"Stephen Davenport" wrote in message
...
How is the WOW network QC'd, do you know?

It's not as far as I'm aware. This was a 'feature' that was requested when
the project was being developed but I've never seen any mention of QC on
either individual items of data or the data record of the station more
generally which is a shame. (Yes I know that there's star rating of the
general station set-up which I suppose is some help, but it isn't QC of the
data itself.) There is an argument that if you have a dense enough mesh of
observations then the mean values will be fairly obvious from eyeballing a
cluster of nearby stations and similarly the outliers will be equally
obvious, but I'd still rather see proper QC.

My guess is that this is a PC decision in wanting the WOW project to be as
inclusive as possible, although I can understand that arriving at a set of
rules to QC the data wouldn't necesssarily be trivial. But it still bothers
me a little that what is being sought as a body of supplementary scientific
data is then apparently not subject to any real scrutiny. Maybe there's a
plan for QC in a subsequent phase of WOW.

John Dann
www.weatherstations.co.uk

"johnd" wrote in message
...

"Stephen Davenport" wrote in message
...
How is the WOW network QC'd, do you know?

It's not as far as I'm aware. This was a 'feature' that was requested when
the project was being developed but I've never seen any mention of QC on
either individual items of data or the data record of the station more
generally which is a shame. (Yes I know that there's star rating of the
general station set-up which I suppose is some help, but it isn't QC of
the data itself.) There is an argument that if you have a dense enough
mesh of observations then the mean values will be fairly obvious from
eyeballing a cluster of nearby stations and similarly the outliers will be
equally obvious, but I'd still rather see proper QC.

My guess is that this is a PC decision in wanting the WOW project to be as
inclusive as possible, although I can understand that arriving at a set of
rules to QC the data wouldn't necesssarily be trivial. But it still
bothers me a little that what is being sought as a body of supplementary
scientific data is then apparently not subject to any real scrutiny. Maybe
there's a plan for QC in a subsequent phase of WOW.


Hello John. There is some very rudimentary QC to flag consistent ridiculous
values and inform the observer. I too was involved from the early stage and
as you say the philosophy is to encourage observations. At some stage
serious work will be done to try and incorporate the obs into NWP and it is
at that stage that rigorous QC should be undertaken. The very high res.
models will reject silly obs anyway. Always difficult though especially in
the more data sparse areas. Is that ob. of 28C in the Scottish Highlands
realistic, for example? Or that report of 60 mm rain on Dartmoor realistic?

Cheers,
Will
--

On Wednesday, August 1, 2012 8:41:12 AM UTC+1, wrote:
"johnd" wrote in message

...



"Stephen Davenport" wrote in message

...

How is the WOW network QC'd, do you know?



It's not as far as I'm aware. This was a 'feature' that was requested when

the project was being developed but I've never seen any mention of QC on

either individual items of data or the data record of the station more

generally which is a shame. (Yes I know that there's star rating of the

general station set-up which I suppose is some help, but it isn't QC of

the data itself.) There is an argument that if you have a dense enough

mesh of observations then the mean values will be fairly obvious from

eyeballing a cluster of nearby stations and similarly the outliers will be

equally obvious, but I'd still rather see proper QC.



My guess is that this is a PC decision in wanting the WOW project to be as

inclusive as possible, although I can understand that arriving at a set of

rules to QC the data wouldn't necesssarily be trivial. But it still

bothers me a little that what is being sought as a body of supplementary

scientific data is then apparently not subject to any real scrutiny. Maybe

there's a plan for QC in a subsequent phase of WOW.





Hello John. There is some very rudimentary QC to flag consistent ridiculous

values and inform the observer. I too was involved from the early stage and

as you say the philosophy is to encourage observations. At some stage

serious work will be done to try and incorporate the obs into NWP and it is

at that stage that rigorous QC should be undertaken. The very high res.

models will reject silly obs anyway. Always difficult though especially in

the more data sparse areas. Is that ob. of 28C in the Scottish Highlands

realistic, for example? Or that report of 60 mm rain on Dartmoor realistic?



Cheers,

Will

--

I doubt whether WOW observations will ever be used seriously. There are too many uncontrollable variables, exposure, instrument/sensor type, calibration, etc, etc, to make the data worthy of use other than "advisory".