Soil Science Society of America Journal 64:2136-2139 (2000)
© 2000 Soil Science Society of America
DIVISION S-6-SOIL & WATER MANAGEMENT & CONSERVATION
A profile cone penetrometer for mapping soil horizons
D.J. Rooney and
B. Lowery
Dep. of Soil Science, Univ. of Wisconsin, 1525 Observatory Dr., Madison, WI, 53706-1299 USA
rooney{at}earthit.com
 |
ABSTRACT
|
|---|
Accurate soil mapping at a fine scale is needed for site-specific farming, research on solute transport, and many other applications requiring detailed analysis of both the depth and thickness of soil horizons. Soil mapping techniques currently used are too costly to address the spatial variability of soil mapping units. We propose the use of a profile cone penetrometer (PCP) to assist in mapping soil properties at a landscape scale. Data collected with the PCP show clear changes in soil properties with depth, and were confirmed using profile descriptions from soil pits as well as 4.3-cm-diam soil cores. Combining three-dimensional PCP data with soil attribute information will provide a rapid and effective means to digitally update soil surveys and improve the efficiency and cost effectiveness of sampling techniques.
Abbreviations: CI, cone index • PCP, profile cone penetrometer
|
INTRODUCTION
|
|---|
SOIL HORIZONS are in part identified by their morphology and partly by properties that differ from those of the overlying and underlying horizons (Soil Survey Staff, 1999). This fundamental principle of soil classification emphasizes the qualitative element used to delineate various strata with reference to their depth and thickness. These vertical spatial designations are then used as guides for assigning additional soil attribute data such as texture, structure, color, or other characteristics that differentiate one horizon from another. It is not the absolute value of an attribute in a given layer that gives it a unique identity, but rather its value relative to the surrounding soil.
Cone penetrometers have been used to measure various soil attributes in the field. Cone index (CI), the force per unit area measured on the cone tip, has been correlated against many specific soil properties. These include bulk density (Carter and Tavernetti, 1968; Pidgeon and Soane, 1977; Ayers and Perumpral, 1982; Henderson et al., 1988), soil strength and compaction (Hooks and Jansen, 1985; Larney and Kladivko, 1989; Lowery and Schuler, 1991), pore size distribution (Vepraskas, 1984), and cementation (Puppala et al., 1995). Many studies have been conducted to examine the effect of soil moisture on CI values (Richards, 1942; Ayers and Perumpral, 1982; Henderson et al., 1988). Conclusions and correlations developed from these studies vary widely depending on the area of study, the cone dimensions and design of the penetrometer cone, and soil map unit. An investigators ability to interpret and predict specific attributes of a given soil depends greatly on these factors though it is evident from the literature that CI values vary from one soil map unit to another depending on their physical properties. To our knowledge, delineation and classification of surface soil strata for mapping applications with a penetrometer have not been the focus of any particular study.
Information regarding the depth and thickness of soil horizons is currently obtained through soil core sampling or, in extreme cases, excavation of pits to enable detailed analysis of soil characteristics and their vertical position. The result is a representative profile description that is used as the basis of comparison for soil samples collected from the surrounding landscape. Attributes of the soil samples and their vertical position are referenced back to the representative profile to determine the boundaries of the designated soil-mapping unit. These iterative processes yield coarse soil maps with limited spatial significance that are best used for general land use and suitability purposes. Intensive land-use applications such as site specific farming and other spatially related land uses require a more detailed and spatially accurate three-dimensional characterization of soil properties. A method is needed to supplement current field soil mapping techniques that increases the efficiency of soil survey techniques as well as the spatial significance of the output for the end user. Thus, the objective of this paper is to introduce the use of a profile cone penetrometer for rapidly developing spatially accurate maps of soil properties with minimal soil disturbance.
|
Materials and Methods
|
|---|
A modified penetrometer design was used for this study that combined attributes from both the American Society of Agricultural Engineers (ASAE, 2000) and American Society for Testing and Materials (ASTM, 1995) standard penetrometer specifications. A cone apex angle of 60° was chosen, as this configuration has been used widely in studies to characterize geologic strata for geotechnical site investigations (Baligh et al., 1980; Robertson, 1990). No sleeve friction measurement was obtained as designated in the ASTM specification. A penetration rate of 5 cm s-1 was used, instead of 3 (ASAE) or 2 cm s-1 (ASTM). It should be noted that CI has been shown to be relatively insensitive to velocity of penetration (Waldron and Constantin, 1970; Anderson et al., 1980). A constant penetration has been shown to be a more important variable than velocity when using a penetrometer to delineate soil properties (Freitag, 1967; Hooks and Jansen, 1985). Push rate was maintained within a standard deviation of less than 0.5 cm s-1.
