I saw a mention here of an event where 45mm fell in ten minutes. This
got me curious; I've been in a similar downpour here in 2000, but had
no gauge to document it. (The one measuring device I had was one of
my cats, who started sprinting the 20 yards from the back gate to the
catflap when it started, and was as wet as a fluffy cat can possibly
get by the time he made it). It was probably the most violent downpour
I've ever seen, and I've been in flash storms in places prone to far
heavier ones than the UK (New Zealand, the eastern US, Turkey).

Is it possible to detect such events by radar, or do we just have to
hope that one happens over a gauge? - given how localized these things
are, the latter would be like adding an asteroid impact counter to your
weather station.

Do we know anything about the processes that produce them?

======== Email to "j-c" at this site; email to "bogus" will bounce ========
Jack Campin: 11 Third Street, Newtongrange, Midlothian EH22 4PU; 0131 6604760
http://www.purr.demon.co.uk/purrhome.html food intolerance data & recipes,
Mac logic fonts, Scots traditional music files and CD-ROMs of Scottish music.

On Wed, 08 Oct 2003 21:35:01 +0000 (GMT), Rodney Blackall
wrote:

In article ,
bogus address wrote:

Is it possible to detect such events by radar, or do we just have to
hope that one happens over a gauge?

It depends on the resolution of the display. 5 km square pixels will
average out "cloudbursts" to 64-128 mm/hr but higher resolution displays,
e.g. 2 km square will be better. However there is another problem; how
many intensity levels are displayed and the colour contrasts used so that
the human eye notices the one extreme pixel out of perhaps 528x528.

I've always preferred a scheme with only eight colours but recycling them
repeatedly with white marking the scale change. A round, violent storm
would look like a slice of Swiss roll.

Looking at the best of the free radars
http://www.metoffice.com/weather/europe/uk/radar/ ,
I see there are 7 colours with a maximum rainfall rate of 16 mm/hour.
The images are small, at 300x300 pixels but even the larger Met Office
subscription images (500x500, 15 min radar) use the same colour key
and I find it hard to pick out individual pixels. Areas of heavy rain
are, however, clearly shown and the images can then be enlarged to
show individual pixels better, in an image editor like Paint Shop Pro.

The best publicly available radar for analyzing cloudbursts, IMHO, is
the Avbrief subscription service which has static images every 15
minutes at 1184 x 1184 pixels They do cover a larger land and sea area
than the MetO images and would be 800x800 if cropped to match. Avbrief
use a 9 colour key, with a highest rainfall rate of 32mm/hour
coloured white. Using a 17 inch monitor at 1024 x 768, I can often
pick out individual pixels coloured purple (16mm/hr) or white (32)
without enlarging the avbrief images. The animation image sizes are a
bit smaller... but you'd never spot individual pixels in an animation.

Presumably it is possible to programme the display software to give some
sort of alarm (flashing pixels?) when a danger threshold is passed.

I'm sure it is, Rodney (but not by me!). With the larger images, it's
easy to spot the downpours anyway.

--
Dave

"Dave Ludlow" wrote in message
...
On Wed, 08 Oct 2003 21:35:01 +0000 (GMT), Rodney Blackall
wrote:
snip
The best publicly available radar for analyzing cloudbursts, IMHO, is
the Avbrief subscription service which has static images every 15
minutes at 1184 x 1184 pixels They do cover a larger land and sea area
than the MetO images
snip

Dave, don't forget the European composite radar image on the MetO aviation
service which in theory should show everything. Although I must admit the
way it's displayed on the web is a bit ropey.

Cheers,
Jon.