Last update: March 23, 2012 11:09:40 AM E-mail Print


Simple techniques for estimating soil compaction


C H Donaldson

Pasture Research

Grootfontein College of Agriculture

Private Bag X529





A soil penetrometer is an instrument used for estimating soil compaction. Two methods of advancing the penetrometer into the soil and of measuring the amount of force required to cause penetration are generally used. When the soil is hard and firm an impact method whereby the penetrometer is driven into the soil by a hammer or falling weight is frequently employed in a more static penetration test the penetrometer is pushed steadily into the soil. This method is used mainly when the soil is moist or soft.

Many hand-held penetrometers are designed to be used in this manner-some of these types are self-recording and no calculations are needed to interpret the results (Anon 1983; Terry & Wilson, 1953).

Many varieties of penetrometers have been designed to give quantitative measurements of soil penetration resistance for a more precise correlation with certain soil physical properties (Davidson 1965). The soil compaction process (Harris, 1971) and other means of measurement (Freitag, 1971) have also been described Most penetrometers can be adapted to many uses The Proctor penetrometer was, for example, successfully used to measure soil compaction or hardness caused by the long term effects of continuous and fixed seasonal grazing treatments (Donaldson & Grossman, 1985) and to study the long term effects of nitrogen and phosphate fertilizer treatments on veld (Donaldson, Rootman & Grossman, 1984)

After considerable field experience with the spring dynamometer (Proctor penetrometer) and hydraulic type penetrometers it became evident that reliable measurements were largely dependent on pushing the penetrometer into the soil at a constant and steady rate. This operation may therefore easily lead to in. consistent results between operators. The aluminium alloy needles of these penetrometers were also found to be too fragile and unsuitable for hard and dry soils. These and other problems, such as availability and expense, could be overcome by using simple homemade instruments. The first instrument that was designed, and referred to as the hammer type penetrometer, was found to be rather time-consuming and also tiring to the arm. These problems could be overcome by developing a second instrument referred to as the rod penetrometer.

The aims of this paper are firstly to describe the construction and operation of two home made penetrometers; secondly, to evaluate the penetrometers on the basis of operator variability; and thirdly, to evaluate the soil compaction data determined with three penetrometers in camps subjected to long term fixed seasonal grazing treatments.



Construction and operation of penetrometers


1. Hammer-t ype penetrometer

This penetrometer (Fig 1(a)) consisted of two basic parts: a steel rod 120 cm long with a diameter of 10 mm and a steel needle, 25 cm long x 6 mm wide brazed to the lower end of the rod, total mass = 1 098 g, the hammer comprising a 27,5 cm long galvanized water pipe (internal diameter at least 16 mm) fixed to a 15 mm galvanized T-piece, giving a total mass of 518 g.



Operation of the hammer-type penetrometer consisted of holding the rod in a perpendicular position with the lower end on the soil between ground layer vegetation.

The hammer was held above the rod with its lower end level with a mark 15 mm from the top end of the rod. The hammer was then allowed to fall a constant distance before it hit the rod. The height of fall was fixed at 30 cm. Soil compaction was measured by counting the number of times the hammer had to be dropped on the rod to force the rod to a pre-determined depth into the soil. A depth of 50 mm was used in these studies.


2. Rod penetrometer

This penetrometer (Fig 1(b)) also consisted of two basic parts a rod and steel needle with the same specifications as the hammer-type penetrometer; a tube 105 cm long by 15 mm (minimum internal diameter) with a flat washer brazed to the lower end of the tube to prevent the tube from sinking into loose soil. The rod was marked at 5 mm intervals in an upward direction when the lower surface of the washer was level with the ground and with the lower end of the rod. The zero penetration reading corresponded with the top end of the tube. A mark was also made at a distance of, for example, 30 cm below the zero mark.

To operate the rod penetrometer the tube was held in an upright position. The rod was then positioned in the tube with the lower 30 cm (down) mark level with the upper end of the tube. This implied that the lower end of the rod was held 30 cm above soil surface. The rod was then released and the depth of penetration was determined by reading from the calibrated rod the value corresponding with the upper end of the tube

Two separate investigations were undertaken in mixed Karoo veld at the Grootfontein College of Agriculture. The soil of the experimental site was of the Glenrosa form derived from sandstone. Soil moisture determinations of the top 7 5 cm soil layer were determined gravimetrically for each treatment at the time of the trials. Six samples were taken at random from each treatment site.


Trial 1

Compaction measurements were made during June 1984 on dry and artificially wetted plots in each of the following adjoining sites:

Site 1 - Veld path subjected to occasional motor traffic;

Site 2 - Overgrazed veld;

Site 3 - Veld protected from grazing for a period of 43 years.

Site 2 and site 3 were approximately 30 m apart with site 1 occupying a position midway between sites 2 and 3 Twenty 2 x 3 m plots were allocated to each site. Sufficient water was ponded on ten plots of each site to ensure moisture penetration at least 30 cm deep. Soil samples for soil moisture determinations were taken 24 hours after wetting.

Soil compaction measurements were conducted by three operators using the rod and hammer penetrometers illustrated in Figure 1 (a) & (b) and a hydraulic type penetrometer. Compaction readings, 3 per plot (30 per treatment) were taken by each operator. Statistical analyses of the results of the penetrometer data for each instrument tested were done as stipulated by a 3 x 3 x 2 factorial design.


