MTC's Winter 2011-12 Forecast

Winter Forecast 2011-12 For Ireland

from M.T. Cranium (Peter O'Donnell)

Methodology

This seasonal forecast is based on Peter O'Donnell's research into natural variations in atmospheric patterns using an extensive data base on both monthly and daily time scales.

Peter's research began in North America around 1980 and was extended to Europe as part of global modelling efforts, with the major advance being a forecast and detailed research program that began in 2005. As a result, this is the seventh winter forecast that Peter has issued for the U.K. and Ireland, and the second one for Boards.ie as well as Irish Weather Online. Global patterns and solar activity are factored into the forecasts, but the primary basis remains a numerical model of temperature variations that can be connected to index values for pattern types.

The model considers various factors that theoretically link the complex solar system magnetic field (which could be described as "space weather" or the state of the solar wind compared to average at various points in time). Then, using a theory that links the solar system magnetic field to earth's upper atmosphere, the model derives a forecast of atmospheric variations from normal. A third and very significant step is to model in an independent set of variations based on lunar-atmospheric tidal effects.

The main point to keep in mind is that the forecast model is basically the sum of many small variations which, if they existed alone, would produce only slight variations in the weather day to day or year to year. But if there are twenty or more such variables all interacting, then the cumulative effects can become considerable. This particular seasonal forecast is fine tuned from "re-analysis" or a look back at recent seasonal forecast efforts and feedback into the research model from a study of timing or intensity errors.

Unlike some in the long-range forecasting business, we make no elaborate claims of high accuracy but we believe that over the past five years this method has produced some of the best results, for example, the severe cold spell of last December.


Overview for winter 2011-12

From a starting point in mid-October, we consider the five month period November to March, with "climatological winter" defined as being the months of December, January and February, whereas "meteorological winter" can, as we saw last year, fall in any of the five months in the Irish climate.

As is now becoming clearer on reliable time frame global forecast models, late October could be a lot colder than the past three weeks have been. That colder trend may fade out for part of early NOVEMBER with another mild spell, but colder weather with some snow may then return in the second half of November. In general this is not expected to be as dramatic as last year, and the severe cold is not expected to continue through most of DECEMBER; instead, there may be roughly equal spells of mild and cold weather with possibly a rather stormy theme and a wide variety of daily weather types that will include some strong winds, some heavy rainfalls and some snow or ice. The output for the Christmas to New Year period suggests a milder interval after some rather cold days about a week before Christmas.

JANUARY may be "the" month of the winter as our model forecast suggests some much colder than average weather through most of the month. The details may include a few brief milder turns which could involve snow to rain type breakdowns, but there are indications of cold spells with northeast or east winds that are often associated with snow in eastern counties, and in particular around Meath and Dublin. as well as Wicklow. The lunar components suggest that the week starting around 8th January could become stormy and cold enough for snow to be dominant by then, although the south could remain in more mixed precipitation. The greatest depth of cold, suggesting higher pressure building up from the east, should come a little after mid-January. More snow seems likely in the period that has delivered in some recent winters, around end of January and first few days of February.

The output then suggests a steady warming trend to produce a rather mild FEBRUARY once that early cold and snow comes and goes, and that scenario suggests some risk of flooding with the thaw and melt. Otherwise February may become a rather settled and almost early spring-like month with some of the model temperature output well above normal.

MARCH looks rather chilly especially towards the end, and could involve a minor return to winter or near winter conditions, so in terms of a seasonal "overview" the dominant theme might be variable but a colder winter than many in the "modern" period, perhaps a winter more typical of the colder climate periods of the past, but either not as extreme as last year, or if everything comes together just right (or wrong, depending on your point of view), a January that may remind us of the December of last winter, with some previews of that in mid-December. Wishing to remain a bit conservative, the monthly temperature forecasts relative to normal are set as:

NOV -1 deg (variable but trending cold)
DEC 0 deg (on balance normal but variable, stormy)
JAN -2 deg (cold, and potentially colder to -4 deg, depending on snow cover)
FEB +2 deg (after a cold start, mild to very mild)
MAR -1 deg (mild to start but becoming rather cool later)

Precipitation forecasting tends to be error-prone just because of greater regional complexity, but with so many changes in regime, it would be sensible to expect some intervals of heavy precipitation and also some two to three week dry spells. Over the research period, winter full and new moons tend to be more stormy than background in the eastern Atlantic, and we'll explore some of the research model details in that regard as we move through the season. The full moon around 10-11 December should produce a significant storm (probably the second of a set) likely moving northwest to southeast given the expected setup, so that could be a time for one of the winter's stronger wind events. The new moon around Christmas is embedded in a mild spell so we are expecting either a foggy inversion with the storm track well to the north, or possibly a mild storm running well north of Connacht around that time, and then there could be a northerly blast to follow around New Years, followed by a volatile period of strong west to northwest flow.

The January full moon energy peak 8th-10th coming around a time indicated to be turning much colder could be the trigger for the most significant weather event of the winter which could be a stormy period involving some heavy snowfalls. Later on, the new moon period seems more likely to drive the storm track well south towards Iberia. If there is a secondary track closer to Ireland this may become a time for snow or ice storms (20th-24th). There are secondary energy peaks that fall roughly five days before these main lunar energy peaks, which sets up an uneven modulation of five and nine or ten days in the likely pressure oscillation, that could show up in a tendency for one week to be settled, then the next week stormy, as a sort of background rhythm to the winter season. The settled weeks are likely to be also the coldest until February, but then the mildest as the flow becomes more southerly or southwesterly. And it appears from the numerical output that these settled periods will tend to fall around mid-months.


Conclusions and Updates

With this time scale, the details are bound to change to some extent and the main point of adding them to basic trends is to give the reader a sense of the range of weather types to be expected. To some extent, timing these exactly is not really the main point for those who want to plan their business ahead of time -- the more conventional forms of forecasting will lead into the active periods, but this overview may give you some idea what to expect in general and we would stress the menu being more important than the order of service. That might not work for a restaurant but the current "state of the art" in seasonal forecasting is frankly well short of being what we could call a fully developed science, no matter what various gurus say in their press releases. For this forecaster, the stimulus of trying to provide some details has a positive effect on the research, and gives us a foundation that is better than just a vague outlook as was and is the custom with large agency forecasts.

Winter 2011-12 looks like being quite an active winter and not lacking in wintry weather types, although not entirely dominated by them either. Readers in more isolated and upland locations might be well advised to have extra supplies on hand at a fairly early stage of the winter in case the early opportunities for cold dig in, but we think your main challenges will come in January this year. In any case, updates to a long range forecast make even more sense than updates to a shorter time scale and so we don't take a "live or die" approach to this early call, as patterns develop and interact, we may well have an adjusted outlook closer to the actual winter season. Not surprisingly, the outlook for the U.K. will tend to mirror this forecast for Ireland but a separate forecast for the U.K. will become available early next week.