Data Summary - possible calculation error

[RayProudfoot]

Winter is attributed to the year in which it starts. That is the meteorological convention. I'll try and find a link that specifically states this. One I know of in passing is this: https://www.metoffice.gov.uk/weather/le ... nter-start

Freddie, that would mean me changing lots of records in my Season Averages page which I'm not prepared to do. It seems illogical to use the year winter starts when most of it occurs in the year it ends. I might be wrong but at least my own records and the php code now match as I've just reverted to what I had before today's update.

I did find that site myself but it doesn't state which year winter records should be stored. Winters of course are described as 1962/63 for clarity but when you only have a single year to store it against it makes more sense to me to store it in the year that has two months of winter rather than the year that only has one. Each to their own I suppose.

I wonder how antipodeans store their summer records?

[Mapantz]

I’ve had a look at the Met Office for guidance on how to store winter records but it doesn’t give a definitive answer.

Winter is attributed to the year in which it starts. That is the meteorological convention. I'll try and find a link that specifically states this. One I know of in passing is this: https://www.metoffice.gov.uk/weather/le ... nter-start

That's exactly what I explained.

Take a look Ray, I wasn't wrong.

It has been like that for as long as records have been kept.

[RayProudfoot]

That's exactly what I explained.

Take a look Ray, I wasn't wrong.

It has been like that for as long as records have been kept.

Sorry but that link does not say definitively which year winter records should be stored against. I know meteorological winter starts on 1 December and ends on the last day of February but please show me where it says it should be stored against the year winter starts.

[RayProudfoot]

@mapantz, I've just looked at your data summary page and you have winter records stored against 2020. But how is that possible if they should be stored against the year winter starts? Winter 2020 should not be populated until December 2020.

Your station's records started on 1 March 2016 and your first winter season average is stored against 2017. So you are using the same rule as me.

[Mapantz]

@mapantz, I've just looked at your data summary page and you have winter records stored against 2020. But how is that possible if they should be stored against the year winter starts? Winter 2020 should not be populated until December 2020.

I use the same one that Mark does, but he updated his version and I haven't.

[RayProudfoot]

I use the same one that Mark does, but he updated his version and I haven't.

Yes, I can see that now. Given we all seem to be doing our own thing I shall stick to what seems best for me. I very much doubt my visitors will spot the difference.

[freddie]

Winter is attributed to the year in which it starts. That is the meteorological convention. I'll try and find a link that specifically states this. One I know of in passing is this: https://www.metoffice.gov.uk/weather/le ... nter-start

Freddie, that would mean me changing lots of records in my Season Averages page which I'm not prepared to do. It seems illogical to use the year winter starts when most of it occurs in the year it ends. I might be wrong but at least my own records and the php code now match as I've just reverted to what I had before today's update.

I did find that site myself but it doesn't state which year winter records should be stored. Winters of course are described as 1962/63 for clarity but when you only have a single year to store it against it makes more sense to me to store it in the year that has two months of winter rather than the year that only has one. Each to their own I suppose.

Having worked in meteorology for 35 years I'm pretty sure I know which year winter is attributed to. I wish I could find a link with the definition! I wasn't actually asking you to change anything - merely stating the meteorological convention.

[RayProudfoot]

Having worked in meteorology for 35 years I'm pretty sure I know which year winter is attributed to. I wish I could find a link with the definition! I wasn't actually asking you to change anything - merely stating the meteorological convention.

Freddie, I mean no disrespect and a link would clear things up. What would be hard is resetting my brain as I've had it hard-wired to accept winter records are stored against the year it ends.

[Mapantz]

I think it is simply down to the fact that the MetOffice use meteorological seasons to keep their data 'tidy' as astronomical seasons don't always fall on the same day. A change to a new year has to happen, and it just happens to be in a Winter month (unless you're in the land of the 'roos)

So, that means that the 3 months of Winter data will be in two different years, and so for example, they'll have their 3 months of data for the 'Winter of 2020/2021'

Here's data from 2019/2020's Winter: https://www.metoffice.gov.uk/binaries/c ... r_2020.pdf

[beteljuice]

BTW .... some cultures in South Asia have six seasons, and in Finland and Sweden seasons are based solely on temperatures above and below freezing, so their seasons begin on different days each year depending on their climate.

[RayProudfoot]

I think it is simply down to the fact that the MetOffice use meteorological seasons to keep their data 'tidy' as astronomical seasons don't always fall on the same day. A change to a new year has to happen, and it just happens to be in a Winter month (unless you're in the land of the 'roos)

So, that means that the 3 months of Winter data will be in two different years, and so for example, they'll have their 3 months of data for the 'Winter of 2020/2021'

Here's data from 2019/2020's Winter: https://www.metoffice.gov.uk/binaries/c ... r_2020.pdf

Look at pages 2-5. Look at the title. Winter 2020! Not Winter 2019. Surely that proves my point. Even the document title says 2020 and not 2019. I rest my case.

[mcrossley]

Here's data from 2019/2020's Winter: https://www.metoffice.gov.uk/binaries/c ... r_2020.pdf

Cross posted with Ray!...
In that PDF, the diagrams are labelled "Winter 2020" though.

And for instance look at the climate data sets, eg: https://www.metoffice.gov.uk/pub/data/w ... ate/UK.txt
where Winter 2019/20 is put into the year 2020 row.



Edit: we need a popcorn smiley

[Mapantz]

Not sure what you're looking at? It's there in black & white

Winter 2019/2020
The following represents an assessment of the weather experienced across the UK during
Winter 2019/20 (December, January and February) and how it compares with the 1981 to
2010 average.

Forget what the diagrams/images/title of document says - they probably have a reason for that. However, the truth of the matter is, the MetOffice will ALWAYS use Winter of 2009/2010 or 2019/2020

It's been like that for.. well, ever! It's well known across the British meteorological community. I'm really not sure why there's such an extensive debate about it? I actually find it rather odd that you would define it any other way.

[RayProudfoot]

Not sure what you're looking at? It's there in black & white

Winter 2019/2020
The following represents an assessment of the weather experienced across the UK during
Winter 2019/20 (December, January and February) and how it compares with the 1981 to
2010 average.

Look at the title on the three maps showing the variations. Winter 2020.

[sfws]

PLEASE NOTE MY CHANGE OF TOPIC TITLE FOR MY POST

I wonder how antipodeans store their summer records?

The BOM in Australia attributes summer to the year in which it ends; for 2019 example see page title in browser for http://www.bom.gov.au/climate/current/s ... mary.shtml and then look at page title in bold when you read actual page. For them 2020 already has summer season completed, they are still in autumn.

I start my years with September, to my mind that is when the weather really changes in a way that I experience, lower sun, shorter days, a feeling that as summer is over it is a new beginning. It is when people say "after such a mild winter, we expected a better summer" or similar rot, yes I think most ordinary people in UK connect each winter to a following summer.

I do follow the convention that each season is 3 month long, that does lead to simplification unlike using any astronomy data.

I don't see a sudden change in climate swapping any other months, yes sometimes the precise weather chooses to change with the month, but not its general feel.

Not from February to March, nor from November to December [I would agree with Freddie (to my mind the Met O web site revamp may make it look friendlier, but it is a lot harder to find what you want than the old design) that is UK standard], nor from December to January, and not September to October (that Steve Loft selected as default for Chill Season start).

With a September start I can easily compare the chill hours and frost seen for gardeners, and get a much better comparison of whether it has been a wet or dry season. That is my choice. I swapped from the paper records of most of my life to computer file based ones in 2009, from then onwards I maintained both daily and monthly summaries, made easier by swapping to Cumulus 1 from September 2010. I first did seasons starting in September, to match my first use of Cumulus. For a couple of years I experimented with different starting months. By 2014 when I had all my data summaries in a database, I settled up on my September knowing it was easy to recalculate season data for past data if I ever regretted using September start. Although I restarted weather recording in my current home just last January, a helpful meteorologist in a nearby village who has monthly spreadsheets available on his web site helped me get data back to September (he responded to my PMs for what was not on his website) so here my web scripts still report by day, by month, and by season starting in September.

I'm not a professional meteorologist, just someone fascinated by weather all my life, and I do what suits me, not what some professional meteorologist says, or what some third party package does. I give advice, but allow freedom for others to disagree. Equally I don't see a need to follow the crowd.