Weather phenomena.

CAPE, Shear, and the Thunderstorm

This document attempts to portray the relationship between CAPE, SHEAR and the resultant THUNDERSTORM.
 First the definitions:
The energy made available by lifting a saturated (i.e. 'cloudy') air pocket until it reaches its level of equilibrium. Assessed on a thermodynamic diagram as the area between the environment curve (Environmental Lapse Rate/ELR) & the parcel path curve.
 SHEAR: Shear in broad terms is given by the change of both wind direction and wind speed in the atmospheric environment through which the pocket (as defined above) is passing: in practical meteorology, it is often defined as the difference between the surface to 500 metre mean wind and the surface to 6 km mean wind, both expressed as vectors.
 Shear will be present in cases where the wind speed increases with height.
[ The Shear (of the horizontal wind-flow) as defined above is pertinent to the generation / enhancement of 'vertical vorticity'; It should not be confused with the shear leading to 'horizontal vorticity', which is usually measured over the lowest 3 km of the troposphere, and which plays a major part in the generation of Helicity. However, this latter quantity is also most important in studies & forecasting of severe local storms.]
 CAPE is important because it defines how vigorous the updraughts within any particular storm-complex potentially are - the stronger the updraught, then all other elements allowed for, heavy rain, large hail, squally winds etc. are on the cards. SHEAR defines what happens to the updraught as it develops, and also governs the interaction between the storm-downdraught and the storm-inflow environment.
 Because of the importance of the link between these two parameters, an attempt at unifying is contained in what is known as the BULK RICHARDSON NUMBER (BRN). This is defined as:

BRN=CAPE / (0.5 * (shear difference)^2)

(the shear difference [ or simply SHEAR ] is as defined above)
 For conceptual purposes, it is simply necessary to remember that:
 BRN is proportional to CAPE and inversely proportional to SHEAR
 For very high BRN values: the shear is too weak to stop the outflow pool of cold (downdraught) air moving quickly away from the parent updraught: new cells may form as the gravity current propagates downstream, but usually well away from the parent - and distinct from it. Also, as the updraught is quasi-zero sheared vertically, the rain shaft falls into a saturated environment (little or no evaporation possible), and there is no potential for a substantial downdraught.
 For moderate BRN values: the shear element (especially speed shear) now plays a crucial part in skewing the updraught, tilting the growing cloudy environment, allowing some or all of the precipitation shaft to fall into unsaturated air - evaporative cooling (plus precipitation drag) will generate a cold downdraught - the drier the air, the greater the potential for accelerating cold/gravity current flow. As the downdraught hits the surface, it splays out, meeting environmental inflow and generating new cells away from (but close by) the parent cell - this is the basis of the multi-cell thunderstorm.
 For low BRN values (but NOT quasi-zero numbers .. see below): Supercell storms may occur (other factors being right). The shear is now strong (and composed of both speed and significant directional components - this latter is most important), and the new/growing cell effectively forms alongside or even within the environment of the 'parent', the whole forming a steady-state system. The situation is complex though and rotation of the developing storm must be present - this is thought to be due to advection (and significant distortion / stretching) of low-level horizontal vorticity into the updraught of the storm environment, but knowledge here is still incomplete, though growing. (It is here that Helicity comes into play .... see elsewhere).
 For very low BRN values: the shear is too strong against very weak CAPE: the developing convective (cloudy) towers are ripped apart and are generally too ill-organised for persistent self-sustaining Cb development. However, note that within the environment of a tropical storm, a low BRN may be associated with organised convective 'streets'.
 The diagram below attempts to 'paint' the idea of linkage between CAPE & SHEAR; note carefully though that there are no values shown - quite deliberate as it must not be thought that there are always 'critical' values between one storm type and another: also, just because you have sufficient CAPE and SHEAR, it is also necessary to have sufficient humidity to sustain the storm through its life, and for the most severe types, an initial 'CAP' (i.e. an inhibition to surface-based convection) must be present to allow the 'loaded-gun' scenario to develop in some way (see elsewhere).
 The abbreviations have the usual meaning (re: METAR/TAF code); TN=tornadic activity possible.
(Remember, SHEAR is made up of both speed and directional components - in individual cases, each should be considered alone, as well as in concert:
The categories are not intended to have defined limits - merely illustrative of the rough combination of CAPE/SHEAR variables which give a particular type.)
Diagram CAPE vs. SHEAR


Conditions for thunderstorms

Five necessary conditions for thunder (as opposed to just an ordinary sharp shower)

  1. Instability through a reasonable depth of the atmosphere: very approximately at least 3000m / 250 hPa
  2. Enough humidity to sustain / feed deep convection through a reasonable depth (at least 100hPa above the freezing level): advection of warm, humid air at low levels greatly assists thunderstorm development.
  3. Cloud tops reaching to a level where the temperature is at least: minus 18°C. (but see 4, below)
  4. A large amount of energy available to be released (CAPE). If the CAPE is large, then the cloud top temperature can be as warm as minus 15°C, and thunder will occur - however, these are the exception rather than the rule.
  5. Something to "kick it all off" - a 'trigger' or initiating action (differential surface heating, orographic ascent, frontal (conveyor) ascent, mass convergence, coastal or sea-breeze convergence, vorticity-driven forcing).

Additional factor to be considered ( determines the subsequent character of the storm ):
* vertical wind shear through the depth of the convective layer.
(For more on this, see the article "CAPE, Shear, and the Thunderstorm")

The approach to forecasting these depends upon whether you are
(a): looking for 'individual' cellular storms, and associated phenomena, e.g. multi-cell, super-cell etc. OR ...
(b): vigorous convective activity embedded in layer cloud, frontal cloud etc.

1. Check for a trough (or a closed upper low) in the upper flow at around 700 or 500 hPa. If the trough is 'broadly' rounded at it's base (or you are dealing with a slow-moving, closed upper low), then convective activity can be expected across a reasonably wide area associated with this trough / upper cold pool; if the trough is moving; is 'sharper' at it's base, with a well-defined axis, then activity is roughly along and on the eastern (or 'forward') side of that axis. [ "eastern side" assumes that the trough is a classical west-to-east moving feature at mid-latitudes - other arrangements can and do occur, for example 'Easterly Waves' in tropical latitudes. ]

2. Once you've decided that the upper air charts are broadly conducive to such activity: use radiosonde ascents to assess items: 1, 3 and 4 above. If the answer is 'yes' to all three, with either expected surface temperatures, or due to forced lifting via dynamic ascent (700hPa vertical motion), orographic ascent (sufficient horizontal flow across hills/mountains) etc., (item 5) .. move onto 3. below)

3. You can assess moisture available via the radiosonde ascents, but they are getting few and far between now. Humidity levels can be inferred from 850hPa output from models.

