Snow and Ice Warning : Saturday(PM)/Sunday 9th/10th Dec - TECHNICAL ANALYSIS ONLY

MJohnston

Villain wrote: »

Harmonie was on the money, like really really accurate 24 hrs out.

Do you know a site that shows the Harmonie for Ireland? Meteociel.fr doesn't

Meteorite58

MJohnston wrote: »

Meteorite, did the cloud clear enough before sunset to get a better snapshot of the snow coverage across more of the island?

The NASSA sat pic just comes out once a day MJohnson, cloud slow to clear south as the day went on.

MJohnston

Damn, still not found any snow accumulation charts anywhere, do we ever get any of those from ME?

ninebeanrows

Hirlam still wants to give a very slight bit of sneachta to parts of the snow starved Greater Dublin region tomorrow AM

Gaoth Laidir

MJohnston wrote: »

Damn, still not found any snow accumulation charts anywhere, do we ever get any of those from ME?

Since the stations went automatic there are no snow recordings from any station except the 4 main airports plus Casement. The northern Irish stations (also automatic) do have automatic ultrasonic snow-depth-sensors so have depths in every hourly report. All snow depths are available in map format on www.weatherobs.com

Gaoth Laidir

Weathercheck wrote: »

Hirlam still wants to give a very slight bit of sneachta to parts of the snow starved Greater Dublin region tomorrow AM

Looks very light, but we'll bank it!

Meteorite58

MJohnston wrote: »

Damn, still not found any snow accumulation charts anywhere, do we ever get any of those from ME?

Ogimet gives some information, good for showing the type of precipitation on an hourly basis. Could only find two stations reporting snow amounts in the reports

http://www.ogimet.com/cgi-bin/gsynres?lang=en&osum=no&state=Irel&fmt=html&ano=2017&mes=12&day=10&hora=18&ord=REV


Current temps

Donegal Storm

That NASA satellite used to do two passes per day, not sure why they've only got one pic now. It should be clear all day tomorrow anyway so we should get a nice pic then

On the models, Arome also did a decent job, the archive thing on meteociel doesn't seem to be working but it got locations and snow depths pretty much bang on from what I remember

MJohnston

Donegal Storm wrote: »

That NASA satellite used to do two passes per day, not sure why they've only got one pic now. It should be clear all day tomorrow anyway so we should get a nice pic then

On the models, Arome also did a decent job, the archive thing on meteociel doesn't seem to be working but it got locations and snow depths pretty much bang on from what I remember

Yeah, I couldn't get that archive working, so I had to leave it off.

MJohnston

Okay, so for a lack of a genuine snow depth chart, I've fudged a bit. I took the ECM 12z from today, the +0h snow depth chart, which while that isn't going to be 100% reflective of reality (especially as the ECM doesn't have particularly high resolution for this chart), it's about as close as I can get to reality and it seems to largely line up with reports we're hearing. I also had to 'subtract' the +0h snow depth chart from the ECM 0z from the 9th (using Photoshop), because there was a lot of lying snow in the north already before yesterday (imperfect but you can see it vaguely enough). This created my 'baseline' for comparison.

Again I'm comparing to the 9th Dec 0z forecasts - in this case Snow Accumulations since +0h until 10th Dec 12pm.



Less effective than the other comparison perhaps, but here are some grades anyway:

• ECM - B?: Here I'm comparing the un-subtracted baseline because it's ECM and ECM being compared. It seems to have mostly done pretty well here, though the forecast chart seemed to have a lower rain-snow boundary height than reality, and over-estimated for the east-coast and Kildare. General amounts seem about right.
• ARPEGE - B+: This is real good, not sure if the stuff in the North actually came through, and it under-estimated the southern extent a little bit. General amounts seem about right.
• WRF - E: Way too much volume, far too large on extents, and too far north compared to reality.
• ICON - F: Total fail, I don't know what happened there, the only thing it got right was Wicklow mountains.
• GFS - F?: Difficult to compare, but it doesn't look very good - seems nearly as bad as the ICON.
• HIRLAM - Seems to largely be the same story as with the precip - the intensity was a bit too high, and therefore got the scale of the event wrong. But it got the positioning basically right, if you look at the higher ends of the scale they do correlate with where the heaviest snowfalls occurred.

Another win for the ARPEGE here, and I think the ECM comes out looking pretty good too. You can take these with a heavier pinch of salt because the baseline picture isn't as reliable.

Hopefully we have more snow events so I can do a few more comparisons, and also for the snow

Gaoth Laidir

Last week, though, the ARPEGE was pretty bad and consistently showed less snow than the others and what actually fell. It may not be good in a NW setup.

One other one to note. The Canadian GEM was shockingly bad both last week (showed no snow whatsoever) and at the weekend (showed some in the north only). I'm surprised at thay as it's normally very reliable and one of MT's favourites.

MJohnston

Gaoth Laidir wrote: »

Last week, though, the ARPEGE was pretty bad and consistently showed less snow than the others and what actually fell. It may not be good in a NW setup.

One other one to note. The Canadian GEM was shockingly bad both last week (showed no snow whatsoever) and at the weekend (showed some in the north only). I'm surprised at thay as it's normally very reliable and one of MT's favourites.

Yeah, I'm tracking these results in a spreadsheet for myself, and I've got a column in there to briefly describe the type of setup, as I figured different models might perform better for different setups. Which means it'll be a long time before there's enough data to spot even a strong trend but still!