WHEEL TRAFFIC EFFECTS ON SOIL COMPACTION

AND GROWTH OF WHEAT*M. Feldman

Member CSAE

Engineering Research ServiceResearch Branch, Canada Agriculture

Ottawa, Canada

K. W. Domier

Member CSAE

Agricultural Engineering DepartmentUniversity of Alberta

Edmonton, Alberta

INTRODUCTION

Soil compaction affects the suitability of a soil for its intended use —crop production. Compaction reducespermeability to water, reduces aeration, and increases mechanicalstrength of the soil — all effects thatmay reduce the quality and quantityof the crop produced.

Compaction is caused by forces applied by machines and animals or resulting from natural phenomena. Before compaction can be predicted andcontrolled, these forces will have tobe described by accurate behaviourequations. Soil forces are the equation inputs, and are difficult to measure; change in compactness is theequation output and can be moreeasily expressed, usually as the absolute volume. However, such equationsare not available, though considerablework has been done. Simple loadingdevices and test methods establishedthat the largest principal stress is notuniquely related to compaction (7).More elaborate tests using pressuretransducers in the soil (7), and a triatrial apparatus (7), (3), failed toformulate accurate behaviour equations. Soil strength values measuredby mechanical seedlings were 3 to 7times the values predicted by theTerzaghi rupture theory model (5).

Similarly, accurate behaviour equations representing plant response tosoil compaction do not exist. A reviewpaper (2) concluded that mechanicalresistance influences growth of rootsand underground shoots in many situations, rather than merely becominga limiting factor only in unusuallystrong soils. The same review suggested that forces required to deform soil

*Master of Science Thesis (same title) inAgricultural Engineering by M. Feldman,University of Manitoba, October 1968.

Contribution No. 181 from EngineeringResearch Service, Research Branch, Canada Agriculture, Central ExperimentalFarm, Ottawa, Ontario.

RECEIVED FOR PUBLICATION

SEPTEMBER 20. 1969

can be estimated by classical soilmechanics theories if plastic and elastic zones of action and modes offailure are recognized and separated.Modes of soil failure due to plantforces are tension, shear without compression, and shear with compression.In other investigations, cotton seedlings exterted thrusts of up to onepound (5), and increased pressure oncorn seedlings reduced growth (8).

While the search for behaviourequations continues, with progressively more discerning methods, empirical measurements of soil compaction and crop response can supplyparameters for better tillage and traction machinery design. Numerousstudies have been made, but onlysome of the more relevant ones willbe mentioned here. Gill and VandenBerg (7) reported that, at equalweight and pull, a tire caused morecompaction than a track, partly dueto greater slippage of the tire. Increased compaction decreased rootpenetration and reduced water infiltration rate. Rosenberg (14) reported that crop growth was relatedto bulk density, penetrometer resistance and oxygen diffusion rate. Tomatoes were more sensitive to soilcompaction than corn, though tomatoes were affected by surface treatments, while corn yield correlatedwith subsurface treatments (17).Minimum tillage improved crop response due to reduced machine traffic (15). In North Dakota (9), overwintering reduced traffic pans.Domier (4) initiated studies to dealwith soil compaction problems inManitoba. Plots were packed bysuperimposed passes with a tractor.Compared to check plots, the treatments reduced yield on early seededplots, but some treatments improvedyield on late seeded plots. There wereno differences between plots seededwith a discer and those seeded witha double disc drill. In general, post-seeding treatment yields were lessthan pre-seeding treatment yields.

In addition to the results reportedby Domier (4), the following problems have been observed on Manitoba clay soils:

1. The emerging crop is often poorerin wheel paths made during harrowing. It is not generally knownwhether these effects carry throughto crop maturity.

2. In some cases, initial crop growthappears better in wheel tracks.

3. With a discer, the path of the lefttractor wheel remains visible evenafter the operation. There is no experience as to whether its effectsare harmful or not.

4. Harmful effects increase as soilmoisture content increases.

The object of this study, therefore,was to examine pre-seeding and post-seeding soil compaction on a wet clayfield during seeding activities bycharacterizing the soil conditions anddetermining the crop response.

