Soil & Tillage Research 127 (2013) 34–44

Contents lists available at SciVerse ScienceDirect

Soil & Tillage Research

journal homepage: www.elsevier.com/locate/still

Using a morphological approach to evaluate the effect of traffic and weather

conditions on the structure of a loamy soil in reduced tillage
Hubert Boizard a,*, Sung Won Yoon b, Joel Leonard c, Sylvain Lheureux a, Isabelle Cousin b,
Jean Roger-Estrade d, Guy Richard b
a
INRA, US1158 Agro-Impact, Estrées-Mons, BP 50136, F-80203 Péronne, France
b
INRA, UR272 Unité de Science du Sol, Ardon, BP 20619, F-45166 Olivet Cedex, France
c
INRA, US1158 Agro-Impact, Pôle du Griffon, 180 rue Pierre-Gilles de Gennes, 02000 Barenton-Bugny, France
d
AgriParisTech, Département SIAFEE, Bâtiment EGER, BP 01, 78850 Thiverval-Grignon, France

A R T I C L E I N F O A B S T R A C T

Article history: A morphological approach and soil porosity have been used to evaluate the effect of compaction and
Received 8 November 2011 climate on the soil structure of a loamy soil in a reduced tillage system. The study was carried out
Received in revised form 25 April 2012 between 2000 and 2006 as part of the long-term ‘‘cropping systems and soil structure’’ experiment
Accepted 27 April 2012
conducted in Estrées-Mons in northern France. Soil hydraulic, mechanical and pore morphological
properties were also measured to characterise the effects on soil structure dynamics. A complementary
Keywords: characterisation of the soil structure was conducted on a microscale.
Soil structure assessment
The method for morphological description of the soil macrostructure was well suited to studying soil
Reduced tillage
Soil compaction
structure dynamics in reduced tillage systems. Results showed that the soil structure in the layer
Platy soil structure without tillage depended in the first instance on compaction intensity. Structural porosity was partly
Soil cracks preserved in the cropping system with little compaction. In contrast just one operation, such as the
Soil structure harvest of sugar beet (Beta vulgaris L.), was sufficient to reduce structural porosity for 5 years even
though no further compaction occurred during this period. Morphological analysis revealed the
evolution of highly compacted zones under the effects of weather conditions. Platy soil structures were
systematically observed in the upper part of the highly compacted zones under the tilled layers, with
cracking slowly penetrating deeper into the soil with time. The structure types observed corresponded to
specific soil properties resulting from the transformation of the soil structure over time. A micro-
morphological assessment was performed to get detailed information about the network of cracks. The
morphological characterisation showed that a visual morphological approach was insufficient for
revealing the entire network of cracks.
ß 2012 Elsevier B.V. All rights reserved.

1. Introduction However, whilst many studies have focused on the origins

The soil structure of the tilled layer of cultivated fields shows
spatio-temporal heterogeneity due to anthropogenic (i.e. tillage
and compaction) as well as natural processes (i.e. climate, root
growth and fauna activity). These processes alter the spatial
arrangement, size and shape of soil aggregates as well as the inter-
or intra-aggregate pore system (Dexter, 1988). Soil compaction is a
major process. It can significantly alter the physical, chemical and
biological properties of soil (Oades, 1993; Hakansson et al., 1995),
with negative environmental impacts such as increasing nitrous
oxide gas emissions (Soane et al., 1995; Ball et al., 1999; Bessou
et al., 2010) or leading to runoff and downslope erosion (Batey,
2009).
* Corresponding author. Tel.: +33 3 22 85 75 12.
E-mail address: boizard@mons.inra.fr (H. Boizard).
behind compaction and its effects (Ball et al., 1997; Smatana et al.,
2010), few have dealt with structure regeneration by natural
agents such as soil biota, roots and climate, although these
processes are considered essential in reduced tillage (Utomo and
Dexter, 1982; Oades, 1993; Taboada et al., 2004). Weather
conditions in particular – through drying–wetting and freezing–
thawing cycles – have a marked influence on the formation of soil
cracks of various kinds and consequently on the physical and
hydraulic properties of the soil (Hussein and Adey, 1998; Rajaram
and Erbach, 1999). The development of horizontal cracks near the
soil surface has often been observed during the transition to
reduced tillage. Alvarez and Steinbach (2009) and Sasal et al.
(2006) observed for example a platy structure in the soil in most
situations in the Argentinean Pampas in no-tillage systems, and
near-surface soil layers in no till for 4 and 6 years also exhibited
this kind of structure in the study by VandenBygaart et al. (1999).
Stengel et al. (1984) studied cracking in swelling clay soils

