I'm a graduate student at University of Manitoba and I'm trying to
define the water properties of a sandy- very stony soil (stones up to
> 80% weight in some layers). I have Campbell CS616-L probes installed
in the soil profile at 4 depths and I would like to couple that with a
measurement of soil water potential. Being the soil very coarse, I
expect the VMC to be almost constant at water potentials <-100/-200
kPa. I was thinking to use Watermark sensors but was advised not to do
so by Campbell people (they say it's not accurate). However the paper
from Thompson et al. (2006) shows that they can be acceptably accurate
even with literature calibration equations (but especially with re-
parameterization from tensiometer data) for the -10 / -150 kPa range.
Another option would be to use Irrometer tensiometers (there are no
many funds to buy anything more expensive), but I don't know how well
they would work in a stony layer.
Finally another option would be to determine the retention curve in
the lab with Pressure plates, but this would mean removing most of the
stone component which would alter the soil field conditions.
I am PhD student in Australia working on similar material on rehabilitated mined land. After considering numerous options, I opted for CS229 matric sensors from Campbell Scientific. They have been used successful on such materials here. The size of the sensors allows you to install them in the fine material between the rock particles/fragments. In my case I have a pair of CS229 and CS616 at 15 and 140 cm depths. CS229 costs about US$200 each, and hopefully it fits your budget.
There are numerous water potential sensors out there. The best way to learn how they perform is to share experiences with those who have used them on similar material and climatic conditions. Some of the sensors are very sensitive to temperature and mineralogy (e.g iron oxides). Some sensors have been reported to give erroneous results or cease logging under temperature extremes.
> I'm a graduate student at University of Manitoba and I'm trying to > define the water properties of a sandy- very stony soil (stones up to >> 80% weight in some layers). I have Campbell CS616-L probes installed > in the soil profile at 4 depths and I would like to couple that with a > measurement of soil water potential. Being the soil very coarse, I > expect the VMC to be almost constant at water potentials <-100/-200 > kPa. I was thinking to use Watermark sensors but was advised not to do > so by Campbell people (they say it's not accurate). However the paper > from Thompson et al. (2006) shows that they can be acceptably accurate > even with literature calibration equations (but especially with re- > parameterization from tensiometer data) for the -10 / -150 kPa range. > Another option would be to use Irrometer tensiometers (there are no > many funds to buy anything more expensive), but I don't know how well > they would work in a stony layer.
> Finally another option would be to determine the retention curve in > the lab with Pressure plates, but this would mean removing most of the > stone component which would alter the soil field conditions.
If you are limited by budget to using an Irrometer I suggest you could install it in a wider augured hole and pack it in sand slightly finer than the surrounding soil.
Provided rain is kept from infiltrating down the hole, ie there is only lateral infiltration, the moisture tension reading should equilibrate with the surrounding soil.
We make robust sensors based on the thermal dissipation principle and we pre-calibrate them for moisture tension in the range 0-60kPa (www.mcburneyscientific.com) . If required we can supply a version with sensitivity optimised for the range 0-20kPa, which might suit your situation. However it comes with built-in logger, solar charger and choice of bluetooth or GPRS, so it won't be cheaper than the Irrometer.
I'd be interested to hear how you get on and what other suggestions you receive.
----- Original Message ----- From: <lucaco...@gmail.com> To: "sowacs: soil water moisture content measurement systems and sensors"
<sowacs@googlegroups.com> Sent: Thursday, March 20, 2008 7:45 PM Subject: Soil water potential
> Hello all
> I'm a graduate student at University of Manitoba and I'm trying to > define the water properties of a sandy- very stony soil (stones up to >> 80% weight in some layers). I have Campbell CS616-L probes installed > in the soil profile at 4 depths and I would like to couple that with a > measurement of soil water potential. Being the soil very coarse, I > expect the VMC to be almost constant at water potentials <-100/-200 > kPa. I was thinking to use Watermark sensors but was advised not to do > so by Campbell people (they say it's not accurate). However the paper > from Thompson et al. (2006) shows that they can be acceptably accurate > even with literature calibration equations (but especially with re- > parameterization from tensiometer data) for the -10 / -150 kPa range. > Another option would be to use Irrometer tensiometers (there are no > many funds to buy anything more expensive), but I don't know how well > they would work in a stony layer.
