Gedday,

As you presumably all know, we have had a cut-off low and split jet configuration sitting over Southeastern Australia over the past week, giving us generally cool, wet-weather with occasional very heavy widespread convective rain.

I was curious as to how this situation was handled by GASP ensemble prediction system, available as a spaghetti diagram plus in other forms via the NMOC chart discussion page.

Choosing last night's analysis as the base analysis, you'll see the surface low has drifted eastward into the Tasman and that we still have a barocline across eastern Australia at about 30S, with a convoluted cutoff extending from the circumpolar vortex up over southeastern Australia.
 

This cutoff pattern can be seen in the current ensemble-spaghetti presentation as shown on the current _0000 hrs ensemble map

I was curious as to how well this convoluted pattern was captured by the ensemble system; so I went to the ensemble archive and downloaded the spaghettis valid at this verifying time.  It is quite interesting. I have put links to a number of them on my web-page.

spaghetti forecasts: +12   +24   +36    +48 +60    +72   +84    +96   +120    +132   +144    +156   +168
 

Going through the sequence, you'll see that from two days ago (the +48 forecast) virtually every member of the ensemble had the 500 hPa cutoff pattern over us.  Extending further back (e.g +84), this still true.
At +120 hours or 5 days about half of the ensemble members have the cutoff, as shown below:

If you jump to +168 or seven days, only about 10 percent of the ensemble had the cut-off; but looking carefully at the spaghetti figure, essentially all of the ensemble had signs of a convoluted/blocking type pattern with low-latitude trough/high-latitude-ridge/low-latitude trough pattern across our longitudes.

The reason for this seems quite simple.  If you look at the starting analysis for the 7-day prog (below), the cutoff low is already there over South-eastern Oz.  So... once you have a block, or highly convoluted pattern like this... it is a stable pattern; and the whole ensemble maintains it... through seven days.

 Hmmm.... I wrote this.... then I wondered, "Is it true?:

 Going back to the ensemble archive page, http://gale.ho.bom.gov.au/bm/internal/daas/ensemble2/index.html,

you can pull up graphs of the spread of the ensemble through the period of the forecast.
So, below, I put the spread chart at 500 hPa for the Australian region  for the current situation (left) versus the equivalent chart for the beginning of the blocking forecast we just discussed (base time 12 Z on the 25th... right-hand chart):  AS you can see they're quite different.

There is no careful analysis here.... just thinking out loud, in the hope that others also will look at the ensemble charts, think aloud and we'll all eventually get more out of them.  Do the ensemble boys out there Mike Naughton and co) keep a time series of the spread; so we can pick the situations where the spread is small and when it is large?

cheers

John McB