RE: Accuracy of capacitance soil moisture ...

[Dale Hardin]

With regard to the measured radius of capacitance sensors, the EnviroSCAN
sensor must be differentiated from most others. Concerning the EnviroSCAN,
Paltineau and Starr, 1997: Real Time Soil Water Dynamics Using Capacitance
Probes, Soil Sci. Aoc. Am. J., Vol 61, report that "99% of the [EnviroSCAN]
sensor's response was obtained within a radial distance of 10 cm from the
wall of the PVC access pipe.  The remainder of the sensor readings extended
to a soil thickness of 18 cm from the access pipe wall."  This greater
radial distance, and consequently greater measured volume, is one of the
characteristics that differentiate the EnviroSCAN from other capacitance
probes.

Note that, since the EnviroSCAN probe is not reported to have a variation in
radius of influence based upon soil characteristics as is the case with the
neutron probe, there will likely be some differences in results when
directly comparing the EnviroSCAN to the neutron probe.

Additionally, the same authors reported that "these sensors may be placed
just 5 cm into the soil, effectively measuring soil water contents at the 0-
to 10-cm depth interval."  Again, this is an improvement over the neutron
probe in that the volumetric water content at shallow depths may be
accurately measured with minimum soil-air boundary effects.

With regard to the disturbance of the soil during the emplacement of most
sensors, the EnviroSCAN sensors are placed within an access tube that is
inserted into the soil by internal augering and driving.  A rubber plug is
used to seal the bottom of the access tube.  This minimizes the alteration
of the soil properties in the vicinity of the sensors.

I will complete these remarks by stating that I am an EnviroSCAN distributor
in Florida.  I have usually refrained from entering these discussions
because of this affiliation, but have decided to participate in a positive
manner wherever I can.

Regards,

Dale Hardin
Agricultural Information Technologies, Inc.


-----Original Message-----
From: owner-sowacs@aqua.ccwr.ac.za [mailto:owner-sowacs@aqua.ccwr.ac.za]
Sent: Saturday, April 10, 1999 4:28 PM
Subject: Re: Accuracy of capacitance soil moisture ...

You are correct in comments about 'error' in measuring soil water.   The
volume of measurement is indeed a critical factor.

I am not aware of actual measured volumes (it is a non trivial job to
'measure radius of influence')  except for the neutron meter which
measures a radius of at least 200mm in wet clay and up to 800mm in dry
sand.   (depends on H content)     I am not aware of formal research to
measure the measurement volume of other sensors.

I believe the following figures are approximately correct.  Capacitive
sensors measure about 10mm from the active surface (may be a rod or a
flat plate depending on design,  TDR about 10mm radius from the antenna
(by the length of the antenna), heat dissapation sensors depend on the
water content measuring up to 30mm from the heat source in wet soil.

Half the problem is to measure the water content in the sensed volume
correctly.  (this is a calibration problem) the other half of the
problem is to know the relationship between that volume and the
surrounding soil.  This is a combination of the problem of changing the
soil next to the sensor during installation and the degree that this
(sometimes very small) volume, is representative of a field.  The usual
assumption is that the larger the volume, the more representative it is
likely to be.


Cliff Hignett


owner-sowacs@aqua.ccwr.ac.za wrote:

> NOTE:  To get off this list, send email to majordomo@aqua.ccwr.ac.za
> with the body of the message containing the line:
> unsubscribe sowacs
>
> Sorry for this late response.
>
> Moshe wrote:
> We may trust the sensor manufacturers ....
>
> I believe that "field calibration" of soil water sensors depends on:
> (a) what volume of soil is sampled in the calibration   exercise - to
> be
> processed by oven drying.
> (b) how the volume sampled in the calibration relates to
> the            volume of
> soil sensed by the instrument - do they         match in both size and
> location?
> (c) how do both volumes relate to the so-called
> "representative elementary
> volume".
>
> This last concept is important but no-one appears to have put numbers
> on
> it, for soil water measurements. The idea is that small samples do not
>
> contain the full range of variability of the property being measured.
> So
> with increasing sample size, a point is reached where the variability,
>
> found from taking several samples, becomes constant. ie the larger
> sample
> contains no further components of variability, which were not all
> always
> present in each of the very small samples.
>
> Moshe wrote:
>
> Water moves and equilibrates within the soil by potential gradients,
> not by
> water content. .......
>
> I agree except that:
>
> The amount of water held in soil at a particular tension depends on
> the
> recent history of wetting and drying, in technical terms, which
> "scanning
> curve" is being followed in the hysteretic relationship between soil
> water
> and soil water tension. This hysteretic behaviour means that, for
> example,
> at a tension of 5kPa (0.05 bar) the water content of a field fine sand
>
> sample could be anything between 0.08 vwc to 0.25 vwc ( This is a
> worst
> scenario but has been recorded). The typical envelope of water
> contents at
> this tension appears to be closer to 3-4% vwc. This band of water
> content,
> corresponding to a given tension, is subject to both internal drainage
> and
> potential crop water uptake.
>
> Hysteresis arises from differences in the wetting behaviour causing
> dry
> soil to re-wet. Under normal conditions, this depends on the intensity
> of
> rainfall/irrigation at the soil surface, or flux at the emitter. The
> higher
> the flux the greater the likelihood that air will be entrapped and
> full
> rewetting of the pore space is prevented.
>
> The consequence for supplemental irrigation is that the amount of
> water
> required to bring the soil back to a "full" field capacity condition (
> no
> entrapped air) will depend on the degree of air entrapment created,
> during
> rainfall re-wetting events. I believe that a tensiometer cannot
> identify
> this, while the soil water sensor can suggest what change of water
> content
> has taken place - relative to a "full" field capacity condition. Of
> course
> this has to have been identified earlier. A colleague in Botswana
> reported
> difficulty in developing a soil water characteristic to interpret
> tensiometer readings. This is the only practical reference to
> difficulties
> arising from hysteresis that I have heard, in the context of
> scheduling.
>
> In arid conditions, and with drip irrigation promoting a complete
> recharging of the pore space during irrigation, then hysteresis is
> probably
> irrelevant.
>
> Can anyone help get the IAEA report posted on our discussion network
> along
> with Martin Shmitz's results?
>
> regards
>
> Martin Parkes
> WasteTrim at Technology Transfer Centre, Alrick Building,
> Mayfield Road, Edinburgh EH9 3JL, SCOTLAND
> tel (44) 131 472 4708
> fax (44) 131 662 4678


--
Cliff Hignett
Soil Water Solutions
45a Ormond Ave
Daw Park
South Australia 5041
pH 61 (08) 8276 7706
WWW.SOILWATER.COM.AU