On Wednesday, 16 December 2015 13:55:27 UTC, Dave Cornwell wrote:
"..The GFS model as we know has a tendency to over-develop LP systems
and if it does decrease in intensity in subsequent runs, it may promote
a more rapid pressure build over Norther Scandinavia than currently
projected."
Do we know? Is this feasible?

Often see stuff of this ilk quoted on weather forums but surely the
people at GFS, ECMF etc aren't stupid. If models consistently have a
tendency to predict something incorrectly surely they would just tweak
the algorithms a bit. I thought that was the whole point of a
mathematical model.

Dave

I was quite shocked when Will mentioned a while ago that the models DON'T consider climatology, for example Lamb Weather Types etc. One of Philip Eden's articles a while ago mentioned that if you are faced with a 50/50 situation while looking at a model you go with the 'average' of what happened before. I think that is where the skill of a professional forecaster comes in.

--

------------------------------

*This email was sent by a company owned by Financial Times Group Limited
("FT Group http://aboutus.ft.com/corporate-information/#axzz3rajCSIAt"),
registered office at Number One Southwark Bridge, London SE1 9HL.
Registered in England and Wales with company number 879531. This e-mail may
contain confidential information. If you are not the intended recipient,
please notify the sender immediately, delete all copies and do not
distribute it further. It could also contain personal views which are not
necessarily those of the FT Group. We may monitor outgoing or
incoming emails as permitted by law.*

"Scott W" wrote in message
...
On Wednesday, 16 December 2015 13:55:27 UTC, Dave Cornwell wrote:
"..The GFS model as we know has a tendency to over-develop LP systems
and if it does decrease in intensity in subsequent runs, it may promote
a more rapid pressure build over Norther Scandinavia than currently
projected."
Do we know? Is this feasible?

Often see stuff of this ilk quoted on weather forums but surely the
people at GFS, ECMF etc aren't stupid. If models consistently have a
tendency to predict something incorrectly surely they would just tweak
the algorithms a bit. I thought that was the whole point of a
mathematical model.

Dave

I was quite shocked when Will mentioned a while ago that the models DON'T
consider climatology, for example Lamb Weather Types etc. One of Philip
Eden's articles a while ago mentioned that if you are faced with a 50/50
situation while looking at a model you go with the 'average' of what
happened before. I think that is where the skill of a professional
forecaster comes in.
=======

Yes indeed. And to re-iterate (in a nutshell), NWP models solve
computationally the Navier Stokes equations on an oblique spheroid for the
general dynamics (4D synoptic evolution). Physics and radiation packages are
used to add in detail of vertical profiles of temperature, precipitation,
cloud and humidity at the gridpoints. Post-processing is used to add in
site-specific detail accounting for unresolved features (such as topographic
detail) and for commercial requirements. Ensemble means of synoptic
variables are used to provide a broad overview of expected evolution with
the ensemble members giving an indication of the uncertainty.
Post-processing of ensemble members will give probability of weather at
specific locations. Nowadays, the Met Office has ensembles at 1.5km
resolution covering the UK out to circa 36 hours ahead.

Will
--
http://www.lyneside.demon.co.uk/Hayt...antage_Pro.htm
Will Hand (Haytor, Devon, 1017 feet asl)
---------------------------------------------

On 16/12/2015 17:13, Eskimo Will wrote:

"Scott W" wrote in message
...
On Wednesday, 16 December 2015 13:55:27 UTC, Dave Cornwell wrote:
"..The GFS model as we know has a tendency to over-develop LP systems
and if it does decrease in intensity in subsequent runs, it may promote
a more rapid pressure build over Norther Scandinavia than currently
projected."
Do we know? Is this feasible?

Often see stuff of this ilk quoted on weather forums but surely the
people at GFS, ECMF etc aren't stupid. If models consistently have a
tendency to predict something incorrectly surely they would just tweak
the algorithms a bit. I thought that was the whole point of a
mathematical model.

Dave

I was quite shocked when Will mentioned a while ago that the models
DON'T consider climatology, for example Lamb Weather Types etc. One of
Philip Eden's articles a while ago mentioned that if you are faced with
a 50/50 situation while looking at a model you go with the 'average' of
what happened before. I think that is where the skill of a professional
forecaster comes in.
=======

Yes indeed. And to re-iterate (in a nutshell), NWP models solve
computationally the Navier Stokes equations on an oblique spheroid for
the general dynamics (4D synoptic evolution). Physics and radiation
packages are used to add in detail of vertical profiles of temperature,
precipitation, cloud and humidity at the gridpoints. Post-processing is
used to add in site-specific detail accounting for unresolved features
(such as topographic detail) and for commercial requirements. Ensemble
means of synoptic variables are used to provide a broad overview of
expected evolution with the ensemble members giving an indication of the
uncertainty. Post-processing of ensemble members will give probability
of weather at specific locations. Nowadays, the Met Office has
ensembles at 1.5km resolution covering the UK out to circa 36 hours ahead.

Will
-----------------------------------------------------------------------
..... which would make the original viewpoint (quoted) unlikely in my
opinion.

Thinking about the GFS in particular, as it goes out for a longer period, has anyone measured its accuracy recently.

It just seems to me that it's been a lot less volatile in its predictions, when compared to previous years, with little in the way of eye candy at 300+ for example.

So is this a reflection of improvements in the modelling, or a reflection of the synoptic situation?

On 16/12/2015 17:55, David Mitchell wrote:
Thinking about the GFS in particular, as it goes out for a longer period, has anyone measured its accuracy recently.

It just seems to me that it's been a lot less volatile in its predictions, when compared to previous years, with little in the way of eye candy at 300+ for example.

So is this a reflection of improvements in the modelling, or a reflection of the synoptic situation?

------------------------------------------------------------------------------------------------------------
It could be both I suppose but I have monitored a couple of the longer
term charts and they have been very accurate at 10 days although of
course as you say, things have been pretty stable in terms of patterns.
In fact I'm surprised a certain person hasn't issued a *forecast* during
the last six weeks.
Dave

On Wednesday, 16 December 2015 18:39:04 UTC, Dave Cornwell wrote:
On 16/12/2015 17:55, David Mitchell wrote:
Thinking about the GFS in particular, as it goes out for a longer period, has anyone measured its accuracy recently.

It just seems to me that it's been a lot less volatile in its predictions, when compared to previous years, with little in the way of eye candy at 300+ for example.

So is this a reflection of improvements in the modelling, or a reflection of the synoptic situation?

------------------------------------------------------------------------------------------------------------
It could be both I suppose but I have monitored a couple of the longer
term charts and they have been very accurate at 10 days although of
course as you say, things have been pretty stable in terms of patterns.
In fact I'm surprised a certain person hasn't issued a *forecast* during
the last six weeks.
Dave

Christmas Day has firmed up on last couple of op runs - and I notice that William Hill have lengthened the odds of a white Christmas in London to 8/1 from 5/1


--

------------------------------

*This email was sent by a company owned by Financial Times Group Limited
("FT Group http://aboutus.ft.com/corporate-information/#axzz3rajCSIAt"),
registered office at Number One Southwark Bridge, London SE1 9HL.
Registered in England and Wales with company number 879531. This e-mail may
contain confidential information. If you are not the intended recipient,
please notify the sender immediately, delete all copies and do not
distribute it further. It could also contain personal views which are not
necessarily those of the FT Group. We may monitor outgoing or
incoming emails as permitted by law.*

On Wednesday, December 16, 2015 at 6:39:04 PM UTC, Dave Cornwell wrote:
On 16/12/2015 17:55, David Mitchell wrote:
Thinking about the GFS in particular, as it goes out for a longer period, has anyone measured its accuracy recently.

It just seems to me that it's been a lot less volatile in its predictions, when compared to previous years, with little in the way of eye candy at 300+ for example.

So is this a reflection of improvements in the modelling, or a reflection of the synoptic situation?

------------------------------------------------------------------------------------------------------------
It could be both I suppose but I have monitored a couple of the longer
term charts and they have been very accurate at 10 days although of
course as you say, things have been pretty stable in terms of patterns.
In fact I'm surprised a certain person hasn't issued a *forecast* during
the last six weeks.
Dave

I nearly referred to those forecasts when posting. Ironically, "if" they were a regular daily feature, with less obsession about 80% accuracy, they would be extremely useful in monitoring not just models, but patterns as well, as I've said before.

On Wednesday, December 16, 2015 at 6:39:04 PM UTC, Dave Cornwell wrote:
On 16/12/2015 17:55, David Mitchell wrote:
Thinking about the GFS in particular, as it goes out for a longer period, has anyone measured its accuracy recently.

It just seems to me that it's been a lot less volatile in its predictions, when compared to previous years, with little in the way of eye candy at 300+ for example.

So is this a reflection of improvements in the modelling, or a reflection of the synoptic situation?

------------------------------------------------------------------------------------------------------------
It could be both I suppose but I have monitored a couple of the longer
term charts and they have been very accurate at 10 days although of
course as you say, things have been pretty stable in terms of patterns.
In fact I'm surprised a certain person hasn't issued a *forecast* during
the last six weeks.
Dave

Con't be bothered. Looking for the change.