I originally made my manual rain gauge from a standard size funnel.
Collected in a per-tared aluminium container, weighed on my wife's quite
accurate cooking scales in gms to give the exact volume, did the maths and
by entering in a spreadsheet got what seemed to be quite accurate results.
This could be a bit tedious in the old days when it used to rain here (;-)
so I bought an electronic tipping gauge.This works well and gives good
results but it presumably only tips with every complete whole mm. (This
model). For example it recorded 3mm last night but this could have been
3.9mm. If it rains again soon it will only take 0.1mm to get 1mm recorded.
I'm not particularly bothered because these are not official figures and it
will even out over time (unless of course what is already in the tray
evaporates before the next event, which will lead to an overall low bias).
Incidentally I am using an electronic gauge purchased from
www.conrad-direct.co.uk which is identical to the Oregon RGR122 (comes from
same factory) but only costs £22-00. It is useful in that it keeps a running
total and the last 9 days individual amounts.

Dave

Someone one another forum recently posted a link relating to rain
gauge accuracy issues:

http://tinyurl.com/zhfw8

(NB This is a link directly to a PDF document so it may be better -
depending on your news reader - to download the file and then open it
in Acrobat rather than trying to open it directly.)

None of the discussion will be new to anyone familiar with rain gauge
use, but it's quite a handy and readable summary of the topic.

John Dann
www.weatherstations.co.uk

I built a tipping bucket type rain a while ago. 0.5 mm of rain per tip which
activates a magnetic reed switch and it is then logged.

I have an official Met Office funnel type rain gauge which I use for all my
rainfall measurements.

When I compare the two I find that the tipping bucket gauge generally
under-records by about 10%.
_______________
Nick G
Exe Valley, Devon
50 m amsl

Nick G wrote:
I built a tipping bucket type rain a while ago. 0.5 mm of rain per tip which
activates a magnetic reed switch and it is then logged.

I have an official Met Office funnel type rain gauge which I use for all my
rainfall measurements.

When I compare the two I find that the tipping bucket gauge generally
under-records by about 10%.
_______________
Nick G
Exe Valley, Devon
50 m amsl



My knowledge of mechanics is non existent but is it not possible to make it tip
at approx 0.46mm and log 0.5mm, thus making it record more accurately than -10%?


--
Gianna

Jonathan Stott wrote:
Keith (Southend) wrote:

That thought has crossed my mind, but the Davis ones tip every 0.2mm,
so it's probably not such an issue. In fact quite often a heavy due
can produce 0.2mm some mornings.

Obviously a tipping gauge will always under-read the precipitation - I
guess more so when the rain is heavier. Where does the water that's
draining through the funnel go while the buckets are tipping?

If we start looking for too great an accuracy, are we not in danger of
only looking at the area covered by the 6" diameter of the bucket? The
level of rainfall may have been different in another part of the garden.


--
Howard Neil
(western end of the Brecon Beacons National Park, at 235 metres asl)

Jonathan Stott wrote:
Howard Neil wrote:

Obviously a tipping gauge will always under-read the precipitation -
I guess more so when the rain is heavier. Where does the water that's
draining through the funnel go while the buckets are tipping?

If we start looking for too great an accuracy, are we not in danger of
only looking at the area covered by the 6" diameter of the bucket? The
level of rainfall may have been different in another part of the garden.

Surely the best solution then is to use several gauges mounted in
different places and take the mean of them all?


But do we need such precision?

--
Howard Neil

Howard Neil wrote:


But do we need such precision?


Probably not, the important thing is that it records accurately overall,
+0.2 for every say 2.0mm is a 10% over read. It is this calibrating
process that is more important.

--
Keith (Southend)
http://www.southendweather.net

"Jonathan Stott" wrote in message
...
Keith (Southend) wrote:

That thought has crossed my mind, but the Davis ones tip every 0.2mm, so
it's probably not such an issue. In fact quite often a heavy due can
produce 0.2mm some mornings.

Obviously a tipping gauge will always under-read the precipitation - I
guess more so when the rain is heavier. Where does the water that's
draining through the funnel go while the buckets are tipping?


From 1979-2000 I used a standard rain gauge to measure rainfall. From 2001 I
began using a Davis AWS as well and noticed discrepancies with the standard
rain gauge readings-instruments next to one another by the way. After
raising this problem on USW and tweaking as advised the readings fell more
into line. However this success only applies to the gentle and occasional
Essex rain for once the rainfall intensity increases the two readings drift
apart with the AWS under reading by about 10%. Now I refer to both gauges
and adjust the Weatherlink database as and when.

Incidentally this problem of under reading has occurred on all 3 Davis rain
gauges I have used over the years

All the best


--
George in Epping, West Essex (107m asl)
www.eppingweather.co.uk
www.winter1947.co.uk

Keith (Southend) wrote:
Howard Neil wrote:


But do we need such precision?


Probably not, the important thing is that it records accurately overall,
+0.2 for every say 2.0mm is a 10% over read. It is this calibrating
process that is more important.


I would agree that accuracy better than +10% is important. My earlier
comment was a little short for what I was trying to say. It was
suggested that several gauges be used and the amounts averaged out. All
that would achieve is an average of inaccurate readings. However, there
will still be natural differences in actual rainfall within a locality
and I don't see extreme precision as being necessary. What I think is
more important is consistency along with reasonable accuracy (say + or -
1%?)

--
Howard Neil

Howard Neil wrote:

I would agree that accuracy better than +10% is important. My earlier
comment was a little short for what I was trying to say. It was
suggested that several gauges be used and the amounts averaged out. All
that would achieve is an average of inaccurate readings. However, there
will still be natural differences in actual rainfall within a locality
and I don't see extreme precision as being necessary. What I think is
more important is consistency along with reasonable accuracy (say + or -
1%?)

I would suggest that accuracy is less important than being able to
compare readings between different sites. In this case, using a standard
exposure would take preference over a slight drop in accuracy. Obviously
the standard exposure would be selected such that any decrease in
accuracy caused by the exposure is minimised.

--
Jonathan Stott
Canterbury Weather: http://www.canterburyweather.co.uk/
Reverse my e-mail address to reply by e-mail