I see that UKMO still don't have the algorithms for the weather warnings
totally debugged as yet.

Tomorrow for Ely (N of Cambridge) there's apparently a yellow warning for
rain (thunderstorm presumably), though the yellow region actually stops some
30-40 miles South of here.

And at 1300 there's 5% probability of rain and no warning, whereas at 1600
there's still 5% probability of rain but a yellow warning. (Perhaps the
probability even at 1600 is really 5% but if the event were to happen then
it would be of ark-requiring proportions?)

Actually I'm not criticising - I'm not sure I could do any better, but it's
interesting that when you start to look at the detail of eg the forecast and
associated warnings the system is clearly still a work in progress.

On May 27, 9:42*pm, "johnd" wrote:
I see that UKMO still don't have the algorithms for the weather warnings
totally debugged as yet.

Tomorrow for Ely (N of Cambridge) there's apparently a yellow warning for
rain (thunderstorm presumably), though the yellow region actually stops some
30-40 miles South of here.

And at 1300 there's 5% probability of rain and no warning, whereas at 1600
there's still 5% probability of rain but a yellow warning. (Perhaps the
probability even at 1600 is really 5% but if the event were to happen then
it would be of ark-requiring proportions?)

===================

That's just what it is. Isolated heavy showers would be the order of
the day this afternoon - so it's quite possible to have only a 5% risk
but for any shower to occur to be particularly heavy (thunderstorms
indeed). And any showers are more likely to be heavy at 16:00 than
13:00, even though the probability of any shower is the same. Your
region ought to be close to the convergence line. And there's little
wind, so if there is a downpour it might be pretty slow to move on.

It doesn't seem unreasonable to me.


Stephen.

"Stephen Davenport" wrote in message
news:1e20c282-2566-4edf-a1e5-
It doesn't seem unreasonable to me.

Very possibly not. Not complaining. as I said, but now that the forecasts
and warnings are more granular I'm taking a little more interest in them.
And there's obviously a process of education as to how to interpret them. (A
good oppportunity I would have thought for UKMO to have created eg a
10-minute video explaining the warning system, but no sign of any such
links, unless I've missed them.)

It's interesting for example that an event at 5% probability is deemed
worthy of a warning. My personal interpretation of 95% probability (ie the
inverse) is 'beyond reasonable doubt'. So, 'beyond reasonable doubt', I can
feel confident that I will _not_ encounter a warning-worthy event this
afternoon. Again, this is not criticism, just trying to reach a personal
calibration of what the warnings might actually mean in reality.

johnd wrote:

"Stephen Davenport" wrote in message
news:1e20c282-2566-4edf-a1e5-
It doesn't seem unreasonable to me.

Very possibly not. Not complaining. as I said, but now that the forecasts and
warnings are more granular I'm taking a little more interest in them. And
there's obviously a process of education as to how to interpret them. (A good
oppportunity I would have thought for UKMO to have created eg a 10-minute
video explaining the warning system, but no sign of any such links, unless
I've missed them.)

It's interesting for example that an event at 5% probability is deemed
worthy of a warning. My personal interpretation of 95% probability (ie the
inverse) is 'beyond reasonable doubt'. So, 'beyond reasonable doubt', I can
feel confident that I will not encounter a warning-worthy event this
afternoon. Again, this is not criticism, just trying to reach a personal
calibration of what the warnings might actually mean in reality.

I certainly agree with the point that you're making. It's not easy to deal with
high impact, low probability events is a way that's easily understandable by
Joe Public. As you rightly point out, in the case that you have highlighted the
thinking is that there is an overwhelming probability that the event will not
occur. Perhaps it would be helpful of the text of the warnings explicitly made
that point in such cases.

If warnings were issued every time a warning threshold had less than 5%
probability, but more than 0% probability of being reached we would be back to
the bad old days of being swamped with warnings.

--
Norman Lynagh
Tideswell, Derbyshire
303m a.s.l.

On May 28, 11:18*am, "Norman" wrote:
johnd wrote:
"Stephen Davenport" wrote in message
news:1e20c282-2566-4edf-a1e5-
It doesn't seem unreasonable to me.

Very possibly not. Not complaining. as I said, but now that the forecasts and
warnings are more granular I'm taking a little more interest in them. And
there's obviously a process of education as to how to interpret them. (A good
oppportunity I would have thought for UKMO to have created eg a 10-minute
video explaining the warning system, but no sign of any such links, unless
I've missed them.)

