SoilUse

and Management
Soil Use and Management, September 2011, 27, 395–403 doi: 10.1111/j.1475-2743.2011.00354.x

Improvements in the visual evaluation of soil structure

R . M . L . G u i m a r ã e s 1 , B . C . B a l l 2 & C . A . T o r m e n a 1
1
Department of Agronomy, Universidate Estadual de Maringá, Av. Colombo, 5790, 87020-900, Maringá, Brazil, and 2SAC Crop
and Soil Systems Research Group, West Mains Road, Edinburgh EH9 3JG, UK

Abstract
Spade methods to visually evaluate soil structural quality (Sq) are simple, quick, cheap and can be used
by farmers, gardeners, consultants and the scientific community. However, European and Brazilian
users of one such method, viz. visual evaluation of soil structure (VESS) which is a development of the
Peerlkamp test, have been concerned about its subjectivity. The method of soil slice break-up and
operator influence on scores have been questioned. Thus, our aim was to make soil scoring by the
VESS technique more objective and thus to revise the scoring guide. We compared scoring with normal
breaking up of the soil slice by hand with scoring after breaking up the slice by dropping (drop shatter)
to make the soil break-up more operator independent. After slice break-up, aggregates were split by
hand and their internal porosity was evaluated to develop the use of visible porosity as an aid to
scoring. This proved inconclusive on its own, so a method of reducing larger aggregates to 1.5–2.0 cm
core fragments and describing their shape and porosity was developed to score soil Sq. Breaking up a
spadeful of soil by hand or by dropping resulted in the same Sq score. The method of reducing
aggregates and evaluation of their shape improved VESS, particularly in the middle range of soil
quality and the revised chart is shown. VESS was sensitive to changes in Sq in layers within the profile
and its use for diagnosing Sq in different layers allows targeted soil improvement by tillage.

Keywords: Soil structure, porosity, aggregates, spade, drop shatter, slice

Introduction
The assessment of qualitative or semi-quantitative indicators
of soil structural quality (Sq) using methods based on the
visual evaluation of soil structure (VESS) has increased.
Amongst these methods the ‘cultural profile’ is widely used
(Gautronneau & Manichon, 1987; Richard et al., 1999),
although it requires specific knowledge and considerable time
and effort. Spade tests require less time and effort than
profile tests. Shepherd (2000) developed the visual soil
assessment which is simple and more objective, but still
requires time and some soil knowledge to complete. To
provide a quick, objective, readily understood and quickly
taught test, visual soil structure quality assessment (VSSQA)
was developed by Ball et al. (2007) based on the Peerlkamp
test (Peerlkamp, 1959). This method consists of extracting a
slice of soil and manually breaking it up along the natural
boundaries between aggregates. Then the size, shape and
Correspondence: R. M. L. Guimarães. E-mail: rachellocks@
gmail.com
Received August 2010; accepted after revision May 2011
colour, roots, pores and aggregates are used to give the soil
slice a score, augmented by comparison with a visual key.
The scores range from 1 (good soil Sq) to 5 (poor). The
method was widely tested in temperate climates giving good
results (Ball et al., 2007). Moreover, Mueller et al. (2009)
showed that this visual method along with others are feasible
tools for providing fast, semi-quantitative assessment of soil
quality and the results correlate well with crop yield. Giarola
et al. (2009, 2010) tested the VSSQA method on sub-tropical
soils and demonstrated its utility for evaluating and
monitoring soil structure and for showing differences between
tillage systems.
However, users of the VSSQA are concerned about the
subjectivity of the method, mainly concerning the difficulties
in breaking up the aggregates along the natural boundaries
by inexperienced operators (Giarola et al., 2009). Also, the
influence of soil water content and texture on the final
results has been questioned as well as the relevance of the
size and position of layers that show poor soil Sq within
the topsoil (Giarola et al., 2010). In the test, assessment of
the morphology and shape of aggregates provides limited
information on the likely functions of porosity inside

ª 2011 The Authors. Journal compilation ª 2011 British Society of Soil Science 395
396 R. M. L. Guimarães et al.

