81 mph gust at Île d'Yeu

[xmetman]

There were gusts to 70 knots (81 mph) for three consecutive hours this morning from Île d'Yeu (07300) as storm Miguel passed by.

I should imagine the observations are done at the lighthouse on what looks a lovely island of the west coast of France. I may have plotted it a few times in my career but never realised it was an island till now!

[Freddie]

Bruce, I noticed that the gust speeds seem a bit at odds with the mean speeds the majority of the time.  So I took a look at the Synops I have access to for 07300.  This is what I have decoded for the period 0000z to 1400z today:

Time     Dir    Mean    Gust

----------------------------

0000     110     12      26

0100     250     10      40

0200     200     09      23

0300     180     13      20

0400     210     21      32

0500     200     30      42

0600     190     33      50

0700     170     29      53

0800     190     40      60

0900     220     33      69

1000     240     20      55

1100     250     17      46

1200     260     15      37

1300     280     21      46

1400     270     19      43

May I ask what is your data source?

Regardless of any discrepancies, it certainly has been a very windy day, and I pity anybody holidaying in that region at the moment.

--
Freddie
Dorrington
Shropshire
115m AMSL
http://www.hosiene.co.uk/weather/
Stats for the month so far: https://www.hosiene.co.uk/weather/statistics/201906JUN.xlsx

[xmetman]

Hi Freddie

I swapped from FSU to OGIMET about eight years ago now.

I can see what you are getting in the 11 UTC ob there are five separate 9 groups! 

The final 91136 group which I'm decoding as the highest gust in the hour 36 MPS is being preceded by a 90730 group.

There seem to be three speeds 20, 24 & 36 in three 91 groups.

Bruce.

[Freddie]

Hi Bruce,

Using the 11 UTC ob for illustration:

91020 is the maximum gust in the 10 minutes before the ob.

90710 91124 is the maximum gust in the hour before the ob - denoted by the 907tt group being 90710.

90730 91136 is the maximum gust in the three hours before the ob - denoted by the 907tt group being 90730.  Note that these groups are in section 5 (after the 555 indicator) not section 3.

So your parsing routine should match on the 907tt 911GG group that occurs in section 3 (after the 333 indicator but before the 555 indicator, if present).

[Colin Youngs]

"I should imagine the observations are done at the lighthouse ... I may have plotted it a few times in my career but never realised it was an island till now!"

The clue is in the word "île" :-)

https://dictionary.cambridge.org/dictionary/french-english/ile?q=%C3%AEle

Colin Youngs

Brussels

[xmetman]

At that time it was only ever a number to me. I wonder now if some assistants ever took any interest in the observations they were plotting. I certainly did and always wanted to analyse and saw them up. So much so I ended up leaving the fax machine running at weekends so I could get some practice in.

I never did very well at French!
Our poor lady teacher could never keep us under control and we nicknamed the weekly double lesson the French Revolution!

[xmetman]

I'll take a look at that code. It's not my favourite area because it's rather complex. My main priority is to get the obs from block 03 decided correctly. As you know some countries are very good but they are fast becoming the exception rather than the rule. I'm beginning to wonder why I download obs for the whole world when I rarely use it.

[Keith (Southend)]

Once you've started doing this and have been doing for years I think the only thing that will stop *us* is the pearly gates. :-)

One thing I would say I have several times contacted Ballester Valor when I have found some anomalies with the decoding from Bufa to Synop Code, so it is possible an error has crept in? Always worth checking they look the same as:

https://weather.cod.edu/digatmos/syn/

Assuming they appear in that dataset.

Keith (Southend)

[xmetman]

Thanks Keith.

You do get kind of hooked after a few years!

I've written to OGIMET on a number of occasions but never get a reply.

I notice that you can now download raw BufR from his site, but there is little extra information in it that's not already in the old style SYNOP after running it through a decoder.

I'm never quite clear who he is or works for or how he gets his data but I would be lost without it!

[xmetman]

We've had a rare thunderstorm earlier this afternoon in our part of the Highlands and it's still a bit on the breezy side across the southeast.

Hopefully these French gusts are OK...

[Keith (Southend)]

This is the email address I have for him:

g...@oxixares.com

Must admit when I emailed a month or so I never got a reply.

Keith (Southend)