What do you consider “Early Spring temperatures”...and "Late Spring temperatures"? (icy, storm)

[nei]

With Ariete's rules, spring began here Mar 1. And winter began around Dec 30.

[Ariete]

Originally Posted by nei

I rather like this system, but it seems excessively prone to being thrown off early spring heat waves (or midwinter warmth). For very variable climates, this system might have issues. For my climate, I'd replace 10°C with 18°C (maybe a degree less?) but keep 0°C.

Yes, you have to keep in mind that the swing in temperatures are so much smaller here, combined with the low sun angle, that it's virtually impossible to get a prolonged spell with means above 5C. My record high in January is 10.0C!
It's also so common to have nights below or near freezing in March and April, that it's really hard to reach means of 10C. Our March record high is 17.5C, and even during that day the low was 0.1C. Next day a similar story: 15.4C high, -0.8C low.
Something like a high of 30C in April... well, I'll let you know when a Scandinavian location records one, will be the first ever.

I didn't mention before that the seasons must have a 'realistical' date when it starts. Winter cannot begin before Nov 1, spring before Mar 1, and autumn before Aug 1 (in Lapland it may start in August).

The seasons are not always carved in stone, as I mentioned earlier, and exceptions can be made regarding a warm spell. April 2008 was another example, the mean temp between Apr 26 and May 4 was 13.6C, but as the temps fell after this below 10C and recovered only from May 22 onwards, so the 'summer' started on the 22nd that year. Ironically, the summer of 2008 turned out to be a cool one.

You can also require that the time period should be longer than seven days. The Swedes require only five days. The 18C threshold works probably in hot summer climates, but maybe the swing between 0C and 18C covers a bit too much ground?

Originally Posted by nei

With Ariete's rules, spring began here Mar 1. And winter began around Dec 30.

They are not my rules, they are the Finnish Met Institute's. But works pretty well, doesn't it? Don't worry about the beginning of winter, it started here on Jan 6 last year.

[nei]

Originally Posted by Ariete

You can also require that the time period should be longer than seven days. The Swedes require only five days. The 18C threshold works probably in hot summer climates, but maybe the swing between 0C and 18C covers a bit too much ground?

Not really, for us it would yield a summer start date of

May 30 in 2012
May 30 in 2011
May 26 in 2010
Jun 25 in 2009*

*If the thershold was 6 days it would be May 25.

A streak of 7 days above 10°C is almost impossible for March (though it happened last year) and not really guarenteed until late April because of variablity and cool nights.

[Ariete]

Originally Posted by nei

Not really, for us it would yield a summer start date of

May 30 in 2012
May 30 in 2011
May 26 in 2010
Jun 25 in 2009*

*If the thershold was 6 days it would be May 25.

A streak of 7 days above 10°C is almost impossible for March (though it happened last year) and not really guarenteed until late April because of variablity and cool nights.

Interesting. That's quite a reliable pattern. I'm also a bit jealous, your 24h means go above 18C on average two weeks after ours go above 10C... I'm not really sure when we have our first 7 day period above 18C, perhaps I don't want to know.

The problem with raising the mean temperature here is that it always pushes the summer forwards, while shortening it much less at the latter end, even if individual high temps are almost always higher in June than in September. In Moscow they could probably use the 15C or even 17C mean to create more realistical lengths of the seasons.

[nei]

For the end of summer (7 days in a rows with a mean below 18°C)

2012: Sep 21
2011: Oct 7
2010: Oct 8
2009: Sep 22

[dunno what to put here]

Summer start dates using the Finnish method at Leeds Weather Centre (1986 - 2003):

1986: 30 April
1987: 13 April
1988: 15 April
1989: 1 May
1990: 29 April
1991: 9 May
1992: 12 May
1993: 16 April
1994: 26 April
1995: 30 April
1996: 20 May
1997: 27 April
1998: 12 May
1999: 29 March
2000: 19 April
2001: 14 May
2002: 20 April
2003: 13 April

[nei]

Ok, here's using 10°F by the Finish method:

2012: Mar 24
2011: Apr 30
2010: Apr 7
2009: Apr 30
2008: Apr 24
2007: May 4

[dunno what to put here]

Using 18C as a methodology, these are the results, only including years that actually met the criteria:

1989: 12 June
1990: 28 July
1994: 29 July
1995: 24 July
1997: 6 August
1999: 2 July
2001: 25 July
2003: 13 July



I'm not sure what could be used for the UK - for Leeds, 13C - 15C could be used.

[nei]

this thread is making look forward to the next couple weeks seeing how much warmer things get. Only a few more weeks. Current forecast shows us not passing 5°C for the next week.

[Mac15]

Well the cold spell keeps getting extended and it looks to have extended to the 24th now, so probably not till april here either.

I wonder if there has ever been a March without double figures before.