> Finally another option would be to determine the retention curve in > the lab with Pressure plates, but this would mean removing most of the > stone component which would alter the soil field conditions.
Luca, you mentioned a paper reporting work of our group with the Watermark sensor. We observed that for a given soil that did not dry rapidly, that an in-situ calibration or published equation (Thomson and Armstrong, Shock) that was re-parameterised for those conditions, gave good results for that soil and those conditions. Two provisos are (i) the above a certain maximum dry value of b/w approx. -4 and -8 kPa the Watermark does not respond, and (ii) what may be particularly important in your stony soil that in rapidly drying soil that the Watermark responds more slowly than a tensiometer. We never tested the hypothesis that specific calibration for rapidly drying conditions would permit reasonably accurate measurement with the Watermark sensor. As referred to by others, ensuring good contact between the sensor and the soil matrix is essential.
Decagon have recently released a new soil matric potential sensor; as I understand it is something of a hybrid between a Decagon Echo sensor and a Watermark. That could be another option; as yet there has been little testing, apart from that of the manufacturer.
Good Luck Rod Thompson University of Almeria, Almeria, Spain
>I'm a graduate student at University of Manitoba and I'm trying to >define the water properties of a sandy- very stony soil (stones up to > > 80% weight in some layers). I have Campbell CS616-L probes installed >in the soil profile at 4 depths and I would like to couple that with a >measurement of soil water potential. Being the soil very coarse, I >expect the VMC to be almost constant at water potentials <-100/-200 >kPa. I was thinking to use Watermark sensors but was advised not to do >so by Campbell people (they say it's not accurate). However the paper >from Thompson et al. (2006) shows that they can be acceptably accurate >even with literature calibration equations (but especially with re- >parameterization from tensiometer data) for the -10 / -150 kPa range. >Another option would be to use Irrometer tensiometers (there are no >many funds to buy anything more expensive), but I don't know how well >they would work in a stony layer.
>Finally another option would be to determine the retention curve in >the lab with Pressure plates, but this would mean removing most of the >stone component which would alter the soil field conditions.
>Any suggestions about that?
>Thank you, >Luca.
ROD THOMPSON Universidad de Almería, Dpto. Producción Vegetal, Edificio CITE-2B, La Cañada 04120 ALMERÍA SPAIN
If you are trying to define water properties for stony soils you should be aware that the stones highly modify the water storage and transmission characteristics of the soil. This is something that is not normally taken into account but the effect could change completely the soil water response. Firstly the volume of soil occupied by stones does not storage water and therefore the volumetric water content is reduced by the volumetric stone content and secondly the hydraulic conductivity is also affected through the tortuosity factor (there are other issues but not going into them now). From our experience a stony sandy soil could behave as a clay soil for higher water tensions and at the same time when water tension is lower it can have a very high value for Ksat.
Note that calibration equations for soil dielectric sensors assume no stone content and the dielectric response is different if you have stones around the sphere of influence of the sensor.
Summing up, soil water dynamics for the fine earth fraction only will not tell you much about the soil water response for a stony soil, especially if the stone content is high. Try also to get a good estimate of the stone content and try to get a higher resolution in depth on your measurements with whatever kind of sensor you use.
Best regards,
Jesus
--
Jesus Fernandez Galvez, PhD Estacion Experimental del Zaidin National Council for Scientific Research, CSIC C/ Profesor Albareda 1, 18008 Granada, Spain Tel: +34 958 181600 Fax: +34 958 129600 Currently, visiting scientist at University of Reading, UK
> I'm a graduate student at University of Manitoba and I'm trying to > define the water properties of a sandy- very stony soil (stones up to >> 80% weight in some layers). I have Campbell CS616-L probes installed > in the soil profile at 4 depths and I would like to couple that with a > measurement of soil water potential. Being the soil very coarse, I > expect the VMC to be almost constant at water potentials <-100/-200 > kPa. I was thinking to use Watermark sensors but was advised not to do > so by Campbell people (they say it's not accurate). However the paper > from Thompson et al. (2006) shows that they can be acceptably accurate > even with literature calibration equations (but especially with re- > parameterization from tensiometer data) for the -10 / -150 kPa range. > Another option would be to use Irrometer tensiometers (there are no > many funds to buy anything more expensive), but I don't know how well > they would work in a stony layer.