aggregates which is probably more relevant to plant
growth. Soil structure is determined by how individual soil
granules clump together and how aggregates are linked.
With larger aggregates, there are poorer conditions for
plant growth as a result of reduction in intra-aggregate
porosity (Shepherd, 2009). In the test, aggregates refer to
structural units which are both naturally formed and
created during slice break-up. Porosity visible to the naked
eye (visible porosity–VP) inside aggregates can be used as
an additional criterion for assessing structure (Roger-
Estrade et al., 2004). These visible pores are responsible for
soil drainage and air fluxes which are important soil
structural functions for agriculture.
We propose a change in name of VSSQA to VESS which is
shorter and more accurate. Our aim was to make the VESS
method more objective by (1) identifying layers of differing
structures, (2) comparing the technique with the drop shatter
technique of Shepherd (2000) with a view to use soil dropping
for breaking up the slice of soil with less operator influence,
(3) developing the use of VP within aggregates and shape as
additional diagnostics and (4) using this information to
produce a revised version of the VESS guide.
Materials and methods
Sites
Seven sites were used, five in Scotland and two in Brazil. The
site descriptions are summarized in Table 1. The climate at
sites A–E is temperate mesothermal with an annual rainfall of
665 mm at site A and 660 mm at North Berwick which is
within a few kilometres of sites B–E. The mean annual
temperature at sites A–E is 9 C. In January, the mean daily
minimum temperature is 1.5 C. As a result, there are
occasional short duration frosts in the winter months which
help the development of a surface tilth. Sites C, D and E
correspond to those used in a transect for identifying the
influence of soil type on emissions of nitrous oxide and nitric
oxide (Skiba & Ball, 2002). Sites F and G have a sub-
tropical, humid mesothermal climate with an annual rainfall
of 1500 mm and are in an area where annual crops are grown
using no-tillage. The mean annual temperature at sites F and
G is 22 C.
Experiments
Visual evaluation of soil structure
The VESS method consists of sampling an undisturbed slice
of soil from ca. 25 cm depth, 10 cm thick and 20 cm wide
using a spade, and manually breaking it up along the fracture
planes between aggregates whenever possible. Breaking up is
by gently pulling the sample apart to expose and create
structural units (Ball et al., 2007). We improved the method
of breakdown of the structural units and clods by subjectively
assessing aggregate strength. We applied progressive force by
first attempting to squash them between the fingers, then in
one hand and then with both hands and comparing with a
visual key attribute scores to the soil structural layers in the
slice. Size, strength, porosity, roots and colour are the main
criteria used to define soil Sq. Factors which increase the
score are difficulty in slice extraction, the presence of larger,
more angular and less porous aggregates, continuous large
worm holes, clustered, thickened and deflected roots and
areas of grey soil and sulphurous smell. We gave emphasis to
the importance of measuring the depth of the slice after
removal and carefully identifying layers of differing structure
after loosening before further breaking up the slice following
Giarola et al. (2010). Figure 1 shows topsoils with one, two
and three layers of different soil structural qualities. The Sq
ranges from 1 (good) to 5 (poor soil structure). For the
purposes of comparison we calculated the weighted average
score for 10 blocks per site for 0 to 10 cm and 10 to ca.
25 cm and 0 to ca. 25 cm to show the differences in scores
between layers. The overall score of a soil is then determined
by multiplying the score of each layer by its thickness and
dividing the product by the overall depth (Ball et al., 2007).
Improving soil slice break-up
Sites C, D and E were used to assess improved soil slice break-
up methods because they were in three texture classes: C

Table 1 Characteristics of the soils used in this study

Soil Sand Silt Clay

abbreviation Soil series Latitude Longitude Location FAO class (%) (%) (%) Drainage status

A Carpow 5627¢N 300¢W Dundee, UK Dystric-fluvic Cambisol 70 20 10 Free

B Macmerry 5555¢N 310¢W Edinburgh, UK Eutric Cambisol 57 28 15 Imperfect
C Kilmarnock 5604¢N 243¢ W North Berwick, UK Stagnic Cambisol 58.9a 15 26.1 Imperfect
D Cauldside 5559¢N 240¢W East Linton, UK Mollic Fluvisol 21.1 28.2 50.7 Imperfect-poor
E Dreghorn 5605¢N 246¢W North Berwick, UK Eutric Cambisol 75.5 10.6 13.9 Free
F São Jorge 2323¢S 5214¢W Paraná, Brazil Acrisol 73.1 4.5 22.4 Free
G Campos Verdes 2327¢S 5200¢W Paraná, Brazil Ferralsol 4 18 78 Free
a
Textures from sites B and C are from Skiba & Ball (2002).

