On 29/01/2015 14:38, John Hall wrote:
In message ,
Scott W writes
After requesting MetO rainfall cards for several stations for February
1991 I've been surprised how much the precipitation varied over quite
a small area.

There was 23mm - 27mm over a wide area though some stations seem to
have recorded far less. Which brings me to a question. Is it a case of
stations missing out or is it down to the skills of the observer in
recording snow? How meticulous was the Met Office's quality control in
1991? Were volunteers trained?
snip

Just one thought, that when it's drifting snow it must be very
difficult. In fact one wonders if it's possible in extreme conditions to
get a meaningful figure at all using a rainguage.

I think that training was done for volunteers but it relied on them be
open to it. I think it's still done but I'm not sure. As far as
variability is concerned, there are many factors that affect this. Even
over a small area there can be quite large differences. It depends on
convective v dynamic precipitation, the exact exposure of the gauge etc.
I remember observing at Kew Observatory in the late 60s. It had snowed
and the snow was quite deep. However, when I did the normal measurement
from the raingauge, there was barely anything in it. If I remember
correctly I used the (empirical) fact that 10cm of undrifted and new
snow equates to about 10 mm of rain. This was confirmed by some other
experimental gauges we had. I think the snow had had just been blown
over the gauge by the strong winds.