0167-1987/$ – see front matter ß 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.still.2012.04.007
 H. Boizard et al. / Soil & Tillage Research 127 (2013) 34–44
(11%, 30% and 53% of clay) in field conditions and showed that the
network of cracks is denser when the clay content is high in a
reduced tillage system.
However, beyond punctual characterisations, it remains diffi-
cult to evaluate the dynamics of changes in soil structure over time.
Boizard et al. (2002) used a morphological approach proposed by
Manichon (1987) called ‘‘profil cultural’’. This method allowed to
evaluate the cumulative effects of cropping systems on the
structure of the ploughed layer under different plough tillage
systems. The change of soil structure over time was evaluated by a
field visual assessment indicator referring to the proportion of
compacted clods (called D) in the tilled layer. This indicator
provided a more detailed description of changes in the soil
structure over time than the mean soil bulk density. The main
advantage of this method is that it takes into account the spatial
variations in soil structure caused by tillage, wheeling and weather
conditions. However since its development in plough tillage
systems in 1982, its applicability to describe the evolution of soil
structure accurately in situations with reduced tillage remains to
be seen.
The objective of this paper is to evaluate (i) the effect of
compaction and weather conditions on the soil structure of a
loamy soil in reduced tillage and (ii) the ability of the
morphological approach to describe properly the changes in the
soil structure over time. The study was conducted between 2000
and 2006 as part of the long-term ‘‘cropping systems and soil
structure’’ experiment in loamy soil in northern France. We have
confined this paper to the effect of compaction and weather
conditions, excluding the effect of earthworms, other soil fauna
and roots.
2. Materials and methods
2.1. Site description and experimental design
The site is located in northern France, at Estrées-Mons (508N
latitude, 38E longitude, 85 m elevation). The soil is an Orthic
Luvisol (FAO classification). The 0–30 cm horizon has a silt loam
texture (19% clay, 76% silt, 5% sand and 1.7% organic carbon) and a
pH of 7.6. The gravimetric soil water contents (SWC) measured at
10, 32, 50, 100 and 1500 kPa were 0.253, 0.229, 0.208,
0.175 and 0.084 g g1 respectively. Water content at field capacity,
measured in the field 2–3 days after excess water had drained
away during the winter season (Hillel, 1971), was 0.24 g g1. The
Atterberg plastic and liquid limits were 0.29 and 0.23 g g1
respectively. The coefficient of linear extensibility (COLE, Gray and
Allbrook, 2002) ranged between 0.06 and 0.07, indicating a
moderate to high shrinking potential.
The experiment was conducted as part of the long-term
‘‘cropping systems and soil structure’’ field experiment (Boizard
et al., 2002). Three cropping systems were compared between
1999 and 2008 with a wide range of soil compaction intensities,
depending on crop rotation and decision-making rules. In 1999, a
new treatment with superficial tillage at a depth of only 6 cm was
introduced into the experiment in order to compare the effects of
annual ploughing and reduced tillage on soil structure evolution.
In 2006 we selected three plots with reduced tillage (plot
numbers 8, 9 and 12) to study the effects of compaction and
weather conditions on the evolution of soil structure and physical
properties of the 6–30 cm layer without tillage. The soil textures
were very similar between the plots 8, 9 and 12 (19.9, 19.1 and
18.7% clay respectively). The crops and main characteristics of the
cultural operations of the three plots are described in Table 1.
Seedbed preparation was performed with a light disc harrow.
Maize (Zea mays L.) and sugar beet were sown using a precision
drill. Pea (Pisum sativum L.), flax (Linum Usitatissimum) and wheat
35
(Triticum aestivum) were sown using a combined rotary harrow
and disc drill. The mean depth of stubble tillage and seed bed
preparation was 6 cm, varying between 4 and 8 cm.
We calculated a compaction index (CI) to characterise the
compaction intensity caused by each cultivation operation (Table
1). The index was described in Boizard et al. (2002) and
corresponds to an expected proportion of severely compacted
zones (D) created under wheel tracks in the 0–30 cm layer of the
plot for each operation. The areas of D zones under wheel tracks is
estimated from an empirical relationship (Roger-Estrade et al.,
2000) between the relative area of D zone created during traffic
and the soil water content at the time of traffic; the compaction
index is calculated for each field operation as the ratio of the areas
of D zones created by the two main wheels of each equipment used
relative to the total area of 0–30 cm layer. In plot 8 (named LC for
low compaction), the risk of compaction was low and the
compaction index (CI) was close to zero except in 2000 and
2001. In contrast, the compaction risk was high in plots 9 and 12,
with the late harvesting of sugar beet and maize in the autumn
during wet periods of the year. But the CI differed between plots 9
and 12. The scheduling of cultivation operations led to variable soil
moisture conditions during harvesting. High compaction occurred
in plot 9 (named HC for high compaction) only once in late autumn
2001 during sugar beet harvesting, whilst repeated compaction
events occurred in plot 12 (named RC for repeated compaction).
These differences allowed comparisons to be made between a
situation without severe compaction, a situation with severe
compaction but only at the beginning of the period studied, and a
situation with severe compaction on several occasions during the
period studied.
2.2. Weather conditions
The average atmospheric temperature between 1999 and 2006
was 11.1 8C and the average annual rainfall was 713 mm. Table 2
shows rainfall amounts and mean temperatures by season. The
first years of the experiment were rather humid, with annual
rainfall between 810 mm and 936 mm in 1999–2002, whilst the
2003–2005 period was quite dry (490 mm, 569 mm and 578 mm
in 2003, 2004 and 2005 respectively). So the annual climatic deficit
evaluated as the differences between rainfall and potential
evapotranspiration (ETP) was higher from 2003 to 2005. The total
number of days on which the air temperature was below 0 8C
varied between 10 days during the winter of 2002–2003 and 23
days during the winter of 2003–2004.
2.3. Characterisation of the soil structure
2.3.1. Morphological characterisation of the field soil structure on a
macro-scale
Dry bulk density was measured after each sowing in the
unwheeled zones using a gamma ray transmission probe (10
replicates per plot) at a depth of 0.125 m, 0.175 m and 0.225 m
from the soil surface (probe with two tubes placed 0.30 m apart).
Total porosity estimated from the bulk density was divided into the
textural and structural pore space.
The structural void ratio (eS) was calculated from field bulk
density ra and textural soil density rt, both measured at the same
water content, as:
rs rs
eS ¼  ; (1)
ra rt
with rs being the particle density.
The textural soil density was measured as a function of soil
water content using 2–3 mm aggregates as described by Monnier
et al. (1973). The initially saturated aggregates were gradually
 36
Table 1
Cultural operations from 1999 to 2006.