ª 2011 The Authors. Journal compilation ª 2011 British Society of Soil Science, Soil Use and Management, 27, 395–403

(a) (b)
Figure 1 Soil after visual evaluation of soil structure evaluation showing (a) one layer, (b) two layers and (c) three layers.
medium, D clayey and E sandy texture. Site C was on a farm in
an area under intensive arable cultivation which was in spring
barley stubble at the time of sampling. Site D had a cover crop
and site E was under Brussel sprouts (Brassica oleracea).
Ten sampling locations were chosen, equally spaced along
a diagonal transect of ca. 50 m length for site C, 10 locations
randomly at site D and at the upper and lower boundaries of
the cropped area at site E to avoid crop damage. At each
location, two slices of soil, ca. 25 cm deep and 10 cm thick
were taken. One slice was broken manually in the usual way
whereas the other was broken by dropping one to three times
from a height of 1 m onto a plastic tray (Shepherd, 2000).
Dropping was repeated only for aggregates >10 cm. As well
as dropping the slice directly onto the tray, we tried three
methods of keeping the slice together whilst dropping it.
These were wrapping in cling film, enclosing in a loose-fitting
plastic bag or covering with a loose-fitting lid. We replicated
the first two methods of keeping the slice together three times
and the last one 10 times. After break-up by dropping, the
structure of both slices was evaluated by VESS. We tried to
avoid further break-up of the soil after dropping. The water
content at sampling was 0.28, 0.24 and 0.14 kg ⁄ kg for sites C,
D and E, respectively.
Aggregate porosity as a diagnostic tool
The visual appearance of the porosity on the outer surface of
the aggregates is already used to assess soil quality (Ball
et al., 2007), but there is no quantification of the porosity on
the inner faces of the aggregates in scoring Sq. At sites C–G
at the same time as the ‘improved slice break-up’ experiment
and after the VESS evaluation, we estimated VP in the
aggregates which resulted from the break-up of the individual
layers. For this evaluation we adopted the classification of
Roger-Estrade et al. (2004) and attributed numbers to the
three types of clods based on their morphology. These
aggregates were split along their failure planes and their inner
ª 2011 The Authors. Journal compilation ª 2011 British Society of Soil Science, Soil Use and Management, 27, 395–403
Visual evaluation of soil structure 397
(c)
surfaces were inspected for pores and roots and allocated to
three classes of VP. These classes were VP1 (C clod) = high
eye-VP and loose roots visible between smaller component
aggregates, VP2 (F clod) = aggregates containing cracks as a
result of weathering and some flattened roots and VP3 (D
clod) = no eye-VP or roots (Figure 2). Five aggregates per
layer were evaluated and the average taken for each layer as
identified during estimation of Sq. This classification is
similar to that of Roger-Estrade et al. (2004) and used by
Peigné et al. (2009).
We found some difficulties in applying this technique to the
Brazilian clay soils, mainly because of the dry weather
restricting slice extraction. Because of this we had to hammer
the spade to sufficient depth to take a spadeful of soil which
contributed to some break-up of the aggregates whilst taking
the slice, but did not affect the porosity inside the aggregates.
We also found it difficult to classify VP at sandy site E because
overall macroporosity was low. We considered an aggregate
broken to reveal uneven inner faces with or without visible
pores to be VP2, and an aggregate broken to reveal flat,
uniform pore-free faces to be VP3. Also, where we found many
fine aggregates, these were considered VP1. If VP2 and 3 were
also present, we estimated the proportions of each type by eye.
Aggregate fragmentation and tensile strength
As an extension to the breaking of the aggregates for
examining VP, we fragmented aggregates using a method
similar to that of Guimarães et al. (2009) to obtain aggregates
of relatively uniform size for preparing samples for tensile
strength measurement. After slice break-up, aggregates
>5 cm were fragmented by selecting representatives from
each layer and progressively breaking them into two. This
involved breaking down larger natural aggregates into hand-
made smaller aggregates of ca. 1.5–2.0 cm in diameter.
We investigated whether there was a relationship between
aggregate or fragment shape and porosity and structural
398 R. M. L. Guimarães et al.

