nad

Hi MIA, thanks for your reply. While reading the science of Discworld by Terry Pratchett, Ian Stewart and Jack Cohen I came across the concept of “Lies-to-children”; when explaining a complicated subject to a child we try to do it by telling them something we think they will understand, we don’t do it to deceive but rather to help. For example, I was told that the atom was like a very tiny universe where the nucleus is like the sun and electrons are like planets orbiting it, and then I was told to forget all that “rubbish” and to go and look at the Bohr model and so on. I found this quote from Terry Pratchett in Wiki which I think explains it much better than I can "Most of us need just 'enough' knowledge of the sciences, and it's delivered to us in metaphors and analogies that bite us in the bum if we think they're the same as the truth." I suspect there have been many stitches put in the rear ends of people that only read summaries. </digression> NAD

For those that like this sort of thing there is a nice table in the Supplemental Material of formulae giving the radiative forcing for some common greenhouse gasses. I hadn't seen them in this sort of ready reckoner form before, sorry if I am treading on old ground. Looking at he latest concentration of Carbon Dioxide I see it is at 407ppm and taking pre-industrial as 280 ppm we get 5.35(ln407 - ln280) = 2.00 W/m2 a nice round number. A doubling of CO2 gives 5.35(ln560 - ln280) = 3.7 W/m2, NAD

Hi MIA, thanks for the link. A tardy response I know, but I don't visit he Blogiverse very often. Regarding the use of parameters, I found the section “Climate Model Development and Tuning” interesting. It informs us that when developing climate models, some processes have to be parameterized either because they are too complex, or they happen over too small a time scale or area, such as clouds. Also there will always be a trade off between model resolution and computer resources; the smaller grid or more time steps that are included, the more the computing requirement increases. Another use of parameters is in establishing what the initial state of the system, much the same as with Numerical Weather Prediction models. All the separate model components such as the atmosphere model, oceans model, sea ice model and land models are finally combined into a climate model. The model then undergoes a tuning phase. It is at this stage you can get such things as compensating errors, where two or more errors combine to give what appears to be the correct answer. I know from experience, although not from developing climate models, that these types of errors can be quite awkward to debug. Here is a quote from the section. >>The requirement for model tuning raises the question of whether climate models are reliable for future climate projections. Models are not tuned to match a particular future; they are tuned to reproduce a small subset of global mean observationally based constraints. What emerges is that the models that plausibly reproduce the past, universally display significant warming under increasing green­house gas concentrations, consistent with our physical understanding. << To me the report seems quite thorough and open when discussing the shortcomings of using parameters in climate models. As to the question of whether they are accurate enough to base policy decisions on, I will leave that decision to others. NAD

While it is quiet here I would like to tie up a few loose ends on my understanding of the MJO and try to move the story on a little and into the subtropics. So what happened last week with the MJO? See the OLR anomalies: We can see the area of long lived convection around the maritime continent and some new convection appearing in the central and Western pacific around 10 to 15 degrees north and south. These seem to be almost sneaking around the area of anomalously high OLR near the dateline and following the areas of highest sea temperatures. A digression: The Coriolis Effect is zero at the equator and gets stronger as you move north or south. The out flowing winds mentioned in an earlier post that flow away from the equator will at some point be affected by the Coriolis force and be deflected to the east. Wind that is flowing close to or along equator will be affected less. This has the effect of channelling some of the wind along the equator, sometimes called a “wave guide”, and the resulting wave is called a Kelvin wave. So what happened to these high level winds that escaped from the equator? Here are the 250mb winds for the same period. Concentrating on the Northern hemisphere, the areas of red and orange around 20-30 degrees north are where the strongest winds are, also called the subtropical jet. To me they look strongest to the north of the strongest convection. It is also interesting that the jet looks enhanced north of the area of convection in the Caribbean. Is the MJO adding a westerly wind component to the atmosphere? A commentator on this thread suggested I take a look at the Forecast Model thread. At first glance it looks as if there is more to learn about human psychology than teleconnections; however, if you have your noise filter turned to maximum, there are indeed some nuggets of wisdom to be found, and I think you can guess which people are contributing them. One thing that I did see was that the MJO is/was forecast to move across the Pacific; maybe the charts posted support this. As ever I am always happy to be corrected.

