March 2015 CET forecasts

[IanR]

 locally it must be below average here, cant believe the cet is above, its felt cold and gloomy every day apart from last saturday but one,  only 7c again here today, like most days  

[damianslaw]

When will the 81-10 CET values be used I wonder? It seems odd we are still using 61-90 averages..

[Nick L]

 locally it must be below average here, cant believe the cet is above, its felt cold and gloomy every day apart from last saturday but one,  only 7c again here today, like most days  

 

It does seem hard to believe. But while the day times have been dull, chilly and depressing, we haven't really had any cold nights. Absolutely detest this weather. Either be properly cold or warm!

[Midlands Ice Age]

It does seem hard to believe. But while the day times have been dull, chilly and depressing, we haven't really had any cold nights. Absolutely detest this weather. Either be properly cold or warm!

 

Expect some downwards adjustments at the end of the month!!

 

I contacted the MO about the way they calculated the CET on there unadjusted daily dataset.

Eventually (on the 4th attempt) I managed to speak to someone who really undertood the issue!

ie the helpdesk tried to be helpful, but didn't understand the issue. 

 

BFTV is indeed correct....

 

EG     The minimum fof the 1st of March is taken from the period 09:00 on the 28th Feb to 09:00 on the 1st March.

The maximum i during the period from 09:00 on the 1st to 09:00 on the 2nd.

 

During times of temperature change or when the temperature is dropping through the breakfast period it can give rise to quite large differences, compared to what is "felt ' .  I reckon that we already have clocked up an adjustment of 0.2C this month.

 

Global Warming on TWO keeps a dataset of all this and publishes at the end of every month.

 

So you can see how the pattern of adjustments changes during the month.

 

MIA

[Summer Sun]

6.3c to the 16th

 

1.1c above the 61 to 90 average

 

0.2c above the 81 to 10 average

Edited March 17, 2015 by Summer Sun

[The PIT]

Cloudy skies overnight kept temps up but we still dropped down to 5.8C.

[Interitus]

When will the 81-10 CET values be used I wonder? It seems odd we are still using 61-90 averages..

 

It's based on WMO convention and if adhered to it won't go to 81-10, the next update will be to 1991-2020. This is the current 30/30 model i.e. 30 years average updated every 30 years. There are plans to keep this method for long term climate trends and add the 30/10 averages too for National Climate Products (they mention 71-2000 as being suitable at the moment!), which will be decided this year -

http://www.stat.washington.edu/peter/statclim/CCl-MG-2011-Doc_10_climatenormals1.pdf

[Summer Sun]

6.3c to the 17th

 

1.1c above the 61 to 90 average

 

0.2c above the 81 to 10 average

[The PIT]

Sunny Sheffield now 5.7C only a clear night will make it drop tomorrow.

[Summer Sun]

6.2c to the 18th

 

1.0c above the 61 to 90 average

 

Bang on the 81 to 10 average

[BornFromTheVoid]

Minimum today is 1.2C while maxima look like reaching the low 10s, so remaining on 6.2C on tomorrows update.

 

After that, the 06z GFS has the CET at:

 

6.3C to the 20th (7.9)

6.3C to the 21st (6.3)

6.2C to the 22nd (4.6)

6.2C to the 23rd (6.0)

6.2C to the 24th (5.1)

6.1C to the 25th (5.1)

6.1C to the 26th (6.0)

6.2C to the 27th (8.2)

6.2C to the 28th (7.0)

 

Looks like we've a good chance of ending up below the 81-10 average, whether we'll drop to the 61-90 average of 5.7C remains to be seen.

[Roger J Smith]

I was following that discussion of the climatological day used for CET. Can anyone confirm, are those times used for the final values or just for the provisionals? In other words, is the end of month adjustment caused by a recalibration from the off-set times quoted to the actual 24-hour periods? Or is it caused by reconsideration of the data with the same time periods employed?

 

Must say, those off-set times are quite unorthodox in climatological record keeping. The standard practice when recording highs and lows for first-order weather stations is to go from midnight to midnight local time or at the very worst to pick a z time that is close to local midnight, for example 06z to 06z for a site in the central time zone of North America.

 

Some of the discussion about adjusting mean temperatures in historical records (for climate change purposes) touches on this. If you look through the older climate records in North America, you find  cases where the minimum temperatures were recorded from previous evening to current evening if you see what I'm saying, maybe something like 6 p.m. yesterday to 6 p.m. "today" and if it then fell another ten degrees from 6 p.m. to midnight that isn't in the data. Neither is the warmer midnight high in some cases, the data have a maximum from 6 a.m. onward. With a strong cold front you can get two consecutive days in older data that look something like this -- day one has max 2 C and min -10 C, day two has max -16 C and min -23 C. The two days don't overlap in other words. Now these are largely compensating errors so the adjustment procedure is fairly restrained.

 

I would have imagined that the final values for the CET data set are adjusted to calendar day and represent the max and min on that calendar day. These provisionals will match during "regular" weather because while distorted they still contain the times of day most likely to produce the extremes. But whenever there is an unusual temperature trend up or down, the provisionals won't be very similar to the final numbers. So we tend to see bigger adjustments in winter than summer and in colder months than in milder months.

[The PIT]

Always been 9:00 as far as I'm aware Rodger to take the readings.

[Midlands Ice Age]

I was following that discussion of the climatological day used for CET. Can anyone confirm, are those times used for the final values or just for the provisionals? In other words, is the end of month adjustment caused by a recalibration from the off-set times quoted to the actual 24-hour periods? Or is it caused by reconsideration of the data with the same time periods employed?