> Finally another option would be to determine the retention curve in > the lab with Pressure plates, but this would mean removing most of the > stone component which would alter the soil field conditions.
Just to specify a bit more on our experiment, we have 32 CS616-L
probes installed in 8 locations at 4 depths, but we can only afford
4/8 water potential sensors (depending on the costs). The soil is very
coarse and stony so it dries quickly, but a perched groundwater tables
is present at spring rising up to the surface. So for a short time
very wet conditions are expected (it's variable year after year
though).
The soil (under grassland) is very much layered, with a rather thick
sandy A horizon underlaid by a coarse sand/stony thick layers and
usually thin fine/very fine sand layers.
Under these conditions I believe that the water potential range where
most of the water movement occurs is the gravitational range (0/-10
kPa).
The site is under freezing conditions for 4-5 months per year.
Considering the replies and the literature on the topic, it seems to
me that:
- the watermark sensors are reasonably accurate in the range of
-10/-200 kPa for soils not drying too quickly (i.e. not too coarse
soils), if a local calibration/re-parameterization is done
- they can withstand freezing/thawing so they can be left during
winter
- low tension tensiometers (i.e. Irrometer LT) should monitor with
accuracy the - 0-40 kPa range
- they should be removed during winter, thus not allowing a monitoring
of the water potential dynamics under freezing.
- the water column may break up often if the soil dries a lot, so they
may require quite a lot of maintenance.
still I believe the LT tensiometers may be the best choice, provided a
good contact with the soil is made. But again I would like to hear
more advices if you have them.
Thank you for helping out a graduate student :-),
Luca.
On Mar 24, 5:29 am, Rodney Thompson <rod...@ual.es> wrote:
> Luca, you mentioned a paper reporting work of our
> group with the Watermark sensor. We observed
> that for a given soil that did not dry rapidly,
> that an in-situ calibration or published equation
> (Thomson and Armstrong, Shock) that was
> re-parameterised for those conditions, gave good
> results for that soil and those conditions. Two
> provisos are (i) the above a certain maximum dry
> value of b/w approx. -4 and -8 kPa the Watermark
> does not respond, and (ii) what may be
> particularly important in your stony soil that in
> rapidly drying soil that the Watermark responds
> more slowly than a tensiometer. We never tested
> the hypothesis that specific calibration for
> rapidly drying conditions would permit reasonably
> accurate measurement with the Watermark
> sensor. As referred to by others, ensuring good
> contact between the sensor and the soil matrix is essential.
> Decagon have recently released a new soil matric
> potential sensor; as I understand it is something
> of a hybrid between a Decagon Echo sensor and a
> Watermark. That could be another option; as yet
> there has been little testing, apart from that of the manufacturer.
> Good Luck
> Rod Thompson
> University of Almeria,
> Almeria, Spain
> At 12:45 20/03/2008 -0700, you wrote:
> >Hello all
> >I'm a graduate student at University of Manitoba and I'm trying to
> >define the water properties of a sandy- very stony soil (stones up to
> > > 80% weight in some layers). I have Campbell CS616-L probes installed
> >in the soil profile at 4 depths and I would like to couple that with a
> >measurement of soil water potential. Being the soil very coarse, I
> >expect the VMC to be almost constant at water potentials <-100/-200
> >kPa. I was thinking to use Watermark sensors but was advised not to do
> >so by Campbell people (they say it's not accurate). However the paper
> >from Thompson et al. (2006) shows that they can be acceptably accurate
> >even with literature calibration equations (but especially with re-
> >parameterization from tensiometer data) for the -10 / -150 kPa range.
> >Another option would be to use Irrometer tensiometers (there are no
> >many funds to buy anything more expensive), but I don't know how well
> >they would work in a stony layer.
> >Finally another option would be to determine the retention curve in
> >the lab with Pressure plates, but this would mean removing most of the
> >stone component which would alter the soil field conditions.
> >Any suggestions about that?
> >Thank you,
> >Luca.