The PCP system used in the study consisted of a 2-cm-diam cone with a 60° apex angle, a 1360-kg load cell (Omegadyne LC101; Sunbury, OH)1
for measuring force of penetration, a string potentiometer (Unimeasure HX-EP; Corvallis, OR) used to measure depth of penetration, and a data-logger (Campbell Scientific 21x; Logan, UT) for recording data. Data were stored in the field using a storage module (Campbell Scientific SM716). The PCP system was connected to a truck-mounted, hydraulic soil-probe device (Giddings #9HD; Fort Collins, CO). The hydraulic probe was mounted on a four-wheel powered, 3855-kg rated truck. A schematic of the components is shown in Fig. 1
.
In situ measurements were made at three locations in a field with soil mapped as a Plano silt loam (fine-silty, mixed, mesic Typic Argiudolls) on the Arlington Research Station, Arlington, WI. The three sites were located within a 1.6-ha field that was planted to corn (Zea mays L.) the past 2 yr. Prior to this, it was planted to alfalfa (Medicago sativa L.) for 5 yr. All the soil in this 1.6-ha area is mapped as a Plano silt loam. The parent material at the site consists of loess overlying glacial lake bed sediment deposits, overlying till or outwash. There is a 1.5-m change in elevation in the field of gently rolling topography. At each of the three sites, six penetration measurements were made and three 4.3-cm i.d. soil cores were collected, as shown in Fig. 2
. Penetration tests were conducted to an average depth of 130 cm at Site 1, 104 cm at Site 2, and 140 cm at Site 3.
After penetration tests were conducted at each of the three sites, a pit was excavated (Fig. 2) so that the face was located between the two parallel rows of penetrations, which enabled detailed analysis of soil properties. Dr. Peter Almond (a pedologist from Lincoln Univ., New Zealand) provided a profile description for each pit and delineated the depth and thickness of each horizon. These horizon designations were compared with the CI values obtained from each pit location (Fig. 3)
. Continuous bulk density (Blake and Hartage, 1986) and water (Gardner, 1986) content values were analyzed in 5-cm increments from each soil core obtained from each test site. Texture analysis was conducted at various intervals with intervals selected on the basis of profile descriptions from each core by the hydrometer method (Gee and Bauder, 1986). Bulk density and water content values were averaged between cores for each site in 5-cm increments and compared along with texture to the mean and standard deviation of CI values. Mean standard deviation and correlation coefficients were calculated by means of a spread sheet.
View larger version (32K):
[in this window]
[in a new window]
|
Fig. 3 Mean and standard deviation cone index values from six penetrations collected at three sites. Horizon delineations and soil descriptions are taken from the profile description
|
|
|
Results
|
|---|
Mean and standard deviation values for each set of six penetrations along with soil profile boundary delineations suggest the three sites have somewhat different soils (Fig. 3). It should be noted that within a given site, the six profile penetration readings are very similar. Most soil horizon boundaries that are established as a result of physical property changes were detected by PCP. Changes in CI magnitude, trend, and/or variability between horizons were observed in the soil profiles tested at all three sites. These CI changes vary in intensity both within the soil profiles at each site as well as between sites as a function of the abruptness of the horizon boundary. For example, at a depth of 68 cm, there is an increase in CI variability between the Bt and Bt2 horizons at Site 2 (Fig. 3). With the exception of the 100-cm depth at Site 1, the boundaries delineated by the profile description correspond to a change in CI variability, or an increase or decrease in mean CI values, or a combination of both (Fig. 3). Examination of the pit at Site 1 a week after its excavation revealed a noticeable structural change occurred at 100 cm. This structural boundary was not discernible from the examination of the fresh pit face. There is also a large difference in bulk density from the 48- to 91-cm (1.42 g cm-3) to 92- to 130-cm depth (1.50 g cm-3). Thus, it is possible that there will be soil profile features expressed in CI readings that are not delineated by a mapper. Likewise, CI values do not always correspond directly to the horizons as delineated by a mapper since, for example, the mapper might delineate horizons based on color, which might not be related to changes in physical properties.