Plot 8 with a low risk compaction (LC) Plot 9 with a high risk compaction (HC) Plot 12 with a repeated risk compaction (RC)

Operation Crop Date SWCa CIb Crop Date SWC CI Crop Date SWC CI

Sowing W. wheat 12/10/1999 25.0 0.17 Sowing W. wheat 29/10/1999 23.5 0.04 Sowing W. wheat 15/11/1999 21.1 0.00
Harvesting W. wheat 13/08/2000 20.9 0.19 Harvesting W. wheat 13/08/2000 21.5 0.21 Harvesting W. wheat 13/08/2000 20.5 0.16
Stubble discing 25/09/2000 14.6 0.00 Stubble discing 25/09/2000 15.3 0.00 Stubble discing 25/09/2000 15.8 0.00

H. Boizard et al. / Soil & Tillage Research 127 (2013) 34–44

Sowing Pea 23/04/2001 24.3 0.10 Preparation S. beet 02/04/2001 23.5 0.04 Preparation Maize 23/04/2001 23.7 0.05
Harvesting Pea 13/08/2001 16.7 0.02 Sowing S. beet 02/04/2001 23.5 0.03 Sowing Maize 23/04/2001 23.7 0.15
Sowing W. wheat 09/10/2001 22.1 0.00 Harvesting S. beet 15/11/2001 22.8 0.67 Harvesting Maize 15/10/2001 20.9 0.22
Harvesting W. wheat 01/08/2002 13.2 0.00 Sowing W. wheat 15/11/2001 22.8 0.01 Chopping Maize 17/10/2001 20.9 0.10
Preparation Flax 17/03/2003 22.8 0.00 Harvesting W. wheat 14/08/2002 17.2 0.15 Sowing W. wheat 17/10/2001 21.8 0.00
Sowing Flax 17/03/2003 22.8 0.02 Preparation Maize 01/04/2003 20.0 0.00 Harvesting W. wheat 01/08/2002 15.9 0.01
Harvesting Flax 07/08/2003 7.6 0.00 Sowing Maize 01/04/2003 20.0 0.03 Preparation S. beet 13/03/2003 22.8 0.00
Stubble discing 25/09/2003 10.7 0.00 Harvesting Maize 01/10/2003 9.2 0.00 Sowing S. beet 13/03/2003 22.8 0.08
Sowing W. wheat 13/10/2003 12.9 0.00 Chopping Maize 13/10/2003 15.4 0.00 Harvesting S. beet 19/11/2003 19.0 0.20
Harvesting W. wheat 03/08/2004 13.6 0.00 Stubble discing 13/10/2003 15.4 0.00 Sowing W. wheat 19/11/2003 14.9 0.00
Stubble discing 03/09/2004 22.3 0.00 Sowing W. wheat 13/10/2003 15.4 0.00 Harvesting W. wheat 03/08/2004 11.7 0.00
Preparation Pea 21/03/2005 21.9 0.00 Harvesting W. wheat 03/08/2004 13.5 0.00 Stubble discing 03/09/2004 20.5 0.00
Sowing Pea 21/03/2005 21.9 0.00 Stubble discing 03/09/2004 22.1 0.00 Preparation Maize 04/04/2005 21.5 0.00
Harvesting Pea 22/07/2005 17.2 0.03 Preparation S. beet 16/03/2005 23.0 0.03 Sowing Maize 04/04/2005 21.5 0.06
Stubble discing 01/09/2005 19.3 0.00 Sowing S. beet 16/03/2005 23.0 0.05 Harvesting Maize 26/10/2005 22.5 0.31
Sowing W. wheat 11/10/2005 18.9 0.00 Harvesting S. beet 08/11/2005 16.0 0.02 Chopping Maize 27/10/2005 22.5 0.20
Harvesting W. wheat 28/07/2006 11.4 0.00 Preparation 08/11/2005 16.0 0.00 Sowing W. wheat 27/10/2005 22.5 0.06
Sowing W. wheat 08/11/2005 16.0 0.00 Harvesting W. wheat 28/07/2006 10.6 0.00
Harvesting W. wheat 28/07/2006 10.4 0.00
a
SWC, gravimetric soil water content (g g1) of the 0–30 cm layer at the time of compaction.
b
IC, compaction index is the ratio of the volume of the D structure created at each operation to the total volume of the 0–30 cm layer of the plot.
 H. Boizard et al. / Soil & Tillage Research 127 (2013) 34–44 37

Table 2
Summary of climate data from 1999 to 2006.

Years 1999 2000 2001 2002 2003 2004 2005 2006

Avg. temp. (8C)

First quarter 5.5 5.5 5.4 6.2 4.6 12.9 4.5 2.9
Second quarter 13.3 13.4 12.5 12.9 14.0 12.5 13.5 12.9
Third quarter 18.4 16.6 16.8 16.7 18.3 17.6 17.3 19.1
Fourth quarter 6.8 8.1 7.5 8.1 6.7 7.0 7.7 9.1

Annual 11.0 10.9 10.5 11.0 10.9 12.5 10.7 11.0

Precipitation (mm)
First quarter 157 132 282 204 136 132 99 144
Second quarter 187 157 198 118 156 100 152 178
Third quarter 233 219 229 168 90 218 213 231
Fourth quarter 254 302 228 308 108 120 114 144

Annual 830 810 936 797 490 569 578 696

P – ETP from April to October (mm)