It's interesting for example that an event at 5% probability is deemed
worthy of a warning. My personal interpretation of 95% probability (ie the
inverse) is 'beyond reasonable doubt'. So, 'beyond reasonable doubt', I can
feel confident *that I will not encounter a warning-worthy event this
afternoon. Again, this is not criticism, just trying to reach a personal
calibration of what the warnings might actually mean in reality.

I certainly agree with the point that you're making. It's not easy to deal with
high impact, low probability events is a way that's easily understandable by
Joe Public. As you rightly point out, in the case that you have highlighted the
thinking is that there is an overwhelming probability that the event will not
occur. Perhaps it would be helpful of the text of the warnings explicitly made
that point in such cases.

If warnings were issued every time a warning threshold had less than 5%
probability, but more than 0% probability of being reached we would be back to
the bad old days of being swamped with warnings.

--
Norman Lynagh
Tideswell, Derbyshire
303m a.s.l.

=======================

Agreed. A graphic does not tell the whole story and explanatory text
would be welcome.

John - a yellow warning is warranted for low probability / high impact
events such as this according to the Met Office warning matrix, found
he

http://www.metoffice.gov.uk/weather/.../warnings.html

Without detailed text, though, one needs to interpret that as low
probability but high impact through noting the 5% probability
(although I note a brief text today).



Stephen.

Actually, hovering over the 'More details' link brings up the Chief
Forecaster's assessment as well.

Stephen.

"Norman" wrote in message
...
johnd wrote:

"Stephen Davenport" wrote in message
news:1e20c282-2566-4edf-a1e5-
It doesn't seem unreasonable to me.

Very possibly not. Not complaining. as I said, but now that the forecasts
and
warnings are more granular I'm taking a little more interest in them. And
there's obviously a process of education as to how to interpret them. (A
good
oppportunity I would have thought for UKMO to have created eg a 10-minute
video explaining the warning system, but no sign of any such links,
unless
I've missed them.)

It's interesting for example that an event at 5% probability is deemed
worthy of a warning. My personal interpretation of 95% probability (ie
the
inverse) is 'beyond reasonable doubt'. So, 'beyond reasonable doubt', I
can
feel confident that I will not encounter a warning-worthy event this
afternoon. Again, this is not criticism, just trying to reach a personal
calibration of what the warnings might actually mean in reality.

I certainly agree with the point that you're making. It's not easy to deal
with
high impact, low probability events is a way that's easily understandable
by
Joe Public. As you rightly point out, in the case that you have
highlighted the
thinking is that there is an overwhelming probability that the event will
not
occur. Perhaps it would be helpful of the text of the warnings explicitly
made
that point in such cases.

Well this morning I was interpreting what was said in the
forecast, showers were considered to be possible from the SE up to
Pennine areas. Unlikely for any given place to get one, but if
you did it would likely be heavy. i.e. 'low probability, high impact'
On a practical level this triggered a 'take brolly to work just in
case' action
I was unlikely to get wet, but If I did I would probably get soaked
especially as I'm hardly going to be wearing a coat in these temperatures.
As it happened there was some cloud build up in the afternoon,
now dispersed. But it *might* have happened.
--
Col

Bolton, Lancashire
160m asl

On May 28, 6:40*pm, "Col" wrote:
"Norman" wrote in message

...









johnd wrote:

"Stephen Davenport" wrote in message
news:1e20c282-2566-4edf-a1e5-
It doesn't seem unreasonable to me.

Very possibly not. Not complaining. as I said, but now that the forecasts
and
warnings are more granular I'm taking a little more interest in them. And
there's obviously a process of education as to how to interpret them. (A
good
oppportunity I would have thought for UKMO to have created eg a 10-minute
video explaining the warning system, but no sign of any such links,
unless
I've missed them.)

It's interesting for example that an event at 5% probability is deemed
worthy of a warning. My personal interpretation of 95% probability (ie
the
inverse) is 'beyond reasonable doubt'. So, 'beyond reasonable doubt', I
can
feel confident *that I will not encounter a warning-worthy event this
afternoon. Again, this is not criticism, just trying to reach a personal
calibration of what the warnings might actually mean in reality.