A comparison between the three test sites demonstrates there is considerable variability present within a soil map unit. Differences are noted with respect to the diagnostic horizons present as well as their depth and thickness. The CI variability, as well as the trend of the mean value at each depth, is also different when comparing the three sites. The shape of the mean CI profile and its corresponding variability are more important parameters than absolute mean CI value when comparing the profiles from each site.
The most distinct difference is observed when comparing Sites 1 and 3 to Site 2. The shape of the penetration profiles are noticeably different, particularly with respect to the Ap horizon and CI variability between 20 and 65 cm. Additionally, penetration refusal occurred in the glacial till encountered at a depth of 95 cm at Site 2, while no glacial till was reached with the penetrations conducted at Site 1 and 3. Large spike penetration resistance values are associated with stones found in the glacial till (Fig. 3). Stones were encountered at 90 cm at Site 2 and 95 cm at Site 3, but there were no stones detected at Site 1. When stones were encountered during a penetration measurement, very large CI values were obtained. In each case, a spike in CI is apparent and easily differentiated from soil effects on the penetration resistance. These spike values make mapping depth to glacial till very simple. Obviously, varying depth to glacial till suggests a change in soil horizonation.
An increase in mean CI value or variability is not necessarily associated with an increase in mean bulk density and variability or a change in water content. Additionally, we did not find that a direct relationship can be observed between grain-size distribution and CI values. It should be noted that the range of values for water content (0.22–0.31 m3 m-3) and bulk density (1.26–1.50 g cm-3) for this soil may have been too small to show large differences in CI as a function of these parameters in some horizons. A cluster correlation between bulk density, water content, texture, and CI indicates little to no direct correlation between these values. The relationship between CI and some parameters was determined for each site as well as all sites combined. We think the lack of correlation between CI and any specific property is because of the small data set. We are conducting other, more detailed studies to establish relationships between CI and soil physical properties.
|
Summary
|
|---|
Although we found no direct relationship between CI and any specific soil physical property from our data set, the combination of properties that are unique to each diagnostic horizon results in a unique CI profile. This site-specific relationship can be used to map soil strata on a landscape scale by collecting core samples at locations chosen by observing changes in the CI profile. The depth within a soil profile to collect samples for laboratory analyses can also be selected more efficiently by observing the boundary depths from the CI profiles.
We have attempted to demonstrate that the PCP may be capable of distinguishing changes in soil physical properties that are associated with soil horizon thickness and boundary delineations. This technique can provide a rapid and spatially accurate method for mapping soil horizons at a landscape scale. We are in the process of using this technique to produce digital elevation model (DEM) based three-dimensional maps of soil profile characteristics on a landscape scale for detailed site characterization applications.
We are conducting more extensive research to study the relationship between CI and specific soil properties within horizons. It is possible that a method can be developed to determine the depth and thickness of subsurface horizons on the basis of physical property changes by calculating break-points in CI parameters such as magnitude, trend, and variability. Once delineated in this way, the CI relationship to soil physical properties can be established within each horizon across the landscape rather than between horizons in a given profile. This method of horizontal comparison of CI for assessing specific properties would be more useful for soil physical property characterization.
The further development of penetrometer systems for the assessment of specific soil properties must be pursued. Systems used for the analysis of texture, moisture, and optical properties are currently being designed and tested in a variety of laboratory and field settings. The ability to deploy sensors or combinations of sensors in the future will enable the optimization of site characterizations for specific soil properties, thus increasing the efficiency and effectiveness of soil mapping.American Society of Agricultural Engineers 2000; American Society for Testing and Materials 1995
|
NOTES
|
|---|
Research supported by CALS Nonpoint Source Pollution Project and USDA-CSREES NRI project no. 95-35108-51534.
