Where have the long fetch northerlies gone?

[Snowyowl9]

14 hours ago, Optimus Prime said:

It's a long fetch northerly, but it isn't particularly cold; even for November/December.

January 1969, the king of northerlies?  

Give it time the cold air will upgrade if it carries on from the north,looking like a 10 day cold spell,apart from a few milder blips Monday.

[Thundery wintry showers]

On 11/9/2017 at 14:21, Interitus said:

An interesting topic, but the first thing to note is that, contrary to the points made so far, the changes to the Arctic sea ice and temperatures will have probably only meant minor differences to the potency of northerlies for the UK. The reason for this is that research into what are known as cold air outbreaks (CAO) or  marine cold air outbreaks (MCAO) shows that the cold boundary layer over the ice is transformed very rapidly into a modified boundary layer over the sea, within just a few hundred miles.

For example data from plane flights for ARKTIS field experiments showed that -

 observed boundary-layer modifications were very variable ranging from 100 to 300 m thick boundary layers with air temperatures between -32 and -22 °C over the ice to thicknesses between 900 and 2200 m and air temperatures between -15 and -5 °C after 300 km fetch over the open water (Brummer 1996)

and that between 300-1000 km the boundary layer reached a steady state (Brummer 1997).

The strength of MCAO can be calculated by comparing potential temperatures of the SST and the 700mb level (Kolstad & Bracegirdle 2008 https://link.springer.com/article/10.1007/s00382-007-0331-0) - as typical for forecasting features of MCAO such as polar lows. The 700mb is chosen as being just above the boundary layer depth noted above, whereas the 900mb temperature for example is quickly modified, whilst 500mb, though used sometimes for polar low forecasts, doesn't capture all MCAO.

The charts below show the regions of strong cold air outbreaks, and their change between 1961 and 2000. It is slightly unintuitive, but the first chart shows the most extreme outbreaks at the 95th percentile, which happens on average roughly 18 days per year at any given location. These are clustered though so the assumption is for them to be within the winter half of the year and 3 days in duration so occurring about once a month on average. The highest values (>0 K/bar) indicate strongest MCAO and naturally occur where cold air sources are closest to the much warmer SST, such as the northern fringes of the Atlantic and Pacific, and also further south the Siberian cold air flowing out towards Japan, shown in the first chart.

The other two charts show that large changes have occurred where ice has retreated exposing new areas of open water, but near the UK and even to the north of Norway, the 95th percentile extreme level has only reduced a little over time -

It is thought that changes in NAO for example have been generally as or more important than reduced sea ice for areas away from the ice, such as the slight increase in extreme outbreaks for the western Atlantic, and that may account for differences around our shores also i.e. fewer northerlies, not necessarily weaker.

 

 

Interesting and well-researched stuff!  I agree with the above concerning the period 1961-2000 (and I think it would still hold if we extended the recent period as far as 2004) but my question would be, would it still hold if we were to compare, say, 1993-2004 against 2005-2016?  I suspect not. 

Although Arctic sea ice reduced over the period 1980-2004, the source region between Jan Mayen, Svalbard and the Russian Arctic only showed a modest amount of warming, and temperatures in the late 1990s/early 2000s were still below the peak that they reached between the 1930s and early 1950s.  In 2004 John L. Daly attempted to disprove the Arctic Amplification theory by pointing to records from "Arctic Rim" stations, and at stations with records going back into the 1940s, such as Ostrov Vize, there was a slight cooling trend between then and 2004.  Given the dynamics of modification over open water discussed above, a warming of up to a few degrees Celsius would indeed be unlikely to make much of a difference to the potency of our northerlies.

But since 2005, while the decline of Arctic sea ice extent hasn't accelerated much, there has been a very marked step-change in the temperatures in that source region.  The mean annual temperature at Ostrov Vize has shot up from a mean of about -14.5C to nearer -10C, and much of that difference comes from the winter half-year when temperature anomalies of 10-15C have been quite common.  In 2012 and 2016 their annual mean was about -6C (8C above the long-term normal), with anomalies of as much as 17C in individual months.  In Svalbard it's a similar story but not quite as extreme.  The anomalous warmth also increasingly appears to be penetrating to higher levels of the atmosphere with rather less cold pooling evident at 850hPa than in the past.

Despite the neutralising effects of the modification over the ocean, I suspect that those sort of anomalies will have been making more of a noticeable difference to the potency of our northerlies since about 2005.  We may well only be talking a difference of 1-2C once the airmass modification dynamics have played out, but traditionally east-coast snow showers have often been only just on the "snow" side of marginal and so a difference of 1 or 2C can be enough to replace snow with a sleety mix.  On the other hand we did get some very potent northerlies, even by long-term historical standards, in November and December 2010, so we might still find that the average difference comes out less than I would expect.

Edited November 26, 2017 by Thundery wintry showers