"Ken Cook" wrote in message
...
Hi, All,
MetO installed a Campbell Stokes sunshine recorder at the Copley Site on 1st
July.
The electronic R&D recorder is set at its most sensitive and will not go any
lower, as that was the best setting under weak sun. As it is on the 10 metre
anemometer pole it records an extra hour of sun from 04Z-05Z which the
Campbell Stokes, on a screen stand, does not capture. The rest of the day
has no shade for either instrument.
To my surprise, as I thought the electronic would be more sensitive, during
partly cloudy conditions, the C/S records more sun than the electronic, so
much so that the C/S is currently running upwards of 10% higher for the
month so far.
I am sure MetO Quality Control will put me right if I am reading the trace
incorrectly when they look at the cards and visit the site. I have read the
Observer's Handbook time and again and am sure I'm doing it right. It does
seem very subjective, however!
Anyway, the half-way figures for Copley for July 2015, up to 09Z 16th:
Mean max 19.7C Mean min 10.0C Mean 14.9C(+0.2C)
Rain 54.3mm(76%)
Wind WSW 7.4KT(-0.2KT)
Sun(E) 94.3hr(59%)
Sun(C/S) 102.5hr
This morning's ob 09Z-09Z Max 18.4C min 5.5C grass -0.1C rain nil sun 15th
10.6hr(8.9hr electronic)
Ken
Copley
Teesdale
http://www.kencook.magix.net/
Hello Ken. I too use the R&D recorder, and find in summer months it can
record quite a bit less than both the CS recorders at Reading Uni and
Maidenhead, and Stephen Burt's KZ recorder. The June figures I have so far
are Wokingham 166 hrs, Maidenhead 224 hrs and Stratfield Mortimer 225 hrs.
There was some local correspondence about the large variation across the
area in May, and I copy below Stephen Burt's thoughts on the matter. I am
unable to adjust the threshold on my R&D any higher as if I do it causes
spurious sun in certain overcast conditions. I have a spare sensor that I
could install in the near future if my figures fall even more relative to
the others. I would expect the R&D to be the more accurate when compared to
the CS in conditions where there is intermittent sun, the effect larger in
summer than winter.
Dear Bernard -
Yes indeed, I still use my R&D (Instromet) sensor - in fact, I have two, one
at 11 m AGL is my original unit (installed 2001), one at 6 m adjacent to my
K&Z CSD3 unit is one of their new units with an infrared filter photodiode
unit, which was installed in November 2013.
The raw totals for all three for May we
Instromet 1 (at 11 m AGL) 173.28 h
Instromet 2 (at 6 m AGL) 168.36 h
K&Z CSD3 (at 6 m AGL) 180.03 h
My 'climatological' totals are the K&Z reading adjusted slightly for horizon
obstructions early and late using the higher Instromet unit, which is almost
unobstructed; this was 186.71 h in May. (The K&Z unit tends to read higher
than the Instromets in summer, although personally I think the Instromet
threshold is closer to 'sunshine' as estimated by 'hard shadows', but even
so all three normally agree within 10% over a month - within the WMO spec.)
I would urge caution in adjusting the threshold point on the Instromet,
simply because it will introduce inhomogeneity into your records. Your
sunshine values have always been quite similar to mine, and I wouldn't do
anything hasty because of one month's disparity. May's sunshine was very
broken: the average spell length* was just 0.12 hours, compared with 0.97
hours in April, and I'm sure that contributed to the relative over-reading
of the university's C/S unit. Roger and I are looking into the over-reading
of the university's K&Z unit - I suspect the threshold is slightly
mis-calibrated, but short of sending it back to K&Z there's no way to get it
checked. I don't think we have a 'spare' unit in a cupboard to swap it out
with while that's done, even if I could persuade our finance officer to
countenance the expense .
I'd keep your eye on it for a couple of months at least before making any
changes. I think the discrepancies across our region last month are partly
real, but mostly instrumental/exposure/calibration differences.
Best regards
Stephen
* Using the logger, I count every time the 'sunshine = yes' voltage
threshold on the K&Z is exceeded during the day, logging down to a
resolution of 1 min. Total sunshine duration / total spells = average length
of sunshine 'spells', an interesting and extremely variable statistic
--
Bernard Burton
Wokingham Berkshire.
Weather data and satellite images at:
http://www.woksat.info/wwp.html
---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus