| |
The Storms of October 2000
|
|
|
Following a very unsettled and wet month, meteorological events
took a turn for the worse towards the close of October 2000. As a
precursor to the main events which affected the British Isles, a
major cyclogenesis event (i.e. formation of a depression) took place
on 26th well to the northwest of the UK, resulting in a deep low
anchored between Greenland and Iceland for the next two days. The
associated weather fronts produced some mild, breezy and
rather wet weather across the UK on 27th, but largely cleared the
country by the end of the day. However the cold front stretched its
way well to the the southwest, and became the "birthplace" for the
first in a series of potent wave depressions which tracked across
the UK during the period Saturday 28th to Monday 30th October. The
parent low and its trailing cold front are vividly depicted on the
Atlantic infrared satellite
image for 0600Z 27th October.
|
|
The first of the "daughter" depressions, or wave depressions as
they are known in the trade, deepened quickly as it crossed the
country during the 28th. The return of the milder "tropical
maritime" air was accompanied by a period of heavy rain in many
places, with up to 8mm being reported in an hour across parts of
the Home Counties. This rainfall is picked out nicely by this
radar image valid at
0700Z, and was forecast quite well by the Met Office's
mesoscale model 19 hours
in advance. The mesoscale model has a resolution of 11km, and
provides a major input to UK local forecasts. During the afternoon
the cold front swept east across Wales and England, accompanied by
gale force winds in places, as the depression continued to deepen.
The infrared satellite image
for 1800Z shows the low in the act of developing. A tornado was
reported from Bognor Regis shortly before 1700Z, resulting in
local severe damage. The heavy rain associated with the front is
illustrated in the radar
picture for 1600Z. Again the
mesoscale model run from
midday on 27th, (at 28 hour range), provided good steer for this
feature. Infact NWP model guidance proved very useful even
before this - the charts issued by our forecasters in our National
Meteorological Centre (NMC) have a large degree of input from
the numerical models, and that issued
early on the 27th was a very successful preview.
The centre of the depression turned towards Northern Scotland, and
brought a spell of gale force winds to much of the north during the
next 12 to 24 hours. The cold front had by now swept into the North
Sea, followed by clearer, chilly conditions for most of the UK
overnight Saturday, and into the first part of Sunday. Indeed many
woke to bountiful sunshine and blue skies on the 29th, but showers
had continued all night close to some western and southern coasts,
thunder being reported from eastern parts of the English Channel
coast. A few showers spread inland during the morning, and it was
cold enough for a little snow to be mixed in over parts of northern
Britain.
|
|
But out to the southwest the situation was already worsening, as a
duo of wave depressions hurtled in from the Atlantic. The
visible satellite image
for 1200Z shows most of the country still in the reasonably fair
weather, although with "dappled" patterns indicative of showers over
some northern areas. But cloud is already encroaching on the
southwest: depression no. 2 is forming over Eire, and no. 3, which
will become the big one, is somewhere near 52N, 30W. The veil of
upper cloud ahead of no. 2 spread very quickly northeast across
Wales and all but the far north of England by mid-afternoon on
the 29th. Rain set in soon afterwards, becoming steadily heavier
and more persistent as dusk approached. By 17Z heavy rain had
spread across much of the southern part of the country as shown
by radar imagery -
and forecast quite successfully by models at the Met Office. The
30 hour forecast from the
preliminary
global model, which provides a primary tool for producing
aviation forecasts amongst other products, depicts the heavy
rainfall and shows the tightly spaced isobars forming a trough
over southern Britain. Even two-and-a-half days before this
storm reached southern Britain, the computer models gave good
warning - as shown in the
chart issued by
NMC forecasters just after midnight on 27th, valid at midday on
29th.
As would be envisaged from the pressure pattern, a period of
strong winds accompanied the rain, with some of the strongest
gusts experienced across Wales and southwest England e.g.
76 kts (87mph) at North Hessary Tor on Dartmoor. The centre of
the depression crossed Wales and the Midlands before heading into
the southern North Sea, still deepening apace. But worse was to
follow. After a temporary lull across most of the south during the
evening, the next wave depression was centred off southwest Ireland
by midnight, and all the signs were that this would be the most
damaging of the three.
|
|
Rain continued to fall across southern areas, a result of the
contrast between mild air close to the south coast and the much
colder conditions across the north, where temperatures in many
parts had failed to reach double figures during Sunday. The
radar picture for 0000Z
on the 30th is typical, and again this was predicted rather
well by the mesoscale model
30 hours in advance. The wind, too, continued to blow strongly
close to the south coast, with only the slightest of lulls
apparent before strengths began once more to increase.
During the second half of the night the new and rapidly deepening
depression crossed Ireland, the Irish Sea, Wales and then on into
southern parts of Northern England. Forecasts for this event were
almost consistently good: the
two-day forecast from the preliminary global model valid at
0600Z on the 30th predicted the intensifying depression not far
from its actual position, and although rather less deep than
actuality, was able to forewarn of widespread gusts between 70
and 90mph over southern areas of the UK. The
12-hour forecast from the
same model, and valid at the same time, was even more successful.