222 108 114 185 367 290 239 181

Number of days with

No rain 94 194 214 213 243 221 222 55
Temp. <0 8C 14 12 16 21 10 23 25 4

dried to zero water content and 2–3 g samples were regularly
taken to determine dry mass and volume. The volume of the
aggregates was measured for each water content in kerosene using
Archimedes’ principle. The textural soil density at given water
content was then obtained from the ratio of the dry mass of the
sample to its volume at that water content. The particle density rs
was determined using a pycnometer (Blake, 1965).
A morphological description of the 0–30 cm layer was carried
out every year, after each sowing date, from a randomly located
3 m wide soil profile perpendicular to the tillage and wheeling
directions. Identification of the tilled layer in reduced tillage
treatments LC, HC and RC was possible because the plots were
ploughed until 1999, and 7 years after ploughing had ceased, the
lower limit of the ploughed horizon was still visible. The tilled layer
was approximately 30 cm in all situations. We used the ‘‘profil
cultural’’ method proposed by Manichon (1987) and presented in
detail in Boizard et al. (2002) to map soil structure variations in this
tilled layer. The different structure types were identified by
exerting lateral pressure with a knife. The highly compacted zones
containing specific features (no visible macropores, a massive
structure and a smooth breaking surface) were delimited in slight
relief on the observation face. Four structure types were identified.
The D structure is characterised by a massive structure and no
visible structural porosity. The D structure is related to severe
compaction events. D zones of soil are found under wheel tracks
when traffic in wet conditions has severely compacted the soil,
destroying the structural porosity. D clods are produced when
these zones are fragmented by ploughing or shallow tillage.
Weathering can also alter the D structure: due to wetting–drying
or freezing–thawing cycles, cracks gradually appear in D clods and
zones. When such cracks are observed, the structure is classified as
F in the original method. Since a specific platy soil structure was
often observed at the bottom of the seed bed during our study, we
distinguished this platy soil structure which we called P from the
remaining F structure. In contrast to the D structure, soil zones
showing visible aggregates with a high level of intra- and inter-
aggregate porosity (whose origin is not weathering) are classified
as G.
The proportion of areas with D, P and F structures was
calculated as the ratio of the area of the structure to the total area of
the tilled layer under the seedbed. The variations in these variables
were investigated by analysing nine profiles in a plot in 1994. The
mean was 28%, the standard deviation 6% and the coefficient of
variation 22%. This last value was used as an estimate of the
precision of the measurement of the percentage of D, P and F
areas.
2.3.2. Laboratory characterisation of field soil structure on a micro-
scale
To examine more precisely the effects of climate on the soil pore
system on a micro-scale, an undisturbed soil cylinder (diameter:
18 cm; height: 10 cm) was collected in the first 30 cm under the
seedbed in the three experimental treatments. Sampling was
conducted in the spring of 2005, just after the sowing of each crop,
on the same date as the soil profiles described. The main axis of
these cylinders was horizontal. After air drying, the cylinders were
impregnated with a polyester resin containing fluorescence dye
which is reactive with ultraviolet light, allowing the porous
network and solid phase to be distinguished easily. The
impregnated cylinders were then cut into nine 2 cm-thick slices.
The two sides of each slice were photographed under UV
illumination using a Nikon Coolpix digital camera. The photo-
graphs were taken with the 8–20 mm zoom fully deployed and in
macro mode. A detailed picture (5.5 cm wide and 4 cm high) was
taken from each cylinder to be scanned at 8–12 cm and 18–22 cm
depth. Finally, each image was 2048 by 1536 pixels.
Image analysis was performed using Optimas 6.5 software
(Media Cybernetics, 1999). A thresholding procedure enabled the
pores and soil matrix to be separated.
The area (mm2), perimeter (mm) and index of circularity
(perimeter/area) were determined for each pore. Cracks were
identified as pores with a perimeter greater than 0.72 mm, an
index of circularity greater than 50 and a main axis length greater
than 2 mm. We calculated the surface density of the cracks and
also arranged them into six classes according to their orientation
from the horizontal axis: 0–308, 30–608, 60–908, 120–1508 and
150–1808.
2.4. Mechanical and hydraulic properties
Three soil hydraulic and mechanical properties – infiltration
rate, shear strength and penetration resistance – were measured
for each structure type (G, D, F, P) by taking morphologically
uniform soil volumes in the three plots. The physical properties of
the G structure type were also measured in a plot of ploughed
tillage treatment (plot 17 named MP) to compare the G structure
type in conventional and reduced tillage. Measurements were
performed following the morphological characterisation of the
38
field soil structure in the same pit in the following five situations
(12 replicates for penetration resistance, and 8 replicates for shear
strength and infiltration rate). A visual examination of the samples
was performed carefully to avoid the presence of earthworm
channels, so that only the effect of climate was being analysed:
– G soil structure for LC treatment (LC-G);
– G soil structure for MP treatment (conventional tillage, for
comparison, MP-G);
– P soil structure in the 6–20 cm depth zone of HC treatment (HC-
P);
– D soil structure in the 20–30 cm depth zone of HC treatment
(HC-D); and
– D soil structure in the 20–30 cm depth zone of RC treatment
(RC-D).
The last two situations were chosen because it was assumed
that the age of creation of the D zones (2001 in HC and 2005 in RC)
was likely to alter soil properties. The gravimetric water content
before measurement varied between 0.18 and 0.19 g g1 and the
soil water content at saturation between 0.27 and 0.28 g g1.
The penetration resistance was measured in situ using a
specially designed needle penetrometer 1.96 mm in diameter. The
needle was driven by a motor with a constant vertical velocity of
10.2 mm min1. The applied strength was measured continuously
using an electronic sensor and the ratio of the maximum strength
value at the time of soil failure to the soil contact area defined the
penetration resistance. The gravimetric soil water content was
recorded at each measurement location and varied little between
0.18 and 0.20 g g1 for the different treatments.
Shear strength was measured both at field moisture and under
close to saturation conditions using a shear vane 19 mm in
diameter and 3.5 mm high. The small area of measurement
allowed sampling zones of uniform structure. After insertion into
the soil, a torque was applied and continuously measured using an
electronic sensor and the maximum torque value at the time of soil
failure was used to calculate shear strength knowing the vane
geometry.
A simple indicator of vertical water transfer capacity was
obtained from the measurement of the infiltration rate in a 5 cm
diameter cylinder and under a 2 cm hydraulic head until a
permanent infiltration regime was obtained. The small measure-
ment area allowed sampling zones of uniform structure and
avoided undesired effects, such as the influence of earthworm
0.4
Structural void rao (m3 m-3)
0.3
0.2
0.1
0
2000
Fig. 1. Change in the structural void ratio at a depth of 8–25 cm over time. Structural void ratio was measured after each sowing in the unwheeled zones (10 replicates per
plot). The measured values are plotted with a confidence interval corresponding to one standard deviation.
H. Boizard et al. / Soil & Tillage Research 127 (2013) 34–44
macropores on infiltration. The flux obtained is different and
higher than the soil saturated hydraulic conductivity because of
the three-dimensional nature of the flow (Reynolds and Elrick,
1990). Although saturated conductivity may be derived from such
an experiment, it was not judged necessary here given that initial
soil water content was fairly constant, thus allowing comparisons
to be made. Steady state infiltration rates were log-transformed
before testing for differences between the different situations.
3. Results
3.1. Change in soil porosity over time
ANOVA results give the F-ratio as significant at a probability of
0.01 for plot, year and plot  year. The mean structural void ratio
(eS) at 8–25 cm depth observed during the whole period (2000–
2006) was 0.20, 0.13 and 0.13 respectively for LC, HC and RC.
Though the variation of eS was high in 2000, eS ranged between
0.18 and 0.21 in 2001. In 2002, eS dramatically decreased to 0.09
for HC and RC. This decrease could be caused by high soil water
content during sowing in spring and harvesting in late autumn
2001, leading to severe compaction (Table 1). In both HC and RC,
the void ratio remained low, slightly above 0.10 from 2003 to 2006
(Fig. 1), and the observed variations were not significant. It is
pointed out that the eS remained low in HC since the cultural
operations were performed in dry conditions from 2003 to 2005.
In 2002, the void ratio in LC also decreased, but to a lesser extent
than HC and RC: eS was 0.135 with a great difference between the
depth of 0.125 m (eS = 0.10) and 0.225 m (eS = 0.17). The decrease
could be caused by high soil water content during sowing in 2001
(Table 1) which is likely to create compaction at the soil surface.
From 2003 to 2006, eS was significantly higher in LC than in HC and
RC (Fig. 1). In LC the compaction was low due to the fact that the
crops were always harvested in the summer, during a period of the
year which is generally dry in this area. For comparison, the void
ratio of the control treatment MP, which had the same crop
sequence as LC, was 0.37 due to loosening and fragmentation
induced by yearly plough.
3.2. Changes in soil structure over time: assessment by visual
examination
Fig. 2 shows changes in the proportion of D, F and P areas over
time in the first 30 cm of soil under the seedbed. The proportion of
LC
HC
RC
2001 2002 2003 2004 2005 2006
Year
 H. Boizard et al. / Soil & Tillage Research 127 (2013) 34–44 39

