Ken Ring

There is an interesting article about him and his methods in this weeks Farmers Journal, worth a look if you can pick it up in your local shop. It'll be up on their website during the week, I'll post a link then.

Article from this weeks Farmers Journal, make of it what you will

http://www.farmersjournal.ie/2009/1024/farmmanagement/crops/feature.shtml?mn=7&sm=7-1

Deep Easterly wrote: »

It did snow in places on the 28th/29th of October last year despite uppers of less than -5, although it was confined to higher elevations in the east (above 80m asl) by the time they did warm as that depression in the above chart bore down upon us. I agree though that uppers need to be below -6c for us in lowland Ireland to see some of the white stuff.

Yep, we got a good covering of snow here last year, the week before Halloween. Only lasted a few hours though.

nacho libre wrote: »

on his site he predicts some significant falls about the second week in Nov, then Nov 16,

now look at this GFS chart(hopefully it won't have changed again by tomorrow)

Ken Ring uses an old folk weather tradition of predicting the weather based on exactly whatever happened 18 years and 11 days ago. That's all there is to it. No magic. Never heard of chaos theory obviously. If it were that simple I think everyone could dispense with supercomputers, numerical prediction models, and most of modern meteorology since around Newton. http://www.sillybeliefs.com/ring.html

Just curious, what would be sufficient evidence to reject Ken Ring's predictions? If it doesn't snow on his predicted dates, would that be enough? Do you think the weather repeats itself like clockwork?

Just for general interest, I can confirm that other people (including me) have done similar research to Ken Ring, from what I understand through limited opportunities to communicate directly, namely, using lunar cycles as one basis for data analysis, as well as other factors that some might think of as "astrological" but which strike me as having a physical plausibility in terms of rotating sectors in the solar system magnetic field, and how that might interact with our atmosphere.

Last winter (on the UK forum net-weather) I collaborated with fellow "lunatic" Blast from the Past, and we came up with a forecast that was widely seen as reasonably good, we called for a cold January and snowfall at times into early February and this certainly verified well. My summer 2009 forecast was not very detailed, I basically made the point that I thought it would be close to average as compared to the "bar-be-que summer" concept, and in fact this was probably closer to the case, what actually happened was that the bar-be-que summer concept did reasonably well for southeast England, and as anyone in Ireland would know better than me, it was decidedly a wet summer in western parts of the UK and Ireland.

So what I'm saying here is that some of these research avenues are promising, I don't think of them as having the full potential to "crack the code" because the identified large-scale signals like El Nino, NAO and PDO are obviously important and not as well understood by any means as they would need to be eventually, so I expect a sort of convergence of different methods in the future towards better or "even better" (being rather generous) long range forecasting.

Which leads us to the inevitable question, what about this winter? Leaning towards this idea -- rather mild most of the time now to Christmas with a few cold intervals, bits of snow, then colder around Christmas, briefly very mild again around New Years, and cold in mid-January. There could be a notable winter weather outbreak then. I suspect February may turn mild again like it did in 2009, perhaps about the same time or even from the start.

I will update that if I change the forecast after a full review (just about ready to publish one on Net-weather now, and so if further discussion changes this outlook, I will change this Irish version as well).

One thing I can say for sure after several years of extensive research and real-time watching is this -- around Ireland, you can count on windy and even stormy conditions more frequently than at random within 24 hours of the winter full and new moons. Unless the whole region is under a blocking high, any other pattern will usually develop a strong low near northwest Ireland at these times. This is, I believe, part of Ken Ring's thinking as well. There are other predictable storm times that I could discuss later, but some of them overlap with full and new moon in December and January more than other times of year. So this makes the storms a little more focused as well as the conventional meteorological factors.

I happen to be one of those people who suspect that the ancient peoples of the British Isles understood things about the connections between astronomy and weather that have sort of faded out of the organized body of knowledge, and that this was one of the motivations they had for building their sky-related monuments. I am not really confident that we could ever prove this conclusively, but the idea that they went to all that trouble just to identify the winter and summer solstice seems a bit daft to me, you could do that pretty easily without building huge stone rings. (Ken Ring, stone ring, hmmm)

M.T. Cranium wrote: »

One thing I can say for sure after several years of extensive research and real-time watching is this -- around Ireland, you can count on windy and even stormy conditions more frequently than at random within 24 hours of the winter full and new moons. Unless the whole region is under a blocking high, any other pattern will usually develop a strong low near northwest Ireland at these times. This is, I believe, part of Ken Ring's thinking as well.

