Soil Science Society of America Journal 66:396-401 (2002)
© 2002 Soil Science Society of America
DIVISION S-1—NOTES
A heavy-duty time domain reflectometry soil moisture probe for use in intensive field sampling
Dan S. Long*,a,
Jon M. Wraithb and
Greg Kegelc
a Northern Agric. Res. Center, Montana State Univ., Havre, MT 59501
b Dep. of Land Resources and Environmental Sciences, Montana State Univ., Bozeman, MT 59717-3120
c Dep. of Industrial and Engineering Technology, Montana State Univ., Northern Campus, Havre, MT 59501
* Corresponding author (dlong{at}montana.edu)
 |
ABSTRACT
|
|---|
Mapping soil water content for site-specific management of farm fields is commonly achieved through grid soil sampling. This effort frequently requires intensive soil coring, which is destructive and time consuming. Precision farming and research can be facilitated if sampling techniques are improved and made cost effective. We designed and constructed a heavy-duty time domain reflectometry (TDR) probe of simple design that fits a hydraulic soil sampling machine. The design is simple enough to be adapted to the TDR equipment and hydraulic sampler that a researcher or practitioner may already possess. The probe consists of a housing and adapter that can be fabricated in a local machine shop from materials that are easily obtainable. Probe schematics and summary soil water maps obtained using the new design are presented. As with conventional TDR probe designs, stony or indurate soils may present particular problems in effective insertion of the probe.
Abbreviations: TDR, time domain reflectometry
|
INTRODUCTION
|
|---|
SOIL WATER is often the most limiting factor affecting crop yields under arid and semiarid conditions. Thus, there is strong interest to characterize spatial patterns in soil water content across farm fields. This spatial information is especially useful for implementing precision agriculture, or site-specific crop management (Robert, 1993). However, the acquisition of site-specific information often requires intensive soil sampling. To characterize the distribution pattern of many soil attributes (e.g., nutrient levels, salt content, and plant-available water), researchers often sample in accordance with a systematic grid that has been imposed within a field (Peterson and Calvin, 1986; Sabbe and Marx, 1987). For soil water, cores may be taken individually at each grid node and analyzed for gravimetric or volumetric water content. Distribution maps may then be created using a mathematical interpolation technique such as kriging (Oliver, 1987; Mulla, 1988; Yates and Warrick, 1987).
Time domain reflectometry is an established and reliable means to measure volume water content in soils (Topp et al., 1980). Advantages relative to other methods include high accuracy and precision, nondestructive nature of the measurements, and lack of calibration requirement in many cases (Or and Wraith, 1999). Several TDR soil water measurement systems are commercially available. Common applications utilize buried probes at various depths and various locations, with wire transmission leads brought to the surface. Unfortunately, the presence of transmission lines on the soil surface may impede routine field operations. Additionally, maximum probe lead lengths under typical conditions of low to moderate salinity and clay contents are 50 to 80 m because of excess signal loss in longer runs. Finally, use of buried probes at multiple locations involves substantial sensor expense. Thus, use of permanently installed probes is generally unfeasible to determine soil water content at numerous spatially distributed grid locations, as may be needed for site-specific management of farm fields.
An alternative to permanent installations is portable hand probes, which allow measurements to be made rapidly at multiple locations. Hand probes are typically 15 to 60 cm in length, but may be difficult to push into soil that is hard or dry. This difficulty may lead to lateral movement during insertion, resulting in incomplete soil contact along the rod lengths. Annular air gaps around the probes will cause inaccurate soil water content readings (Ley et al., 1994). A heavy-duty hand probe with sliding hammer to facilitate insertion is manufactured by Soilmoisture Equipment Corp. (Goleta, CA). However, this probe incorporates an impedance mismatch along the transmission path immediately before the start of the rods, to facilitate signal analysis by the manufacturer's proprietary software. The influence of this impedance change on the resulting TDR travel time signal is not compatible with analysis software of many other manufacturers. Additionally, the probe was not designed for the labor-intensive nature of extensive grid soil sampling.
