Comparison of the effect of long-term tillage
and crop rotation on physical properties of a soil
C. CHANG and C. W. LINDWALL
Research Station, Agriculture Canada, P. O. Box 3000, Main, Lethbridge, AB, Canada T1J4B1. Contribution 3878831. Received
1 August 1988, accepted 9 May 1989
Chang, C. and Lindwall, C. W. 1990. Comparison of the effect of
long-term tillage and crop rotation on physical properties of a soil.
Can. Agric. Eng. 32: 53-55. The recent development of cost-effective,
non-residual herbicides has resulted in considerable interest in minimum
tillage systems as an alternative to the conventional tillage practice of
frequent cultivation. This study was conducted to compare the longterm effects of conventional tillage and no-till on various soil-water
properties within the tillage depth (0-60 mm) and below the tilled layer
(90-120 mm). Parameters measured include saturated hydraulic con
ductivity (HC), saturation percentage (SAT), plant-available waterholding capacity (PAWHC), large air-filled porosity (LAP), and bulk
density (BD) of the soil. Most of the significantly different effects of
long-term tillage or crop rotation treatments in soil properties were
observed in the 30- to 60-mm depth. After 10 yr of no-till continuous
cropping, there was a trend for this layer to have a lower HC and
PAWHC and higher BD than soil from the conventionally tilled treat
ments. None of the soil properties, however, approached values that
would limit crop production. Tillage treatment effects on HC, BD,
PAWHC and LAP in the surface 30-mm layer and below the tillage
zone (90-120 mm) were not significantly different.
INTRODUCTION
Tillage systems are traditionally designed to provide weed con
trol, incorporate crop residues, fertilizer and chemicals, and
prepare a suitable seedbed. However, increased awareness of
soildegradationresulting from oxidation of organic matter (Dormaar 1983), wind and water erosion, soil compaction and soil
salinity has stimulated interest in developing alternatives to the
traditional summerfallow-crop system employed on the Cana
dian Prairies. The recent development of cost-effective, nonresidual herbicides has spawned considerable interest in no-till
systems as an alternative to traditional intensive cultivation.
Reduced and no-till systems conserve more crop residual
cover, reduce soil erosion losses (Hayes and Kimberlin 1978),
and save time and energy (Frye 1985) without sacrificing yield
losses of cereal grains compared with conventional systems in
southern Alberta (Lindwall 1978; Carefoot and Lindwall 1981)
and elsewhere (Phillips 1969; Anderson 1976; Unger and Wiese
1979). However, some soil physical properties which affect crop
of summerfallow-winter wheat rotation and continuous winter
wheat under conventional tillage and no-till on saturated
hydraulic conductivity, saturation percentage, plant-available
water-holding capacity, air-filled porosity and bulk density of
soil within the tillage depth (0-30 and 30-60 mm) and immedi
ately below the tilled layer (90-120 mm).
MATERIALS AND METHODS
The plots were established in 1976 on a Dark Brown Chernozemic clay loam soil at Lethbridge with crop rotations as main
plots and tillage treatments as subplots in a split-plot design. The
size of the subplots was 2 x 40 m. Two winter wheat rotations
were established, continuous winter wheat (WW) and winter
wheat-fallow (WF). The tillage treatments were conventional
tillage (CT) and no-till (NT). The CT treatment (tilled at a depth
of 90 mm or less) utilized a heavy-duty cultivator with 400-mm
sweeps for the initial seedbed preparation, and a rod weeder and
packer combination for the final seedbed preparation.
The CT-WF rotation normally required 4 to 5 tillage operations
for effective weed control and seedbed preparation; the CT-WW
rotation required two to three operations. In the NT-WF rota
tion, atrazin at 0.7 kg ha-1 was applied afterharvest to provide
residual control of weeds through part of the subsequent sum
merfallow season. For the remainder of the summerfallow season
and prior to seeding, weed control was maintained with para
quat at 0.56-0.84 kg ha-1 plus bromoxynil at 0.46 kg ha"1.
