I've been banging the drum about it for years, and proved it on here over the course of several years (must have been approaching 10 years ago now). Latterly the ECM themselves proved it, and as we've seen this year it's still very much a thing.
The issue is the lack of aeroplane data, as they fly in the jet stream. With lower amounts - typically less than half the normal amount on Christmas Day - and even with increased satellite observations, it means the models can't get a grip on the jet. As that affects both things lower down (the position and strength of blocks) and higher up (via wave-breaking into the higher stratosphere) it can have big knock-on effects. And in our part of the world 9 times out of 10 it makes the models show phantom blocks, which then slowly vanish as data comes back online. It's no coincidence that the ECM's zonal wind chart showed a mean reversal on the midnight 26th Dec set, then ever since it's backed off slowly.
The ECM's report is here, and it's well worth reading even if you're ambivalent about the whole data loss issue.
https://www.ecmwf.int/en/about/media-centre/news/2020/drop-aircraft-observations-could-have-impact-weather-forecasts
I had a big spat with White Meadows over it (he refused to listen to any evidence whatsoever and kept making snide comments whenever anyone asked about data loss), so I blocked him years ago. For all I know he's still going on about it now!
Originally Posted by: Retron