In between these two forecasts there was some weakening of the
forecast centre, and a related tendency to keep it a little further
south, as shown by the
mesoscale model forecast one day ahead, again valid at 0600Z. But
in such cases the forecaster is able to use continuity, as well
as output from models run at other centres, to make the best
judgment on the outcome. And in this case NMC forecasters
maintained warnings throughout. Indeed warnings for this storm
were issued as early as the start of Friday, as shown by the
forecaster interpretation of model output issued at the
three-day range.
As is most often the case the strongest winds were experienced
to the south of the track of the centre, whilst to the north, most
of Scotland, Northern Ireland and parts of northern England
escaped with much calmer conditions. Wind speeds reached a peak in
the early hours along the South Wales coast, touching 84kts
(96mph) at Mumbles. During the next few hours the strongest winds
transferred east, The Isle of Portland in Dorset reporting a gust
to 81kts (93mph) at 0600Z. The strongest winds accompanied the
cold front which reintroduced cooler conditions once more,
passing through London between 0600 and 0700Z, and clearing into
the North Sea by around 0800Z. Langdon Bay near Dover registered
a gust of 76kts (87mph) between 0600 and 0700Z, and by 1000Z
this site had been reporting gusts to 60kts (69mph) or more for
16 hours! At inland locations typical gust strengths were
50-70kts (57-80mph). Some structural damage accompanied these
gusts, with trees brought down and roads and railways blocked.
The saturated ground would not have helped as far as the trees
were concerned, and many trees were still bearing enough leaves
to significantly increase the impact of the gusts. The spatial
variation of the gusts experienced over a 15-hour period ending
at 0900Z on 30th is shown on this
map.
Rain had fallen all night, and by 9am 24-hour totals of between
25 and 50mm were commonplace, with locally 75mm and more.
Consequently local flooding was a widespread source of disruption
during the morning rush hour on 30th. In a number of areas recent
rains had already brought the situation close to breaking point,
and these renewed heavy rains resulted in much more serious
flooding problems, in areas as far apart as Kent, Wales and
Yorkshire. The 24 hour rainfall totals are shown on this
isohyet chart,
i.e. lines of equal rainfall amount.
The northwest of the home counties, together with much of the
South Midlands and East Anglia experienced slightly lesser gusts
and rainfall figures, but another area of severe weather
developed as the depression crossed Northern England, immediately
to the south of the centre. The storm was still developing rapidly
at this time with mean sea level pressure at its centre falling
towards 950hPa. The warm and cold fronts formed a sharp apex
between which was located a narrow band of warmer air from further
south. Temperatures at some locations rose some 5 to 7 degrees
within the hour. But as the cold front crossed, the temperature
falls were even more dramatic. The depression had by now developed
a wrap-around of thick cloud and heavy precipitation to the west
and northwest of the centre, as is often observed during this stage
of an intense development. The
0900Z radar image
shows just how intense the rainfall was. This precipitation,
falling into the already chilly air, was able to lower
the temperature still further by evaporative cooling. And so the
precipitation within this band turned first to sleet and then to
snow, and swept eastwards just to the west of the retreating
depression. Locally the temperature was observed to fall as much
as 10 degrees within one hour, as the wind veered and strengthened
and a short spell of quite heavy sleet and snow set in.
Several routes were affected, notably in the Pennines. The
numerical prediction models were poorer in this respect, with
the predicted areal extent of wintry precipitation much less
than was observed - this failing may be linked to a tendency to
underpredict the rate of deepening referred to below. As for the
winds, Lincolnshire bore the brunt of the second area of severe
gusts, with 80kts (91mph) recorded at Donna Nook on the coast
between 0900 and 1000Z, and a sustained wind speed of 51kts
(58mph) at Cranwell at 0800Z.
The potent depression then moved out into the North Sea,
deepening further during the morning, and reaching an estimated
depth close to 941hPa by early afternoon. The intensity of
the centre is graphically depicted by the
15-hour forecast for
the mesoscale model, valid at 1500Z. It is fair to point out
that the scale of the deepening after 0600Z was underpredicted
by most of the model runs. Wind strengths over the sea increased
still further, leading to sustained winds in excess of hurricane
force. The storm finally began to lessen in intensity as it neared
the south of Norway.
|
| How did the NWP models do? |
|
The sequence of events was very well-previewed by the numerical
models at the Met Office from the 26th October onwards. By this
stage the model solutions were providing a consistent
picture of a sequence of potent depressions, and by the end of
26th, the warning flag had been raised as to the potential
severity of the third in the sequence of storms, which had just
appeared in the latest runs. Although there were some differences
between successive runs in the tracks and relative intensities
of the three depressions, each new run confirmed that this
would be a period of exceptionally stormy weather, and greatly
helped the forecasters (who use the model forecasts as their
prime input of information) to issue high quality and
confident warnings of an extreme event.
|
|
|
|