Fig. 2. Change in the proportion of D, F and P areas over time measured after
sowing (m2 m2). The proportion of D, F and P areas were calculated in the first
30 cm of soil under the seedbed. The measured values are plotted with a confidence
interval corresponding to one standard deviation.

D areas was higher in HC and RC than in LC. In LC, pea sowing in

2001 was performed in wet conditions and the compaction index
(CI) was 0.10 with SWC at 24.3%, close to field capacity (Table 1).
Compacted volumes under wheel tracks were visually observed
after pea sowing, mainly at the top of the 0–30 cm layer of soil. In
the following years, a decrease in the proportion of these D areas
was observed. A photograph of LC in 2006 showed that the G
structure type was dominant (Fig. 3a).
In HC the proportion of D areas greatly increased in 2002 due to
high soil water content during sugar beet harvesting in late
autumn 2001 and CI was 0.67. Little compaction occurred between
2003 and 2006 and the proportion of D areas slowly decreased
from 60% to 40%. An inverse trend was observed for the platy
structure, which was systematically observed from 2003 onwards,
with the proportion of P zones reaching a maximum of 23% in 2006.
The P structure types were observed systematically at the top of
the compacted volumes (D structure types) with a dense network
of horizontally oriented large cracks (Fig. 3a and b). The upper part
of the compacted volumes in the soil is gradually replaced in HC by
soil volumes exhibiting a platy structure (Fig. 4).
Fig. 3. (a) Photographs of the cultural profiles in 2006. (b) Photograph of a soil
volume with a platy structure (P-structure type) located at the top of a D zone.
In RC, the initial proportion of D and F areas is high (73%) and
remains between 50% and 70%. The decrease in the proportion of D
areas was counteracted by the compaction events in 2001 during
maize harvesting (CI = 0.16), 2003 during sugar beet harvesting
(CI = 0.20) and 2005 during maize harvesting (CI = 0.31). A platy
soil structure developed between 2003 and 2005 and disappeared
in 2006 (Figs. 3a and 4).
3.3. Morphological characterisation of cracks using image analysis
Fig. 5 contains pictures which are representative of the soil
structure for each plot. In LC, the porosity was lower at 8–15 cm
40 H. Boizard et al. / Soil & Tillage Research 127 (2013) 34–44
Fig. 4. Changes in thickness and location in the profile of the platy soil structure over time.
than at a depth of 20–27 cm, and the pores appeared to be larger
and more connected at depth. In HC and RC, the opposite was
observed: porosity was more developed and more connected at the
top of layer than at the bottom. In HC, large horizontal cracks can
easily be identified in the top layer, whereas vertical cracks with a
lower width can be distinguished at the bottom.
These qualitative observations are confirmed by the results of
image analysis (Fig. 6). The cracks were more numerous in LC,
which has the higher void ratio, especially at the 20–27 cm depth.
The much greater porosity at depth is due to surface compaction
during pea sowing in 2001 which reduced the number of
macropores. No preferential orientation can be observed.
The crack density is significantly higher in HC (1.12 cracks cm2)
than in RC (0.48 fissure cm2) (Fig. 6). Compaction in RC during
harvesting in 2005 probably reduced the number of cracks. In HC
there are numerous, mainly horizontally oriented cracks at the 8–
15 cm depth, which suggests a pronounced effect of weather
conditions only evident after several years without compaction.
Image analysis on thin sections revealed a platy structure containing
both macrocracks visible to the eye and a high density of finer cracks,
not detectable by the field description of the soil structure. At the 20–
27 cm depth, image analysis also revealed that many of the cracks
observed in the lower part of the samples were oriented vertically.
3.4. Consequences on soil mechanical and hydraulic properties
Shear strength at field water content (0.18–0.19 g g1 soil
depending on the situation) differed significantly between the
different types of soil structure (p = 0.001). The shear strengths of
HC-D and RC-D samples were higher than those of P-structure
type, meaning that the formation of the soil platy structure caused
a decrease in shear strength because of the failures that appeared
in the soil matrix. Measurements showed significant differences in
resistance to penetration, depending on structure types (p = 0.001).
The resistance to penetration was very high in HC and RC in the P
and D structures due to severe compaction and was significantly
lower in the G structure (Table 3). An ANOVA performed on the log-
transformed value of the infiltration rate showed differences
between structure types (p = 0.037).
Large and significant (p = 0.037) differences in the steady state
infiltration rate were observed between the different situations
analysed (Table 3). Values obtained in G zones were significantly
higher than in P and D zones, although very variable, which is
consistent with the higher porosity. However no significant
difference in steady state infiltration rate was observed between
the P soil structure and D soil structure despite differences in
porosity. The preferential horizontal orientation of cracks in the
 H. Boizard et al. / Soil & Tillage Research 127 (2013) 34–44 41

