On Sun, 14 May 2006 10:03:39 +0100, Jonathan Stott wrote:

Obviously a tipping gauge will always under-read the precipitation - I
guess more so when the rain is heavier.

I think it's actually more accurate are reasonable rates. Less chance of
evaporation emptying a nearly full bucket. I know my gauge will not
register anything for the "heavy mist" type rain no matter how long it
goes on for.

Where does the water that's draining through the funnel go while the
buckets are tipping?

Into one of the buckets, either the one about to empty or into the empty
one depending on the precise timing of the drop. The buckets flip (well
see saw) *very* quickly, at least they do in my gauge.

--
Cheers
Dave. pam is missing e-mail

Where does the water that's draining through the funnel go while the
buckets are tipping?

As far as I can tell with mine as soon as the first bucket tips the second
bucket is immediately in place to catch the rain draining through. it just
rocks backwards and forwards on a pivot.If you meant ultimately - it just
runs out the bottom!

Dave, and thanks for the interesting replies.

On Mon, 15 May 2006 11:23:08 GMT, "Dave.C"
wrote:

As far as I can tell with mine as soon as the first bucket tips the second
bucket is immediately in place to catch the rain draining through. it just
rocks backwards and forwards on a pivot

This is certainly the case with recent Davis type tipping gauges - the
two buckets are essentially just one continuous bucket with a thin
vertical partition at the centre point. So if the rain is not dripping
into one bucket then it's dripping into the other. There are therefore
no drops lost while the bucket tips in the sense of drops falling
outside the buckets.

However, what does in theory cause small errors is that extra drops
may continue to fall into the bucket that is already tipping out of
the way and so do not contribute to the new tip.

Although this effect may seem unlikely, the buckets probably tip only
relatively slowly. Just before they tip the buckets will obviously be
close to being in balance and so one more droplet in the heavy bucket
will only just tip the balance. I've seen estimates of 0.5-1 second
for the tip to pass the midway point, though I've never seen any hard
evidence and it wouldn't surprise me if it were typically either
slower or quicker than this.

Anyway, just doing some back of the envelope sums:

Assume heavy rainfall of say 12mm/hr = 0.2mm/min or 1 tip/min for a
0.2mm increment gauge

One 0.2mm tip for a Davis gauge would require - if my maths is correct
- 4000ul (that's microlitres) of accumulated rainfall. So how big is a
single drop? A standard laboratory rule of thumb is 50ul but can vary
considerably depending on eg the size of the nozzle, especially for
small/tiny nozzles. But for the sake of convenience let's call it 40ul
so on this basis maybe 100 droplets could be needed for one bucket tip
- close to 2/second. So it's not inconceivable that somewhere up to
2-3% under-recording could result during heavy rainfall from
over-filling each bucket. And if the droplet size were bigger than my
estimate then it would still be possible for a droplet to land on the
partition wall mid-tip and split between the two buckets.

(Can you tell I had a few idle minutes at lunchtime, though I'm not
guaranteeing my maths?)

John Dann
www.weatherstations.co.uk

I've had two 0.2 mm TBRs logged to my AWS for several years now, and
they typically under-record between 1 and 3% compared to my adjacent
Met O standard 5 inch gauge. This is probably about the best one can
expect as these are professional class instruments (both belong to the
Environment Agency) and are regularly calibrated. The slight error most
probably results from some loss due to differences in wetting of the
funnels, and in evaporation of the part-filled bucket contents.

As a matter of interest, three months back I purchased one of the
wireless 1 mm capacity tipping bucket gauges referenced in the first
post on this thread. I hasten to add I use this only for
distant-reading purposes, not climatology! After careful calibration, I
found the 1 mm tip was surprisingly accurate for an inexpensive
instrument (within 4%, which was as accurate as I could measure the
volume of water poured in while I counted the tips). Daily totals show
good agreement with the checkgauge when more than 2-3 mm falls, but
where it loses out are small falls ( 0.5 mm or so), especially a day
or two apart, where they simply evaporate and are lost to the record.

Monthly totals from both gauges compared with my standard gauge have
been (making slight adjustments to bring them all to 09-09h UTC
terminal hours):

March - checkgauge 50.9 mm (=100%), TBR 1 49.7 mm (97.6%), Conrad gauge
40 mm (79%)
April - checkgauge 34.3 mm (=100%), TBR 1 34.3 mm (100%), Conrad gauge
30 mm (87%)

I would recommend one of these little instruments, even if you have an
AWS. They are great for watching thunderstorm rainfall tick up whilst
enjoying the lightning in the comfort of the conservatory, or for a
quick glance on getting up in the morning how much (if any) rain has
fallen overnight. But _not_ recommended for climatological purposes!

Stephen