An effective TDR probe designed to obtain site-specific field-scale information should facilitate insertion into hard and dry soils, and should minimize the effort required to obtain soil water measurements at multiple locations. To reduce costs and increase flexibility, the probe should also be adaptable to the TDR equipment that a researcher or practitioner may already possess. Therefore, we designed a simple, heavy-duty TDR probe adapted to a hydraulic assist that will allow researchers to more effectively measure soil water contents within farm fields. We detail the construction of this device and illustrate its potential utility using measurements obtained from a dryland agricultural field in northern Montana.
|
Materials and Methods
|
|---|
Probe Design and Construction
The heavy-duty TDR probe has three main components: the bottom housing, the top housing, and the adapter coupling (Fig. 1)
. The bottom and top housings were constructed from 11.4-cm (4.5-in) diam. Delrin plastic rod (DuPont Corp., Wilmington, DE). Delrin is an acetal resin plastic offering high strength and good electrical insulating properties. In addition, this material can be easily fabricated with hand tools and production machinery. The bottom and top housings are coupled by means of coarse threads. Unscrewing the housings allows access to the internally mounted rods and attached coaxial cable. A two-rod design was used in accordance with the Soilmoisture Equipment Corp. "Slammer" probe (Goleta, CA). The probe consisted of two stainless steel rods 9.53 mm (0.375 in) in diam., each having an exposed length up to 60 cm and rod center separation of 5.1 cm (2.0 in). This geometry (i.e., ratio of rod separation to rod diameter of 5.33) is consistent with criteria (Knight, 1992; Ferre et al., 1996) to minimize measurement susceptibility to lateral spatial variations in water content, as well as to localized zones of compaction that may develop immediately surrounding the rods during installation.
View larger version (56K):
[in this window]
[in a new window]
|
Fig. 1. Exploded diagram of the heavy-duty TDR probe: adapter coupling (A), top housing (B), and bottom housing (C).
|
|
Two holes in the bottom housing (Fig. 2)
were countersunk to accept heavy-duty waveguide rods that vary in length from 20 to 60 cm and are purchased from Soilmoisture Equipment Corp. A narrow groove in the bottom housing accepts an O-ring to seal the internal surfaces when the housings are screwed together. The channel cut into the wall between the flat-bottomed cavity in the center of the bottom housing and rod openings permits the routing of coaxial cable signal or ground leads to each rod. Cable leads are attached to the rods by means of small screws after a small hole is drilled and tapped in the upper end of each waveguide rod.
The top housing is 127 mm (5 in) in length (Fig. 3)
, which provides sufficient distance of separation between the adapter and waveguides. Without this much separation, the steel adapter will interfere with the electromagnetic TDR signal that is propagated along the waveguide rods. The top housing provides an internal upper surface that touches the ends of the rods when the two halves are screwed together. This surface counters the upward force on the rods when the probe is pressed into the soil. In addition, the top housing provides an external upper surface to connect the adapter to a commercially available hydraulic core sampler (Fig. 4)
. The adapter was welded from iron-square tubing and steel-round stock. The square tubing fits onto the Kelley bar of a Giddings soil sampling machine (Fort Collins, CO) and is disconnected in the same way as regular soil coring tubes are. However, the size and shape of the adapter is flexible and may be easily adapted to other hydraulic samplers. A center hole is drilled through the center of the top housing to rout the coaxial cable to the outside. This cable exits the top of the probe through a hole drilled into the side of the adapter. Four tapped holes placed in the top of this housing allow the probe to be attached to the adapter.
Cost for the Delrin plastic was $78.00. Fabrication of the housings and adapter by a local shop cost $200. Therefore, total cost for a single probe, excluding the waveguide rods and coaxial cable, was <$300. Coaxial cable is about $0.82 m-1, with cost of a BNC connector to interface probe lead and TDR cable tester about $5.00. The heavy-duty TDR probe rods we used are about $125 to $170 per pair from Soilmoisture Equipment Corp., depending on length. We used a 6-m coaxial cable for our initial application, which was sufficient to connect the probe at rear of pickup with cable tester inside the cab (see Fig. 4). We used the probe with our existing TDR cable tester (Model 1502B, Tektronix Inc., Beaverton, OR) and data logger (Model 21X, Campbell Scientific Inc., Logan, UT). A removable metal storage cabinet was installed in the rear seating area of the truck to secure the cable tester and data logger. Power was supplied from a 12V deep cycle marine battery.
Confirming Probe Performance and Mapping Soil Water Content
To confirm the measurement characteristics of the heavy-duty TDR probe, a simple laboratory calibration exercise was conducted. The probe was vertically inserted into a 60-cm deep and 15-cm diam. column of Flathead fine sandy loam soil (coarse-loamy, mixed Pachic Udic Haploborolls) enclosed in polyvinyl Cl (PVC) pipe. Known volumes of water were incrementally added to the top or bottom of the soil column, then allowed to equilibrate for 1.5 to 2 h, with longer times for dryer conditions. The apparent dielectric constant of the soil was measured using TDR after each stepwise addition of water. The Ledieu et al. (1986) linear relationship was used to relate measured soil apparent dielectric constant to volume water content. This calibration relationship provides results similar to the Topp et al. (1980) polynomial equation in the water content range 
= 0 to 
0.5, which covers the range of interest for most agricultural soils. Time domain reflectometry-measured volume water contents were compared with those calculated using known soil and incremental water volumes.
A contour map of soil water content was produced by taking a single reading at each node of a regular diagonal grid within one 20-ha dryland wheat field near Havre, MT on 1 Jun. 2000. Each TDR measurement was georeferenced with use of a GPS receiver with ±1-m positional accuracy. The geostatistical procedure of kriging was used to interpolate soil water contents for unsampled locations and to produce contour maps of soil wetness. Kriging estimates were computed from best-fit variograms and a weighted linear combination of eight surrounding observations.
|
Results and Discussion
|
|---|
To confirm measurement characteristics of the TDR probe, volumetric water contents measured using TDR were compared with those calculated using known incrementally added water volumes. Results (Fig. 5) provided confirmation of suitability similar to conventional probes. Measured water contents were somewhat lower than calculated for the range greater than about = 0.25 (Fig. 5). This may have been a result of (lateral) water distribution pattern within the soil column, experimental error in measuring water volumes added, or inaccurate TDR signal analysis. Each of these are potential sources of error using conventional TDR probes as well. Utility of the Topp et al. (1980) and Ledieu et al. (1986) calibration relationships for the Flathead soil had been previously confirmed using smaller TDR probes (Das et al., 1999). The relationship between measured and calculated water contents illustrated in Fig. 5 is clearly acceptable and similar to that expected using TDR. Our goal was merely to confirm, using a simple and rapid calibration procedure, that the large probe functions similarly to those used in conventional TDR applications.
Using the heavy-duty probe with the Giddings machine, we spatially characterized the soil water content in the top 30-cm at the agricultural site at Havre (Fig. 6)
. Less than 5 min was required per reading when using the hydraulically inserted heavy-duty TDR method. This sampling speed was increased compared with the conventional method of measuring soil water content from soil cores, which is time and labor intensive. This includes extracting cores from sampling tubes, placing samples into bags, drying soil samples, and determining weight loss upon oven drying for 48 h. By reducing the sampling time, this proposed method allows the researcher to increase the number of samples that can be taken within a field. Thus, spatial patterns in soil water content can be resolved in greater detail within fields, for a similar or smaller time investment, than with conventional methods.
View larger version (43K):
[in this window]
[in a new window]
|
Fig. 6. Distribution of soil water content in top 30-cm as derived from measurements with the heavy-duty TDR probe at sampling locations (denoted by x).
|
|
The heavy-duty TDR probe is relatively inexpensive, but its use requires the availability of expensive TDR cable testing equipment. In comparison, cost for a Giddings standard slotted sampling tube (5.08-cm diam., 91-cm length) with tube bit and adapter is only about $200. Nevertheless, the TDR approach provides quick and reliable measurements of soil water content, which lends to intensive sampling of farm fields. Greater initial equipment expense may lead to lower overall cost when amortized over extensive field sampling. The probe design is simple enough that it can be readily adapted for use with many TDR instruments and analysis software, and with other hydraulic core samplers. Disadvantages include the fact that TDR measurements may not be possible in soils with high clay contents or high salinity, and stony or very hard soils may inhibit probe insertion even with hydraulic assist. Finally, the design is limited to sampling of surface soil layers. For most soils, 60-m probe length probably represents a practical limit because of consideration of probe insertion and signal loss. For applications where knowledge of soil wetness at deeper depths is required, alternative methods such as neutron moisture meters, acquisition of soil cores, or use of buried TDR probes may be more appropriate.
|
ACKNOWLEDGMENTS
|
|---|
The Montana Wheat and Barley Committee and Montana Agricultural Experiment Station provided funding for developing the heavy-duty TDR probe and purchasing TDR cable testing equipment. The authors thank Mr. Jeffery Whitmus, MSU Agricultural Research Technician, for outstanding technical support.
|
NOTES
|
|---|
Contribution of the Montana Agric. Exp. Stn., Montana State Univ.-Bozeman, Journal No. 2001-44.