Hi @Snowy Hibbo, Thank you, I had not seen that website from the University of Tsukuba, http://gpvjma.ccs.hpcc.jp/TIGGE/tigge_MJO_score.html it contains a wealth of information. You can scroll through the years between 2007 and 2013 and see that the verification statistics improve over the years, which I see as a vindication of the work put into this area. It was not my intention to imply that the GCMs and NWPs do not do a splendid job in their representation of the MJO; it was more to say that the developers feel that they could do even better. For example snip >> The representation of the Madden–Julian oscillation (MJO) is still a challenge for numerical weather prediction and general circulation models (GCMs) because of the inadequate treatment of convection and the associated interactions across scales by the underlying cumulus parameterizations << https://journals.ametsoc.org/doi/abs/10.1175/JAS-D-14-0120.1 But yes, in my post the words “struggle to accurately predict ...” does come over as saying that the models do a bad job. Thank you for pointing that out, after all I am here to learn.

I realise that the thread is eager to move on from the basics of things like the MJO; however I would like the opportunity to complete the musings from my first post by looking at how the MJO moves and whether we can predict its movement? Why does the MJO travel from west to east? I have seen a number of theories over time but for me the most convincing is that of coupled planetary waves. This is not as frightening as is sounds. Here are a couple of images one showing a schematic diagram cross section through the area of convection and the second an OLR anomaly chart from my first post. The MJO cycle starts in the West Indian Ocean accompanied by a burst of westerly winds, as the air passes over the warm ocean it is heated and convection is created as shown by the cross section. As a consequence of this convection the air rises and the air near the surface it is replaced with air from nearby, thus the air is said to be converging at low levels. Higher up the air diverges and flows away from the convection it then converges with air coming from the opposite direction; this converging air sinks and diverges at the surface and completes the cycle. On the westerly side of the convection the sinking air causes the convection to be suppressed, this can be seen by the area of brown colour near the equator at about 70 degrees east (a ridge). On the eastern side of the convection we find the easterly trade winds, they are close to the area of high OLR anomaly around the dateline, these are drawn into the convection and enhance it. So we have the convection being nibbled away on the western edge and added to at the eastern edge. It is this that gives the impression of movement west to east. You could imagine it as a giant heat engine travelling around the equator against the prevailing wind. The waves referred to at 70 east and the other obvious waves further west are the planetary waves, in this case Rossby waves. These waves are important to the weather at our latitude. You can read more here from the Met Office. http://onlinelibrary.wiley.com/wol1/doi/10.1002/qj.49712656902/abshttp://onlinelibrary.wiley.com/doi/10.1002/qj.49712656902/abstract Can we predict the movement of the MJO? I believe the MJO phase charts are created using these wind flows specifically at 200hp (high level outflow) and 850hpa (low level inflow), in conjunction with the OLR charts. Here is the one day OLR anomaly chart for Monday. As you can see on a daily timescale it has a lot of “noise” and looks chaotic, I suspect that integrating this with the MJO phase chart would present some challenges. The Global Circulation Models (GCM) and Numerical Weather Programmes (NWP) struggle to accurately predict the movement and strength of the MJO. If you use a search engine to look for MJO and GCM you will see some gnashing of teeth over this, you may also notice that it is an active area of research. I came across this recently https://ams.confex.com/ams/97Annual/videogateway.cgi/id/37087?recordingid=37087&uniqueid=Paper302240&entry_password=987428 from the American Meteorologist Society it gives some insight as to why the speed of propagation is important. You might want to skip the first 4 or 5 minutes. What we do know is that the MJO historically travels predictably at about 4 to 7 meters per second, say 9 MPH, which with a bit of arithmetic gives approximately 8000 /(24 * 9) or about 37 days to circle the earth, if it goes to form it becomes a useful for predicting in the medium to long term time range. We also know what it is currently doing will have an effect on our weather in perhaps a week or two, making it a useful medium term prediction tool. As mentioned earlier, it is the Rossby wave action that goes on to travel through the sub tropics and then on past our latitude that has a profound effect on our weather. I’ve gone on longer than I intended, so I’ll stop. If I have made any howlers I apologise and am happy to be corrected. Posting on public forums is not my natural habitat, so I would like to publicly thank Tamara for her supportive PM.