4. If the answer to all five conditions is 'yes' then thunder is likely. To decide between individual cellular storms, and the variety of multi-cell, supercell etc., see "How does a single-cell shower differ from a multi-cell thunderstorm, or even a 'supercell'?".

(NB: if 3. and 4. above are not met, or only marginal, then a TCU/+SHRA is more likely than CB/TS etc.)

Modern-day forecasting centres often don't work out the steps above: they use NWP output to view forecast ascents, output cloud top temperatures, heights, etc. In operational forecast offices, we have computer output indicating the vigour of the showers, plus cloud top temperatures (for convective types), explicit Convective top/depth forecasts and forecast ascents: we use all these to define where/whether thunder will occur. Not all (if any) of these output are available on the Internet.

Another approach, favoured in North America, is to use instability indices: some of these can be seen on the plotted output from the University of Wyoming ... see the links from the article on thermodynamic diagrams.

Dew, Hoar Frost, Rime & Glaze

Dew and Hoar Frost

On clear, chilly nights grass, twigs and other surfaces (e.g. car windscreens) are strongly cooled with a net loss of heat driven by strong, uninterrupted outward radiation. (If there were clouds present, or fog were to form, then these would absorb, then re-radiate heat back to the surface, slowing or cancelling the fall of temperature. A brisk breeze would also stir the lower atmosphere, mixing warmer air above the surface with that on the surface - this too would offset surface heat loss.) Provided that this 'loss' of heat is not made up from elsewhere (e.g. upward flux from the soil, emission of heat from adjacent buildings etc.), the temperature of such objects will fall below that of the adjacent air. As soon as the surface is cooled below the saturation temperature (dew point), water vapour will begin to condense onto these surfaces (either vapour from moist soil surfaces diffusing upwards to the chilled object, or downward mixing of vapour from the over-lying atmosphere): if the air temperature remains above 0degC, this water will remain liquid as drops of dew. [ NB: the products of dew formation should not be confused with the guttation process, where water is exuded from the tips of plants, usually under conditions of warm, moist soil; these droplets are usually quite large and concentrated at the extremities of growing plants, and careful examination will avoid misattribution.]

If the temperature is much below 0degC, the droplets of dew will begin to freeze since, unlike droplets condensing in the free atmosphere, they will contain some freezing nuclei from the hard surfaces. As soon as a speck of ice has formed, further ice crystals will grow directly onto the ice nucleus without the intermediate formation of water droplets (sublimation, or direct deposition from vapour to solid). Thus, on a cold morning, the grass, vehicle surfaces etc., is covered with a feathery structure of loose ice crystals, or hoar frost, which reflect light from all their surfaces and so appear white.

Glazed Frost (or Freezing Rain)

Where rain has formed in relatively warm air at the cloud level and falls through a layer of very cold air near the ground, the droplets may become 'super-cooled'. When the droplet strikes the cold ground, aircraft in flight etc., it begins to freeze; however, the heat released on freezing warms the droplet up to around 0degC, and allows the water to remain liquid and spread out: only when more heat is removed from it (to the object upon which it has impacted), will the entire droplet freeze, forming a layer of clear ice. In this way, a rapid build of clear ice (or glaze) is achieved, with dramatic consequences for transport, power/communication lines, safety of the public etc. (It is clear ice because the delayed freezing of the droplet allows air to be expelled - if the freezing were to be sudden, the trapped air makes a frozen small droplet look cloudy - see Rime Ice.)


Cloud and fog droplets are very much smaller than rain droplets and when, on a very cold, foggy night super-cooled droplets impact upon objects which have been cooled below 0degC, they immediately begin to freeze in the same way as for Glaze. However, because of the small size, the heat released on freezing can be quickly conducted away either to the object or to the air, so that the whole droplet will freeze very rapidly and will not spread out as in the case of Freezing Rain. As more and more cold fog droplets strike the windward side of the object they will build up a mass of porous ice or rime, with many air spaces between the solidified droplets. It is because rime contains these air spaces that it appears white in contrast to glazed frost which is clear, solid ice with little or no trapped air.


El NiƱo and NW Europe

EL NIÑO: Summary of some published articles relating to links between variability of the ENSO patterns, and weather in NW Europe.

0. Outline:

During the early part of 1998, when the warm-phase ENSO event, commonly referred to as 'El Niño' was at its peak, there was much discussion in the newsgroup as to just what, if any, the effect on the weather would be/was over the UK and elsewhere in Europe. This note does not break new ground, but attempts to pull together references to articles, publications etc., that deal with the subject, so that readers can form their own opinions. However, at some points in this note, I have interjected my own (strictly personal) comments, and to clarify these passages, I have italicised them in square brackets so they are clearly not part of the original referenced work. Bear in mind that I do not claim to be a specialist in this subject, or indeed in climatology as such, but like others, I have a keen interest in the wider workings of the atmosphere. This note was updated in autumn 2006.

The note concentrates upon effects found (or suspected) during the European winter (December - February, or January - March in some studies). This is not to say that there are no effects at other times: indeed some researchers have found tentative signals during the summer, but the over-whelming body of evidence at present (2006) is that where there is a detectable effect, it is most pronounced in our winter. Closer to the seat of the 'disturbance' of course, then the effects will be more pronounced throughout a greater part of the year.

1. What is 'El Niño'?

El Niño is the name now usually given to the phenomena of large-scale and long-lasting weakening of the northeast / southeast convergent trade winds and consequent warming of the surface layers (due to reduced evaporative cooling/reduced cold-water upwelling) in the eastern and central equatorial Pacific Ocean. These phenomena occur rather irregularly: roughly every 2 to 7 years. The event typically lasts 12-18 months, and can be detected by 'swings' in the Southern Oscillation, an inter-annual see-saw in tropical sea level pressure between the eastern and western hemispheres, thought to be produced in large part by the aforesaid changes in focus of warm / cold anomalous water. During a marked (warm-phase ENSO) El Niño, notably higher than average mean sea level pressures occur in the western tropical Pacific and Indian Ocean regions, and significantly lower mean sea level pressures develop in the south-eastern tropical Pacific. The event is credited to the year of first observance of the warm (or cold) water anomaly: thus the 1997/98 event is usually noted as the El Niño of 1997. [ based on the notes obtained from the U.S. Dept of Commerce, NOAA, 'TOGA-TAO' web site.]

