Model Output Discussion - Spring Has Sprung

[danm]

 

[ANYWEATHER]

 Downburst yes Albedo effect......

[Scorcher]

 Downburst Hasn't affected things here today. We rather unexpectedly reached 20C this afternoon (I was expecting the first 20C tomorrow rather than today).

If the 850 hPa temps are high enough it doesn't matter much about the ground not being bone dry.

[danm]

Very quiet in here tonight….

The ECM is a bit of a corker for next week:

[WYorksWeather]

All looking generally solid again after the wobble yesterday - all models have a decently strong high at day 7.

GFS, ECM, GEM, UKMO in that order:

 

ECM probably the pick of the bunch for warm and dry, but all looking reasonably dry and quite warm, though GEM probably showery in the north.

To bring it onto a different topic - I've noticed quite a few people wondering why temperatures aren't running very high at the surface based on the 850hPa temperatures. I think part of this is orientation of the high and the still damp ground, but I've not noticed anyone yet mention model bias.

In this instance, it is an important factor. Using the GFS as an example, in most situations it under-predicts, but where the ground is excessively dry, it over-predicts temperatures. In most instances, the UK is subject to underestimation of temperatures, with the only recent exception that comes to mind being the July 2022 record-breaker, which some models overdid because of the excessively dry ground - a few runs even shortly beforehand had 41-42C.

But outside of that, let's just take today, for instance. Here is the high-res UKV model output for this afternoon from the morning 03z run:

You can see that none of these charts feature a max above 20C. And yet, the Met Office had a provisional maximum of 22.1C at Santon Downham, partly due to just a few cloud breaks at the right time.

Other models are typically worse for this than the UKV, which is normally only out by a degree or so. ECM can even be 3C or 4C too low sometimes. Another pattern is that temperatures tend to be revised up closer to an event or even on the day. The afternoon UKV, actually issued retrospectively, did do better, with a raw max of 21C, which is close enough.

Another pattern is that temperatures get revised up typically closer to an event. This was something I noticed in the run up to the September 2023 heatwave in particular - a lot of model output initially was translating 15-20C 850hPa air with cloudy early mornings and clear afternoons into temperatures only around the 26-29C range. As we got closer to the event, those figures got revised up, and we ended up reaching 33C on the hottest day and recording seven consecutive over 30C.

Again, the current warm spell is not in the same category, but if we happen to record one or two days in the mid-20s when we're only 'supposed' to record low 20s, it really wouldn't surprise me very much.

[danm]

A cracking set of runs this morning. Once we get the weekend trough out of the way, it's high pressure all the way through the rest of next week.

GFS:

 

ECM:

 

UKMO:

 

GEM:

Edited May 2 by danm

[Howie]

 danm

Finally!

[Atmogenic]

I hope my SAD disappears for a while if all these good runs verify 

[Addicks Fan 1981]

 Atmogenic we just haven't had the jet stream in our favour, been blowing right across us rather than skirting to the north and steered towards iceland.   

Also we are meant to have the final warming of the stratosphere but it appears to be quite sluggish from what I looked at on the JMA stratosphere site.   

It wouldn't take much for the jet stream to move north and I think they'll be some twists and turns with the model output in the forthcoming weeks.   

[Sun Chaser]

 danm I don't want to be negative but I have so little trust in any kind of good projected outlook on the models that I'm sure that either it will disappear by the time it gets to day 6, or the high sinks and we all end up with a lovely cloudy westerly setup again!

[ANYWEATHER]

A bit premature to be thinking the predicted settled spell next week will be plain sailing without out any twists and turns! I like everyone else need some warmth and sunshine, and no rain for a few weeks, but anyway looking forward to something of a settled spell , without the caveat of overcast and chilly weather which still can plaque many in early May.

Edited May 2 by ANYWEATHER

[Addicks Fan 1981]

 Sun Chaser it all depends on the jet stream and where that is situated, at the same time signals lead the models too.   

i would suggest to people who post here to keep our eyes on world climate service on X as well as judah cohen, MetRyan96 and what they post.   

Also it also depends on the pace which we go into la niña in too and the prevailing winds.   We need the teleconnections on our side.

[danm]

 Sun Chaser well all we can go on is what the models are showing. Yes, could all go t**s up, but this has been trailed for some time and currently has good, cross model agreement. But yes, nothing is guaranteed until we're at a few days out. 