PROCEDURE

The Treatments

The experiment was carried out ona three-acre plot of Scanterbury,Morris and McTavish clay soil (moderately drained) on the Glenlea Research Station, University of Manitoba. Soil moisture content was 26%(0 to 3-inch depth) during preseed-ing treatments and 32% (0 to lVfe-inch depth) to 37% (2 to 3-inchdepth) during post-seeding treatments. Pre-seeding compaction treatments were performed while seedingthe plot, in one direction, with a tractor and discer equipped with seedingand fertilizer attachments. Manitouwheat was seeded at 77 lbs. per acre,with 11-48-0 fertilizer applied at 73lbs. per acre. The following day, post-seeding compaction treatments wereadded at right angles to the directionof seeding at the same time as theplot was harrowed. In this way, pre-

CANADIAN AGRICULTURAL ENGINEERING, VOL. 12, No. 1, MAY 1970

seeding compacted, post-seeding compacted, and non-compRcted areaswere specifically located over the entire plot. The non-compacted areas,providing comparisons to evaluatetreatment results, received both tillage operations but no wheel traffic,so represented the intended soil environment.

The left wheel of the tractor pulling the discer during seeding accomplished the pre-seeding compactiontreatments. Treatment levels wereobtained by varying the rear tractorwheel slip rate between 7 and 30 percent. The low slip level was attainedby pulling the discer alone. For increased slip, the required drawbarpull was obtained by hitching a loadtractor to the drawbar of the headtractor in addition to the discer. Acable was used for hitching, to allowthe load tractor to travel to the rightof the discer.

A two-wheel trailer, fitted withspecial tires (9.00 x 15 treadless racing slicks) and loaded with cast ironweights applied the post-seeding compaction. A tractor towed the trailer,and the trailer, in turn, towed a setof spike tooth harrows. The tractor'sswinging drawbar was pinned to oneside so that the racing tires on thetrailer did not follow the tractorwheels.

The racing slicks provided the mostpractical means of applying a known,uniform packing pressure. With sucha low ply, flexible tire the pressuredistribution applied to the soil wasquite uniform and essentially equalto the inflation pressure (7), (18).Three treatment levels of 15, 27 and40 psi were chosen. These approximate the range of contact pressuresmeasured on the lug face of apowered tractor tire (16). Trailerload was changed along with inflation pressure of the racing tires according to the manufacturer's loadand inflation pressure recommendations.

Characterization of theSoil Conditions

The soil conditions resulting fromthe treatments were characterized bymeasuring four types of soil parameters. These were bulk density,shear strength, penetrometer resistance, and oxygen diffusion rate.

Dry bulk density, measured by the

sand cone method, was used as adirect measure of soil compactness.For each observation, a small excavation was dug between the one- andtwo-inch levels, below the soil surface. The excavation volume wasmeasured, using uniform Ottawasand, and the oven dry weight of theexcavated soil was obtained.

A 2-inch square shear box, imbedded and pulled horizontally with asmall spring scale, provided themeans of measuring cohesion. Cohesion is the remaining component ofshear strength when normal force iszero, according to the well knownMohr-Coulomb Theory. The measurements were made at one depth in thesoil, approximately midway betweenthe seed and the surface. This parameter was chosen for measurementsince compaction can increase soilstrength, impeding shoot emergenceand root growth (2), (7).

Penetrometer measurements weremade with a 30-degree cone loweredor raised by a hand crank. The soilforce resisting the cone was counteracted by an automotive valve spring.Through a linkage to the pen holder,penetration moved the pen verticallyand spring deflection moved it horizontally, relative to the stationarypaper holder. A nylon tip pen recorded the locus. The force-deflectioncurve was established and convertedto a cone index-deflection curve. Coneindex is the resisting force per unitcone end area. The penetrometer,used by many other researchers (1),(6), (9), (13), (17), provided aquick and easy device to obtain relative comparisons of different conditions. However, no correlation withany physical soil property has beenestablished.

The oxygen diffusion rate apparatus used was quite similar to thatdescribed in detail by other reports(10), (11). Potential was appliedacross a porous cup and.the platinumelectrodes, all in the soil, and the resulting current recorded. The magnitude of the current depends upon therate of oxygen flux at the electrodesurface. This, in turn, is related to therate of oxygen diffusion through thesoil to the electrodes that representplant roots. Since compaction reducesporosity, oxygen availability is a separate factor influencing plant growththat should be measured when assessing soil compaction.

CANADIAN AGRICULTURAL ENGINEERING, VOL. 12, No. 1, MAY 1970

Determination of the Crop Response

Crop response was measured by theoven dry weight of above groundgrowth, and plant population. Thisinformation was obtained from random samples made approximatelyevery two weeks as weather permitted. Vegetative growth provides anearlier evaluation than grain yieldand potentially a more accurate measure of response by considerably reducing the time available for intervening weather effects. Vegetativegrowth rate measurements have beendescribed by Loomis (12).

RESULTS AND DISCUSSION

Pre-Seeding Compaction Treatments

Crop growth was approximately thesame across all slip rates (Figure 1),showing no evidence that slip rate influenced crop growth. Similarly, sliprate did not influence plant population, with the number of plants approximating the theoretical numberof seeds planted (22 to 25 plants perft.2). This could be expected, sincenone of the four soil parameters (bulkdensity, cohesion, penetration resistance and oxygen diffusion rate) coulddetect any differences in soil compactness between the different sliprates. (Plant population and soil parameter data are not shown). If thetractor wheel slippage affected compaction, it seems that the discer destroyed these effects.

The difference observed on the pre-seeding compaction treatments wasearlier emergence, compared to non-compacted locations. As shown inFigure 1, this meant more growth; thepooled mean dry weight of growth onthe pre-seeding treatments, for thesecond growth sampling, exceededthe mean of the non-compacted observations by over 40%. Observations

CROPGROWTH, PRE-SEEDING COMPACTION TREATMENTS

& 5

4-

0 3

1 2

NONCOMPACTED

Seeded June 3

* Sampled June 18

. Sampled July 4 and 5

Figure 1. Two samplings of crop growth onthe noncompacted and" pre-seeding treatment areas.

of local farmers' fields confirmed generally earlier emergence in depressions due to pre-seeding wheel tracks.This occurence seems to be onesource of the observation that somewheel tracks result in beneficial cropresponse.

The reason for earlier emergencewas shallower seed coverage in the

MEAN PENETROMETER CURVES,POST-SEEDING COMPACTION TREATMENTS

3MeanValues at lj - Inch Depth

15 psi 27psi 40 psi74 83 91

50 100

CONE INDEX (psi)

Figure 2. Mean penetrometer resistance before seeding, and on the non-compacted andpost-seeding treatment areas.

OXYGEN DIFFUSION RATE,POST-SEEDING COMPACTION TREATMENTS

Soil Moisture: 34 To43%

TREATMENT PRESSURE (psi)

Figure 3. Mean oxygen diffusion rate onthe noncompacted (0 psi) and post-seedingtreatment areas.

SOIL SHEAR, POST-SEEDINGCOMPACTION TREATMENTS

5T

Eoz

If

0.95

3.52

Soil Moisture: 36 to 43 %

Difference Not Statistically Significant

10 20 30

TREATMENT PRESSURE (psi)

40

Figure 4. Mean soil shear strength on thenoncompacted (0 psi) and post - seedingtreatment areas.

10

path of soil displaced by the discerfrom the pre-seeding wheel track. Toillustrate, seedlings from adjacentareas were compared 10 days afterseeding. The treatment area seedlingstarted one inch below ground, andgrew a total of 414 inches, while thenoncompacted area seedling started2Y2 inches below ground, and wasonly 314 inches long. It seems thatless soil was available in the wheeltrack for coverage due to verticalcompaction and lateral displacementof the soil by the tractor tire.

Post-Seeding Compaction Treatments

The penetrometer results and oxygen diffusion rates showed that compactness increased for each incrementin tire-soil contact pressure for thepost-seeding treatments. At a depthof 1% inches, each 12 psi incrementin contact pressure increased the coneindex by 8 or 9 psi (Figure 2). Thesame increments in contact pressurereduced oxygen diffusion rate by3 x 10_8 to 5 x 10"8 gm/cm2 sec(Figure 3). The first increment inpressure increased cohesion by 0.7psi (figure 4), but there was no evidence of any differences in cohesionbetween the medium and high pressure treatments. Bulk density measurements (Figure 5) were not sensitive enough to detect any differencesin the mean densities for the threecompaction levels. All four soil parameters reflected greater differencesbetween non - compacted and lowpressure (15 psi) treatments than between the low and high (40 psi)pressure treatments.

Increasing soil-tire contact pressurereduced the weight of vegetativegrowth produced. The reduction wasdue to both slower growth rate(Figure 6) and fewer plants (Figure7.) The plant populations on the 27psi and 40 psi treatments were lessthan on the lower pressure treatment(Figure 7) and on the noncompactedarea (shown as 0 psi), and less thanthe assumed seeding rate. The meanpopulation on the 40 psi treatmentwas about 15 plants per ft.2 comparedto nearly 20 on the 15 psi treatment.The final sampling, taken eight weeksafter seeding when the plants hadheaded out shows that growth on thehigh pressure treatment was approximately 15 gm/ft.2 less than on the 15psi treatment and that growth increased by 55% on the 15 psi treatment, but only 33% on the 40 psi

treatment, between July 22 and 28(Figure 6).

With one exception, there was noevidence of different mean growthor plant populations between the lowpressure treatment and the non-compacted location. There was more dryweight on the non-compacted areason the third sampling (July 22,Figure 6).

BULK DENSITY, POST-SEEDING COMPACTION TREATMENTS

I.OOt

^.80|J>-.60t~

ZS-40

2.20

.00

SoiI Moisture: 36 to 43%* Differences Not Stotisticolly Significant

0 10 20 30 40

TREATMENT PRESSURE (psi)

Figure 5. Mean bulk density on the non-compacted (0 psi) and post-seeding treatment areas.

CROP GROWTH, POST-SEEDING COMPACTION TREATMENTS

a,b, etc. Identify Means NotStatistically Significant

z<

C 20

o

Z 10 -

10 20 30 '40

TREATMENT PRESSURE (psi)

Figure 6. Four samplings of crop growthon the noncompacted (0 psi) and post-seeding treatment areas.

PLANT POPULATION,

POST-SEEDING COMPACTION TREATMENTS

• MEANS OF 4 SAMPLINGS

I CONFIDENCE INTERVAL (OC - 5%)ASSUMED NO. SEEDS PLANTED

UPPER AND LOWER LIMITS

r=r=r=.

10 20 30

TREATMENT PRESSURE (psi)

Figure 7. Mean plant population on thenoncompacted (0 psi) and post-seeding treatment areas compared to the assumed number of seeds planted.

CANADIAN AGRICULTURAL ENGINEERING, VOL. 12, No. 1, MAY 1970

When the third growth samplingwas taken, additional observationswere made in the wheel tracks of thetractor towing the post-seeding compaction equipment. This provided anopportunity to compare the effects ofcommon farm tires with the treatment effects. The tractor was equipped with 18.4-34 tires, inflated to 16psi. The data are not shown, butgrowth in the tractor wheel pathswas equivalent to that in the 27 and40 psi treatments.

CONCLUSIONS

1. Pre-seeding compaction, abovethe seed depth, due to rear tractorwheel slip rates up to 30%, apparently was destroyed by the discer.The main result of the wheel trackwas a strip of shallower seed cover,likely due to the wheel compressingand laterally displacing enough soilto reduce the effective depth of thediscer blades in its path. Under theconditions of this test, the reducedseed depth gave early emergence andthe best plant growth; response underdrier conditions would need investigating.

2. Post-seeding traffic increasedsoil compactness as applied pressurewas increased. Contact pressures exceeding 15 psi decreased emergenceand growth rate of wheat on clay soiljust dry enough for field operation.The growth differences remained evident at the headed out stage. Withfurther investigations to establish theeffects at other moisture contents,limiting soil moisture contents andcontact pressures could eventually beestablished that would minimize cropdamage.

3. This study and others point outthe need for more precise information on the ideal soil environment required by various plants. More effortcould be expended in this direction.

4. Of the soil parameter measuringdevices, the penetrometer was themost useful for determining relativesoil compactness. Field data wasquickly and easily obtained. Automatic recording, continuous with depth,was a distinct advantage. Modifications to improve accuracy and establishment of the effects of soil conditions and soil moisture contents areneeded.

Bulk density determination by thesand cone method was neither a convenient nor an accurate field mea

surement. Better field methods of determining bulk density are requiredsince it should be a direct measurement of compactness.

The shear box was not as desirablea method for measuring soil compactness as the penetrometer. It was moredifficult to use and was not as sensitive. Consideration of improved devices to determine shear strength infuture studies should depend onwhether a basic relationship betweenshear strength and plant strengths canbe formulated.