For more on definitions, terminology.....
and for some notes on the Southern Oscillation etc......

[ The name El Niño is long established in the local dialect of the peoples of the west coast of Peru and Ecuador, and simply refers to the annually occurring warm/southward flowing current of water that (a) led to a much reduced fish catch in the region, sometimes with no fish catch at all, and (b) seemed to coincide with the Christmas-tide period, hence THE Child/El Niño. Only within the latter part of the 20th century has the name been attached to the unusually strong warmings and total devastation of fish stocks etc., and moreover, the name is now synonymous with the entire Pacific-wide (and adjacent areas) event during which the waters of the equatorial east Pacific warm significantly above normal values. El Niño, and the Southern Oscillation are coupled via the acronym: ENSO, and strictly an El Niño event should be called a warm ENSO event (or warm phase of ENSO). We now have La Niña, and it is not clear whether this has just been thought up (i.e. the female equivalent of El Niño), or was long established. I suspect the former. It must be clearly understood that the La Niña event refers only to markedly colder waters, not just a return to a 'normal' pattern.]

2. What are the known direct effects?

From: Reference (2)

J. Bjerknes (Reference (12), is noted as suggesting a teleconnection between the shifting of the equatorial Pacific zone of convection and a change in the upper level circulation at mid-latitudes in the North Pacific/North America and more tentatively, over the North Atlantic. He analysed the 1957-1958 winter [NB: only one event] and is quoted thus in that paper: .. "a positive temperature anomaly in the tropical Pacific (extending over about 90 deg from South America to the mid-Pacific) strengthens the mid-latitude zonal wind system within this sector; the associated negative pressure anomaly ( a deeper Aleutian Low ) in the extratropical cyclone belt anchors the phases of the prevailing stationary waves in the upper westerlies so that a succession of positive and negative stationary wave anomalies appears downwind over North America/Southwestern Greenland and the north-eastern Atlantic/north-western Europe."

[Although only based initially on the one case study, these direct effects as regards the Pacific/North America region have now been accepted by-and-large, to the extent that the altered flow patterns/strength aloft over the North American region can be related to the incidence of 'anomalous' or extreme events in these areas; the suppression of the Atlantic hurricane activity (*), and altered precipitation patterns across the USA are two examples often quoted. See also Reference (5) for another re-statement of the direct and accepted link between ENSO swings and extra-tropical latitude circulations.]

[(*)=for more on how the various ENSO patterns affect Atlantic (and other basin) hurricane activity, see the FAQ: HURRICANES, TYPHOONS, AND TROPICAL CYCLONES at:]

[ However, even before this work, Sir Gilbert Walker* early in the 20th century had demonstrated connections of at least a general nature....]
[* head of the Indian Meteorological Service in the 1920's, and responsible for the theory behind the 'Walker circulation']

From: Reference (6) it is well known that Walker, in his work on predictability of the Indian Monsoon, attempted to find links within datasets he had access to coupling events right across the equatorial regions, from the east Pacific to central Africa. His work resulted in a rather complex series of 'indicators' although the one that has come to us as the most useful is based on a noting of the difference from the average of mean sea level pressures at two stations: Darwin and Tahiti. Walker's work was simplified by an Australian meteorologist, Troup, and what is now known as the Southern Oscillation Index (SOI), can also be found listed as the Troup Index.

When the msl pressure is anomalously high in the monsoon regions of SE Asia / north Australasia and it is lower than average in the equatorial east Pacific, then this defines a warm-event ENSO (-ve value), and vice-versa for a cold-event ENSO.

Walker in fact used data-sets covering the entire globe (such as they were in the 1920's), and went so far as to postulate connections globally and of course this is the subject with which we are most interested.

[It is important to note here that Walker, and Troup, were looking for broadscale, mass-related connections, not local/regional scale 'weather' effects. Indeed, given the data sparsisty/integrity and more importantly, the short records that Walker in particular had to work with, it would be very surprising if they had found direct connections in terms of local weather. That was Walker's original aim: the seeking after a method of long-term forecasting of the onset/intensity of the Indian Monsoon. However, modern reappraisal of the record seems to suggest that whatever connection there may be between the Monsoon over India and phase-changes of ENSO do not appear very reliable, since researchers have found that warm-phase ENSO years are associated with both particularly wet and notably dry years. It seems to be accepted by modern researchers in the field that the warm/cold switch in the ENSO signal is coupled in some way with the timing/intensity of the SW (or Asian) monsoon, but that there are other factors at work which offset the signal. There are no doubt complicated feed-back mechanisms at work here which may still not be understood fully. Certainly Walker did not achieve his primary aim of finding the answer to predictability of the Monsoon: something that is still sought after today.]

From: Reference (4) [ Most researchers now accept some strong, directly coupled mechanisms between ENSO events (warm-neutral-cold), and weather patterns in the immediate equatorial Pacific region: indeed, so confident are climatologists of the direct links that routine forecasts, based on fluctuations in the equatorial Pacific, are produced by NOAA for North America & other 'Pacific Rim' areas.]

From: Reference (5) As noted earlier, as the trade wind flow weakens, upwelling in the eastern Pacific decreases: this is because the wind blowing across the surface of a body of water exerts a drag on the top-most layers, deflected by the Coriolis effect in such a way that divergence of the surface layers occurs, requiring the aforementioned upwelling of sub-surface waters to preserve a balance - however this water is cold, relative to the top-most layers. Thus, if the upwelling is significantly reduced, sea surface temperatures tend to rise, and not just due to the absence, or reduction of cold water injection, but because the amount of evaporation also decreases (due to reduced wind velocity). Further, in a mature El Niño, the warm water of the western and central Pacific spreads eastwards (as there is now a reduced fetch across the surface so the Pacific waters attempt to achieve a level) and the combined effect of movement of the warm water eastwards, and local warming in the east mean that the atmospheric convergence zone (which will be most vigorous where the waters are warmest) migrates towards the central Pacific ... the subsidence over South America decreases to such an extent that a new convective region is allowed to appear here, associated with intense/heavy rain and floods in the normally arid countryside. It is now accepted that at least on this regional scale, the eastward displacement of the Indo-Australian enhanced convergence zone results in large-scale drought conditions in Australia and Indonesia [hence all the fuss about the forest fires/smoke etc.]: and, conversely, locally intense rainfall and hurricane formation in the central Pacific, where such phenomena are normally rare. [ These effects are now accepted such that no further qualification is needed. However, it is worth pointing out that each warm ENSO event, and indeed its opposite, will vary both in timing and intensity, and so the consequent effects will not be the same from event to event. There seems to be some, almost hysterical, assumption that every incidence of anomalous warming in the central/east equatorial Pacific will lead to the same level of 'mayhem' across the region...this is not so.]