[Addicks Fan 1981]

 danm like I said before it'll be to do with whether we have the zonal winds recovery and other different variables that could alter things.  It's a real shame that the teleconnections posters who do a deeper dive are not around to shed some light on the goings on.   

What we need as I have stated clearly is the delay in the development of la niña and some WWB activity on the hovmoller plots.  Also wonder if we have had mountain torque activity as well as that can alter things too?   

I did also notice lately that the MJO forecasts have trended from very active to very quiet.    

[Rain All Night]

17 hours ago, Rain All Night said:

0z ensemble means, out to Thu 16 May (day 15)

There seem to be more ensemble members than before featuring Atlantic lows riding in over the top of the high, especially from 12th May onwards. The signal for the high to move west into the Atlantic seems weaker now than a few days back.

0z deterministic runs, on Wed 8 May (day 7)

UKMO and ICON show little lows getting in the way of the ridge pushing up from the south, which is an outcome that seems to have been averaged away in the ensemble means presented above.

0z ensemble means, out to Fri 17 May (day 15)

The high looks as though it could stay in place over the UK throughout next weekend, which would be a real blessing. After that, the means blur out any clear signal, but the high perhaps now looks more likely to sink than to retrogress.

0z deterministic runs, on Thu 9 May (day 7)

9th May is the one day that has been producing eye-catching ensemble mean charts for ages... will it now deliver?

[*Stormforce~beka*]

 Atmogenic Gosh me too!!!!!

 Sun Chaser I checked the beeb site and it shows partly cloudy every day so I'm not holding out much hope as much as tehse models look great

[Rain All Night]

This just about qualifies as model output as it's the T+0h analysis from the hourly GFS... the view over Europe is quite striking today, both at the surface and at 850hPa:

Colours = temperature, white lines = wind

https://earth.nullschool.net/#current/wind/isobaric/850hPa/overlay=temp/orthographic=0.00,55.00,2077

[Scorcher]

 Rain All Night Fascinating image. You can see why the south has been cooler today while the best temperatures have been from North West England northwards and the west of Scotland.

Most of the warmest air from Germany has missed the south and has impacted the UK north of the Midlands. The east is much cooler because of the strength of that E/NE wind off the North Sea.

[danm]

26 minutes ago, Scorcher said:

Most of the warmest air from Germany has missed the south and has impacted the UK north of the Midlands. The east is much cooler because of the strength of that E/NE wind off the North Sea.

...and mainly because it's cloudy. In East Anglia, just north of the cloud band, it's 23c. 

[Scorcher]

 danm well yes but East Anglia is also in the warmer band of air. Considerably warmer 850 hPa temps there than south and west of London at the moment.

[LetItSnow!]

GFS 12z shows that by the 8th we do look to have a nice ridge over the country and with rather average but pleasant uppers of around 5 C, I reckon this will easily see sunny spells and temperatures into the low-twenties.

Interestingly, after this on both the 6z and the 12z the GFS takes the trough in the mid Atlantic  and powers it up into a big low pressure system giving perhaps a mini thundery breakdown around the 11th and throwing us back into unsettled conditions around day ten. Now, of course this is a long way out in forecasting terms but it does show that caution needs to be had over ideas of a long settled spell beginning. It’s probably not a far fetched idea as warm SSTs in this region may fuel trough development to our south-west. 

[Scorcher]

 LetItSnow! When does the GFS not do that in deep FI though?

It almost always shows a dramatic breakdown after 7/8 days. Seems almost a default setting for the model for our part of the world.

[LetItSnow!]

 Scorcher I’m simply showing what the model is showing up to day 10 (and never beyond) as I usually do and that there was a trend between the two runs, plus why due to SSTs it may not be entirely unfeasible for unsettled and/or further thundery episodes. The model may very well be wrong and we could enter a longer dry spell. I only talk about what is there in as unbiased a way as possible. Just like how I talked about the warm pattern in early April. The GFS has had a better year this year at predicting changes to be fair.

If it’s any consolation I think May will be warm/very warm and thundery.

Edited May 2 by LetItSnow!

[viking_smb]

 danm thank goodness, sunshine 

[SunnyG]

On 30/04/2024 at 07:45, Penrith Snow said:

ECM says no to high pressure next week with a chilly North Westerly by day 8.

Big difference with MetO and GFS which has high pressure in control for most of next week.

Surely ECM is wrong?

Who wants to bet the ECM is right? Because we all know how it's gonna end...