Later in the study, glyphosate at 0.46 kg ha-1 was often sub
stitutedfor paraquat. All herbicides were applied in water at 112-
224 L ha" . In most years, 2,4-D at 0.56 kgha~l was used for
control of winter annual weeds in the crop and where required.
Diclofop at 0.71 kg ha-1 was used to control wild oats.
In the WF rotations, 55 kg ha-1 of mono-ammonium phos
phate (11-48-0) was appliedwith the seed and ammonium nitrate
(34-0-0) at 90 kg ha "was broadcast early in the spring. Simi
larly, in the WW rotation 55 kg ha"1 of 11-48-0 was applied
with the seed, and 150 kg ha"1 of 34-0-0 was broadcast in the
spring. Winter wheat (Triticum aestivum cv. Nor star) was
seeded at a rate of 67 kg ha"1 at a depth of approximately
production, suchas bulk density and porosity, may be favorably
or unfavorablymodified by tillage (Douglas et al. 1979; Ghuman
and Lai 1984; Pagliai et al. 1984; Cassel and Nelson 1985). A
40 mm with a high-clearance, three-rank hoe drill equippedwith
20-mm wide furrow openers on a 200-mm row spacing.
review of the literature indicates that the effects of tillage on the
lected at depths of 0-30, 30-60 and 90-120 mm from each plot
of three replicates in all treatments in 1985, after seeding and
soil physical properties are related to soil type, type of tillage
equipment, tillage depth, soil conditions such as moisture con
tent at the time of tillage, and climatic conditions. If alternative
tillage systems are to be adopted to the extent needed to control
soil erosion effectively and improve soil and water conserva
tion, it is necessary to examine their long-term effects on soil
physical properties and potential implications for crop growth.
The objective of this study was to compare the long-term effects
CANADIAN AGRICULTURAL ENGINEERING
Two undisturbed soil cores (53.5 mm in diameter) were col
after harvest. Soil cores were collected between the grain rows.
Obvious wheel track areas were avoided. The 60- to 90-mm
depth was not sampled because it was assumed that the tillage
(tilled to 90 mm) effects on the soil at this depth interval would
be similar to those on the 30- to 60-mm depth. Soil bulk den
sity (BD), water retention at moisture potentials of 0, -20 and
-1500 kPa by pressure plate (U.S. Salinity Laboratory Staff
53
1954), particle size distributions (Day 1965) and saturated
hydraulic conductivity (HC) (Sommerfeldt et al. 1984) were
determined in the laboratory.
Plant-available water-holding capacity (PAWHC) is defined
as the soil water held between the water potentials of —20and
-1500 kPa. Porosity of large size pores (large air-filled
porosity, LAP) was defined as the percent volume of air-filled
pores at the soil water potential of —20 kPa. For each depth
interval, an analysis of variance was performed to test the
significance of the tillage and crop main effects and the crop
by tillage interaction. Differences between tillage and crop
treatment means were evaluated for significance by Tukey's
test.
Table I. Effects of sampling time on soil physical properties
BD
HC
SAT
PAWHC
(%)
LAP
Time
(cmh"1)
(Mgm-3)
(%)
(%)
Spring
1.87
Fall
1.80
1.09
1.14
13.05*
9.22
31.69
33.74
Spring
0.75
1.35*
54.77*
9.70
23.93
Fall
1.01
1.22
57.45
9.54
27.68
Spring
1.36
1.31
54.83
10.20
22.59
Fall
1.06
1.27
56.87
7.56
26.72
Depth 1 (0-30 mm)
61.20
60.32
Depth 2 (30-60 mm)
Depth 3 (90-120 mm)
* Significantly different at the 0.05 level.
RESULTS
Soil texture, sandy clay loam to clay loam, was uniform
throughout the study area. The sand and clay contents at each
depth interval were not statistically different among the plots.
However, it was observed that the sand content decreased and
the clay content increased with depth. The average sand con
tent ranged from 40.5 to 45.1 %, the clay content ranged from
28.1 to 31.0% and the silt content from 26.2 to 29.5%.
