In response to Cliff's question on neutron meters

[Rick Allen <AllenRic@cc.usu.edu>]

In response to Cliff's question on neutron meters:
>
>Cliff Hignett writes
>
>Question for Rick Allen
>
>Our experience with a wide range of neutron moisture meters was that most of
>the machine variation and moderate variation in access tube geometry could
>be accounted for by expressing the counts as a fraction of the counts in a
>container of water with the same access tube size and type.  (with some non
>linearity between metal and plastic tubes) .  Did the 'shoot out' canvass
>this possibility?
>

We did not attempt to "cross reference" difference meters by using the water
standard.  I am sure that this would remove the vast majority of differences
between probes.  We were merely noting that even when expressed as a count
ratio  (counts in the ground divided by counts in the shield suspended above
the ground) that different meters respond differently even when from the
same manufacturer.

One can probably transfer meter calibrations for one specific type or size
of tube to another type or size of tube by exchanging tubes in your water
standard and comparing ratios.  However, one of the signficant factors
impacting the calibration curve is the size of the tube relative to the
geometric distance of the median location of neutron thermalization from the
detector.  An unsaturated soil will have a larger distance as compared to
pure water.  Therefore, I would anticipate finding that a tube in water
would show more sensitivity to tube diameter than would a tube in the field.  

The other parameter that is of concern with large access tubes is that the
source/detector does not stay in the center of the tube, but slides along
the wall of the tube, according to the plumb of the tube and according to
"curls" in the cable.  This effect is largely nonreproduceable from tube to
tube due to randomness.

By the way, Cliff, I appreciate very much what you had to say concerning
error reduction when using neutron data to predict ET.  We all need to be
better statisticians when reducing and applying our data.

Also, I appreciate the encouragement from Pierre Moutonnet on the utility of
an applications manual for the neutron meter.

Rick Allen
Utah State Univ.