From: Reference (5) As well as these latitudinally-bounded (to the equatorial region of the Pacific-rim) effects, the movement of the Pacific convergence zone (with its associated moisture/momentum/heat mechanisms) alter the upper tropospheric atmospheric circulation over the north Pacific, and it is suggested, a large part of the globe [ see, for example, Reference (7)]. This is because of the effect that these alterations have on stationary waves that tend to form in the upper troposphere (Rossby waves), which in turn leads to the displacement of the alternating zones of semi-persistent low and high pressure systems coupled to the 'loops' in the wave-train leading away from the central/north Pacific towards high latitudes and then eastwards across North America. A consequence of this altered pressure pattern is the southward diversion of the upper jet, and therefore of the associated frontal systems in the North Pacific and high rainfall in the western and southern USA.

[This effect too seems to meet with general acceptance. Indeed, mean maps of zonal velocity/u-component wind at jet-stream altitudes in the NE Pacific do show higher values when the equatorial east Pacific is warmer than average, and this should be expected from Polar Front jet-stream theory: a greater degree of warmth through an atmospheric column will lead to higher pressures aloft, relative to poleward locations, and lead, if only temporarily, to an enhancement of upper level jet strength, and downwind propagation of the jet, with a consequent displacement of its associated entrances/exits. Further, altered jet-strength across the Rockies, will lead to altered CAV trajectories which in turn will lead to a displacement and altered configuration of the downstream long-wave trough. All these changes too seem to be accepted as 'direct' effects of the warm-phase ENSO. However, with time during any event, there will tend to be a 'levelling-out' , so later in a winter season, the effects should be less detectable, but this may not be the case for a significantly warm/cold event. Further, higher latitude effects will also impinge on this relatively simple story, which may or may not be directly coupled to ENSO, but perhaps be tied more to the weather of the previous season. It is therefore dangerous to couple every anomalous event even in the immediate Pacific area, to ENSO variation. ]

3. What are the suspected effects, direct or indirect?

From: Reference (6)

Walker in his paper, suggested that relationships exist between other geographically remote regions: indeed his concern, as already stated, was to find some large-scale methodology of relating what we now call the Southern Oscillation to weather/climate changes elsewhere. He specifically noted a suspected link between msl patterns [ he would not have been able to comment upon the upper-air aspects ], in the Pacific and the North Atlantic. It is worth quoting Walker directly thus ..... ' there are swayings on a much smaller scale between the Azores and Iceland and between the areas of high and low pressure in the North Pacific Ocean .... and that all appeared to be related: .... there is a marked tendency for the highs of the last two swayings to be accentuated when the pressure in the Pacific is raised and that in the Indian Ocean is lowered '... [i.e. a cold-ENSO event - however, it is not clear to me what Walker was talking about here: does he mean 'high' in terms of pressure, or 'high' in terms of extreme? Nevertheless, the fact that this 'swaying' (or I suppose we would now say 'oscillation') in pressure pattern distribution was noted early in the 20th century, and the fact that we have (see later) good statistical evidence for such alterations, not only at the surface but aloft, means we have to accept the broad-scale links between ENSO variability and patterns over the NE Atlantic, particularly during strong ENSO episodes.]

From: Reference (5)

This reference accepts that the climate of other ocean basins (outwith the Pacific) can be altered by El Niño events, and notes the eastward shifting of the convergence zone over Indonesia (during a warm ENSO event) which is suspected to have an impact on the Indian Summer Monsoon through a weakening of its land-sea circulation. This is thought to lead to a failure of the monsoon rains across India. [ However, as pointed out above, this link is disputed by some.] This reference also states that Africa is affected to a significant degree. The author notes that an indirect link has been found between severe drought years in the Sahel and El Niño events ( from Reference (13)), and also with drought frequency in southern Africa, and states that this is probably the result of circulation modifications above the Atlantic and Indian Oceans respectively."

[ It is difficult to refute these statements, however it is pertinent to ask about the integrity of the data-sets used to establish the suspected links, particularly the length of the series, and note the difficulty of obtaining long-period accurate data-sets of sea surface temperature in data-sparse areas. It is also worth noting the primary effect of the warm ENSO pattern is achieved because of direct effects due to changes in evaporation/upwelling etc., and these effects cannot of themselves impact upon areas further west. The effects must be indirect, involving perhaps changes in the upper easterly flow, changes in TUTT(*) orientation/location, and perhaps delayed relative to the primary region affected.]

[ (*)=Tropical Upper Tropospheric Trough: An upper level weakness at low latitudes - to see their importance to tropical forecasting, particularly for tropical cyclone formation, see the FAQ: HURRICANES, TYPHOONS, AND TROPICAL CYCLONES at: ]

[ For this Section, see also References (8) & (9) ]

4. Some notes summarising articles relating to possible effects over Europe

From: Reference (3) & Reference (14)

These authors have no doubt that ENSO variation has a direct effect upon the synoptic climatological patterns of the NE Atlantic/European region. It is further stated that the effect is largest/best detected in the winter months of January and February (& for the second paper also March) following the year of a warm or cold event. [ However, careful reading of Ref. (3) and we have....] " There is more variability between individual warm event winter months, whereas the response to cold episodes is relatively uniform."

[This article goes on to say that model results confirm the teleconnections between varying ENSO events and the northern hemispheric extra-tropical circulation. However, it is well known that such models are crude at best, particularly where ocean/atmosphere interactions are required ... however, such as they are, they do confirm ideas expressed, and accepted earlier.]

"Not unexpectedly, the largest response is observed in the NH winters following the year of the event, in particular in January and February.".... "Warm events are associated with highly variable winters; that is, there are large differences observed in the high winter season (January and February) of one episode to another one. Cold events, however, appear to produce a more uniform response...."

