True, but with Arctic warming and general reduction in sea ice and so in cold air generated in Arctic, it is all related, so just projecting ahead past levels and characteristics of variability is unlikely, in my view, to be valid. Also, thermodynamically, it is quite hard to maintain a colder Arctic as the world warms—the Arctic is a small percentage of the area (despite Mercator-based misimpressions)--and so what goes on elsewhere will carry into the Arctic.
There was a good study at WCRP which showed that much of the inter annual variability is wind related, as ice is moved towards the Atlantic in certain years.
I don't know if It's in print yet.
Veli has made this point before, and the modelling seems to support his view.
This is worrying, as It's a non thermal process. SRM can't predictably change the wind, so unless we can rapidly rebuild the ice to a continuous mass, it will tend to get blown away when the winds dictate.
Just to note, however, that we really do not have a good sense of how big or small variability can be at this melting trend continues—variability is very unlikely, in my view to be much of a saving influence on the decadal scale unless some strong cooling influence results—whether from a major volcanic eruption, lots more sulfate pollution on the global scale, or climate engineering. With world warming, it is hard to have the Arctic go very far or very long in the opposite direction.
I'm with Stoat, Ken Caldeira, David Keith, Alan Robock and others who see this "emergency" effort to rush cloud intervention in the Arctic on behalf of sea ice (and indirectly seabed methane) as undermining the case for a serious push on geo-engineering options, impacts and policy issues. You're getting headlines and the attention of factions in Parliament now, but just wait until the variability kicks the other way.