Hello David, thank you for your kind reply, suitably encouraged I will have a go at this year. The OLR anomalies for this month show. Even though most of the tropical convection in the Pacific is in similar areas to last year, we can see the cooler water east of the dateline having an effect. In my opinion the basics are the same, with waves heading east and poleward, but perhaps not as clear cut as last year. The convective precipitation chart for the same time shows: Intuitively one would think that the most convective activity would take place over the warmest water. Here are the Sea Surface Temperatures for the period. You could make the case that the precipitation has indeed taken place over the warmest water which is just west of the dateline. The MJO seems to have been pegged to the Maritimes and Indian Ocean, I’m not sure why this has been the case, could it be that once established the outflow from this convection suppresses storm formation further east? The idea of “my thunderstorms are bigger than yours” coupled with the strong trade winds might have been responsible. Much of the documentation that I have seen regarding the MJO has a bias towards the North American continent, and with such a large ocean to the east, rightly so. I would think that any standing wave set up by persistent sea surface temperatures will be modified by any large area of deep tropical convection, and that the nearer to home that convection is the larger the effect. You may have noticed from the precipitation charts that this year there has been enhanced convection in the South Atlantic and the Caribbean. What effect has this had on the weather, below is the 500mb anomaly chart for the same time. There looks to be enhanced subtropical ridges around the globe, perhaps showing what the GWO advertises for low AAM in phase 3? The Azores and Bermuda highs look noticeably pumped up. From a parochial point of view I think that a closer look at what is happening in the tropics of the South Atlantic, Caribbean and possibly tropical South America may help our understanding.

Last winter I wanted to take a look at teleconnections, luckily I still have some of the images I used saved on my computer. My approach was to look at the Outgoing Longwave Radiation (OLR) charts provided by the Earth System Research Laboratory (ESRL). Specifically I used the OLR anomaly charts in the hope that they would show where any areas of deep tropical convection were. See here https://www.esrl.noaa.gov/psd/map/clim/olr.shtml The OLR anomaly charts show the difference from normal in the amount of infrared radiation (heat) being emitted to space from any particular location on earth. Here is the chart for 12 December 2016 through 10 January 2017. The blue and red show where there is less radiation than normal leaving the earth, probably due to the radiation being emitted from high cloud tops, this is where I assume that tropical convection had been more frequent than normal. The brown and black areas show where more radiation than normal was leaving the earth, probably due the sky being cloudless and the radiation being emitted from the surface. You could possibly conclude that the chart shows bands of different shading (blue or brown) moving to the right and poleward emanating from the areas of convection. The following chart shows the anomaly in convective rainfall for the same period. The yellow and green colours show where there was more rainfall than normal. I think it matches fairly well with the OLR charts. Here are the 500hpa anomaly charts for the same period. This is perhaps pattern matching rather than science, but for what it’s worth I feel from looking at this particular thirty day period there may well have been a connection between tropical convection and the actual weather patterns. If I recall correctly the MJO was in fairly low amplitude in the Indian ocean/Maritime continent. I find it curious that there was quite a lot of convection in the south Atlantic and a corresponding area of high pressure to the right and poleward, which just happens to be our neck of the woods. I hope my wittering has not cluttered up the thread, feel free to delete if it is

Modified Newtonian dynamics: MOND

In this document http://www.ldeo.colu...er_battisti.pdf Seager and Battisti present some evidence that there are two stable atmospheric circulation patterns in the North Atlantic, a warmer pattern, and a colder pattern. They propose that temperature and rainfall patterns during the colder pattern may be explained by a more zonal and southerly displaced jet stream. They also describe how the warmer pattern is self-reinforcing. Heat released to the atmosphere from the North Atlantic Drift (NAD) helps maintain low pressure near Iceland. This in turn deflects the jet stream to a more SW to NE flow, which then helps to maintain the NAD. I am curious how a flip from one state to the other occurs. Does it have to be as dramatic as, a shift in global teleconnections, or a shutdown of the thermohaline circulation? Could it perhaps be something simpler? For example: If the two patterns are self reinforcing, would after a period of time with the alternate pattern, make the future more predisposed to this new pattern? If perhaps by chance, we are have several years of a southerly tracking and zonal jet stream, is it more likely that subsequent years will continue in this pattern?