"Len Wood" wrote in message
...
On Wednesday, 24 February 2016 12:24:17 UTC, Norman Lynagh wrote:
On Wed, 24 Feb 2016 03:55:46 -0800 (PST),
wrote:
I was just thinking about why a strike by the Met Office is always doomed
to failure before it starts, and that's because so many of the various
Met Office systems are fully automated, and as long as they don't fail,
either through a hardware or software fault they will continue to run,
possibly indefinitely barring a complete comms failure.
It's true operational staff add value to products that are already
produced by the various NWP models, but I also imagine that a lot of
forecast data is transmitted without it ever being seen by anyone, let
alone a forecaster. To get the maximum effect from a 24 hour strike these
operational systems would need be to shut down before the staff left the
building and somehow I can't see that happening.
Years ago when we went on strike the temperature and rainfall wasn't read
at 0900 and 2100 on the day of the strike, but now we have AWS, so the
observational and climate data is safe.
From home I can see weather radar images and satellite imagery from the
Met Office web site, I can see observational and forecast data from the
Met Office DataPoint site, I can see UK SYNOPs courtesy of FSU & OGIMET,
I can see the latest TAFs & METARs courtesy of the ADDS service. BBC TV
weather is probably showing me the latest NWP on TV. In comparison to
what a lot of the outstations received in the way of data in the 1970's
and 1980's that's a wealth of information enough to run a forecast
office.
I think that all the strike will achieve is highlight to the public that
the Met Office computer systems that are in place are so resilient that
they hardly need people at all, so why do we need so many? Obviously some
systems such as gale warnings and alerts aren't fully automatic at the
moment, but I should imagine work on automating these systems are in the
pipe-line.
In the foyer of the Met Office in a few years time, you will probably
find a plaque above the door that reads:
"Will the last person to leave the Met Office please turn out the
lights - but for God's sake don't turn off the servers or the
super-computer!"
ECMWF runs with almost no operational staff. The model just keeps on doing
its
thing with only a systems minder keeping an eye on it. Most of the staff
there
are involved in developing model upgrades.
--
Norman Lynagh
Tideswell, Derbyshire
303m a.s.l.
http://peakdistrictweather.org
--------------------------------------------------------------------------
If the hourly forecast for today is correct, then job done.
No need for weather savvy staff.
This is UKMO philosophy.
Trouble is we know on this ng that the NWP cocks up sometimes.
And the general public are more aware of forecast cock ups these days with
the app on their smart phones.
So the research boys and girls better get it right in the next few years and
not just blindly think that more computing power will improve the forecasts.
+++++++++++++++++++++++++
They will Len. NWP has made great advances over the years and will continue
to do so. The 1.5Km model will seem very coarse in the 2020s when models
will become linked. There will be environmental models linked to atmosphere
and ocean models. Atmosphere/ocean are already linked. River flows will be
modelled automatically and account will be taken directly of sea spray on
visibility with no need for crude post-processing algorithms. Modelling is
the future and it will do it all with increasing accuracy, I have no doubt
whatsoever about that. Observation power will also increase with arrays of
satellites operating at ultra small wavelengths and at very high vertical
resolution. No time scale on all this but perhaps in our now restricted
lifetimes Len. Our grandchildren will know the weather 24 hours ahead to an
astonishing accuracy!
Will
--
" Some sects believe that the world was created 5000 years ago. Another sect
believes that it was created in 1910 "
http://www.lyneside.demon.co.uk/Hayt...antage_Pro.htm
Will Hand (Haytor, Devon, 1017 feet asl)
---------------------------------------------