> ROD THOMPSON
> Universidad de Almería,
> Dpto. Producción Vegetal,
> Edificio CITE-2B,
> La Cañada
> 04120 ALMERÍA
> SPAIN
Hello everybody, This is Ajay from South Australian Research and Development Institute at Adelaide. We are planning to setup a weighing Lysimeter facility with 4-5 year old citrus for root zone solute studies. To monitor the plant water use, sap flow systems are being proposed for use. There are numerous technologies available, could somebody who have compared various sap flow systems suggest which one could be more suitable for this study. Ajay
If you are trying to define water properties for stony soils you should be aware that the stones highly modify the water storage and transmission characteristics of the soil. This is something that is not normally taken into account but the effect could change completely the soil water response. Firstly the volume of soil occupied by stones does not storage water and therefore the volumetric water content is reduced by the volumetric stone content and secondly the hydraulic conductivity is also affected through the tortuosity factor (there are other issues but not going into them now). From our experience a stony sandy soil could behave as a clay soil for higher water tensions and at the same time when water tension is lower it can have a very high value for Ksat. Note that calibration equations for soil dielectric sensors assume no stone content and the dielectric response is different if you have stones around the sphere of influence of the sensor. Summing up, soil water dynamics for the fine earth fraction only will not tell you much about the soil water response for a stony soil, especially if the stone content is high. Try also to get a good estimate of the stone content and try to get a higher resolution in depth on your measurements with whatever kind of sensor you use. Best regards, Jesus -- Jesus Fernandez Galvez, PhD Estacion Experimental del Zaidin National Council for Scientific Research, CSIC C/ Profesor Albareda 1, 18008 Granada, Spain Tel: +34 958 181600 Fax: +34 958 129600 Currently, visiting scientist at University of Reading, UK Quoting lucaco...@gmail.com:
> I'm a graduate student at University of Manitoba and I'm trying to > define the water properties of a sandy- very stony soil (stones up to >> 80% weight in some layers). I have Campbell CS616-L probes installed > in the soil profile at 4 depths and I would like to couple that with a > measurement of soil water potential. Being the soil very coarse, I > expect the VMC to be almost constant at water potentials <-100/-200 > kPa. I was thinking to use Watermark sensors but was advised not to do > so by Campbell people (they say it's not accurate). However the paper > from Thompson et al. (2006) shows that they can be acceptably accurate > even with literature calibration equations (but especially with re- > parameterization from tensiometer data) for the -10 / -150 kPa range. > Another option would be to use Irrometer tensiometers (there are no > many funds to buy anything more expensive), but I don't know how well > they would work in a stony layer.
> Finally another option would be to determine the retention curve in > the lab with Pressure plates, but this would mean removing most of the > stone component which would alter the soil field conditions.
> Any suggestions about that?
> Thank you, > Luca.
> />
Dr. Ajay Kumar Upadhyay Senior Scientist (Soil Science) National Research Centre For Grapes, PB No. 3, Manjari Post, Solapur Road, Pune - 412 307 Maharashtra, India
Phone: 91-020-26914245/ 26915573 Fax: 91-020- 26914246 E mail: ajayk_upadh...@yahoo.com
--------------------------------- Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now.
I would like to suggest some approach to monitoring soil-moisture in you "gravely" soil, that may not be very appealing. However, using any device with a porous sensing elements, in my opinion, may not give you realistic results. This is because for sensing elements to be "equilibrated" with the soil-moisture, they must be in hand-and-glove tight contact with the soil. Gravel, will not make such contact possible.
My suggestion, with its limitations, is to take undisturbed soil cores and develop a moisture-tension curve for your soil using a tension table (very inexpensive device that you can build your self). This could be used as a calibration curve for your soil, and what you need to do, thereafter, is to take undisturbed cores and determine the moisture content of these cores. Using the developed moisture-tension curve, you can relate the moisture status of the soil to its moisture content. I realize the limitations of this approach, but under the circumstances described above and those listed in your early e-mail, it could give you some useful information.
Besides the sap flow you might consider measuring stem/trunk diameter change.
Joe Henggeler
University of Missouri
Portageville, MO
________________________________
From: sowacs@googlegroups.com [mailto:sowacs@googlegroups.com] On Behalf Of Ajaykumar Upadhyay Sent: Monday, March 24, 2008 6:06 PM To: sowacs@googlegroups.com Subject: Re: Sap flow unit
Hello everybody, This is Ajay from South Australian Research and Development Institute at Adelaide. We are planning to setup a weighing Lysimeter facility with 4-5 year old citrus for root zone solute studies. To monitor the plant water use, sap flow systems are being proposed for use. There are numerous technologies available, could somebody who have compared various sap flow systems suggest which one could be more suitable for this study.