Many sources quote from earlier work (e.g. Reference (10) and Reference (11)) which found significant pressure and temperature anomalies due to the mid-latitude response noted earlier downstream of an intensified (for a warm event) Aleutian Low [this is noted elsewhere, and generally accepted] and a coupled low-pressure anomaly over the North Atlantic which reveals a pattern similar to the North Atlantic oscillation. [And which latter has been strongly correlated to weather patterns across the British Isles/NW Europe sector. It is also worth a re-emphasis that Walker picked this up over 60 years earlier, and reproduced at Reference (6). The pedigree for this idea, from Walker, through Bjerknes to latter-day researchers is therefore sound, and as the data have been tested over varying time-sets etc., it would be churlish not to acknowledge the effect on the North Atlantic/NE Europe upper air region from upstream anomalies in the North Pacific. ]

[ Interestingly,Reference (2), contains the following statement....] " Finally, it should be mentioned that this ENSO response analysis .... serves as an interpretation after the effect. The prognostic value and dynamic interpretation needs further investigation; Europe is farthest away from the key regions of ENSO so that it is not surprising if at times the influence of the oscillation is superseded by other effects" .... "For example, the extratropical wintertime atmospheric circulation also is sensitive to certain extratropical sea-surface temperature anomalies and not only to anomalies associated with El Niño". However, Reference (14), notes that later work has 'firmed-up' the link, and finds that ENSO events are well correlated with NAOI values provided the north Pacific sea-surface temperature anomalies reinforce the broadscale Northern Hemispheric pattern: this occurs roughly in 4 cases out of 5.

From: Reference (1)

This author draws on other published work referenced here-under, but also interprets the response in terms of circulation types across the British Isles, using Lamb's Weather Types (LWT). He notes that the A-type appears to be more sensitive to the prevailing ENSO than the C-type, with the former showing positive anomalies during cold events and negative anomalies during warm events. This accords well with statements made in the papers which concentrated more upon effects across 'central' Europe. However, the article goes on to state that although C-type patterns occur on average more frequently during warm-ENSO events, the percentage of correct signals was notably lower than for the A-type: only 50% (C) compared with 68% (A).

[ This would appear to confirm the doubts expressed about the predictive ability of the warm-ENSO event as noted elsewhere. Further doubts are expressed later .... there is a figure of C-type frequencies during a warm ENSO extreme and this does identify the enhanced prevalence in mid-February which corresponds to the singularities previously detected in the European Grosswetter (Reference (3)). The pattern for the A-type during a cold-ENSO extreme is less clear with below average frequencies in early-mid January, and above average prevalence during most of February ].

The author performed a statistical analysis on these data and this gives rise to the following statement... " In other words, the A-type winter frequencies display significant anomalies during ENSO extremes (particularly in February) which are sensitive to the ENSO type. The signal for the C-type, however, is less conclusive.".....(and further....) " These findings support those of (Reference (3) who suggested that warm extremes are associated with highly variable winters, whilst cold extremes tend to produce a more uniform response, with less variation between episodes, and therefore a higher predictability in long-range forecasting.

Reference (14) is a useful updating of the above ideas (to 2006): Using reconstructions back to the 1500's, they appear to show that provided the events are 'strong' (well-defined & long-lasting), then the tendency is as follows:-

Strong "El Niño" (-ve SOI) event: -----> -ve NAOI (i.e. tending to blocked pattern)
Strong "La Niña (+ve SOI) event: -----> +ve NAOI (i.e. tending to mobility)

However, this study confirms other findings that even for 'strong' events, roughly a fifth show no correspondence to the above: it appears that conditions (of SSTA) in the northern Pacific must act to enhance the forcing of the warm/cold ENSO patterns for downstream effect in Europe to be marked. It is most important to understand that whilst there is a potential link, it is not an unvarying one.

5. Summing up, and the Q/A issued with the FAQ.

So, what do we know, or more importantly, what do we think we know?

Strong signals are there, and, as for the 1997 and 1982 events, global effects have been found in datasets from widely separated regions of the globe, and, according to researchers working in Europe, can be found in the relative frequency of broad weather types (e.g. anticyclonic versus cyclonic) between cold and warm forcing. However researchers point out the variability for each individual event, and make the point that the coupling is picked up as a bias towards one type or another in a long-term record.

For a region as far away from the seat of the disturbance as ours is, most researchers make the point that events closer to home will either significantly modify the effect, or perhaps completely swamp any effect such that only a residual signal is to be found by careful statistical analysis. To predict any one winter/spring season over the British Isles/NW Europe on the basis of any one event in the equatorial east Pacific is problematical; however, it is true to say that solid links are being found in the historical record between strong events in the equatorial Pacific & the circulation in the North Atlantic / European region.

This is the Q/A that is published in the long and html versions of the FAQ:

Q. What impact does 'El Nino' have on the weather over Europe?

A. The 'El Nino' phenomenon, or more strictly the warm El Nino -Southern Oscillation (ENSO) event is coupled closely to remarkable shifts in weather patterns in the immediate Pacific basin, and adjacent areas: e.g. parts of North America. For example, it is clear that the altered distribution of warm/cold water across the equatorial Pacific is the primary reason why excessive rain can fall in places like Peru, and a general deficit of rainfall is experienced in Indonesia, parts of Australia and the Philippines. There is also an accepted link between a less-than-'normally' active Atlantic hurricane season and the notably warm event that characterises what has come to be called, THE El Nino.

It is becoming clear from recent studies that we can now rule out the 'No Effect' case: this leaves us with two options -

(a) There IS an effect, but it is on a scale that is dwarfed by regional variations closer to home, e.g. long-term thermal inertia in SST distribution in the N. Atlantic, or continental/oceanic temperature differences across the North America - North Atlantic - Eurasian 'super-region'.

(b) There is a direct, and marked effect that leads to verifiable modification of the weather types across the NE Atlantic/European - Mediterranean region.

(a) appears to be the most likely if we take the year overall; indeed, even in studies published which set out to prove the link between warm/cold ENSO regimes, and impacts over Europe, caution is always advised relating to local/regional scale modification.

(b) is climbing higher in the 'probability' stakes, at least if the 'winter' season only is considered. There are an increasing number of studies published that show a direct link between a warm ENSO season, and, for example, altered rainfall/temperature anomalies across west/central Europe. No lesser person than J.Bjerknes postulated in 1966 that altered activity in the equatorial Pacific appeared to significantly alter the strength/orientation of the PFJ over and downwind of the NE Pacific, which in turn must have at least some effect on the long-wave structure downstream. This appears to have been accepted in later studies & developed further using datasets going back over two centuries or more.

also, have a look at the following sites:

[WMO home page]

[NOAA/TOGA-TAO site for real-time data, advisories, further definitions etc.]