The organic matter content of this soil (0 to 150 mm) was
22.4 g kg"1.
The analysis of variance indicated that the crop-rotation-bytillage interaction was not significant except at the 2nd depth
interval. At this depth interval, the saturated HC and satura
tion percentage (SAT) of the soil for the CT-WF treatment
(1.22 cmh-1 and 59.3%) were significantly greater than those
of the soil from both the NT-WW (0.45 cmh"1 and 54.1%)
and NT-WF (0.89 cm h_1 and 53.4%) treatments. SoilBD for
NT-WW (1.32 Mg m~3) was significantly greater than that of
the soil from the CT-WW (1.28 Mg m-3) and CT-WF
(1.24 Mg m~3) treatments. The porosity (LAP, pore radius
Table II. Effects of crop rotation on soil physical properties
HC
BD
(Mgrn"3)
Treatment
(cmh-1)
wwt
1.77
WF
1.90
1.13
1.12
WW
0.70*
1.30
WF
1.06
1.27
WW
1.20
WF
1.21
21.30
31.28
SAT
PAWHC
(%)
(%)
LAP
(%)
Depth 1 (0-30 mm)
59.47*
62.05
11.43
10.85
31.23
34.20
Depth 2 (30-60 mm)
55.87
56.36
9.46
25.09
9.77
26.52
Depth 3 (90-120 mm)
55.67
56.43
8.72
9.00
24.13
25.18
tWW, continuous winter wheat; WF, 2-yr rotation (winter wheat
fallow)
* Significantly different at the 0.05 level.
Table III. Effects of tillage treatments on soil physical properties
HC
BD
SAT
PAWHC
(%)
(%)
LAP
greater than 7.44 pm) for the soil from CT-WF (29.9%) was
significantly greater than that of soil from NT-WF (23.2%).
Treatment
(cmh-1)
(Mgm"3)
Even though the treatment effects were not statistically signifi
NTt
cant in other depth intervals, similar trends were found for these
properties.
CT
1.64
2.03
1.08*
1.17
The BD values at 30 to 60 mm (depth 2) over all treatments
were significantly higher and the SAT values were significantly
lower in the spring than in the fall. The PAWHC was signifi
cantly greater in the spring than in the fall for the 0- to 30and 60- to 90-mm depths (depths 1 and 3). The other proper
ties (Table I) were not significantly different between the spring
and the fall samples.
The effects of crop rotations on the measured soil physical
properties were mostly nonsignificant (Table II). Only SAT at
depth 1 and HC at depth 2 were significantly greater in WF
than in WW. However, similar trends for those two properties
were found for the other two depth intervals. The tillage effects
on the soil properties measured were more evident than the crop
rotation effects, especially in the tillage zone. The BD was
smaller for the soil with NT than with CT in the 1st depth
interval; however, the result reversed at the 2nd depth interval
(Table III). The HC and SAT of CT were greater than that of
NT
0.67*
1.32*
52.73*
9.42
23.39
CT
1.09
1.26
58.50
9.81
28.23
NT at the 2nd depth interval. The CT treatment, cultivated to
a depth of 90 mm or less, and NT did not affect the soil proper
ties studies differently in the 90- to 120-mm depth.
There was visible evidence that deposition of sediment had
occurred on the west side of the NT plots as a result of erosion
from neighboring fields.
54
(%)
Depth 1 (0-30 mm)
60.91
60.61
10.61
11.66
32.99
32.67
Depth 2 (30-60 mm)
Depth 3 (90-120 mm)
NT
1.03
CT
1.39
1.29
1.30
55.14
56.56
8.77
8.96
23.94
25.37
tNT, no till; CT, cultivated
DISCUSSION
Evidently, the machinery traffic associated with tillage opera
tions did not affect the soil properties studied at the 90- to
120-mm depth. The soil compaction problems reported by
others (Campbell et al. 1974; NeSmith et al. 1987), where the
soil was cultivated under semi-humid or humid climatic condi
tions, were not evident in this study. Furthermore, compaction
curves of Larson et al. (1980) showed that the virgin compres
sion increased with increasing soil moisture content in the poten
tial range of —100 to —5 kPa. In our study, the average soil
moisture content rarely approached —20 kPa at the times of
tillage. If the soil immediately below the tillage depth had been
compacted from tillage, the freezing and thawing cycles during
the winter could have alleviated some of the effects (Krumbach
CHANG AND LINDWALL
and White 1964; Wittsell and Hobbs 1965; Larson and Allmaras
1971).