If you are trying to define water properties for stony soils you should be aware that the stones highly modify the water storage and transmission characteristics of the soil. This is something that is not normally taken into account but the effect could change completely the soil water response. Firstly the volume of soil occupied by stones does not storage water and therefore the volumetric water content is reduced by the volumetric stone content and secondly the hydraulic conductivity is also affected through the tortuosity factor (there are other issues but not going into them now). From our experience a stony sandy soil could behave as a clay soil for higher water tensions and at the same time when water tension is lower it can have a very high value for Ksat.
Note that calibration equations for soil dielectric sensors assume no stone content and the dielectric response is different if you have stones around the sphere of influence of the sensor.
Summing up, soil water dynamics for the fine earth fraction only will not tell you much about the soil water response for a stony soil, especially if the stone content is high. Try also to get a good estimate of the stone content and try to get a higher resolution in depth on your measurements with whatever kind of sensor you use.
Best regards,
Jesus
--
Jesus Fernandez Galvez, PhD Estacion Experimental del Zaidin National Council for Scientific Research, CSIC C/ Profesor Albareda 1, 18008 Granada, Spain Tel: +34 958 181600 Fax: +34 958 129600 Currently, visiting scientist at University of Reading, UK
> I'm a graduate student at University of Manitoba and I'm trying to > define the water properties of a sandy- very stony soil (stones up to >> 80% weight in some layers). I have Campbell CS616-L probes installed > in the soil profile at 4 depths and I would like to couple that with a > measurement of soil water potential. Being the soil very coarse, I > expect the VMC to be almost constant at water potentials <-100/-200 > kPa. I was thinking to use Watermark sensors but was advised not to do > so by Campbell people (they say it's not accurate). However the paper > from Thompson et al. (2006) shows that they can be acceptably accurate > even with literature calibration equations (but especially with re- > parameterization from tensiometer data) for the -10 / -150 kPa range. > Another option would be to use Irrometer tensiometers (there are no > many funds to buy anything more expensive), but I don't know how well > they would work in a stony layer.
> Finally another option would be to determine the retention curve in > the lab with Pressure plates, but this would mean removing most of the > stone component which would alter the soil field conditions.
> Any suggestions about that?
> Thank you, > Luca.
> />
________________________________
Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now.</a
That is an interesting discussion about water in stony soils. In spite of that Im not measuring soil water potential in stony soils. We evaluated infiltration and saturated hydraulic conductivity (with single ring and Guelph Permeater) in stony horizons of petroplinthite Plinthosols and Plinthosols only with soft plintite. We are also evaluating soil water retention curves from those soils but as in the pF 4.2 we analyzed the water hold using sieved soil and other points of low tension (pF 1, 1.8, 2, etc) we did directly from "undisturbed" samples in stell cylinders (some with petroplinthite inside), I ask if someone has experience in interpreting those curves with bias caused by presence of stony in soil water as disucssed by Jesus. I thank in advance any help
Wenceslau Teixeira Embrapa Amazônia Ocidental
> Hello Luca
> If you are trying to define water properties for stony soils you > should be aware that the stones highly modify the water storage and
transmission characteristics of the soil. This is something that is not normally taken into account but the effect could change
> completely the soil water response. Firstly the volume of soil > occupied by stones does not storage water and therefore the > volumetric water content is reduced by the volumetric stone content and
secondly the hydraulic conductivity is also affected through the tortuosity factor (there are other issues but not going into them now). From our experience a stony sandy soil could behave as a clay soil for higher water tensions and at the same time when water
> tension is lower it can have a very high value for Ksat.
> Note that calibration equations for soil dielectric sensors assume > no stone content and the dielectric response is different if you have
stones around the sphere of influence of the sensor.
> Summing up, soil water dynamics for the fine earth fraction only > will not tell you much about the soil water response for a stony soil,
especially if the stone content is high. Try also to get a good estimate of the stone content and try to get a higher resolution in depth on your measurements with whatever kind of sensor you use.