... and of course, a search of the WWW will throw up many active sites dealing with El Nino.

Reference (1)

DATE: August, 1993
VOLUME, ISSUE NUMBER ETC: Volume 48, pp234-239
TITLE: 'Evidence of ENSO in the synoptic climate of the British Isles since 1880'
AUTHOR(S)/EDITOR(S): Robert Wilby
AFFILIATION(S): Geography Department, University of Loughborough.

Reference (2)

PUBLICATION: International Journal of Climatology
DATE: January/February, 1992
VOLUME, ISSUE NUMBER ETC: Volume 12, pp25-31
TITLE: 'Climate anomalies in Europe associated with ENSO extremes'
AUTHOR(S)/EDITOR(S): Klaus Fraedrich and Klaus Müller
AFFILIATION(S): Institut für Meteorologie, Freie Universität, Berlin.

Reference (3)

PUBLICATION: International Journal of Climatology
DATE: January/February, 1990
VOLUME, ISSUE NUMBER ETC: Volume 10, pp 21-31
TITLE: 'European Grosswetter during the warm and cold extremes of the El Niño/Southern Oscillation.'
AUTHOR(S)/EDITOR(S): Klaus Fraedrich
AFFILIATION(S): Institut für Meteorologie, Freie Universität, Berlin.

Reference (4)

DATE: February, 1998
VOLUME, ISSUE NUMBER ETC: El Niño update, No.4
TITLE: 'El Nino update'
AUTHOR(S)/EDITOR(S): None directly credited.

Reference (5)

DATE: January, 1990
VOLUME, ISSUE NUMBER ETC: Volume 45, pp 2-8
TITLE: El Niño and the Southern Oscillation
AFFILIATION(S): School of Environmental Sciences, University of East Anglia.

Reference (6)

BOOK: Monsoons
DATE: 1995
CHAPTER: The variable and interactive monsoon
AUTHOR(S)/EDITOR(S): Peter J. Webster (ed: JS Fein and PL Stephens)
PUBLISHERS: John Wiley and Sons.


Other relevant articles not surveyed in detail:

Reference (7)

PUBLICATION: Monthly Weather Review
YEAR: 1981
VOLUME, ISSUE NUMBER ETC: Volume 109, pp785-812
TITLE: 'Teleconnections in the geopotential height field during the Northern Hemisphere winter'
AUTHOR(S)/EDITOR(S): Wallace, J.M., and Guetzler, D.S.


Reference (8)

YEAR: 1983
VOLUME, ISSUE NUMBER ETC: Volume 222, pp 1195-1201
TITLE: 'Meteorological aspects of the El Niño/Southern Oscillation.'
AUTHOR(S)/EDITOR(S): Rasmusson, E.M., and Wallace, J.M.

Reference (9)

PUBLICATION: Journal of Climatology
YEAR: 1988
VOLUME, ISSUE NUMBER ETC: Volume 8, pp 67-86
TITLE: 'A detailed examination of the extratropical response to tropical El Niño/Southern Oscillation events'.
AUTHOR(S)/EDITOR(S): Hamilton, K.

Reference (10)

PUBLICATION: Monthly Weather Review
YEAR: 1981
VOLUME, ISSUE NUMBER ETC: Volume 109, pp 1150-1162
TITLE: 'The Southern Oscillation. Part I. Global associations with pressure and temperature in northern winter'.
AUTHOR(S)/EDITOR(S): Van Loon, H., and Madden, R.A.

Reference (11)

PUBLICATION: Monthly Weather Review
YEAR: 1981
VOLUME, ISSUE NUMBER ETC: Volume 109, pp 1163-1168
TITLE: 'The Southern Oscillation. Part II. Associations with changes in the middle troposphere in the northern winter'.
AUTHOR(S)/EDITOR(S): Van Loon, H., and Rogers, J.C.

Reference (12)

YEAR: 1966
VOLUME, ISSUE NUMBER ETC: Volume 18, pp 820-829
TITLE: 'A possible response of the atmospheric Hadley circulation to equatorial anomalies of ocean temperature'.
AUTHOR(S)/EDITOR(S): Bjerknes, J.

Reference (13)

YEAR: 1986
VOLUME, ISSUE NUMBER ETC: Volume 320, pp602-607
TITLE: ' Sahel rainfall and worldwide sea temperatures.'
AUTHOR(S)/EDITOR(S): Folland, C.K., Palmer, T.N. and Parker, D.E.

Reference (14)

PUBLICATION: Dynamical Climatology
YEAR: 2006
TITLE: ' ENSO influence on Europe during the last centuries. '
AUTHOR(S): Brönnimann, S., Xoplaki, E., Casty, C., Pauling, A., Luterbacher, J.

Martin Rowley: 24 OCT 2006

Noctilucent Clouds

Noctilucent clouds (NLC)

[ also known as Polar Mesospheric Clouds (PMC) ]

These clouds are cirrus-like in appearance but unlike cirrus (which occur at altitudes between 3 to 18 km, depending upon latitude band), NLC occur in the upper atmosphere between 80 and 85 km, preferentially around the mesopause ( which has a mean altitude of 82 km ) and are associated with temperatures colder than -80degC, perhaps with values close to -100degC.

They are normally only visible at night (hence the name) and during the summer months in each hemisphere. The sun must be between 6 and 16 arc-degrees below the local horizon.

Although there has been some debate surrounding the composition, they are now known to be composed of ice crystals (formed on solid particulate debris - see below), which act to reflect light radiation shining from the sun, which is well below the local horizon.

There is still doubt about the origin of the nuclei of such clouds: it was long thought that the catalyst for such displays derived from debris injected high into the atmosphere from the massive eruption of Krakatoa in 1883 (August). Modern thinking tends to believe that the particles (which become coated with water-ice) originate from outside the earth's atmosphere which then descend slowly through the upper atmosphere. Very low temperatures are required as the amount of water vapour at such high altitudes is so low that highly sub-zero temperatures are required to achieve saturation.

NLC are highly variable in frequency from year-to-year & within each season, probably because the temperature is similarly variable at these high altitudes, but there must also be a 'control' depending upon the numbers of available sublimation nuclei available: this in turn points to possible connection with the occurrence, intensity etc., of periodic meteor showers. There is some link with sunspot minima & also a possible coupling with man-made pollution of the atmosphere. However, it should be remembered that NLC observation depends on visual spotting and that depends on the presence or absence of 'normal' (i.e. tropospheric) clouds & the availability of trained observers to take advantage of good observing conditions.

