Snow Run

Same here now: snow but not settling. You'll be fine up there as it looks from the radar as though it'll get heavier later. You always get more than us... was sliding all over the place when I took my daughter to Beacon on Weds morning!

Same here in Groombridge. Drove up the hill to go to TW at 6.30 to find it snowing heavily. By the time I drove home an hour later there was a decent covering and it was still coming down. Got home to rain - only 3 miles away!

And way out into FI, sliders under a Norwegian Sea block for those who missed out on the inital easterly

Thanks, John. It would follow by definition that there would be NAO and AO index values for every model run (and current/past conditions). From what I've seen, those most often quoted are from NOAA. I have a rough idea that they are from the Azores to about Iceland for NAO, and from there to the pole for AO, but I'll do a bit more reading up!

Hi SSIB I’m not sure if I’ve misinterpreted your comment, but this isn’t my understanding of the NAO. I may well be wrong, as my experience is from deterministic mathematical modelling of structures (FEA), as opposed to NWP. However, I’ve always understood that the NAO (and also AO, Arctic Oscillation) is a statistical index that shows how mean atmospheric pressure varies between two specific points, either current conditions or model output, ie it’s not a model input as a boundary condition. So what I’m trying to say is that either a positive or negative NAO are something that the model output is indicating, and not something they would pick up on. For instance, if the model output is showing a Greenland HP and lower pressure further south, then the corresponding forecast NAO index for this would certainly be negative, but I don’t think that this model result would be fed back into the starting conditions in any way. But I'm happy to be corrected!

Temp 3.9 degC Dewpoint -2.3 degC Just hoping for a nice dusting across Ashdown Forest for my run in in the morning. Failing that, a freeze-up so that I don't have to clean my shoes again!

A good point with regard to the GFS dealing with blocked situations: why would they change it? Perhaps the performance of the model over the US mainland is more important to them (I wonder if there are verification stats, say, for each side of a line of longitude in mid-Atlantic), although would it not need to deal effectively with Pacific weather systems encountering high level blocking as well? I work with numerical structural modelling (finite element analysis, FEA), which follows essentially the same philosophy as NWP. You often find that different programs deal better with certain structures than others. For instance, some are good at detailed layered composites, whereas others might be best at modelling something like oil rig pipework. As such, I was interested in Ian's comment that the JMA holds some respect in the UKMO, possibly due to Japan being in a very similar geographical position with respect to location in the sea just off a large land mass. It's all down to initial boundary conditions and algorithms for FEA - does the same hold true for NWP? Perhaps this should be asked in the Tech Model Discussion thread, but since we're in a lull before the next output...