mikeocarroll

Was -3.0c air temp here last night (min at 6.41), but this a big frost hollow. The heaviest frost of the season so far with pretty much all surfaces white, and didn't melt in the shade until about 10.30-11ish. Meto forecast shows cloud reaching here by this evening though so unlikely to be as cold tonight. If the cloud stays away though then it could fall even lower than last night.

2.9c here - has fallen 5c in the last 2 hours since dusk. Max today was 12.3c.

Summary for October 2011 - near Llandrindod Wells, Mid Wales Temperature (°C): Mean (1 minute) 11.2 Mean (min+max) 11.1 Mean Minimum 7.1 Mean Maximum 15.1 Minimum -2.6 (20th) Maximum 26.8 (1st) Highest Minimum 15.3 (2nd) Lowest Maximum 9.9 (26th) Air frosts 5 Rainfall (mm): Total for month 123.6mm Wettest day 21.9mm (24th) Rain days 20 Pressure (mb): Maximum 1031.0 (14th) Minimum 987.3 (24th) Quite wet here especially in the last week of the month. Some cold nights under high pressure, but otherwise mild.

The picture is from November 2010 according to the caption on the website. This may be obvious to most but wasn't to me - I thought it was today when I first saw it - so just to clear it up in case anybody is as daft as me.

That's a staggering amount of rain showing as falling just off the Pembrokeshire coast, a slight eastward shift would lead to some serious flooding. For the moment the radar seems to be showing the heaviest rain in the Irish sea and the far west of Cornwall.

As far as I can see the last time there were 4 winters with CET of 3.5 or less (3.5 was the 2009 winter CET, 2010 & 2011 were lower) was 1808-11, though it happened quite frequently in the 17th & 18th centuries.* However having had 3 such winters in the last 3 years, there is as good a chance as there's been in a long time that such a 4-in-a-row will occur. If you flip an unbiased coin 4 times in a row it is unlikely that you will get 4 heads. But if you have already got 3 heads, then the 4 in a row is a 50% chance. Weather & climate don't work quite so neatly, but I just wanted to demonstrate how probability works. *note, to actually answer the question, there were 4 winters below average 1962-5, certainly below the 1971-2000 average, and I think probably the rolling average at the time, but not sure (note: this sentence has been edited since original post)

Min of -2.6c here last night, -0.3c currently. It's absolutely still out and a heavy frost on the ground. Looks like warming up over the next few days without too much interesting weather wise, but some moderate wind and rain possible on Monday.

Yep had quite a few showers here today, 7.5mm of rainfall so far it total. Current temp 7.9c, overnight low was 0.0c shortly before dawn when we had clear skies for a bit. Could be slightly colder tonight if things settle down as the high pressure comes in.

An interesting evening's weather ahead. It's been quite breezy here this afternoon but not quite gale force - I'm sure it's been a lot windier further north and west. There's been some heavy precip over Snowdonia earlier and a band of heavy precip at the moment roughly from Llandovery to Builth in mid Wales (just missing me to the south). Should be quite a drop in temps and some more precip earlier - front has just reached Anglesey & temps at Pentraeth have fallen from 13.3c to 10.0c in the last 10mins, should fall a lot more over the next hour or so. It's 11.2c here at the moment and light rain, should be another 2-3 hours before the front reaches.

That's the thing, for such a small area Wales has a huge variety of climate. It's strange (/frustrating) here to compare with other areas as it's usually quite a bit colder than most. If it wasn't for observable signs such as frost & ice I'd think that my equipment was faulty, as for example the -0.9 yesterday morning is much colder than most other people reported across these forums. I've only been here (i.e. in my current location) since Feb so haven't much experience, but the upper wye valley here seems to be a cold place - most montly min temps for Wales are set in this part of Powys (though mostly not as cold as neighbouring Shropshire in England) - the cool must pool down from the surrounding hills, and I suppose a couple of hundred of metres of altitude at the valley floor helps low temps. Will be interesting to observe this winter anyway. Cheers for posting your forecast above, in my experience of (mostly lurking on) netweather, there's always demand for as many local forecasts as possible! It will be quite a shock to the system next week I think.

4.9c here but falling more slowly than it was a couple of hours ago. Looks set to be another cold night - again not forecast.

I'm no expert but I think there is a possibility of a ground frost in the M4 corridor next Thurs/Fri night as high pressure builds in after the cool depression. Air frost less likely but not impossible. It's a long way off to try and say though, should have a better idea in another 2-3 days. Looks like it - It's exciting to see things getting more wintry!

Yes looks like October could be a very warm first half followed by a cool second half - certainly temps of -0.9c here currently will bring down the October mean from the 14.0c it was up to the end of yesterday. Anybody else got a decent frost this morning?

Well unsurprisingly runs showing more significant cold now look unlikely to verify, still looks likely (though far from certain) for a colder spell next week though to make things feel a bit more autumnal.

ECM now backing cold plunge for the middle of next week.... would be great synoptics for real cold in winter but certainly could deliver first snow over Welsh mountains and hills maybe down to 300-400m if verified. Still FI at the moment though so we'll have to wait and see.

Well it looks like getting colder next week - in fact GFS 12z show a decent northerly next week that would potentially bring some hill snow in Wales if it verified, and some very wintry temperatures generally. However other models and other runs aren't showing this at the moment so chances of verification are low. As for winter generally I'm sceptical of any claims to forecast a season ahead...

Actually the Met Office researchers conclude that the low solar activity is responsible for warm temps in Canada & Greenland as well as the Eastern Med - see the link here http://www.bbc.co.uk/news/science-environment-15199065 which is linked to from the Met Office blog as an "accurate" report of the research: ""The key point is that this effect is a change in the circulation, moving air from one place to another, which is why some places get cold and others get warm," said Adam Scaife, one of the researchers on the paper, who heads the UK Met Office's Seasonal to Decadal Prediction team." I'm not saying that the solar activity theory is necessarily right, but it is a theory that has enough to it to be taken credibly and warrant further study.

Is it possible that temperature maxima are broken more frequently in recent years due to an increase in weather stations, making it more likely that a particular hot spot will be recorded? Is there simply more data collected than say 100 or even 50 years ago? Does this affect other extremes? I don't think that this is the whole story but it may be a factor - or it may not be - I haven't got any data on data collection... I think it's also worth bearing in mind how many pieces of weather data are collected, for example if 40 pieces of data are collected in each month (max, min, lowest max, highest min, rainfall 1 hr/24hr/month etc. etc. I'm sure you can get to 40), then if you multiply this by 12 months then there are 480 different records to be broken, and so given that we have less than 400 years of reliable statistics, then you would expect a record each year. And once you start sub-dividing into a given 7-day or 30-day period then this will become even more frequent as there are huge numbers of permutations of such days each year, it is only to be expected that some records will be broken frequently until we have several thousand years' worth of data. It's like in cricket, England are always setting the "highest third wicket partnership against the West Indies at Edgbaston" or something equally specific, but the "highest partnership for any wicket against any opponent ever" is somewhat rarer.

Unusual month temperature wise in Llandrindod Wells, Mid Wales, with the lowest minimum temp (2.4 degC) occurring just 5 hours into the month on 1st September, and the highest maximum (26.8 degC) occurring with just 10 hours remaining of the month, at just before 2pm on 30 September. The wrong way round for an autumn month!

Overnight low here of just 2.4C, shame it wasn't quite in August. Clear & sunny now & 10.9C & rising quickly.

Light graupel here - nought on radar though so not really hopeful of it intensifying or turning to proper snow.