This sounds like a testable hypothesis. Pick a site with a good long data record, say Malin Head or Valentia. Look up the times of the new and full moons, and get the wind averages 24 hours either side of it. Compare them to the constant long-term average, and you have a result. I’m sure if the findings were significant people would be interested. Personal observation can be tainted with subjective errors like confirmation bias, etc. You have to look at the data.

M.T. Cranium wrote: »

I happen to be one of those people who suspect that the ancient peoples of the British Isles understood things about the connections between astronomy and weather that have sort of faded out of the organized body of knowledge

Sounds reasonable to me.

nacho libre wrote: »

i think the more important question is what do you think? I never said he was the oracle of delphi and that we should dispense with tried and trusted scientific methods of predicting the weather. I started the post as a bit of fun as we head into the winter season. Now on to his predictions, his prediction of snow in late Feburary and for the end of December is a safe bet because as we know it frequently snows around these periods in any given winter. Were he to be right about snowfall in November i'd say that was pause for consideration given his predictions for last summer and of an indian summer were largely spot on. However, you would dismiss this as pure luck or coincidence. His predictions to you are old wives tales, that if they do happen to come true on a consistent basis are just down to what can be deduced from studying patterns in our climate over time. don't, Octo, worry your collegaues in Met Eireann won't be going out of business anytime soon

I think you’re attributing a greater level of sophistication to him than he deserves. He simply goes back 18 years and 10/11 days, the Saros cycle. That’s it, there’s no more to it. This has been researched and apparently if you look hard enough at rainfall data over a long enough period with enough decimal places you’ll pick out a pattern – but then again there are many other such cycles aswell with similar magnitudes of influence. But they’re a very minor influence. This is astrology.

So, you still haven’t answered – what would cause you to dismiss his forecasts as no greater than random for the time of year? What would have to happen for you to say he’s just a chancer?

Octo, you may be interested to know that I have done that sort of numerical analysis for temperature and precip data in eastern North America where there appears to be another "singularity" or what I call timing line related to lunar events. The conclusion drawn was that there is a signal of about 2-3 C deg and a 3-5 times random amount of precip. I did not have access to the necessary wind or pressure data but from those signals I would expect to find perhaps a doubling of random wind speed and a 3-5 mb pressure drop signal.

This comes from a large data set that probably includes many stronger cases (direct hits), some counter-examples, and some neutral or flat cases. In other words, these induced lows are scattering over a wide area that happens to coincide with a timing line or zone, and so more likely than not will create a signal, with some counter-examples.

So I would from recent observations of real time weather and looking back at analogue cases, expect that Malin Head in particular would show pretty strong signals and I would encourage somebody at met.ie or here with access to data to look into it. I have done some limited numerical analyses on the CET data for analogues to recent winters, and found temperature peaks that are similar to the above. More anecdotally, the Daniel Defoe storm in 1703 was at new moon, and the Jan 31 1953 windstorm was very shortly after full moon (which in my tables is timed at 0001 GMT 30 Jan 1953, and was a lunar eclipse too).

In fact, my analysis has also separated out a different set of lunar events that operate on the 27.3-day "sidereal" period (the Moon's motions against the fixed sky background) as opposed to the 29.5-day "synodic period" of full and new moons. The most significant energy peak in the shorter cycle is what I call northern max, the position of the full moon in December, where it achieves its highest declination and crosses the galactic equator. Also southern max, the opposite position, shows up as a strong peak. These events as you can visualize, move ahead of full and new moon progressively through the winter and by mid-summer the northern max is at new moon.

So in April and October, known to be windy months, the northern max events can be the strongest, and these will fall roughly five or six days after the full moon (in October) or new moon (in April). The warming influence of this northern max windy period may be a reason for the folklore advice to plant spring crops when the crescent new moon is visible (this is 2-3 days before northern max). I don't think any of these things are more than statistical indices embedded in a much more complex "atmospheric system" or machine, but they seem to be significant -- how significant would depend on how predictable you could make them after studying all the variables involved. The moon's orbit is fairly complex, it takes different paths through the sky over an 18.6 year cycle of declinations, and this cycle seemed to be part of what the ancients were concerned with in building their observatories.

Intriguingly, the "Chaco culture" people of northwest New Mexico who flourished around 800 AD built analogous observatories although on a much smaller scale. They also seemed to identify the range of lunar declination as significant. Perhaps it just interested them for esoteric reasons, but there are weather cycles that have been derived from studying the 18.6 year cycle, and more complex interactions with lunar perigee (an independent 8.85 year cycle). Somebody called David Dilley has some interesting research where he shows correlations between long-term temperature fluctuations and the strength of lunar "syzygy" events; these again are more statistical than hard-core predictive indicators (in my view at least).

