Experiences with using CS616-L Soil Moisture Content Sensors in Sandy/gravel soil

[michael....@gmail.com]

Hello all,

Michael here, I am a graduate student at the Galway Mayo Institute of
Technology in Galway, Ireland. I am researching combined heat and
moisture movement in the Irish soils surrounding underground heat
exchangers for heat pumping. I have installed temperature and moisture
sensors in profiles surrounding heat exchager pipes in the ground, to
establish the temperature and volumetric water content gradients near
the pipes.

I have used Campbell Scientific CS616-L Water Content Reflectometers
but I wired them to National Instruments FP-CTR-502 field point
modules which read voltage or current. This poor choice of field point
module was due to a communication error with National Instruments,
with the help of a circuit board supplied by national instruments it
was possible to get this system working however I am not confident of
the data at all and wish to change the set-up. I am thinking of
purchasing some more CS 616-L sensors with the recommended Campbell
data logger this time.

My questions are:

Has anyone used the complete Campbell System to measure soil moisture
content?
If so;
Are you satisfied with the system?
Are you aware of a better method in existance?
Site specific calibration is important, in brief what method would you
recommend?
What effect does the presence of small stones/ gravel in the vicinity
of the sensor have on the results?

Thanking you all in advance for any suggestions.

Michael Greene,
Department of Mechanical and Industrial Engineering,
Galway-Mayo Institute of Technology,
Dublin Rd.,
Galway,
Ireland.
michael....@gmail.com

[morteza doosti]

Dear Michael

I want to answer your questions to some extent how I can, I wish it will be usefull.

1. I didn't use campbell system but I have some good sugestions:

The time domain reflectometer is a system wich sends a electromagnetic signal to soil, and however if there isn't any iron or other yonic principles (water dielectric=81) water is the most effective on this signal (water dielectric may change by tempreture amoun 2 or 3 points), when signal recive to water molucols the rise time of signal will clopse and view a decrese about some micro seconds, and it is the point you can find water content by this.

If you need other information feel free about call me.

Morteza Doosti

Irrigation Engineering

---

From: "michael....@gmail.com" <michael....@gmail.com>
To: sowacs: soil water moisture content measurement systems and sensors <sow...@googlegroups.com>
Sent: Tuesday, October 21, 2008 2:39:38 PM
Subject: Experiences with using CS616-L Soil Moisture Content Sensors in Sandy/gravel soil

[Chris Rüdiger]

Hi Michael,

I have used both the CS616 and the respective loggers from Campbell (I
think the CR-10 or so, but it's been a while...). They have been
running since late 2002, so in terms of the lifetime of the sensors, I
am not unhappy. Some of the loggers failed, though, and we had to
replace them. But this wasn't much worse than with other systems.
So much for the hardware, the 616s are generally set to give you a
frequency, rather than the voltage. I am not sure how much the voltage
actually changes with changes in the moisture content, or at all for
that matter. What basically happens is that a pulse is sent down the
rods and the return time will change with the amount of water in the
soil, as the higher conductivity of the water attenuates the pulse
frequency. So, I am not surprised that you have issues with your
data.
Campbell provides you with a standard calibration equation, which is
ONLY valid for silty loam. You will have to calibrate the sensors for
each soil type, as they are quite sensitive to changes in the particle
size distribution and also the soil temperature. I have undertaken
some extensive calibration efforts for various soil types here in
Australia (based on the method suggested by Western and Seyfried,
2005, I further developed their approach which was designed for the
CS615), so I can help you with some suggestions, if you need it (just
get into touch with me via email). Just as a recommendation: if you
use them, then install them horizontally, never vertically, there may
be some significant effects, due to the changes in the soil horizon,
and the integration over the 30cm.
As an alternative, you can use Thetaprobes, they are easy to handle,
and give you the voltages, rather than frequencies, so you could keep
your loggers. I know that they are in use by the French Meteorological
Office, and they are quite happy with them, because they are less
sensitive to soil type and temperature. Though they have some issues
when closer to saturation.
The presence of stones and gravel depends and what you want to get out
of the information. The guys at the USDA remove the fraction of rocks
larger than 20mm, I think, and then adjust their measured volumetric
soil moisture to the soil moisture in the soil fraction. Stones can
have an impact, if for example, you have a rock sitting between the
two sensors (and also touching them). But to quantify this effect is
more than just difficult.

Well, I hope that helps and don't hesitate to contact me, if you have
any further questions.

Chris

> michael.h.gre...@gmail.com

[Luca Coppi]

Hi Micheal

I'm using them in a sandy/gravelly soil in southeastern Manitoba. I'm quite satisfied about them, they are producing stable results (no sudden changes/noise), just for one year though. However, as every instrument they have their disadvantages:

- they are fixed in one location: if the location is not representative of the field conditions or there are factors affecting it (i.e. animals burrows), the results may be skewed.

- the installation is important: you can use rod-guides that Campbell sells to create holes and install the instruments there without the need to hammer them in a stony soil and risk to damage them. However this may cause air pockets to be created, thus affecting the readings. There shouldn't be any air between the rod and the soil because the air has a very low dielectric constant (the variable for which the sensors are calibrated for) while the water has a very high value.

- calibration is also required. I'm currently collecting soil samples at certain time intervals and measuring the gravimetric mositure content, to be compared to the sensor readings. However, in a gravelly soil gravimetric moisture is also tricky (the best way is to separate the stones from the wet sample and include them with a subsequent calculation; there are several references in literature). Additionally the way you collect the soil samples is important: if you take loose samples, you may lose the water trapped in macropores which may be the most in a coarse soil. So you may want to take core samples if you can.
You can also do a calibration taking soil from the field and installing the sensors in the lab at various moistures (you can weight the whole container so to have a more precise determination of gravimetric moisture).

Theta probes are also used, or you may also look into neutron probes or capacitance probes.

Luca.

2008/10/24 morteza doosti <morteza...@yahoo.com>

--
Luca Coppi
PhD Candidate
Ellis Building, Department of Soil Science
University of Manitoba, Winnipeg, Manitoba
Canada, R3T 2N2
Phone- Work: +1 204 474 8153
Phone- Home: +1 204 477 5200
Phone- Cell: +1 204 291 0877
e-mail: luca...@gmail.com
          umc...@cc.umanitoba.ca
------------------------------------------------------------------------------------------------
SoilScience-Gradstudents mailing list
SoilScience-...@lists.umanitoba.ca
http://lists.umanitoba.ca/mailman/listinfo/soilscience-gradstudents
-------------------------------------------------------------------------------------------------

[Priyantha B Kulasekera]

Hi Michael
As explained by several experienced scientists earlier, CS 616 is a reliable instrument. I had used it to continuously monitor water content data for the last three years without much problem. However, they need to be field calibrated for better results. One of the easy methods is to calibrate it against the TDR readings, which gives a fair estimation of water content when the equation developed by Topp et al. is used.

Good luck with your research.

Priyantha Kulasekera
Grad Student-University of Guelph.

---

From: Luca Coppi <luca...@gmail.com>
To: sow...@googlegroups.com
Sent: Monday, 27 October, 2008 16:33:41
Subject: Re: Experiences with using CS616-L Soil Moisture Content Sensors in Sandy/gravel soil

[Josh H.]

I work for a company that sells and installs the CS616 on occassion.
I would highly recommend the Campbell Dataloggers. The new systems
such as the CR800 and CR1000 are very reliable and will meet your
requirements. If you want to talk to an expert on the CS616 send an
email to Dave Meek at Campbell Scientific he is one of the soil
support guys. dave...@campbellsci.com

Thanks,
Josh

> michael.h.gre...@gmail.com