(please forgive me is this question is naive - I've gone from not knowing
anything about METARs to my current knowledge in about 40mins whilest doing
other things as well)

In a typical METAR reading there can be more than one entry for the cloud
conditions, ie: ... ... ... ... SCT025 BKN040 ... ... which would indicate
scattered cloud at 2500ft + broken cloud at 4000ft (I think). My question
is: is there a limit to how many cloud entries there can be?

Jim
--
Twitter:@GreyAreaUK
"[The MP4-12C] will be fitted with all manner of pointlessly shiny
buttons that light up and a switch that says 'sport mode' that isn't
connected to anything." The Daily Mash.

"Jim" wrote ...
In a typical METAR reading there can be more than one entry for the
cloud
conditions, ie: ... ... ... ... SCT025 BKN040 ... ... which would
indicate
scattered cloud at 2500ft + broken cloud at 4000ft (I think). My
question
is: is there a limit to how many cloud entries there can be?


.... yes: I was hoping someone with more up-to-date knowledge would
pitch in as I'm a bit rusty on the latest changes now, but what I've
written will give you a flavour of the instructions, and also give you
some idea of the difficulty of using METARs for the purpose intended.

The selection of the layers to be reported are as follows:-
1st group: the lowest layer of any amount.
2nd group: the next layer, which covers *more* than 2 oktas (i.e. SCT
or greater).
3rd group: the next higher layer covering more than 4 oktas (i.e. BKN
or greater).
Additional group to report CB or TCU if not already reported as above.
(Reg 15.9.1.4)

Order is from lowest to highest.

BUT .. and this is important for decoding/interpretation purposes ...
from Reg 15.9.1.1, " If there are no clouds below 1500m (5000ft) or
below the highest minimum sector altitude, whichever is greater, no
Cumulonimbus and no restriction on vertical visibility and the
abbreviations CAVOK and SKC are not appropriate, then the abbreviation
NSC shall be used."

Now, there have been changes to this: for example, 'SKC', sky clear,
is *not* to be used [though I still keep seeing it] and I believe (but
can't find the reference) that *any* CB (and possibly TCU)
irrespective of cloud base should be reported ... my old page for the
FAQ is still on my web site ....

http://booty.org.uk/booty.weather/me...TAR_decode.htm

but the relevant data should also be in the FAQ he-

http://weatherfaqs.org.uk/node/197

.... and *military* airfields in the UK at least *will* include cloud
layers above 5000ft & may have different rules to suit the RAF
operations.

.... however, to return to what I interpret as being your original
intention, using METAR reports to infer the 'state of the sky' is now
more or less impossible.

When I started coding such, we put cloud layers etc., in according to
the SYNOP rules, thus 3CU20, 5SC56, 8CI75 etc. That was changed to the
'FL' style of notation (thus 3CU020, 5SC060, 8CI250), then the 'oktas'
bit was smudged out and non-significant types omitted (SCT020, SCT060,
OVC250), and this quickly got overtaken by the 'nothing at/above
5000ft rule', so that only the first cloud group was reported in my
example, and even then if the VIS was above 9 km, CAVOK would be used
etc., etc.

If your intention is to use METARs to describe the state of sky ... it
can't be done! You can never be sure that all the cloud data is there
to accurately decide whether the sky is genuinely clear, or simply
that there is no significant cloud below 5000ft.


Martin.

--
Martin Rowley
West Moors, East Dorset (UK): 17m (56ft) amsl
Lat: 50.82N Long: 01.88W
NGR: SU 082 023

On 2010-03-26, Martin Rowley wrote:
is: is there a limit to how many cloud entries there can be?


... yes: I was hoping someone with more up-to-date knowledge would
pitch in as I'm a bit rusty on the latest changes now, but what I've
written will give you a flavour of the instructions, and also give you
some idea of the difficulty of using METARs for the purpose intended.

The selection of the layers to be reported are as follows:-
1st group: the lowest layer of any amount.
2nd group: the next layer, which covers *more* than 2 oktas (i.e. SCT
or greater).
3rd group: the next higher layer covering more than 4 oktas (i.e. BKN
or greater).
Additional group to report CB or TCU if not already reported as above.
(Reg 15.9.1.4)

Order is from lowest to highest.