The actions of frequent freezing and thawing cycles under
the Chinook conditions of southern Alberta (Longley 1967;
Grace 1987) could be the reason that the measured physical
properties of the surface 30 mm were virtually unaffected by
tillage treatment, even under NT conditions. Tillage with the
heavy-duty cultivator should have loosened thesoil, butthefinal
seedbed preparation with the rod-weeder and packer combina
tion would have re-compacted the surface soil layer. As a con
sequence, the bulk density of the surface layer of soil under
CT was greater than that of soil from the NT plots. Also, the
higher porosity of large pores under CT compared with that
of NT could explain the trend for higher saturated HC values
under the CT treatment than under NT.
The effects of tillage on selected soil physical properties such
as HC, BD, SAT, and LAP were more evident at the 30- to
DORMAAR, J. F. 1983. Chemicalpropertiesof soil and water-stable
aggregates aftersixty-seven years of cropping to spring wheat. Plant
Soil 75:51-61.
DOUGLAS, E., E. McKYES, F. TAYLOR, S. NEGI, and G. S. V.
RAGHAVAN. 1979. Unsaturated hydraulic conductivity of a tilled
clay soil. Can. Agric. Eng. 22:153-161.
FRYE, W. W. 1985. Energy requirements in conservation tillage.
Pages 15-24 in Proc. Southern Region No-till Conference, Griffin,
GA 16-17 July.
GRACE, B. 1987. Chinooks. Chinook 9:52-56.
GHUMAN, B. S. and R. LAL. 1984. Water percolation in tropical
Alfisol under conventionalplowing and no-till systems of management.
Soil Till. Res. 4:263-276.
HAYES, W. A. and L. W. KIMBERLIN. 1978. Pages 53-48 in W. R.
Oschwald, ed. Crop residue management systems. Am. Soc. Agron.,
Madison. WI. Spec. Publ. No. 31.
JONES, C. A. 1983. Effect of soil texture on critical bulk densities
for root growth. Soil Sci. Soc. Am. J. 47:1208-1211.
KRUMBACH, A. W., JR. and D. P. WHITE. 1964. Moisture pore
60-mmdepth thanat the0- to 30-mm and90- to 120-mmdepth
space, andbulkdensity changes in frozen soil. SoilSci. Soc. Am. Proc.
intervals. The tillage operation loosens the soil in the 0- to
90-mm depthbut final seedbed preparation and packing do not
28:422-425.
compact the30- to 60-mm layerto theBDlevel of thatobserved
with NT for this layer. Therefore, the soil from the 30- to
60-mm layer from the CT plots had higher HC and SAT and
lower BD and, to a lesser degree, higher LAP and PAWHC.
The higher HC of soil from the 2-yr crop rotation (WF), com
pared with WW, could be attributed to the greater quantities
of decayed roots which created more channels for water
conduction.
In conclusion, most of the significandy differenteffectsof longterm tillagetreatments were observed for the soil from the 30to 60-mm depth. After 10 yr of NT treatments, the HC and
PAWHC were significantly lower and the BD was higher at
the 30- to 60-mm depth than in the CT regime. However, none
of the soil properties approached values that would limit crop
production as indicated by the yield of winter wheat (Zentner
et al. 1988) and other researchers (Jones 1983; Letey 1985).
Soil physical properties in the depth of 0-30 mm and
90-120 mm (below the tillage zone) were not significantly
differentamong tillage and crop rotation treatments in this study.
