Essentially, the easterlies that we stopped getting in the 90s were regarded as extensions of the Siberian High, albeit looking at reanalysis charts it was often a polar cell involved rather than direct action from the Siberian High itself. This polar cell was invariably supported by upper heights caused by disturbance to the polar vortex - what caused that in the first place, though, is up for discussion!
I still remember in the mid 90s, Philip Eden wrote in Today, "cold high H will sink southwestwards across Scandinavia" and lo, a few days later we had a textbook midwinter easterly. Similarly in 1991, the models picked up on the pending cold a week or more out, as I remember our local (TVS) weatherman going on about it.
The key with most of those easterlies seems to be having a polar high in situ, then relying on an upper ridge moving eastwards across the Atlantic linking up to it - we need zonality, of sorts, to kickstart the whole thing. Once the link is made, a zonal flow deflected southwards of normal is required to "prop up" the blocking high, without it the high quickly slides away SE'wards.
Looking at the models, we're getting upper ridges moving eastwards easily enough (it is, after all, the natural flow of things), but there's nothing for them to link to as the polar cell is further towards the Pacific side of the Arctic. To use an anology, for an easterly we effectively want a pair of cogs meshing together. We're at the point where the two cogs have been moved a bit too far apart, so one is spinning (the Atlantic cog) but there's nothing for it to mesh with. Whether that polar cell ever gets close enough this winter remains to be seen...
Originally Posted by: Retron