On Friday, July 26, 2013 2:24:28 PM UTC+1, Richard Dixon wrote:

These aren't statistical forecasts though are they? They're realisations of ensemble members of GloSea5, or that's at least what I thought - so they'll have their own teleconnection patterns within them but any common biases that develop within them that go against reality will cock the lot up.

=======

Yes, I realise that. Or at least that is also my belief. But to what extent *do* they account for teleconnections? Do they apply lag? And so forth.

I'm more concerned that the text forecast is, without wishing to sound rude, a little glib. But no doubt you get more bang for your buck if you are a paying client.

Stephen.

On Friday, 26 July 2013 14:47:43 UTC+1, Stephen Davenport wrote:

Yes, I realise that. Or at least that is also my belief. But to what extent *do* they account for teleconnections? Do they apply lag? And so forth.

I'm assuming that these are tailored snippets from the aforementioned GloSea5 model so given that its global will have at least a representation of teleconnections - although how teleconnections respond dependent on model resolution could have an influence maybe?

To me spread of the forecast (as per July which was just normal spread about the mean which was very similar to climatology) if anything implied that the weather for the month simply wasn't going to have any skill - it could be warm, it could be cold. It's when you start to see clustering in the ensembles that point to specific synoptic setups that things get interesting.

e.g. - the MO have already shown forecast skill for a couple of the anomalously cold winters where the influence of the stratosphere on our weather is - at least that's my contention - more relevant. Given the relatively less dynamic/kinematic large scale conditions in summer, I'd reckon that it's rather a thankless task trying to make sense of monthly forecasts in these months.

Richard

On Friday, July 26, 2013 3:06:38 PM UTC+1, Richard Dixon wrote:
Given the relatively less dynamic/kinematic large scale conditions in summer, I'd reckon that it's rather a thankless task trying to make sense of monthly forecasts in these months.

==========

This is a is a very true thing to say indeed! Summer is a bugger, and I refer you back to the near uselessness of a weak ENSO teleconnection, and my puzzlement as to why the text cited that as part reasoning for the July forecast. But your penultimate paragraph raises a relevant point as well.

There is one strong teleconnection for this summer, though, which may (or may not!) have helped me out; and a couple of really good analogues. The dynamics have been very close so far this year to a couple of particular years; and cutting through the complexities of the teleconnection analogues has been a help. TheNH sset-up in late spring gave some decnt clues (as they might for my late autumn "guess" - but that's another story!) Sometimes you have to get rid of the trees to see the wood.

Stephen.

On Friday, 26 July 2013 15:26:11 UTC+1, Stephen Davenport wrote:

There is one strong teleconnection for this summer, though, which may (or may not!) have helped me out; and a couple of really good analogues. The dynamics have been very close so far this year to a couple of particular years; and cutting through the complexities of the teleconnection analogues has been a help.

How do you define an "anaologue" in your work? Similar years with similar indices or similar two-dimensional analogue structures?

Richard

On Sunday, July 28, 2013 1:32:54 AM UTC+1, Richard Dixon wrote:

How do you define an "anaologue" in your work? Similar years with similar indices or similar two-dimensional analogue structures?

========

A bit of both.

For the former we are developing some good tools for combining and weighting indices. The problem is climatology, of course. Sometimes they can filter to only a few useful years, especially when you consider that stratospheric signals, for example, are not so discernible before the satellite age.

Stephen.

On Friday, 26 July 2013 15:26:11 UTC+1, Stephen Davenport wrote:
On Friday, July 26, 2013 3:06:38 PM UTC+1, Richard Dixon wrote: Given the relatively less dynamic/kinematic large scale conditions in summer, I'd reckon that it's rather a thankless task trying to make sense of monthly forecasts in these months. ========== This is a is a very true thing to say indeed! Summer is a bugger, and I refer you back to the near uselessness of a weak ENSO teleconnection, and my puzzlement as to why the text cited that as part reasoning for the July forecast. But your penultimate paragraph raises a relevant point as well. There is one strong teleconnection for this summer, though, which may (or may not!) have helped me out; and a couple of really good analogues. The dynamics have been very close so far this year to a couple of particular years; and cutting through the complexities of the teleconnection analogues has been a help. TheNH sset-up in late spring gave some decnt clues (as they might for my late autumn "guess" - but that's another story!) Sometimes you have to get rid of the trees to see the wood. Stephen.

Isn't that a carbon isuue?