Jump to content
Cold?
Local
Radar
Snow?

March 2018 C.E.T. forecasts and optional EWP contest


Recommended Posts

It seems that very little adjustment was made to the final CET figure this month. Back to my previous question... I'm sure this has been explained previously but I can't remember - does anyone know why the final figure tends to be adjusted downwards?

Link to post
Share on other sites
  • Replies 275
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Popular Posts

Now that the dust seems to have firmly settled, it's worth reiterating the daily record for the 1st (-3.8C). Not only was it a daily record, but also the fourth coldest March day on record. To put tha

-3.5 to the 1st 8.1 below the 61 to 90 average 8.5 below the 81 to 10 average ________________________________

My impression is that a large part of the reset from provisional to final CET is to remove values for max and (more often) min that are outside the boundaries of the 24-hour calendar day and to introd

Posted Images

9 minutes ago, Stargazer said:

It seems that very little adjustment was made to the final CET figure this month. Back to my previous question... I'm sure this has been explained previously but I can't remember - does anyone know why the final figure tends to be adjusted downwards?

A small reduction to off set UHI effect is their excuse.

  • Thanks 1
Link to post
Share on other sites
59 minutes ago, DAVID SNOW said:

A small reduction to off set UHI effect is their excuse.

Is this actually true? I've never seen anything official on the matter. I could have sworn someone on this forum once contacted the Met Office and found out that the provisional data are collected from different sites to the finalised data, serving as a good guide as the month progresses. Perhaps I'm going mad.

Link to post
Share on other sites

My impression is that a large part of the reset from provisional to final CET is to remove values for max and (more often) min that are outside the boundaries of the 24-hour calendar day and to introduce midnight minima which always form a part of the normalized calendar day temperature data. The more often this has to be done, the further down the adjustment will take the CET. If we have a very cloudy month with little day to day temperature change (as was often the case despite that one big drop) then the provisional data will usually be close to the final data. 

I also think there may be a policy of consulting more numerous locations that adjusts the final value. As to eliminating the UHI, the stations they select are probably based on that criteria but I don't think they adjust further than this simple choice of location, which is the step in which they factor out UHI influences. 

This month demonstrated the other little-known principle that the end of month adjustment must not work in my favour. ;)

 

  • Thanks 4
Link to post
Share on other sites

Whilst I'm a bit sad that March 2018 wasn't as cold as it could have been, I am happy that we finally achieved some back to back cold proper months.  2 so far.  We have to go back to the "big 6" of "January through June 2013 for the next last period of proper cold back to back months.

Edited by Lettucing Gutted
  • Like 4
Link to post
Share on other sites

Months below the 1971-2000 average

Mar 2018: 4.9 (-1.4)
Feb 2018: 2.9 (-1.3)
Dec 2017: 4.8 (-0.3)
Nov 2017: 6.8 (-0.1)
Sep 2017: 13.5 (-0.2)
Aug 2017: 15.6 (-0.4)

What causes clustering? 

Sep 2015: 12.6 (-1.0)
Aug 2015: 15.9 (-0.3)
Jul 2015: 15.9 (-0.6)
Jun 2015: 14.0 (-0.1)
May 2015: 10.8 (-0.5)

  • Like 4
  • Thanks 1
Link to post
Share on other sites

Looks QBO related when you line them up in clusters every 27 months or so

Dec 2010 to Mar 2013 to Jun 2015 to Sep 2017

however going back further you get off this cold track (e.g. June 2006).

 

  • Thanks 4
Link to post
Share on other sites

The spreadsheet is ready to be downloaded as soon as 4.9c is confirmed.

Only one person got this spot on - Kentish Man.

Edited by J10
  • Thanks 3
Link to post
Share on other sites

Can't see it being Urban heat island because if you listen to the forecasts the adjustment is a lot more than 0.xx degrees.

  • Thanks 1
Link to post
Share on other sites
39 minutes ago, snowray said:

Were there any adjustment down in the end or did we finish up without any adjustments at 4.9c in the end?

Looks like 4.9c to me https://www.metoffice.gov.uk/hadobs/hadcet/cet_info_mean.html

Not sure if this is before adjustments or if this is confirmed at 4.9c though ?

Edited by Dancerwithwings
  • Like 2
Link to post
Share on other sites
4 hours ago, The PIT said:

Can't see it being Urban heat island because if you listen to the forecasts the adjustment is a lot more than 0.xx degrees.