Fig. 5. Image analysis of thin vertical sections: image is 5.5 cm wide and 4 cm high. The white zones correspond to the pore space and the black zones to the matrix.

platy structure explains their limited contribution to vertical flow.
The G-structure type was significantly different for the three
properties between conventional and reduced tillage.
4. Discussion
4.1. Soil structure porosity
Soil structure porosity in a reduced tillage system was shown to
depend greatly on compaction intensity. Structural porosity was
partly preserved below shallow tillage depth (6 cm) in the LC
cropping system with little compaction (Fig. 1). This means that
structural porosity can be preserved without tillage even in a
loamy soil if interventions are made when the compaction risk is
low (Fig. 3). In contrast, in HC and RC, severe compaction greatly
reduced porosity and increased the proportion of D zones (Figs. 2
and 3) and the effect of compaction was long lasting: structural
porosity remained low (close to 10%) over the next 4 years in HC,
although there was little additional compaction between 2003 and
2006 (Table 1). This contrasts with results obtained under
ploughing on the same site where compacted zones were shown
to disappear after only 18 months (Boizard et al., 2002). The natural
recovery of porosity is therefore much slower in reduced tillage
systems. Soil must therefore be protected from severe compaction
by using appropriate rules for determining the timing of
interventions or/and controlled traffic.
42 H. Boizard et al. / Soil & Tillage Research 127 (2013) 34–44

D zones remained fairly high. However, the visual assessment

method allowed a detailed analysis of compaction after each
cultivation operation and the evolution of the compacted volumes
over time through the description of D zones and their localisation
(Fig. 3). Moreover the visual assessment method actually allowed
the progressive development of a platy soil structure to be
followed from the top of compacted zones to greater depths, which
suggests that freezing–thawing or wetting–drying cycles may
drive its extension. It also showed that a platy structure like this
can suddenly disappear when a compaction event transforms P
zones back into D zones. Compared with other visual structure
assessment methods such as the visual evaluation of soil structure
(Ball et al., 2007) or the drop-shatter method (Shepherd, 2009), the
‘‘profil cultural’’ method is more time consuming, but allows a
better understanding of the effects of tillage, compaction and
weather conditions on soil structure dynamics over time.
Nevertheless the laboratory characterisation of soil structure at
fine scale reveals complementary information: (i) a platy structure
containing both macrocracks visible to the eye and a high density
of finer cracks <0.5 mm, not detectable by the field description of
the soil structure and (ii) the fact that many of the cracks observed
in the lower part of the samples were oriented vertically. Since this
study focused on the effects of compaction and weather conditions,
we avoided burrows of earthworms (macropores >1 mm). Vertical
pores caused by plants were probably not seen because of the small
size of pores not being visible with this method. Considering that
cracks and the macropore network play an important role in the
functioning of the soil and root access to the subsoil (Capowiez
et al., 2003; McKenzie et al., 2009), a major challenge in being able
to model the long-term structural dynamics in reduced tillage
systems is to characterise cracks and the macropore network more
effectively.

4.3. Platy structure

A platy soil structure has been observed by several authors in

Fig. 6. The number of cracks per cm2 and their directions observed in 2006. Cracks
were identified as pores with a perimeter greater than 0.72 mm, an index of
circularity greater than 50 and a main axis length greater than 2 mm. The angle is
relative to the horizontal.
4.2. Visual assessment
The visual assessment method yielded results consistent with
results from porosity analysis. The proportion of highly compacted
zones was higher in HC and RC than in LC and although there was
little compaction in HC between 2003 and 2006, the proportion of
untilled situations on soils with a low swelling capacity
(VandenBygaart et al., 1999; Pagliai et al., 2003; Sasal et al.,
2006). However, the origin of this type of structure is still under
discussion. Some authors attribute its development to freezing–
thawing and/or shearing produced by the action of tyres during the
compaction process (VandenBygaart et al., 1999). Sasal et al.
(2006) and Alvarez and Steinbach (2009) observed a platy soil
structure in the Argentinean Pampas where frost is rare, which
suggests that freezing–thawing cycles are not a necessary
condition. The results of this study tend to confirm the role of
climate since horizontal cracks were observed developing from the
surface to depth, and seldom deeper than 10 cm, but they also
seem to indicate that compaction is a necessary condition for platy
structure development. In this study, both freezing–thawing and
wetting–drying cycles occurred (Table 2), but further studies are

Table 3
Shear strength, resistance to penetration and infiltration rate measured for each structure type number of replicates was 8 for shear strength and infiltration, and 12 for
penetration.