Possible observations of noctilucent clouds: C.J. Butler. 'Weather' (Royal Meteorological Society): May, 2006.
Meteorological glossary: Meteorological Office. London, 1991.
Web site:

The Pembroke Dangler

Unstable polar maritime airstreams in autumn and winter typically give showery conditions, particularly to exposed coastal areas. These showers can spread well inland where there is little shelter from the sea, such as those which travel through Cheshire towards the midlands in a NW airstream. However, in the case of the ‘Pembroke Dangler’, more is involved than simply exposure to an onshore wind.

The term was coined by Jon O’Rourke, a regular contributor to USW. It is used to describe a line of showery precipitation which forms in a NNW to NNE airflow, typically from Pembroke to Cornwall. This often results in frequent showers, or even more persistent rain, in quite a narrow band. When this occurs a narrow strip can receive quite a high rainfall total, whilst the rest of the south west is virtually dry.

It is by no means an unusual phenomenon, occurring several times a year, particularly in autumn and winter. The long fetch across ‘warm’ water from the Irish Sea to Cornwall, provides the right conditions for showers to form. The shape of the Pembrokeshire Peninsula has the effect of concentrating this unstable northerly flow, in the same way as an obstruction in a stream would concentrate the flow around it. The resultant line of often heavy showers is a convergence zone.

The long and narrow Cornish peninsula protrudes into this line of precipitation, and often has the affect of intensifying it further. Below are just 2 examples from 2008.

1. 14th December 2008

Radar imageThe rainfall radar image shows the situation at noon. There was little change for several hours. In Penzance around 23mm of rain, together with some sleet in the heavier bursts, fell in the 12 hours from 03:00 - 15:00. Central and eastern Cornwall were mainly dry during this period. The synoptic chart is shown below.








Synoptic chart



2. 30th November 2008.

On this occasion the wind direction was NNE, with depressions centred in the Baltic and over central France. By mid morning the line of precipitation was clearly visible on the radar image, and it persisted for many hours. The wind direction meant this convergence line passed just to the west of Cornwall. It was mainly dry in western Cornwall, but decidedly wet on Scilly. Rainfall radar images and associated synoptic chart are shown below.

Radar image

Synoptic chart

Tropospheric Ducting

The speed of a radio wave in the atmosphere is determined by the dielectric property of the air. This property depends on the pressure, tempertaure and humidity of the air. In general as we move upwards through the atmosphere the pressure decreases and temperature falls. This means that the dielectric property changes with height and allows a slight increase in the speed of a radio wave as we move upwards through the atmosphere. This in turn means that if a radio wave moves away from the earth at an angle less than 90 degrees, then the upper part of the wave travels faster than the lower part. Therefore even under normal conditions this can in effect bend, or refract, the wave back down to earth.

The normal rate of change of dielectric constant with height refracts the wave so that it follows a curved path of about 1.3 times the radius of the earth. Therefore, we typically can receive signals which are 1.3 times further than we can see by line of sight.

Tropospheric ducting occurs when we get a sharp rate of change in the dielectric constant as we move upwards through the atmosphere. This occurs when we get a rapid increase of temperature and arapid decrease in humidity (dew-point) with height

If we look at a vertical profile of the atmosphere showing ducting potential, we can see that there is a sharp increase in temperature (an inversion), coupled with a sharp fall in dew-point (indicating a fall in humidity). The duct occurs below this inversion in the yellow shaded area.


Under these conditions we now have the radio wave bent back towards the earth. However, the radio wave can then reflect back of the earth and become refracted again to return earthwards once more. This can sometimes occur a number of times with little attenuation but some fading. The result can be long distance reception of radio waves that would normally have been far beyond the radio horizon.

Typical conditions required for a good duct to occur are:

  1. An increase in temperature by 3C or more per 100ft.
  2. A rapid decrease of RH (dew-point) with height.

The depth of the duct required for varying wave-lengths is:

  1. 50ft for wavelengths around 3cm (approx. 1000MHz)
  2. 600ft for wavelengths around 1m (approx. 300MHz)

Typical meteorological conditions which can be favourable for ducting are:

  1. Warm dry air over a cooler surface, especially a cool sea
  2. Surface cooling under clear skies overland
  3. Anticyclone (high pressure) or developing high pressure ridges with a cold surface
  4. Sea breezes undercutting warm air overland
  5. At fronts with a strong thermal contrast
  6. In cold downdraughts associated with cumulonimbus clouds (indicated by heavy showers or thunderstorms)

To decide wether there may be potential for ducting then first consult the Met Office forecast synoptic charts. If they are showing hints of high pressure building or a weak ridge crossing the area then there could well be potential.
Other sources of information can come from radiosonde ascents. These show the vertical profile of temperature and dew-point up through the atmosphere. They can be displayed in varying ways, but typically we are looking for a profile which looks like the first diagram on this page. That is a sharp increase in tempearature coupled with a sharp drop off in dew-point. As these diagrams often have a vertical scale measured in pressure rather than height here is a short conversion table covering the heights we would be interested in looking at (these are ICAO pressure/heights and have a standard MSLP (Mean Sea Level Pressure) of 1013.2hPa. Therefore MSLP's different to this would affect the heights):

Pressure in hPa Approximate height in metres (feet)
1013.2 0 (0)
1000 111 (364)
950 540 (1773)
900 988 (3243)


Here are some links to sites where these ascents can be found:

University of Wyoming (simply click on a location)

Finally, I sometimes find that a good hint to a duct occuring can be seen by viewing rainfall radar output. As these operate at a high wavelength and there is some automated correction to filter out incorrect rarar returns they may not always be suitable. Nevertheless, if you look at the site when you think conditions may be favourable and see very strong returns where it should be dry then either the forecast is very wrong, or there is a duct. You can view the Met Office radar system via the BBC site.

Good duct hunting!

Why fronts die

Some notes to explain the processes involved in the weakening (or 'frontolysis') of otherwise active frontal zones.


For significant precipitation to occur, two conditions need to be satisfied.

  1. There must be a mechanism to produce upward motion through a deep layer of the troposphere.
  2. There must be sufficient moisture in depth - and for mid-latitude baroclinic systems (i.e. fronts), this must extend for some way above the zero degree (or freezing) level - i.e. clouds should have supercooled water droplets and ice crystals coexisting.