If you go to the Hadley Centre CET home page, it says at the top that the data is adjusted to allow for urbane warming, and that a -0.2 correction is applied to the mean.

  • Thanks 2
Link to post
Share on other sites

 

Now that 4.9c has been confirmed, here are the figures.

Monthly (and Seasonal)

Kentish Man leads after getting it spot on with 7 players 0.1c out.

image.thumb.png.313af99eaefc620ee9fa22a395b3bb6d.png

The top 10 in the seasonal comp is shown below with vizzy2004 and chrisbell-nottheweatherman finishing the top 3.

image.thumb.png.fc9684b4d1eb45e4943ab2f4eb2246dd.png

Overall

Norrance retains the lead with Man With beard up to 2nd and Dkeane3 up to 3rd.

image.thumb.png.3b88a43bdb3f218f3d230202eb9e77c9.png

Mar 2018 CET.pdf

March 2018 CET.xlsx

 

 

Edited by J10
  • Like 1
  • Thanks 4
Link to post
Share on other sites

Just 0.1C out, although I was hoping for a downward adjustment!  However, although we've had colder Marches, I think March 2018 was the most snowy I've personally ever experienced.

Edited by Don
  • Like 1
Link to post
Share on other sites
3 hours ago, DAVID SNOW said:

If you go to the Hadley Centre CET home page, it says at the top that the data is adjusted to allow for urbane warming, and that a -0.2 correction is applied to the mean.

Ah yes, thank you. To quote,

"Since 1974 the data have been adjusted to allow for urban warming: currently a correction of -0.2 °C is applied to mean temperatures.",

Assuming that this is only applied after all the data have been collected, this explains, at least in part, the origin of the corrections (which tend to be between 0.0C and -0.4C, i.e. centred on -0.2C).

Edited by Relativistic
  • Like 1
  • Thanks 1
Link to post
Share on other sites

In case of difficulty accessing the files, I have attached the snapshot figures taken from the spreadsheet.

Any comments with downloading files or requested formats would be appreciated. :)

Monthly Scores

image.thumb.png.a14280ba52e7740d62ee3b3924d4b811.png

Overall Scores

image.thumb.png.2a17dd309c64f851aa6d4716c936df1d.png

  • Like 2
  • Thanks 3
Link to post
Share on other sites
2 hours ago, Don said:

Just 0.1C out, although I was hoping for a downward adjustment!  However, although we've had colder Marches, I think March 2018 was the most snowy I've personally ever experienced.

Yes it will certainly be remembered as a good March for snow along with 2013. Though that month had a much lower CET of 2.7c. But unlike then the snow came much earlier in the month this year and affected more of southern England than then as well from what I can gather. One things for sure though both those months have proved beyond doubt that even though March is technically the first month of meteorological Spring, you can still get very cold and snowy weather that late, even these days.

Link to post
Share on other sites

Here is the latest thrilling update of

r O b o T i  ___ 4 Ca S T e r S

They walk among us !!

___ Past months ____

____________DEC _________________________ JAN ________________________ FEB

"Forecaster" _ Error _ Pts (range) _ rank (range) __ Error _ Pts (range) _ rank (range) __ Error _ Pts (range) _ rank (range)

1987-2016*__ +0.2 __ 89.4 - 93.4 __ 6 to 9 (88-17)>--0.6 __66.7 - 74.1 _ 15 to 19 ______ +1.9 _ 31.2 - 35.1 _ 51 to 54

1981-2010 __ --0.2 __ 89.4 - 93.4 __ 6 to 9 ______ --0.9 __ 50.0 - 55.6 _ 25 to 28 ______ +1.5 _ 48.1 - 49.4 _ 40 to 41

Consensus __--0.6 __ 71.0 - 74.9 __20 to 23 _____--0.9 __ 50.0 -55.6 _ 25 to 28 _______+1.1 _ 57.1 - 58.4 _ 33 to 34

* becomes 1988-2017 from JAN onward.