Structural types Shear strength (kPa) Shear strength (kPa) Resistance to penetration (kPa) Infiltration (mm/h)
at 0.18 SWC at 0.27 SWC at 0.19 SWC at 0.18 SWC
Mean value  sd Mean value  sd Mean value  sd Mean value  sd

MP-G 33  5a 19.5  3 1319  437a 549  177a

LC-G 52  8bc 20  6 2905  573b 284  332b
HC-P 44  7b 14.5  4 4518  432c 86  82c
HC-D 57  11c 24  12 5211  949cd 65  52c
RC-D 55  8c 23  9 5340  879d 58  46c

p value 0.001 0.09 0.001 0.001

The results are mean (n = 5)  standard deviation. Means followed by the same letters are not significantly different according to the Newman–Keuls test at 5%.
 H. Boizard et al. / Soil & Tillage Research 127 (2013) 34–44
needed to analyse the respective effects of each with greater
precision. Similar studies in soils with different climate conditions
and shrinking potential may help understand the drivers behind
platy soil structure development.
4.4. Physical properties of different structure types
Some soil properties of the different structure types were also
measured for comparison purposes only. Results showed that soil
properties evolved simultaneously with soil structure changes.
Shear strength, for example, decreased with platy soil structure
development because horizontal cracking introduced planes of
weakness, whilst vertical penetration resistance remained unaf-
fected. The creation of horizontal cracks did not help vertical water
transfer and thus infiltration rates in the platy soil structure were
not different from those in the D structure. Work done in the past
on the characterisation of the properties of the D-structure type
(Guerif et al., 1994; De Leon, 1991; Coutadeur et al., 2002)
indicated that the D structure has permanent properties, such as
high resistance to penetration whatever the soil type or
geographical situations. Tardieu and Manichon (1987) showed
that root density in maize depended on the spatial variability of the
D structure and that the root depth attained at maturity depended
on the degree of compaction. It has been found that tilled layers
with a high proportion of D-structure types exhibited lower yield
potentials (Leterme et al., 1994). It would be interesting to test if
the same permanent character holds for the properties of the platy
structure, because it would then allow useful information to be
derived about soil functioning from a visual characterisation of the
soil structure in no-tillage or superficial tillage systems.
5. Conclusion
Boizard et al. (2002) showed that the ‘‘profil cultural’’ method is
well suited to taking into account the spatial variations of the soil
structure caused by tillage, wheeling and weather conditions in
plough tillage systems. In this paper, the method was applied in
combination with a porosity analysis in order to study the change
in soil structure in a reduced tillage system. The morphological
approach allowed a detailed analysis of compaction at each
cultivation operation and change in the compacted volumes over
time through the description of D zones and their localisation. The
main results obtained from this study can be summarised as
follows:
 the natural recovery of porosity is much slower in reduced tillage
systems, however structural porosity can be preserved by using
appropriate rules for determining the timing of interventions
and/or controlled traffic;
 the development of a platy soil structure at the soil surface seems
to be closely associated with both the presence of compacted
zones and subsequent weathering by the climate;
 the combination of the morphological approach and the
measurement of physical properties showed that the structure
types observed corresponded to specific soil properties resulting
from the transformation of the soil structure over time. The
evolution towards a more porous soil structure did not always
result in improved soil properties, particularly where the
creation of horizontal porosity dominates; and
 a micro-morphological assessment, performed to obtain detailed
information about the network of cracks, showed that the visual
morphological approach was insufficient for revealing the
network of cracks in its entirety. A major challenge is to produce
a better characterisation of cracks and the macropore network in
reduced tillage for a more effective description of the functioning
of soil and root growth.
43
Acknowledgements
The authors are grateful to P. Regnier, B. Chauchard, F. Mahu
and D. Boitez for their technical assistance and the Domaine INRA
Mons for supporting the long-term experiment. The English text
was edited by Claire Tarring. This work was conducted with
financial support from the Picardy Region, the French National
Research Agency (ANR) as part of the Agriculture and Sustainable
Development Programme, ‘‘ANR-05-PADD-013, DST’’ and the
French Ministry of Environment and Sustainable Development
(MEDD) as part of the Gessol2 programme.
References
Alvarez, R., Steinbach, H.S., 2009. A review of the effects of tillage systems on some
soil physical properties, water content, nitrate availability and crops yield in the
Argentine Pampas. Soil & Tillage Research 104 (1), 1–15.
Ball, B.C., Campbell, D.J., Douglas, J.T., Henshall, J.K., O’Sullivan, M.F., 1997. Soil
structure quality, compaction and land management. European Journal of Soil
Science 48 (4), 593–601.
Ball, B.C., Scott, A., Parker, J.P., 1999. Field N2O, CO2 and CH4 fluxes in relation to tillage,
compaction and soil quality in Scotland. Soil & Tillage Research 53 (1), 29–39.
Ball, B.C., Batey, T., Munkholm, L.J., 2007. Field assessment of soil structure quality—
a development of the Peerlkamp test. Soil Use and Management 23, 329–337.
Batey, T., 2009. Soil compaction and soil management—a review. Soil Use and
Management 25 (4), 335–345.
Bessou, C., Mary, B., Leonard, J., Roussel, M., Gréhan, E., Gabrielle, B., 2010. Modelling
soil compaction impacts on nitrous oxide emissions in arable fields. European
Journal of Soil Science 61 (3), 348–363.