There is no shortage of independent research into these lunar-atmospheric interactions, and solar system magnetic field variations, but I would say that they remain outside the mainstream and until somebody either breaks through the code of silence or comes up with an irrefutable predictive system that everyone can see is like tidal forecasts in accuracy, there is not going to be much acceptance in the profession of meteorology. That won't stop us from trying, it's sort of one of those holy grail situations. But there are so many other pathways to follow, numerical cycles based on entirely different causes for one thing.

Anyway, if somebody wanted to look at data, I would suggest, take a minimum of 50 years of data just for January and December, this is when the full moon and northern max are close enough together to form a unified signal, and you would have a little more than 100 data points (sometimes there are three full moons in 62 days, I think this coming winter is one example). I for one would be very interested to see the results, but I'd be willing to bet there was a signal of passing low pressure around the time of full moon, probably forced to a point of T+2 days because there is often a sort of trailing wave closer to Ireland when the strong low goes well north, so the expected temperature signal would be a gradual rise from T-3 days to a peak around T+1 or T+2 days. The new moon signal, I would expect to be similar but on a lower trend line. Finding historic full moon dates is relatively simple, but I could provide those.

I may try to do it then, I was focused on getting the CET data into a workable file on my computer, there's 237 years of daily data there and over 350 years of monthly data that I already have on file. But that's no good for these daily-scale events.

If somebody else is curious enough to look at it, let me know, I would welcome any help I can get, there are more tasks you could imagine in this research than time to do them, realistically this could use a larger research group and the budget, but this is the catch-22, to get the budget you have to do the groundwork that the budget would speed up ... oh well, nobody said this would be easy.

I can only find daily data going back about a year on met.ie, is there a link to longer-term daily historical data?

Would also comment, I don't perceive it as a "conspiracy" against alternative methods, it's more like a perception that this has been disproven for all time, which is actually not the case, it's more like there has never been a proof accepted. Some have been submitted, but they don't come out of the process with any result at all, or they are just deemed to be "insufficient" proof without much structural guidance as to what would meet the test of sufficient proof.

I would say there are strong indications that some of these approaches are valid. They are far from being the whole story. That's where they bog down, correlations may give some encouragement, but even a correlation of .71 only explains half the variance of a climate model.

Anyone here ever check out joe bastardi on

www.accuweather.com ?

Go to uk/ie then go to his euro blog.

Interesting to say the least.

The theory I am developing with my research is based on a sort of interference pattern that the moon is postulated to cause in the atmosphere, aligned with the earth's magnetic field. As a result of this process, there are apparently a number of "timing lines" that would be longitude lines if they ran north-south, but they tend to run more northwest-southeast from the North Magnetic Pole to the South Magnetic Pole. In the northern hemisphere, you could visualize where these timing lines were by taking a line from west of Iceland through Ireland and France into the central Med, then connecting that to the NMP (which you could locate at 80N 100W for simplicity). Then draw eight more at equal distances as if the sphere of the earth had the north magnetic pole as its north pole.

Actually the system was developed first from North American data which is why I numbered the timing lines from "one" running through the Great Lakes and east coast of the U.S., eastward. This places Ireland near timing line three in the system. Meanwhile timing line two would be just east of Newfoundland.

I have done a lot of research and observation to try to flesh out this theory over more or less the past thirty years (unfortunately not with any research funding so it has gone slower as I needed to work at other things). There are still considerable gaps in my observational experience over eastern Asia but I have become convinced of the reality of the system from timing line 7 in the central Pacific through to timing line 4 in eastern Europe. As to postulated timing lines 5 and 6, these would run through parts of Siberia, China, Japan, etc, and I have less data or observational time invested there.

As to the southern hemisphere connections, they seem to work out as well, with the basic idea being that the timing lines feed across the equator at quite a diagonal and then straighten out more as they head for the south magnetic pole (which is offset somewhat from being exactly opposite the north magnetic pole). This places timing line one from east Africa through the central Indian Ocean; the French antarctic research station at Kerguelen appears to be just east of timing line one. The timing line through Ireland (3) runs across the equator south of India and emerges into the southern westerlies near eastern Australia.

It's an intriguing system because of its non-linear correspondence to lat and long (the meteorological equator is similar to what this predicts, generally slightly south of the equator in the Pacific and north in the Indian Ocean, but an operational model would probably look somewhat refined from a simple nine-wave interference pattern.)

Anyway, you can see that I had already done some observational work (mostly on positions of low pressure centres) to work out where timing line three might be. Malin Head is probably about six hours downstream from it, which should mean that signals are just about coincident with astronomical events there (as most lows form downstream from their upper level support).