Taking each necessary condition in turn:

1. Upward motion:
In the atmosphere, vertical (upward) motion, i.e. ascent, is achieved in two ways.
(a) Release of convective potential. [ Showery type ]
(b) Dynamic/warm advection forcing. [ Mass ascent type ]

Type (a) is often a factor in frontal rain situations, and can be the means of enhancing precipitation within a frontal belt, but it is type (b) that we are primarily concerned with here, and which I will discuss.

Dynamic forcing is achieved by, for example, ascent forward of a short-wave trough; broad scale ascent in the warm/entrance or cold/exit regions of a strong and well defined jet stream; or a jet streak moving quickly through the main jet stream aloft. It can also be achieved by strong/focused warm advection, indeed in many frontal situations, it is usually a combination of the two (dynamic + thermal warm advection) that needs to be considered.

2. Moisture availability:
There needs to be a source of moisture of course, but perhaps more importantly for significant rainfall/snowfall, there needs to be a continuous supply of moisture, via the low level warm 'conveyor' model. In this model, warm/humid air enters a system at low levels ahead of the driving upper trough, and rises slantwise upwards and either forwards (forward sloping front ===> split upper cold ====> dynamically weak surface cold/kata front), or rearwards (rearward sloping front ====> dynamically intense surface cold/ana front), cooling by adiabatic expansion, and feeding/maintaining the thick layer cloud structure in the process.

Any change that cuts off the supply will cause the front to die; precipitation will form from cloud water/ice droplets, which are not replenished, and the total water (vapour+liquid+ice) content will fall - leading to a decrease in precipitable water content.
(Note that the warm conveyor model also allows for the transfer northwards -- in the northern hemisphere -- of warm air and westerly momentum. These are also important for maintenance of a well defined front, with vorticity and thermal contrast/baroclinicity considerations.)

Looking now at a typical example of cold frontal events, where the cold front is very active over NW British Isles, depositing large amounts of rainfall/snowfall, but by the time it gets to the south of England, it is relatively weak, despite having a good 'Norwegian model' structure across the surface front: wind shift/temperature drop etc.

*** Active: In the NE Atlantic, a well marked upper trough is moving eastwards with strong ascent on its forward side, and with the low level flow from the SSW, with a long maritime fetch, gathering and transporting copious amounts of water vapour off the relatively warm North Atlantic, the combination of upward motion, plus ready availability of vapour into the cloud decks, produces marked rainfall western and northern areas, enhanced over mountains by orographic forced lifting, and encouraging the release of embedded convective potential - if present. The driving upper trough is relatively sharp.

*** In-active: By the time the front arrives across SE Britain, the driving upper trough has 'relaxed' away north-eastwards, removing the source of strong upward motion. At the same time, the act of relaxation implies descent (rising contours), which in turn leads to subsidence warming through a great depth of the troposphere, warming of the cloud, and evaporation of cloud layers, diminishing the rainfall amounts, and squashing the significant sub-zero degC cloud potential.
In addition, and equally importantly, with no ascent through a deep layer, there is no means to replenish the moisture content as the humidity is used up in the precipitation process. A combination of descent (due trough relaxation), and non-resupply, leads to a decaying frontal structure -- the front 'dies'. All we are left with is the low level cloud structure: thick stratocumulus and stratus, with thinning altocumulus and altostratus on top, where precipitation forms primarily by coalescence - insufficient for 'proper' rain, but can produce 'drizzly' conditions. However, even these 'weak' fronts can have some elements of embedded instability to catch out the unwary forecaster!

Now, why does all this happen. Upper troughs don't maintain their character for ever. In the particular case under consideration, the trough is confluent -- this means that the wind speeds in the leg ahead of the trough axis were markedly stronger than those in the leg behind the trough. Under the right conditions, such troughs lose identity quickly. Warm air ahead of a following Atlantic system floods forward into the trough, with warming out of the cold air that caused the trough to form, and maintain its shape. Warm the air out in this case, and the trough loses its shape - it relaxes. Whether a trough relaxes or not depends upon the relative position of one trough to another, the exact arrangement of isotherms in the westerly flow, and developments upstream in the wave-train: hence the interconnectivity of synoptic meteorology.

This process can usually be seen as a 'splitting front'. When the trough is sharp, the rain area is usually closely associated with the surface frontal position. As the trough runs ahead and relaxes ("over-runs" the surface front, hence an 'over-running trough'), the upper cloud moves more quickly than the surface front/associated low level cloud, with incursion of dry/descending air over the top of the original front. On satellite imagery, particularly Infra Red (IR), a 'split' can be seen: the high/cold cloud of the upper front (ana-front) bearing the rain shears away eastwards, as the greyer/warmer cloud of the trailing cold front follows slowly behind -- often lying from NE to SW -- slowing further as the mass descent leads to rising pressure, and the gradient across the front dies - this cold front, the one drawn on surface charts, is now a cold kata front.

The dew point/air temperature across the front can be still marked, but all it really marks now is the low level discontinuity. It has no relevance to the upper structure, where the real action is, and which has gone marching away east, or east-northeastwards into the general upper flow. The best way to judge how potentially active a front is is to look at thickness (TTHK) charts.

There are other rules developed over time, before the advent of NWP models, which are still useful. For example, consider the lower-tropospheric flow as characterised by the 700 hPa wind direction. If the flow is between 180 and 220 degrees, larger amounts of rain are likely than if the flow is from 230 or further to the west. This ties in with the relaxing upper trough model: sharp troughs have a high southerly component to the 700 hPa wind; relaxing troughs have an increasingly westerly component.

Why a particular train of troughs relaxes before reaching UK/near continent, or whether such troughs do the reverse, and sharpen, is all tied up with the broad-scale flow, and consideration of the jet patterns: strength, direction, location and variability. These are governed by large-scale patterns of oceanic/continental temperature contrasts, and inter-action of airflow with significant mountain chains: e.g. the Rockies. If we change the temperature patterns of the globe, we will change the factors governing the jet stream, and perhaps change the distribution of mid-latitude rainfall; but that would be the subject of another missive.

In passing though, the southeast of England has always suffered from highly variable rainfall, and analysis of fronts crossing the UK shows that of the two types (ana/active and kata/weak), the latter is (and perhaps always has been), most common. This may have something to do with frictional effects as fronts try to cross the land mass of the British Isles -- leading to disassociation of the lower and upper frontal structures. Certainly split fronts account for more occasions of the cold frontal type than many people (including meteorologists) realise ... its just that you don't hear about them on the BBC!