Blake, G.R., 1965. Particle density. In: Black, C.A. (Ed.), Methods of Soil Analysis,
Physical and Mineralogical Properties, Including Statistics of Measurement and
Sampling. American Society of Agronomy, Madison, Pt. 1, pp. 371–373.
Boizard, H., Richard, G., Roger-Estrade, J., Dürr, C., Boiffin, J., 2002. Cumulative
effects of cropping systems on the structure of the tilled layer in northern
France. Soil & Tillage Research 64 (1–2), 149–164.
Capowiez, Y., Pierret, A., Moran, C.J., 2003. Characterisation of the three-dimen-
sional structure of earthworm burrow systems using image analysis and
mathematical morphology. Biology and Fertility of Soils 38, 301–310.
Coutadeur, C., Coquet, Y., Roger-Estrade, J., 2002. Variation of hydraulic conductivity
in a tilled soil. European Journal of Soil Science 53, 1–10.
De Leon, F., 1991. Morphologie, propriétés physiques et conditions de formation des
éléments structuraux de la couche labourée. PhD thesis, INA-PG, Paris.
Dexter, A.R., 1988. Advances in characterization of soil structure. Soil & Tillage
Research 11, 199–238.
Gray, C.W., Allbrook, R., 2002. Relationships between shrinkage indices and soil
properties in some New Zealand soils. Geoderma 108 (3–4), 287–299.
Guerif, J., De Leon Gonzales, F., Stengel, P., 1994. Variability of internal structure of
clods in relation to soil conditions prior compaction, effect on soil strength. In:
Jensen, H.E., Schjonning, P., Mikkelsen, S.A., Madsen, K.B. (Eds.), Proceedings from
the 13th International ISTRO Conference on Soil Tillage for Crop Production and
Protection of the Environment, vol. 1, Aalborg, 24–29 July, pp. 543–548.
Hakansson, I., Ouwerkerk, C., van Soane, B.D., 1995. The effects of soil compaction on
physical, chemical and biological factors in the environment. ISTRO Workshop,
Melitopol, Ukraine, 25 August 1993. Soil & Tillage Research 35 (1–2), 111–113.
Hillel, D., 1971. Soil and water: physical principles and processes. In: Physiological
Ecology: A Series of Monographs, Texts and Treatises, Academic Press, New York
and London, pp. 162–165.
Hussein, J., Adey, M.A., 1998. Changes in microstructure, voids and b-fabric of
surface samples of a Vertisol caused by wet/dry cycles. Geoderma 85 (1), 63–82.
Leterme, P., Manichon, H., Roger-Estrade, J., 1994. Analyse intégrée des rendements
du blé tendre et de leurs causes de variation dans un réseau de parcelles
d’agriculteurs du Thymerais. Agronomie 14, 341–361.
McKenzie, B.M., Bengough, A.G., Hallett, P.D., Thomas, W.T.B., Forster, B., McNicol,
J.W., 2009. Deep rooting and drought screening of cereal crops: a novel field-
based method and its application. Field Crops Research 112, 165–171.
Manichon, H., 1987. Observation morphologique de l’état structural et mise en
évidence d’effets de compactage des horizons travaillés. In: Monnier, G., Goss,
M.J. (Eds.), Soil Compaction and Regeneration. Balkema, Rotterdam, The
Netherlands, pp. 39–52.
Monnier, G., Stengel, P., Fies, J.C., 1973. Une méthode de mesure de densité
apparente de petits agglomerates terreux. Application à l’analyse des systèmes
de porosité du sol. Annales Agronomiques 24, 533–545.
Oades, J.M., 1993. The role of biology in the formation, stabilization and degradation
of soil structure. Geoderma 56 (1-4), 377–400.
Pagliai, M., Marsili, A., Servadio, P., Vignozzi, N., Pellegrini, S., 2003. Changes in some
physical properties of a clay soil in Central Italy following the passage of rubber
tracked and wheeled tractors of medium power. Soil & Tillage Research 73 (1-2),
119–129.
Rajaram, G., Erbach, D.C., 1999. Drying stress effect on mechanical behaviour of a
clay–loam soil. Soil & Tillage Research 49 (1–2), 147–158.
Reynolds, W.D., Elrick, D.E., 1990. Ponded infiltration from a single ring. I: analysis
of steady flow. Soil Science Society of America Journal 54 (5), 1233–1241.
44 H. Boizard et al. / Soil & Tillage Research 127 (2013) 34–44
Roger-Estrade, J., Richard, G., Boizard, H., Boiffin, J., Caneill, J., Manichon, H., 2000.
Modelling structural changes in tilled topsoil over time as a function of cropping
systems. European Journal of Soil Science 51, 455–474.
Sasal, M.C., Andriulo, A.E., Taboada, M.A., 2006. Soil porosity characteristics and
water movement under zero tillage in silty soils in Argentinian Pampas. Soil &
Tillage Research 87 (1), 9–18.
Shepherd, T.G., 2009. Visual Soil Assessment, Field Guide for Pastoral Grazing and
Cropping on Flat to Rolling Country, 2nd ed., vol. 1. Horizons Regional Council,
Palmerston North, New Zealand, ISBN: 978-1-877468-75-9, 119 p.
Smatana, J., Macak, M., Demjanova, E., 2010. The influence of different tillage
practices on soil physical characteristics. Research Journal of Agricultural
Science 42 (3), 315–319.
Soane, B.D., Ouwerkerk, van, C., 1995. Implications of soil compaction in crop
production for the quality of the environment. ISTRO Workshop, Melitopol,
Ukraine, 25 August 1993. Soil & Tillage Research 35 (1–2), 5–22.
Stengel, P., Douglas, J.T., Guerif, J., Goss, M.J., Monnier, G., Cannell, R.Q., 1984. Factors
influencing the variation of some properties of soils in relation to their suit-
ability for direct drilling. Soil & Tillage Research 4, 35–53.
Tardieu, F., Manichon, H., 1987. Etat structural, enracinement et alimentation
hydrique du maı̈s. II: croissance et dispostion spatiale du système racinaire.
Agronomie 7, 201–211.
Taboada, M.A., Barbosa, O.A., Rodriguez, M.B., Cosentino, D.J., 2004. Mechanisms of
aggregation in a silty loam under different simulated management regimes.
Geoderma 123 (3–4), 233–244.
Utomo, W.H., Dexter, A.R., 1982. Changes in soil aggregate water stability induced
by wetting and drying cycles in non-saturated soil. Journal of Soil Science 33 (4),
623–637.
VandenBygaart, A.J., Protz, R., Tomlin, A.D., Miller, J.J., 1999. Tillage system effects on
near-surface soil morphology: observations from the landscape to micro-scale in
silt loam soils of southwestern Ontario. Soil & Tillage Research 51 (1–2), 139–149.