Received for publication November 10, 2000.
|
REFERENCES
|
|---|
- Das, B.S., J.M. Wraith, and W.P. Inskeep. 1999. Nitrate concentrations in the root zone estimated using time domain reflectometry. Soil Sci. Soc. Am. J. 63:1561–1570.[Abstract/Free Full Text]
- Ferre, P.A., D.L. Rudolph, and R.G. Kachanoski. 1996. Spatial averaging of water content by time domain reflectometry: Implications for twin rod probes with and without dielectric coatings. Water Resour. Res. 32:271–279.
- Knight, J.H. 1992. Sensitivity of time domain reflectometry measurements to lateral variations in soil water content. Water Resour. Res. 28:2345–2352.
- Ledieu, J., P. de Ridder, P. de Clerck, and S. Dautrebande. 1986. A method of measuring soil moisture by time-domain reflectometry. J. Hydrol. 88:319–328.
- Ley, T.W., R.G. Stevens, R.R. Topielec, and W.H. Neibling. 1994. Soil water monitoring and measurement. Pacific Northwest Coop. Ext. Bull. No. PNW0475. (Available on-line at http://cahe.wsu.edu/infopub/pnw0475/pnw0475.html#anchor1601641). (Verified 24 Oct. 2000).
- Mulla, D.J. 1988. Estimating spatial patterns in water content, matric suction, and hydraulic conductivity. Soil Sci. Am. J. 52:1547–1553.[Abstract/Free Full Text]
- Oliver, M.A. 1987. Geostatistics and its application to soil science. Soil Use Manag. 3:8–20.
- Or, D., and J.M. Wraith. 1999. Soil water content and water potential relationships. p. A53–A85. In M. Sumner (ed.) Handbook of soil science, CRC Press, Boca Raton, FL.
- Peterson, R.G., and L.D. Calvin. 1986. Sampling. p. 33–52. In A. Klute (ed.) Methods of soil analysis. Part 1. 2nd ed. Agron. Monogr. 9. ASA and SSSA, Madison, WI.
- Robert, P.C. 1993. Characterization of soil conditions at the field level for soil specific management. Geoderma 60:57–67.
- Sabbe, W.E., and D.B. Marx. 1987. Soil sampling: spatial and temporal variability. p. 1–14. In J.R. Brown (ed.) Soil testing: Sampling, correlation, calibration, and interpretation. SSSA Spec. Publ. 21. SSSA, Madison, WI.
- Topp, G.C., J.L. Davis, and A.P. Annan. 1980. Electromagnetic determination of soil water content: Measurements in coaxial transmission lines. Water Resour. Res. 16:574–582.
- Yates, S.R. and A.W. Warrick. 1987. Estimating soil water content using cokriging. Soil Sci. Soc. Am. J. 51:23–30.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
D. A. Robinson, C. S. Campbell, J. W. Hopmans, B. K. Hornbuckle, S. B. Jones, R. Knight, F. Ogden, J. Selker, and O. Wendroth
Soil Moisture Measurement for Ecological and Hydrological Watershed-Scale Observatories: A Review
Vadose Zone J.,
February 25, 2008;
7(1):
358 - 389.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. S. Lamhamedi, L. Labbe, H. A. Margolis, D. C. Stowe, L. Blais, and M. Renaud
Spatial Variability of Substrate Water Content and Growth of White Spruce Seedlings
Soil Sci. Soc. Am. J.,
December 2, 2005;
70(1):
108 - 120.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. A. Robinson, D. A. Robinson, S. B. Jones, J. M. Wraith, D. Or, and S. P. Friedman
A Review of Advances in Dielectric and Electrical Conductivity Measurement in Soils Using Time Domain Reflectometry
Vadose Zone J.,
November 1, 2003;
2(4):
444 - 475.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
