On Dec 8, 11:42*am, Richard Dixon wrote:
On Dec 8, 11:24*am, Teignmouth wrote:

Dawlish you are always ramping global warming with your hottest month
or 3rd hottest month on record etc etc etc, or is it your twin brother
posting on your behalf?

It's nice to have the other side of the fence view now and again.

My two-penneth. Was lucky enough to see Tim Palmer talk at a meeting a
month or so ago now. Probably the best weather/climate speaker there
is IMHO. Anyhow - he showed a very nice couple of slides that counter
the oft-quoted argument: "If we can't forecast the weather beyond
about 5 days, how can we forecast the climate"? He went on to show the
ideas of chaos, attractors, and how the atmosphere flips between
various states. His point is that chaos won't necessarily influence
longer-term trends and forecasts, even though it is inherent in the
results. His point is that the flip-flop of chaos into various states
will change and thus the atmosphere under climate change is likely to
remain in one of the states more often than pre-climate change.

The way I see this is that, although we might, for example flip to
more hot summers, this doesn't rule out the "flop" of winters like
this one (or the one last year). I also think personally there are
shorter-timescale modulations that lead to what we've seen in the past
3-5 years of quieter winters that modulate this change.

I also believe that there is more value in seasonal forecasts than
some place on here - which is why so much research goes into them. As
per an earlier post, a vastly improved stratospheric resolution has
improved the longer-term forecasts the Met Office provide, so it
doesn't surprise me that they stuck their neck out in their month
ahead forecast on the web going back way when if they saw a strong
signal for cold. Much like the climate modelling flip/flop comments
related to Tim Palmer's talk above - I believe chaos ruins the chances
of forecasting minutiae in the short-to-medium range up to 7-10 days,
but not the chance of forecasting *trends* over seasons.

Richard

It's interesting. I'd agree with your comments from your listening to
Tim Palmer. Chaos certainly doesn't ruin the chances of forecasting
*trends* over seasons. Be nice to look back in 20 years and see if the
factoring in of the improved stratospheric resolution improves longer-
term forecasting. I hope it does and it is good that lots of research
is seemingly occurring. I hope the MetO, in 20 years time, is much
more open about it's forecasting accuracy at all ranges too. It has to
be remembered that the MetO leads the world in this and still gets
seasonal forecasts hopelessly and publically (until last year) wrong
on a regular basis - hence no seasonal forecasts in public from the
MetO any more.

It's been a source of frustration that seasonal forecasting accuracy
has not improved a great deal over the last 20 years, nor over the 20
years before that. I remember listening to the Chief Forecaster from
the MetO. back in 1982, in notloB (OK, Monty Python refs are in
today!), strangely enough, talking about his confidence that we would
be able to forecast what the seasons ahead would be like with real
accuracy, by the early years of the 21st century and him also being
frustrated about the slow pace of improvement in seasonal forecast
accuracy over the previous few decades.