> Jesus Fernandez Galvez, PhD > Estacion Experimental del Zaidin > National Council for Scientific Research, CSIC > C/ Profesor Albareda 1, 18008 Granada, Spain > Tel: +34 958 181600 > Fax: +34 958 129600 > Currently, visiting scientist at University of Reading, UK
>> I'm a graduate student at University of Manitoba and I'm trying to
define the water properties of a sandy- very stony soil (stones up
> to >>> 80% weight in some layers). I have Campbell CS616-L probes > installed >> in the soil profile at 4 depths and I would like to couple that > with a >> measurement of soil water potential. Being the soil very coarse, I
expect the VMC to be almost constant at water potentials <-100/-200 kPa. I was thinking to use Watermark sensors but was advised not to
> do >> so by Campbell people (they say it's not accurate). However the > paper >> from Thompson et al. (2006) shows that they can be acceptably > accurate >> even with literature calibration equations (but especially with re-
parameterization from tensiometer data) for the -10 / -150 kPa
> range. >> Another option would be to use Irrometer tensiometers (there are no
many funds to buy anything more expensive), but I don't know how
> well >> they would work in a stony layer.
>> Finally another option would be to determine the retention curve in the
lab with Pressure plates, but this would mean removing most of
I am a PhD student at the University of western australia, investigating the ecohydrology of rahabilitated mine sites. One of my objectives is to determine the spatial variability of saturated hydraulic conductivity of the sites. To do that I need a method that will aloow me to make a lot of point measurements is the shortest time possible. To this effect I am using the well permeameter method. However, preliminary results show that the hydraulic conductivity of the material is very high such that a a column/resevoir (1.5m long) filled with water will only allow me to take a total of 10 readings at an interval of 10 secs, meaning that the column/reservoir empties within 2 minutes.
Does any member of the group have a suggestion on how I can obtain reliable results from this method?
> If you are limited by budget to using an Irrometer I suggest you could > install it in a wider augured hole and pack it in sand slightly finer than > the surrounding soil.
> Provided rain is kept from infiltrating down the hole, ie there is only > lateral infiltration, the moisture tension reading should equilibrate with > the surrounding soil.
> We make robust sensors based on the thermal dissipation principle and we > pre-calibrate them for moisture tension in the range 0-60kPa > (www.mcburneyscientific.com) . If required we can supply a version with > sensitivity optimised for the range 0-20kPa, which might suit your > situation. However it comes with built-in logger, solar charger and choice > of bluetooth or GPRS, so it won't be cheaper than the Irrometer.
> I'd be interested to hear how you get on and what other suggestions you > receive.
> best wishes
> Terry McBurney
> ----- Original Message ----- > From: <lucaco...@gmail.com> > To: "sowacs: soil water moisture content measurement systems and sensors" > <sowacs@googlegroups.com> > Sent: Thursday, March 20, 2008 7:45 PM > Subject: Soil water potential
>> Hello all
>> I'm a graduate student at University of Manitoba and I'm trying to >> define the water properties of a sandy- very stony soil (stones up to >>> 80% weight in some layers). I have Campbell CS616-L probes installed >> in the soil profile at 4 depths and I would like to couple that with a >> measurement of soil water potential. Being the soil very coarse, I >> expect the VMC to be almost constant at water potentials <-100/-200 >> kPa. I was thinking to use Watermark sensors but was advised not to do >> so by Campbell people (they say it's not accurate). However the paper >> from Thompson et al. (2006) shows that they can be acceptably accurate >> even with literature calibration equations (but especially with re- >> parameterization from tensiometer data) for the -10 / -150 kPa range. >> Another option would be to use Irrometer tensiometers (there are no >> many funds to buy anything more expensive), but I don't know how well >> they would work in a stony layer.
>> Finally another option would be to determine the retention curve in >> the lab with Pressure plates, but this would mean removing most of the >> stone component which would alter the soil field conditions.
As an electronics technician in the UMass Amherst Civil and Environmental Engineering Department, I worked with many faculty and students and various methods of determining soil conductivity.
Shawn Kelley's "Permanent Conductivity Points" (PCP) approach to determining the electroconductivity of saturated soils in monitoring wells might be applicable to your research. Shawn's PCP design consisted of pairs of brass screws mounted an inch apart every two feet along PVC pipes within the monitoring wells. The number of conductivity points depended on the depth of the given monitoring well. Pairs of wires attached to the brass screws permitted monitoring at terminal blocks using a garden variety conductivity meter.