I can't guarantee it, but I imagine that Ken Ring has found some if not all of these factors and worked them into his forecasting. He may be ahead of me in terms of finding successful analogues. However, whether or not he does, I do not think the Moon is the only player in this system. I also use the timing lines to track what are postulated to be "atmospheric components" of solar system magnetic field sectors. For whatever reason (probably the variable strength of the geomagnetic field) this system works from the premise that when the earth is in a field sector, the effect shows up over timing line one (in eastern North America). The directional sense of this theory is that as the earth moves through field sectors, the effects move eastward. Therefore, any effects noted on timing line three would be (approximately) 80 degrees behind the earth in the solar system and falling further behind unlss they are retrograde in which case they are catching up to the earth and moving west.

You can see how a system like this could evolve into a long-range forecasting system but my numerical analysis has (rather frustratingly) yielded so many distinct components to this model that while correlation is starting to become encouraging, monthly or seasonal forecast accuracy has a long way to go yet.

As I have been working on this pretty much continuously (when possible) for most of my adult lifetime, I don't know if I am going to get it to some point where I can usefully pass it on soon (as I am now 60 and change and could drop off who knows when) but I have established some working contacts with one or two people a generation younger to put them through the same intellectual torture that I have given myself over this time frame.

I understand that Ken Ring is also quite a crusty character and believe me, this sort of research will try the patience of just about anyone, there are many leads and also many dead ends, and the scale of the research challenge is clearly enormous.

I will keep working on this winter pressure series until I can show the results in a week or so. As I believe was explained earlier, I theorize that these waves would decouple into two or three separate series after late January and the summer analogue of reinforced waves might be much less impressive since the storm track then runs further from Ireland (a study of pressure in western Iceland might show similar results).

I've hinted therefore that the full and new moon are not the only lunar "events" that are postulated to cause low pressure on the timing lines. The two major "alternates" happen to fall at full and new moon in late December and are not far separated from them any time from early December to late January. This is why I have used a data set that starts seven days before the full moon in the time period Nov 16 to Dec 14.

Last note here, there is a further complication that arose from detailed study of data from the Great Lakes region (my data base is extensive for Toronto which runs 1840-present). That complication is that timing lines are seen to oscillate east-west over a certain amplitude (about half a wavelength) in ways that may be predictable ... and this may explain a scatter of data or events concerning the exact time when low pressure crosses any given timing line. So this study is not likely to show more than a statistical tendency; individual cases are determined by several other factors.

I've had the odd beer, in fact I had a very odd one last night.

^^^^^
very intresting reading, not that i understand everything, but not to worry

I've finished the data entry and done some analysis on it, which is leading to some other concepts that I want to investigate before I report on it completely.

Also, I think I should start a new thread that is more topical to this research and not have it mixed up in the Ken Ring discussion, which got hijacked by this discussion. So look for a thread in a day or two that will be more suitable for further discussions of this concept of lunar-atmospheric interactions as they apply to Ireland (and specifically Malin Head).

Just to give a few glimpses of what the data show, for the 35 winters ending last winter (1974-75 to 2008-09) I found a regular pulse of atmospheric pressure having a period of about 16 days and an amplitude of about 10 mbs (from troughs of about 1004 mbs to peaks of about 1014 mbs on average), and the troughs coincided with full and new moon although with the longer period, this overlap was best in the late December and January part of the data set (which was designed to cover all December and January data as well as required extra data from late November or early February to keep the data centered on the lunar and not the solar-year calendar).

You'll see these details when I post graphs and more extensive discussions. I also looked at halves of the data for consistency. There was very good correlation from the half of data with the earlier full moon dates and the half with later full moon dates. So the timing of the lunar events in the calendar year does not affect the process noticeably. There was also good correlation of winters with higher overall pressure, with winters of lower overall pressure. In other words, whether it's a blocking pattern or a stormy winter, or just an average one, these pressure waves show up at about the same amplitude, just shifted up or down with the overall pressure trend.

Overall, I concluded that there was a significant process going on in the atmosphere to cause a peak of storminess around these winter full and new moons. There was a tendency to higher pressure especially in the period just before the January new moon, which is a time where the moon moves south across the equator and towards its "southern maximum declination" -- this pressure peak rapidly gives way to a trough at the time of the January new moon (I can say January new moon because the range of possible dates given the setup of my data is about 31 Dec to 30 Jan).

As to the longer cycle of 16 days, this may be explained by some offset process that a longer series of data covering the whole year would reveal, or it may be due to a gradual retrograde annual pressure wave embedded in the system. Since the period was expected to be 14.8 days, slowing it by one "pulse" per calendar year would lengthen it to about 16.2 days.

Of course, I have no idea yet what the data would show for other parts of the year. In any case, I would say, wait until I have time to post the whole analysis in a separate thread, and then we could discuss this interesting result further.