 

________ Current Month __________________________ Contest year averages to date (4 mo)___

"Forecaster" _Mar Error_Mar Points (range)_rank (range) // mean abs error_mean points_mean rank

1988-2017 ___ +1.9 _____ 20.8 - 22.1 ____ 60 to 61 ________ 1.15 ______52.0 - 56.2 __ 33 to 36

1981-2010 ___ +1.7 _____ 26.1 - 26.1 ____ 57 to 57 ________ 1.075_____ 53.4 - 56.1 __ 32 to 34

Consensus ___--0.3 _____ 78.9 - 85.5 ____ 12 to 17 ________ 0.725_____ 64.3 - 68.6 __ 23 to 26

__________________________________________________________________

Our consensus continues to improve vs the random "normal" values for recent 30-year intervals as most forecasters pick up signs of an impending significant negative anomaly in both February and now for March. The errors are scored as in the contest, among those of similar magnitude. In March, no forecaster picked 6.6 (the 1981-2010 normal value) but one had 3.2, a similar error in the opposite direction, and that forecast's rank and points are used. For the more recent normal 1988-2017 one forecast matched its value (6.8) and another was a similar amount below (3.0) the outcome (4.9). Meanwhile, six forecasters were among the group with the same absolute error (only 0.3 deg) as our consensus, three matching it and three a similar distance above the outcome. Eleven forecasts were better than our consensus.

If these robot forecasters were in the contest, they would be ranked about where their mean rank lies, possibly a bit higher but not much (can't directly compare because there are other score variables which I don't track for them such as accuracy points). In general, none of these are performing as well as they did in the previous year when they all ended up in the top ten. 

The average errors (not absolute) of the three so far come out to -0.2 for consensus (our bias is cold), +0.5 for 1981-2010 and +0.85 for 1988-2017. The normals are proving to be unreliably warm in the past two months, they were closer in the first two. 

___________________________________________________________________________________________________

THE NEXT EXCITING EPISODE OF rObOTi C __F or EC asT e Rs may occur about this time next month. Find something more important to do. 

"We will be back."

Link to post
Share on other sites

The EWP for March 2018 is 109.0 mm (just posted).

I will score the forecasts and post updates. 

The top five scores are:

MIDLANDS ICE AGE ___ 10.0 (112.5 mm)

ROGER J SMITH _______ 9.8 (115.5 mm)

JONBOY ______________ 9.6 (120.0 mm)

WEATHER26 ___________9.5 (96.0 mm)

NORRANCE ___________ 9.3 (95.0 mm)

The rest of the scores basically follow the forecasts down to the lowest one (27.0 mm) then pick up two that were far higher than Jonboy's 120 mm (200 and 357 mm).

I won't be able to post these scores very soon today, look for them by this evening (in the scoring thread, will post a link when I report here on the top five annually after March).

Edited by Roger J Smith
Link to post
Share on other sites
1 hour ago, Roger J Smith said:

The EWP for March 2018 is 109.0 mm (just posted).

I will score the forecasts and post updates. 

The top five scores are:

MIDLANDS ICE AGE ___ 10.0 (112.5 mm)

ROGER J SMITH _______ 9.8 (115.5 mm)

JONBOY ______________ 9.6 (120.0 mm)

WEATHER26 ___________9.5 (96.0 mm)

NORRANCE ___________ 9.3 (95.0 mm)

The rest of the scores basically follow the forecasts down to the lowest one (27.0 mm) then pick up two that were far higher than Jonboy's 120 mm (200 and 357 mm).

I won't be able to post these scores very soon today, look for them by this evening (in the scoring thread, will post a link when I report here on the top five annually after March).

Wow..

Thanks for that Roger and your hard work.

I didn't think I had a chance going so high.

Improved by one place (we was robbed!) last month.

 Look for me  in last place this month!!

 MIA

 

Edited by Midlands Ice Age
  • Like 1
Link to post
Share on other sites

The full results of the EWP contest can now be found here:

https://www.netweather.tv/forum/topic/75427-cet-competition-the-scores-ongoing/?page=8

This is the top 12 for the annual contest after March ... 