Trial 2

Compaction measurements with the rod, hammer and hydraulic penetrometers were conducted during April 1984 in dry soil on five paddocks subjected to the following long term (43 years) fixed grazing treatments:

(1) total exclusion from grazing; (2) spring (Sept/Oct) grazing; (3) summer grazing (Nov, Dec, Jan & Feb); (4) autumn grazing (March/April); (5)winter grazing (May, June, July & Aug). The compaction of an animal path in a heavily grazed paddock adjoining the experimental paddocks was also determined. Ten sampling sites were systematically selected across the width of each paddock (and foot path) – five compaction measurements with each penetrometer were taken at each site giving a total of 50 samples per paddock.



Trial 1

The soil compaction (penetration resistance) measurements obtained by the three operators for the rod hammer and hydraulic penetrometers are presented in Table 1.



There were highly significant differences in the soil compaction data between the three sites, between the dry and wet soils and the site x soil moisture interaction for each respective penetrometer.

For each penetrometer the soil compaction of the dry and wet soils of site 1 were greater than the soil compaction of the dry and wet soils of site 2 and site 3. The mean moisture percentages of the wetted soils of sites 1,2 and 3 were 14,8%, 15,4% and 16,1 % respectively - these values did not differ significantly. The mean moisture percentages of the dry soils of 2,9%, 3,0% and 3,3% for sites 1,2 and 3, respectively, were also very similar. As regards the effect of soil moisture, the dry soils gave a significantly higher compaction value than the wet soils (Table 1). The highly significant interaction between sites and soil moisture means that the three sites responded similarly to operators but that the sites were not equally affected by the dryness or wetness of the soil. This finding can also be explained from Table 1 by calculating the ratios in the soil compaction of dry or wet soils for each penetrometer. Unlike for wet soils the ratios of the soil compaction measurements of dry soils for the three sites were very similar for the rod (1; 1,7; 2,5), hammer (1; 1,6; 2,1) and hydraulic penetrometers (1; 1,7; 2,4). It is therefore suggested that soil penetrability measurements be made in dry soils in order to decrease the effect of variable soil moisture on the accuracy of the compaction readings.

The most important feature of the data (Table 1) was the non-significant differences in soil compaction between operators. This finding therefore suggests that anyone of the penetrometers can give similar and consistent results when operated by different workers.


Trial 2

Soil compaction was greatest on the paddocks of the autumn grazing and summer grazing treatments and least on the winter grazing and exclosure treatments - these differences were highly significant (P ≤ 0,01) for all penetrometers tested (Table 2).



There were strong significant correlations between the mean monthly rainfall and the soil compaction of the four grazing treatments for the rod (r = -0,94), hammer (r = 0,94) and hydraulic (r =0,89) penetrometers (Table 2). The vital role played by the amount of rainfall falling during the period of grazing on soil compaction is well illustrated by comparing the grazing treatments receiving different amounts of precipitation but having the same stocking rate and stocking density, viz., autumn grazing versus spring grazing and summer grazing versus winter grazing (Table 2). The data is therefore in agreement with the general principle that the more often soils are wetted to relatively high moisture levels, the greater the chance of maximum soil compaction from animal treading.

The highly significant (P ≤ 0,01) correlations between the compaction readings of (a) the rod and hammer penetrometers (r = -0,93); (b) the rod and hydraulic penetrometers (r = -0,91); and the hammer and hydraulic penetrometers(r = 0,99) provide further evidence of the similar trends in the compaction results obtained with the three penetrometers (Table 2).

The problem of comparing the compaction data obtained from penetrometers which express the soil compaction in different units could be overcome by relating the compaction data to the compaction of a common benchmark such as the animal path (Table 2). A notable feature of the compaction data when expressed as a percentage deviation from the animal path benchmark (Table 2) is the very similar results obtained with the rod and hydraulic penetrometers. The percentage deviation data obtained from the hammer penetrometer were on the average 14 percentage units below those of the other two penetrometers. Results (with different percentage values and threshold limits) similar to those obtained by using the animal path as a benchmark can be expected by expressing the compaction readings of the paddocks as a percentage deviation of the soil compaction in the exclosure treatment (Table 2) or from longstanding untrod fence-line benchmarks (Donaldson & Grossman, 1984).

The soil penetrometers tested provided an adequate measurement of soil compaction for trials of this nature. The penetrometer cannot replace other known techniques for determining soil compaction and soil density. The instruments described in this paper are however simpler, quicker and less expensive than other methods and may therefore deserve attention.



Anon., 1982. Do-it-yourself compaction check. Fmrs. Wkly. (Bloemfontein), July 23; 50-52.

DAVIDSON, D. T., 1965. Penetrometer measurements. p.472-In C.A. Black et al (ed). Methods of soil analysis. Part 1. ASA Monograph series. AM. Soc. Agron. Madison U.S.A.

DONALDSON, C.H., ROOTMAN, G. T. & GROSSMAN, D., 1984. Long-term nitrogen and phosphorus application to veld. L Grassl Soc. Sth. Afr. 1:27-32.

DONALDSON, C.H. & GROSSMAN, D., 1985. Soil compaction of grazed paddocks and untrod fence-line areas as measures of site condition. (Unpublished paper).

FREITAG, D.R., 1971. Methods of measuring soil compaction. p. 47-105. In K.K. Barnes et al (ed). Compaction of agricultural soils. Am. Soc. Agric. Eng., St. Joseph, Mich.

HARRIS, W.L., 1971. The soil compaction process. p.9-46. In K.K. Barnes et al (ed). Compaction of agricultural soils. Am. Soc. Agric. Eng., St. Joseph, Mich.

TERRY, CW & WILSON, H.M., 1953. The soil penetrometer in soil compaction studies. Agric. Eng. 34:831-834.



Karoo Agric 3 (7)