Standardising Sensors - Where to start!

[Cliff Hignett]

[Apologies: my comment placed at the beginning of the "Standardising Sensors
made it appear as if I (Bruce) had originated it: it was in fact a post from
tomrinaz@juno.com (Tom A Reynolds) who put forward the idea.]


Cliff Hignett writes

I have just joined sowacs and seem to have missed the conversation which
generated a  reply by 'Bruce'.   But after some 30 years working with the
neutron meter and with authorship of part of a book on the topic I thought I
should comment.

The 'standard' proposed by Bruce - comprising 500litres of saturated fine
sand would be fine as far as it goes, but the NMM is not sensitive to water,
it is sensitive to hydrogen.  As clay content increases so does the amount
of H which does not get released when the soil is dried at 105 degrees C.   

The idea of a 'standard soil environment'  for cross referencing neutron
moisture meters was canvassed in Australia some 25 years ago when NMM's were
much less uniform in performance than they are now.   What was proposed (and
used for a while) was a series of 6x 20 litre (5 gallon) containers filled
with a clay, a loam and a sand, each  at saturation and air dry condition.
Yes, I know 20 litres is not big enough for an NMM calibration point in near
air dry condition  - but we needed a 'standard' in which all machines would
behave the same that did not require a football field to store it.

For other devices, the problem with 'standard' material calibration is that
there is no sensor (to my knowledge) which works in all envirnoments and
that many are highly non linear requiring a minimum of three calibration
points.   For salt sensitive sensors, (eg capacitive types)  this would have
to be multiplied by a range of salinity levels.   For devices like the TDR
which will not register water held in isolated pockets  (eg clay aggregates
in a sand matrix) the problem gets even more complex.  

For anyone who is interested, I have a mixture I use to calibrate sensors
which will remain friable up to 0.5cc/cc water content.This makes it easy
to successively  bury  sensors one at a time in a material with a fixed,
reproduceable density and water content.    As it relies on addition of
small particles of very high water holding capacity material to a matrix of
low WHC material, it DOES NOT work with TDR's and may not work with heat
flow types, but I have not had problems with capacitive, conductivity, or
heat capacity types.

Cliff Hignett

>[Bruce comments: This looks like a worthwhile initiative to follow up on -
>how about using sowacs to electronically pin-up the procedures and standards?]
>
>There seems to be some lack of attention to the notion that sensors must
>be standardized.  By this I mean, a benchmarking methodology which
>............
Cliff Hignett CPSS CPAg
CSIRO Land and Water
PMB 2 Glen Osmond 
South Australia 5064

ph (08)8303 8459
fx (08) 8303 8551
ah(08) 8276 7706