 

Must say, those off-set times are quite unorthodox in climatological record keeping. The standard practice when recording highs and lows for first-order weather stations is to go from midnight to midnight local time or at the very worst to pick a z time that is close to local midnight, for example 06z to 06z for a site in the central time zone of North America.

 

Some of the discussion about adjusting mean temperatures in historical records (for climate change purposes) touches on this. If you look through the older climate records in North America, you find  cases where the minimum temperatures were recorded from previous evening to current evening if you see what I'm saying, maybe something like 6 p.m. yesterday to 6 p.m. "today" and if it then fell another ten degrees from 6 p.m. to midnight that isn't in the data. Neither is the warmer midnight high in some cases, the data have a maximum from 6 a.m. onward. With a strong cold front you can get two consecutive days in older data that look something like this -- day one has max 2 C and min -10 C, day two has max -16 C and min -23 C. The two days don't overlap in other words. Now these are largely compensating errors so the adjustment procedure is fairly restrained.

 

I would have imagined that the final values for the CET data set are adjusted to calendar day and represent the max and min on that calendar day. These provisionals will match during "regular" weather because while distorted they still contain the times of day most likely to produce the extremes. But whenever there is an unusual temperature trend up or down, the provisionals won't be very similar to the final numbers. So we tend to see bigger adjustments in winter than summer and in colder months than in mYes, the monthly CET values are calculated in the same way.  The daily values as calculated are simply averaged to get the monthly figures - so each individual day's value is obtained as before, and then all the days in a month are averaged to get the monthly value.  This is done for Maximum and Minimum temperatures, and then the mean temperature is a simple average of maximum and minimum.ilder months.

 

Roger....

 

Yep , it is confusing and you are correct that  a rapid change of temps towards 09:00 around the Midland CET area does appear to cause problems with the unadjusted  daily CET's. The daily unadjusted tmes  are as I showed above (and BFTV suggested) an average of 2 days of max and min  data.  

 

 

As to monthly corrected CET dataset I asked and received the following reply --------

 

"We have spoken to our National Climate Information Centre and they have confirmed as follows

 

Yes, the monthly CET values are calculated in the same way.  The daily values as calculated are simply averaged to get the monthly figures - so each individual day's value is obtained as before, and then all the days in a month are averaged to get the monthly value.  This is done for Maximum and Minimum temperatures, and then the mean temperature is a simple average of maximum and minimum."

 

 

 

So at that point,  on my 4th Email, I gave up! Can anyone translate the message!!

 

The METO help desk didn't really understand any problem  and just sent back old references to the sites and times as per 30 years ago.

.

It was only when I got to the National Climate Information Centre via the helpdesk that I obtained the correct  above info and then I raised the issue of the monthly corrected dataset.

 

My question was about the daily unadjusted figures and they eventually had answered that, but the above on adjusted was not clear to me.

 

MIA.

[Evening thunder]

I thought they always used the 9am-9am meteorological day or 21:00-09:00 for minima because that's what was used before the times of automatic recording made it easier to record at midnight.

 

I presumed they keep it like that for consistency, otherwise you would not be able to compare max/min temp data to older records. Makes sense really.

 

In fact I thought it was the standard UK system for official stations? 

[Summer Sun]

6.2c to the 19th

 

0.9c above the 61 to 90 average

 

Bang on the 81 to 10 average

[damianslaw]

Difficult to call where the CET will end up by month's end. Quite a topsy turvy outlook, next few days average at best and a little below early next week with a projected average/slightly above average spell later in the week. Best bet a finish not too far off the 81-00 average, but can't see a below average 61-90 average now.. It will be the third month in a row very close to the 81-10 average.. a very average start to the year.. getting a bit bored of the word 'average', mind last year I was tired of the word 'warmer'.

Edited March 20, 2015 by damianslaw

[Summer Sun]

6.2c to the 20th

 

0.9c above the 61 to 90 average

 

Bang on the 81 to 10 average

[damianslaw]

Probably see a bit of a drop by middle of next week, perhaps creeping below 6 degrees, thereafter a very slow climb perhaps, upper limit perhaps 6.5 degrees, but with downward adjustments somewhere in the low 6s would be a good bet. Still 10 days to go though..

[Weather26]

It's certainly looking like to it could go either way as to whether we end up below average or not.

Edited March 22, 2015 by Weather26

[Summer Sun]

6.3c to the 21st

 

0.9c above the 61 to 90 average

 

0.1c above 81 to 10 average

[summer blizzard]

 

It's certainly looking like to it could go either way as to whether we end up below average or not.

 

 

Certainly can't see the 1981-2010 average of 6.6C being beaten, especially after corrections.

[Summer Sun]

6.2c to the 22nd

 

0.9c above the 61 to 90 average

 

0.1c below 81 to 10 average

[BornFromTheVoid]

Minimum today is 5.3C while maxima look like reaching the low 10s, so an increase to 6.3C is likely on tomorrows update. 

 

After that, the 06z GFS has the CET at:

 

6.2C to the 24th (5.2)

6.2C to the 25th (4.7)

6.2C to the 26th (5.5)

6.1C to the 27th (5.9)

6.2C to the 28th (7.2)

6.3C to the 29th (9.4)

6.5C to the 30th (11.0)

6.6C to the 31st (12.1)

 

A very mild final 3 days has a big impact on the overall CET. Whether that comes off or not remains to be seen. I'd put the likely range before corrections at 6.1C to 6.8C, so 5.6C to 6.8C after.

[Summer Sun]

6.2c to the 23rd

 

0.8c above the 61 to 90 average

 

0.1c below 81 to 10 average