Campbell Scientific also offers a Time Domain Reflectometry (TDR) module and various probes, but they are only 30 cm long.
4$2 2EFLECTOMETER The TDR100 Time-Domain Reflectometer is the core of the Campbell Scientific Time Domain Reflectometry system. The TDR100 (1) generates a very short rise time electromagnetic pulse that is applied to a coaxial system which includes a TDR probe for soil water measurements and (2) samples and digitizes the resulting reflection waveform for analysis or storage. The elapsed travel time and pulse reflection amplitude contain information used by the onboard processor to quickly and accurately determine soil volumetric water content, soil bulk electrical conductivity, rock mass deformation or user-specific, time-domain measurement. Up to 16 TDR100s can be controlled using a single Campbell Scientific datalogger. A 250-point waveform is collected and analyzed in approximately two seconds. Each waveform can have up to 2,048 data points for monitoring long cable lengths used in rock mass deformation or slope stability.
For years, I suggested the Tektronix 1502 series of TDR equipment, a balun and some 300 Ohm twinlead for continuous conductivity monitoring along the length of the cable. Looks like someone else finally had the same idea! Good luck with your efforts!
*** Use of instrument and interpretation of waveforms requires training ***
1) Description:
Time-Domain Reflectometry (TDR) is a technology that was initially developed to allow service personnel to locate damage in buried communication cables. A microwave signal is applied to the coaxial cable and reflected by discontinuities (breaks or shorts) in the cable back to the source, where a video display converts the time delay to distance and graphs a profile of the cable. The apparent distance to the point of damage can be measured and service personnel instructed where to dig. It was found that the apparent distance to the cable break varied from the actual distance depending on the dielectric constant of the surrounding soil, which is directly proportional to soil moisture content.
Soil scientists began using TDR technology in reverse, employing wave guides (broken cables) of known length to deduce soil moisture via several polynomial equations. These wave guides are generally constructed from stainless steel welding rods, which once installed, can be left in place indefinitely. This allows for precise, repetitive, safe, and relatively non-destructive soil moisture measurement at almost any sampling frequency. There are now instruments on the market designed specifically for soil moisture measurement, e.g. Soilmoisture Equipment Corporation (SEC) Trase instrument, which employ automated waveform interpretation, etc. We generally prefer to use the Tektronix Corporation 1502 series instruments, which were designed for cable testing, due to their record of field durability and the amenability for operator interpretation. This interpretation, however, requires a certain amount of operator training to minimize subjectivity and maximize comparability of measurements taken by different operators.
*** Be certain that you are adept at the TDR procedure before taking measurements ***
2) Project Specific Notes:
There are several methods for the construction and installation of wave guides. Some studies employ vertically oriented rod pairs, 5 cm apart, to which test leads are attached directly. This method integrates soil moisture content from soil surface to the lower depth of the rods. On some studies, currently the gradient plots, the TDR rods are oriented parallel to soil horizon and connected to coaxial cable extending above ground. This method, which allows measurement of discrete soil horizons, requires more intensive installation procedures and detailed waveform interpretation. Currently, soil moisture measurements utilizing the TDR technique are made bi-weekly on the Gradient and Riparian projects.
3) Sampling Equipment:
a) Tektronix 1502 Cable Tester with charged battery pack b) spare battery pack c) coax to twin-lead test cable with balun Test cable construction from Tektronic to TDR rod or buried cable: (1) in-line surge surpressor (2) RF cable assembly 6 ft., RG-59 BNC, 75 ohm coaxial cable with BNC connectors for general-purpose communications and test instrument applications (3) TV/RF adapter Adapts male BNC to fit female F-type jack (4) Indoor/Outdoor matching transformer 75 ohm coax/300 ohm twin lead for connecting 75 ohm downlead to 300 ohm antenna (5) two alligator clips d) pack with spare test cable and parts e) field notebook and pencil
4) Sampling Procedure:
a) pull on Power button on lower right hand side of instrument panel b) be certain Vp is set at .99c (2 knobs to left of power button full clockwise) c) set distance to meters in setup menu d) Noise Filter should be one setting clockwise from "HORIZ" e) Dist/Div setting should be ~0.5 m f) connect test cable to instrument panel g) move cursor ("<> position" knob) to the right until end-of-cable inflection is on left side of screen h) set zero at end of cable(s); determine end of test cable by shorting lead by touching alligator clips together; when measuring horizontally installed waveguides with buried cable, attach one clip to inner lead of coaxial connector and the other to outer (shielding) lead, determine end of buried cable by change in waveform. Place cursor at end of cable(s) point; turn Noise filter knob one notch clockwise; distance display should now read "0.00m". i) attach clips to TDR rods for vertically installed rods j) move cursor to endpoint inflection determined by observing change in waveforms k) read and record apparent distance l) continue with measurements; machine must be reset if turned off m) turn instrument off when driving between sites n) plug in for recharge when finished o) for the riparian study, the tdr values need to be entered in the "riptdr.xcl" file.
