amounts of carbon dioxide stored in the soil. Your response is the suggestion that plants can adapt to high carbon-dioxide environments, just as they have adapted to high-radiation environments. Even if that's true, what bearing does it have on the effect that plant growth can have on the level of microbial activity (and the consequent release of more carbon dioxide from the soil)?