I certainly agree with the point that you're making. It's not easy to deal
with
high impact, low probability events is a way that's easily understandable
by
Joe Public. As you rightly point out, in the case that you have
highlighted the
thinking is that there is an overwhelming probability that the event will
not
occur. Perhaps it would be helpful of the text of the warnings explicitly
made
that point in such cases.

Well this morning I was interpreting what was said in the
forecast, showers were considered to be possible from the SE up to
Pennine areas. Unlikely for any given place to get one, but if
you did it would likely be heavy. i.e. 'low probability, high impact'
On a practical level this triggered a 'take brolly to work just in
case' action
I was unlikely to get wet, but If I did I would probably get soaked
especially as I'm hardly going to be wearing a coat in these temperatures..
As it happened there was some cloud build up in the afternoon,
now dispersed. But it *might* have happened.
--
Col

Bolton, Lancashire
160m asl

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

And it did happen in this part of SE England. There was a thunderstorm
with torrential rain over Rochester, Kent, at about 16:30.

Stephen.

Stephen Davenport wrote:

On May 28, 6:40*pm, "Col" wrote:
"Norman" wrote in message

...









johnd wrote:

"Stephen Davenport" wrote in message
news:1e20c282-2566-4edf-a1e5-
It doesn't seem unreasonable to me.

Very possibly not. Not complaining. as I said, but now that the forecasts
and
warnings are more granular I'm taking a little more interest in them. And
there's obviously a process of education as to how to interpret them. (A
good
oppportunity I would have thought for UKMO to have created eg a 10-minute
video explaining the warning system, but no sign of any such links,
unless
I've missed them.)

It's interesting for example that an event at 5% probability is deemed
worthy of a warning. My personal interpretation of 95% probability (ie
the
inverse) is 'beyond reasonable doubt'. So, 'beyond reasonable doubt', I
can
feel confident *that I will not encounter a warning-worthy event this
afternoon. Again, this is not criticism, just trying to reach a personal
calibration of what the warnings might actually mean in reality.

I certainly agree with the point that you're making. It's not easy to deal
with
high impact, low probability events is a way that's easily understandable
by
Joe Public. As you rightly point out, in the case that you have
highlighted the
thinking is that there is an overwhelming probability that the event will
not
occur. Perhaps it would be helpful of the text of the warnings explicitly
made
that point in such cases.

Well this morning I was interpreting what was said in the
forecast, showers were considered to be possible from the SE up to
Pennine areas. Unlikely for any given place to get one, but if
you did it would likely be heavy. i.e. 'low probability, high impact'
On a practical level this triggered a 'take brolly to work just in
case' action
I was unlikely to get wet, but If I did I would probably get soaked
especially as I'm hardly going to be wearing a coat in these temperatures.
As it happened there was some cloud build up in the afternoon,
now dispersed. But it might have happened.
--
Col

Bolton, Lancashire
160m asl

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

And it did happen in this part of SE England. There was a thunderstorm
with torrential rain over Rochester, Kent, at about 16:30.

Stephen.

.....and, of course, here is the real problem. It did happen in Rochester but it
didn't in Bolton. In that situation, what is the real value of a forecast that
says it might happen in both places. With due deference to the late Prof. Allan
Murphy - a weather forecast has no value in itself. It gains value through its
ability to influence the decisions made by the users of the forecast. Perhaps
Col took a sensible precaution by carrying an umbrella. But how many people
were caught out unprepared in Rochester?

I'm very pleased that the science is still a very long way from being able to
say, with certainty, that today there will be a thunderstorm in Rochester but
there won't be one in Bolton. Heaven forbid that we should ever reach that
level of capability

--
Norman Lynagh
Tideswell, Derbyshire
303m a.s.l.

On May 28, 9:38*pm, "Norman" wrote:

I'm very pleased that the science is still a very long way from being able to
say, with certainty, that today there will be a thunderstorm in Rochester but
there won't be one in Bolton. Heaven forbid that we should ever reach that
level of capability

--
Norman Lynagh
Tideswell, Derbyshire
303m a.s.l.- Hide quoted text -

- Show quoted text -

What a dreadful, subversive thing for a professional
weatherman to say, Norman. Well done - I agree entirely.

Tudor Hughes, Warlingham, Surrey.