1 Mention of company or product name does not constitute endorsement by the Univ. of Wisconsin-Madison to the exclusion of others. 
Received for publication May 10, 1999.
|
REFERENCES
|
|---|
- American Society of Agricultural Engineers. 2000. ASAE standard: ASAE S313.3. Soil cone penetrometer. p. 832–833. In Agricultural engineers yearbook of standards. ASAE, St. Joseph, MI.
- American Society for Testing and Materials. 1995. ASTM standard D5778-95. Standard test method for performing electric friction cone and piezocone penetration testing of soils. p. 576. In Annual book of ASTM standards. West Conshohocken, PA.
- Anderson G., Pidgeon J.D., Spencer H.B., Parks H. A new hand-held recording penetrometer for soil studies. J. Soil Sci. 1980;31:279-296.
- Ayers P.D., Perumpral J.V. Moisture and density effect on cone index. Trans. ASAE 1982;24:1169-1172.
- Baligh M.M., Vivatrat V., Ladd C.C. Cone penetration in soil profiling. ASCE, J. Geotech. Engr. Div. 1980;106:447-461.
- Blake G.R., Hartage K.H. Bulk density. In: Klute A., ed. Methods of soil analysis. Part 1, 2nd ed Madison, WI: ASA, 1986:364-366 Agron. Monogr. 9..
- Carter, L.M., and J.R. Tavernetti. 1968. Influence of precision tillage and soil compaction on cotton yields. Trans. ASAE 10:65–67, 73.
- Freitag D.R. Penetration tests for soil measurement. Trans. ASAE 1967;10:750-753.
- Gardner W.H. Water content. Agron. J. 1986;9:493-544.
- Gee G.W., Bauder J.W. Particle-size analysis. In: Klute A., ed. Methods of soil analysis. Part 1, 2nd ed Madison, WI: ASA, 1986:404-407 Agron. Monogr. 9..
- Henderson C., Levett A., Lisle D. The effects of soil water content and bulk density on the compactibility and soil penetration resistance of some Western Australian sandy soils. Aust. J. Soil Res. 1988;26:391-400.
- Hooks C.L., Jansen J.J. Recording cone penetrometer developed in reclamation research. Soil Sci. Soc. Am. J. 1985;50:10-12.
- Larney F.J., Kladivko E.J. Soil strength properties under four tillage systems at three long-term study sites in Indiana. Soil Sci. Soc. Am. J. 1989;53:1539-1545.[Abstract/Free Full Text]
- Lowery B., Schuler R.T. Temporal effects of subsoil compaction on soil strength and plant growth. Soil Sci. Soc. Am. J. 1991;55:216-223.[Abstract/Free Full Text]
- Pidgeon J.D., Soane B.D. Effects of tillage and direct drilling on soil properties during the growing season in a long-term barley mono-culture system. J. Agric. Sci. 1977;88:431-442.
- Puppala A.J., Acar Y.B., Tumay M.T. Cone penetration in a very weakly cemented sand. J. Geotech. Engr. 1995;121:589-600.
- Richards S.J. A soil penetrometer. Soil Sci. Soc. Am. Proc. 1942;6:104-107.
- Robertson P.K. Soil classification using the cone penetration test. Can. Geotech. J. 1990;27:151-158.
- Soil Survey Staff. Soil taxonomy: A basic system of soil classification for making and interpreting soil surveys. Washington, DC: USDA-SCS Agric. Handb. 436. U.S. Gov. Print. Office, 1999.
- Vepraskas M.J. Cone index of loamy sands as influenced by pore size distribution and effective stress. Soil Sci. Soc. Am. J. 1984;48:1220-1225.[Abstract/Free Full Text]
- Waldron L.J., Constantin G.K. Soil resistance to a slowly moving penetrometer. Soil Sci. 1970;109:221-226.
This article has been cited by other articles:
|
|
|
|
 
E. Ben-Dor, D. Heller, and A. Chudnovsky
A Novel Method of Classifying Soil Profiles in the Field using Optical Means
Soil Sci. Soc. Am. J.,
June 18, 2008;
72(4):
1113 - 1123.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