BUT .. and this is important for decoding/interpretation purposes ...
from Reg 15.9.1.1, " If there are no clouds below 1500m (5000ft) or
below the highest minimum sector altitude, whichever is greater, no
Cumulonimbus and no restriction on vertical visibility and the
abbreviations CAVOK and SKC are not appropriate, then the abbreviation
NSC shall be used."

Now, there have been changes to this: for example, 'SKC', sky clear,
is *not* to be used [though I still keep seeing it] and I believe (but
can't find the reference) that *any* CB (and possibly TCU)
irrespective of cloud base should be reported ... my old page for the
FAQ is still on my web site ....

http://booty.org.uk/booty.weather/me...TAR_decode.htm

but the relevant data should also be in the FAQ he-

http://weatherfaqs.org.uk/node/197

... and *military* airfields in the UK at least *will* include cloud
layers above 5000ft & may have different rules to suit the RAF
operations.

... however, to return to what I interpret as being your original
intention, using METAR reports to infer the 'state of the sky' is now
more or less impossible.

When I started coding such, we put cloud layers etc., in according to
the SYNOP rules, thus 3CU20, 5SC56, 8CI75 etc. That was changed to the
'FL' style of notation (thus 3CU020, 5SC060, 8CI250), then the 'oktas'
bit was smudged out and non-significant types omitted (SCT020, SCT060,
OVC250), and this quickly got overtaken by the 'nothing at/above
5000ft rule', so that only the first cloud group was reported in my
example, and even then if the VIS was above 9 km, CAVOK would be used
etc., etc.

If your intention is to use METARs to describe the state of sky ... it
can't be done! You can never be sure that all the cloud data is there
to accurately decide whether the sky is genuinely clear, or simply
that there is no significant cloud below 5000ft.

Thank you very much for an extremely detailed reply :-)

The actual 'sky conditions' aspect is ultimately unimportant, as this is
simply a programming excercise for me. The current challenge will be
to accurately decode a European METAR (as I gather the US ones are slightly
different).

If I ever get it to the stage where it might be of use to people then I'll
release it into the public domain with source code. It'll be a Mac project.

Again, thank you very much for your time.

Jim
--
Twitter:@GreyAreaUK
"[The MP4-12C] will be fitted with all manner of pointlessly shiny
buttons that light up and a switch that says 'sport mode' that isn't
connected to anything." The Daily Mash.

On 2010-03-26, Martin Rowley wrote:
snip
The selection of the layers to be reported are as follows:-
1st group: the lowest layer of any amount.
2nd group: the next layer, which covers *more* than 2 oktas (i.e.
SCT
or greater).
3rd group: the next higher layer covering more than 4 oktas (i.e.
BKN
or greater).
Additional group to report CB or TCU if not already reported as
above.
(Reg 15.9.1.4)

Order is from lowest to highest.

BUT .. and this is important for decoding/interpretation purposes
...
from Reg 15.9.1.1, " If there are no clouds below 1500m (5000ft) or
below the highest minimum sector altitude, whichever is greater, no
Cumulonimbus and no restriction on vertical visibility and the
abbreviations CAVOK and SKC are not appropriate, then the
abbreviation
NSC shall be used."

Now, there have been changes to this: for example, 'SKC', sky
clear,
is *not* to be used [though I still keep seeing it] and I believe
(but
can't find the reference) that *any* CB (and possibly TCU)
irrespective of cloud base should be reported ......

.... I've found it! Indeed, the current regulation states that *any* CB
or TCU should be reported, irrespective of cloud base height. I'm
pretty sure also that if a thunderstorm is reported (e.g. TS), then a
'CB' group *has* to be entered whether the cloud can be 'detected' or
not, either visually or otherwise; there can be occasions for example
where thunder is heard from 8 oktas Stratus!

Martin.




--
Martin Rowley
West Moors, East Dorset (UK): 17m (56ft) amsl
Lat: 50.82N Long: 01.88W
NGR: SU 082 023