right.
you know, tracking that s/w's spacing back in the flow ..it's not even yet over the American owned sounding domain. I'm sure we're not exactly purchasing sounding data from Russia heh ...if they are even plumbing the skies over Siberia at all in the first place.
Anyway, it's all based on assimilation. Although satellite sounding. Still, it's almost like the models assume something should be there in the flow in that spatial-temporal/geometric region, then propagating the assumption along ...at difficult to manage, ludicrous speed. Perhaps if there was better physically realized, direct sampling the handling might be more stable.
I mean I suspect something like this is going on. Look at the Euro's last 4 cycles ... left to right, the most recent to 24 hours ago ( 6 hours apart). On off on off... That's been rather predictable just within that small range. Something perhaps added than missing every other run