LARSON, W. E. and R. R. ALLMARAS. 1971. Management factors
and natural forces as related to compaction. Pages 367-427 in K. K.
Barnes et al. ed. Compaction of agricultural soils. Am. Soc. Agric.
Eng. Monogr., St. Joseph, MI.
LARSON, W. E., S. C. GUPTA, and R. A. USECHE. 1980. Com
pressionof agricultural soils from eight soil orders. SoilSci. Soc. Am.
J. 44:450-454.
LETEY, J. 1985. Relationship between soil physical properties and
crop production. Pages 277-294 in B. A. Steward, ed. Advances in
soil science. Vol. 1. Springer-Verlag, New York, NY.
LINDWALL, C. W. 1978. Zero tillage for winter wheat production.
Pages 21-22 in G. C. R. Croome and N. D. Holmes, eds. Research
Highlights-1978. Agriculture CanadaResearch Station, Lethbridge AB.
LONGLEY, R. W. 1967. The frequency of chinooks in Alberta. The
Alberta Geographer 3:20-22.
NESMITH, D. S., D. E. RADCLIFFE, W. L. HARGROVE, R. L.
CLARK, and E. W. TOLLNER. 1987. Soil compaction in double-
cropped wheat and soybeans on an Ultisol. Soil Sci. Soc. Am. J.
51:183-186.
PAGLIAI, M., M. LA MARCA, G. LUCAMANTE, and L.
GENOVSES. 1984. Effects of zero and conventional tillage on the
length and irregularity of elongaged pores in a clay loam soil under
viticulture. Soil Till. Res. 4:433-444.
REFERENCES
PHILLIPS, W. W. 1969. Dryland sorghum production and weed con
trol with minimum tillage. Weed Sci. 17:451-454.
ANDERSON, C. H. 1976. Eco-fallow in the northern great plains of
Canada. Pages 33-45 in Conservationtillage. Proc. Great Plains Work
shop. 10-12 Aug. 1976. Ft. Collins, CO. Great Plains Agric. Council
SOMMERFELDT, T. G., G. B. SCHAAUE, and W. HULSTEIN.
Publ. 77.
Sci. 64: 256-272.
CAMPBELL, R. B., C. C. REICOSKY, and C. W. DOTY. 1974.
UNGER, P. W. and A. F. WIESE. 1979. Managing irrigated winter
wheat residues for water storage and subsequent dryland grain sor
ghum production. Soil Sci. Soc. Am. J. 43:582-88.
Physical propertiesand tillageof Paleudultsin the southeasterncoastal
plains. J. Soil Water Conserv. 29:220-224.
1984. Use of Tempe cell, modified to restrain swelling, for determi
nation of hydraulic conductivity and soil water content. Can. J. Soil
CAREFOOT, J. M. and C. W. LINDWALL. 1981. Winter wheat
U.S. SALINITY LABORATORY STAFF. 1954. Diagnosis and
minimumtillage. Pages 65-66, in L. J. L. Sears, K. K. Krogman and
T. G. Atkinson, eds. Research Highlights-1981. Agric. Can. Res. Sta.
improvement of salinealkali soils. U.S. Dep. Agric, Washington, DC.
Lethbridge, AB.
CASSEL, D. K. and L. A. NELSON. 1985. Spatial and temporal
WITTSELL, L. E. and J. A. HOBBS, 1965. Soil compaction effects
on field plant growth. Agron. J. 57:534-537.
ZENTNER, R.P., C.W. LINDWALL, and J.M. CAREFOOT. 1988.
Economics of rotations and tillage systems for winter wheat produc
variability of soil physical properties of Norfolk loamy sand as affected
by tillage. Soil Till. Res. 5:5-17.
DAY, P. R. 1965. Particle fractionation and particle size analysis.
Handb. 60.
tion in southern Alberta. Can. Farm. Econ. 22:3-13.
Pages 545-567 in C. A. Black, ed. Methods of soil analysis. Part 1.
Agronomy No. 9. Am. Soc. Agron., Madison WI.
CANADIAN AGRICULTURAL ENGINEERING
55