Data from the TDR measurements are converted to dielectric constant from the square of apparent length divided by actual length of waveguide, then to percent soil moisture via a polynomial equation:
% H20 = [-5.3 x 10e-2] + [(2.92 x 10e-2) (D)] - [(5.5 x 10e-4)(De2)] + [(4.3 x 10e-6)(De3)]
where: D = dielectric constant e = exponent
This material is based upon work supported by the National Science Foundation under Cooperative Agreements DEB-9632854[3] (Text Version)[4] & DEB-0218001[5] (Text Version)[6].
Any opinions, findings, conclusions, or recommendations expressed in the material are those of the authors and do not necessarily reflect the views of the National Science Foundation.
Shawn P. Kelley PO Box 3373 Amherst, MA 01004 Email: skel...@ecs.umass.edu Telephone: 413-695-1816 (H), 413-577-1242 (O) ___________________________________________________________________________ _____________________________ EDUCATION & University of Massachusetts, Amherst, MA HONORS Ph.D. in Geotechnical Engineering Civil and Environmental Engineering Department, September 2003 University of Massachusetts, Amherst, MA M.S.C.E. in Environmental Geotechnology Civil and Environmental Engineering Department, May 1997 University of Massachusetts, Amherst, MA B.S.C.E., Civil and Environmental Engineering Department, May 1994 GPA: 3.14 Dean's list: Fall 1993, Spring 1994 CERTIFICATIONS: E.I.T. Certified (Massachusetts No. 15724), March 1995 O.S.H.A. 40 hr. Hazardous Waste Training (Certification No. 9020) Nuclear Soil Density Testing Certified AWARDS: William M. Boyer Civil Engineering Award ? May 1994 Civil and Environmental Engineering Department, University of Massachusetts Amherst Student Service Award ? May 1994 College of Engineering, University of Massachusetts Amherst ADSC-IAFD Academic Scholarship ? September 2000 ADSC: The International Association of Foundation Drilling, Industry Advancement Fund Service Award ? December 2000 Geotechnical Engineering Group, University of Massachusetts Geotechnical Engineering Fellowship ? December 2002 Geotechnical Engineering Group, University of Massachusetts HONOR SOCIETY: Chi Epsilon, National Civil Engineering Honor Society Member PROFESSIONAL Geotechnical Engineer, GeoDesign, Inc., Middlebury, CT and GeoDesign, Inc., Windsor, VT EXPERIENCE July 2001? January 2003 ? Performed numerous geotechnical engineering site inspections on projects involving shallow foundation subgrade preparation, soil fill placement and compaction, test pit excavating, and pile driving. ? Performed soil-drilling inspection on numerous projects, particularly the State of Connecticut I- 95 New Haven Harbor Crossing Improvements Project.
...
I am a PhD studet investigating ecohydrology of rehabilitated mined land in Western Australia. I am using Theta probes (Delta T) hooked to a Campbell Scientifc datalogger (CR1000). When I did a test run of the system with the theta probes in their casings, I got quite variable voltage readings ranging from -24 to about -68 mV (see attachment). These readings appear unstable until about the 3rd and 4th readings. What is more worrying is that when I placed two of the theta probes in water (probes # 5 and 9) in the 2nd and 4th reading, the readings didn't show any response. An earlier calibration of the probes using a series of soils with different moisture contents showed an excellent relationship between voltage and moisture content. The probes gave a voltage reading of 1100 mV in water. However, apart from the theta probes, the other sensors (air temp, rel. hum, matric sensors) seem ok.
Could someone please assist me troubleshooting the problem.
|