The oxygen diffusion rate apparatus was a good measurementmethod, provided a sufficient numberof observations were obtained. Oxygen diffusion rate measurements weremore time consuming than the othermethods. Improvement of the apparatus to gather data faster and reducevariability in the readings is needed.The oxygen diffusion apparatusshould be included in soil compaction studies since various factors affecting plant growth should be measured separately in any one study.

5. Plant growth and plant population were good measures of crop response. However, the presumptionthat yield correlates with vegetativegrowth was not clearly verified in theliterature.

ACKNOWLEDGEMENTSThe authors wish to acknowledge

the financial support provided by theCanada Department of Agriculturefor this project.

REFERENCES

1. Arndt, W. and Rose, C. W. 1966.Traffic compaction of soil and tillage requirements. J. Agr. Eng.Res. 11(3): 170-187.

2. Barley, K. P. and Greacen, E. L.1967. Mechanical resistance as asoil factor influencing the growthof roots and underground shoots.Advance, in Agron. 19: 1-43.

3. Chancellor, W. J. and Korayem,A. Y. 1965. Mechanical energybalance for a volume element ofsoil during strain. Trans. Amer.Soc. Agr. Eng. 8(3): 426-430.

4. Domier, K. W. 1963. A preliminary look at soil compaction onOsborne clay soil. Proa, SeventhAnn. Man. Soil Sci. Meeting: 1-10.

5. Drew, L. O., Garner, T. H. andDickson, D. G. 1967. Seedlingthrust vs. soil strength. Amer.Soc. Agr. Eng. Paper No. 67-612.

CANADIAN AGRICULTURAL ENGINEERING, VOL. 12, No. 1, MAY 1970

6. Freitag, D. R. 1968. Penetrationtests for soil measurements. Trans.Amer. Soc. Agr. Eng. 11(6):750-753.

7. Gill, W. R. and Vanden Berg, G.E. 1967. Soil dynamics in tillageand traction. Agriculture Handbook No. 316, US Govt. PrintingOffice, Washington.

8. Henry, J. E. and Johnson, W. H.1969. Corn germination efficiencies as affected by soil pressureand temperature. Trans. Amer.Soc. Agr. Eng. 12(1): 141-144,149.

9. Kucera, H. L. and Promersberger,W. J. 1960. Soil compaction — aNorth Dakota problem? N.D.Farm Res. Bui. 21(7): 11-15.

10. Lemon, E. R. and Erickson, A. E.1952. The measurement of oxygen diffusion in the soil with aplatinum microelectrode. Soil Sci.Soc. Amer. Proc. 16: 160-164.

11. Letey, J. and Stolzy, L. H., Part1; Birkle, D. E., Letey, J. Stolzy,L. H. and Szuszkiewicz, T. E.,Part II; Stolzy, L. H. and Letey,J., Part III. 1964. Measurement ofoxygen diffusion rates with theplatinum microelectrode. Calif.Agr. Exp. Sta.: Hilgardia 35(20):545-576.

12. Loomis, W. E. 1953. Growth anddifferentiation in plants. Ames,Iowa: The Iowa State CollegePress.

13. Lyles, L. and Woodruff, N. P.1963. Effects of moisture and soilpackers on consolidation andcloddiness of soil. Trans. Amer.Soc. Agr. Eng. 6:273-275.

14. Rosenberg, N. J. 1964. Responseof plants to physical effects ofsoil compaction. Advance, inAgron. 16: 181-196.

15. Swamy Rao, A. A., Hay, R. C.and Bateman, H. P. 1960. Effectof minimum tillage on physicalproperties of soils and crop response. Trans. Amer. Soc. Agr.Eng. 3 (2): 8-10.

16. Trabbic, G. W., Lask, K. V. andBuchele, W. F. 1959. Measurement of soil-tire interface pressures. Agr. Eng. 40: 678-681.

17. Van Doren, D. M., Jr., Brown, W.N. and Johnson, W. H. 1964. Soilcompaction and crop growth.Ohio Agr. Eng. Sta., June.

18. Vanden Berg, G. E. and Gill, W.R. 1962. Pressure distribution between a smooth tire and the soil.Trans. Amer. Soc. Agr. Eng. 5(2):105-107.

11