Rank __ FORECASTER ______________ Dec _ Jan _ Feb _ Mar ____ TOTAL ___ last month

_01 ___ JONBOY ___________________ 9.2 __ 7.8 __ 6.4 __ 9.6 ____ 33.0 ____ _ 02

_02 ___ SINGULARITY _______________5.2 __ 7.2 __ 9.8 __ 7.7 ____ 29.9 ____ _ 05

_03 ___ VIZZY2004 __________________9.0 __ 9.0 __ 8.9 __ 2.6 ____ 29.5 ____ _ 01 

_04 ___ STEWFOX __________________7.0 __ 9.4 __ 4.0 __ 8.9 ____ 29.3 ____ _ 06

_05 ___ DKEANE3 ___________________7.2 __ 7.6 __ 5.2 __ 9.1 ____ 29.1 ____ _ 07

_06 ___ STEVE B ___________________ 7.4 __ 8.2 __ 3.5 __ 8.8 ____ 27.9 ____ _ 11

_07 ___ NORRANCE _________________7.0 __ 2.2 __ 8.4 __ 9.3 ____ 26.9 ____ _ 14

_08 ___ MIDLANDS ICE AGE __________ 3.4 __ 3.3 __ 9.8 __10.0 ___ 26.5 ____ _ 19

_09 ___ DR (S) NO __________________ 9.8 __ 8.8 __ 1.2 __ 6.2 ____ 26.0 ____ _ 08 

_10 ___ CHRIS BELL NOT THE WxMAN_10.0 __6.8 __ 2.6 __ 6.4 ____ 25.8 ____ _ 10

_11 ___  BORN FROM THE VOID ______ 6.4 __ 9.8 __ 3.5 __ 5.7 ____ 25.4 ____ _ 09

_12 ___ POLAR GAEL _______________ 4.6 __ 9.4 __ 8.6 __ 2.6 ____ 25.2 ____ _ 03

 

There was one minor error carried over from the annual table after February, the total score of DR(S)NO failed to include the 1.2 for February and thus the ranks from 8th to 12th in the previous table were out of order, in case anyone looking at their previous rank notices a change (in that range only). DR(S)NO had been tied 11th with Syed2878 but that rank should have been 12th, Dr(S)No moved up to 8th and anyone who was 8th to 10th went to 9th, 10th and 11th in the adjusted Feb rankings.

I have backed this up with an excel file and did not find any other previous errors in addition and I hope there aren't any new ones, check your scores though.

The rest of the scoring can be seen at the link above. 

  • Like 1
  • Thanks 4
Link to post
Share on other sites

Over there in the scoring thread (link above in EWP reports) I have posted a table showing ranks of all forecasters who enter both CET and EWP contests, so you can compare directly (your rank for CET is against those who enter both, about one-quarter of the field is temperature-only, you'll see your rank too in a discussion section).

These are just the ranks for total points to make the comparison more exact. 

Your overall rank in the CET may be slightly different as other factors are used to derive a total score there. 

But what's interesting for anyone who enters both contests is that there are cases of much different outcomes in the two contests, a few people seem to be really good at the precip and fair to middling (or lousy) at temperatures, and vice versa. Others have a similar skill level in both (some good, some middling, some poor).

You might want to take a look and see what it tells you about your forecasting because if you're good at one and not the other, it might be a case of faulty assumptions about something you are predicting correctly if you get my drift. 

  • Like 2
Link to post
Share on other sites
On 04/04/2018 at 11:24, Roger J Smith said:

The EWP for March 2018 is 109.0 mm (just posted).

I will score the forecasts and post updates. 

The top five scores are:

MIDLANDS ICE AGE ___ 10.0 (112.5 mm)

ROGER J SMITH _______ 9.8 (115.5 mm)

JONBOY ______________ 9.6 (120.0 mm)

WEATHER26 ___________9.5 (96.0 mm)

NORRANCE ___________ 9.3 (95.0 mm)

The rest of the scores basically follow the forecasts down to the lowest one (27.0 mm) then pick up two that were far higher than Jonboy's 120 mm (200 and 357 mm).

I won't be able to post these scores very soon today, look for them by this evening (in the scoring thread, will post a link when I report here on the top five annually after March).

Down as 119mm here

HadEWP_Mar2018.gif

Link to post
Share on other sites
3 minutes ago, Weather-history said:

Down as 119mm here

HadEWP_Mar2018.gif

Confirmed by the daily figures

https://www.metoffice.gov.uk/hadobs/hadukp/data/daily/HadEWP_daily_qc.txt

119.1mm which makes its the 8th wettest March on record for England and Wales

 

177.5 1947

160.4 1981

134.7 1818

127.0 1979

120.0 1919

119.7 1914

119.5 1836

119.1 2018

  • Thanks 1
Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...