VP 1
High eye-visible porosity and loose roots are
visible between smaller component
aggregates

VP 2
Aggregates are usually rounded and contain
cracks and some flattened roots, few
eye-visible pores

VP 3 Figure 2 Appearance to the naked eye and

Aggregates are angular and sharp-edged with description of visible porosity (VP) in the
no eye-visible porosity inner face of aggregates progressively
fragmented by hand.

score. Thus, we sampled 15 slices of soil in Scotland with sites, block extraction was easier although more effort was
high clay content and 15 with high sand content (soils D and needed in the clay soils in dry conditions. At all sites, there
E). Five slices per treatment were sampled. The treatments were at least two layers; the weighted averages for 0–10 and
were a non-tracked area, a tracked area and a native forest. 10–25 cm are given in Figure 3. The scores for soil G (oxisol)
From each layer of different soil quality, 40 aggregates were were better than soil D, the similar heavy clay soil in
sampled and aggregates that represented the average shape of Scotland. Also the scores for the coarser-textured soils (C, E
each Sq were photographed for inclusion in a chart. After and F) were better than for the clay ones.
evaluation of aggregate shape, we applied an indirect test of
tensile strength. The aggregates were passed through 13.5 and
Improving soil slice break-up
19 mm diameter sieves and the fraction between them was
selected. These size ranges were chosen partly because of ease The method of dropping the slice of soil which most revealed
of handling and measurement and partly because these are the structural units was the normal Shepherd drop shatter
the sizes normally produced by soil tillage (Imhoff et al., method without wrapping the slice. Wrapping the slice in
2002). Twenty aggregates were separated from each layer of cling film or in a tight fitting plastic bag sometimes
slices as identified by VESS giving a total of 1100 aggregates compacted the slice by 10% and restricted break-up.
and crushed in individual tests. The indirect tension test was Confining the sample may have prevented break-up and may
carried out using an electronically controlled loading frame have compacted the slice because of the high water content.
with an electronic load cell with a capacity of 10 kN coupled
to a computer for data acquisition and storage. The effective
diameter of each aggregate was calculated following Watts & 0–25 cm
5 4.6
Dexter (1998) and the tensile strength was calculated 0–10 cm
following Dexter & Kroesbergen (1985). 4.0 10–25 cm
3.7
4 3.5 3.4
3.1 3.2 3.0
2.9
Statistical analysis 2.8
2.5
VESS (Sq)

3 2.5 2.6 2.5

2.4
Regression analyses were done using the mean values of the
variables, and linear coefficients were subjected to the t-test 2
using GenStat 11th edition (Payne et al., 2008). The
differences between variables were compared by confidence
1
interval (P < 0.05) (Gabriel, 1987).

0
Results SOIL C SOIL D SOIL E SOIL F SOIL G

We had difficulty in using the VESS methodology in high Figure 3 Weighted average of soil structure quality scores for 0–25,
clay content areas in Brazil (soil G) because of the marked 0–10 and 10–25 cm depth by visual evaluation of soil structure
cohesion, mainly as a result of the dry weather. At the other (VESS) for soils C–G (P < 0.05).

ª 2011 The Authors. Journal compilation ª 2011 British Society of Soil Science, Soil Use and Management, 27, 395–403

Putting the sample in a tray with a loose-fitting lid helped to
contain the sample after dropping, but prevented accurate
identification of the depth of layers with different structure.
Breaking up the soil by hand as recommended gave similar
results for VESS as breaking up the soil by dropping
(Figure 4). Thus, the hand break-up of the slice, if done as
recommended, gives a result comparable to the method of
break-up by dropping. Nevertheless, the user might need
some training in breaking up the aggregates along their
natural failure planes (boundaries). We noticed that soil of Sq
4 and 5 did not usually break-up after dropping although Sq
1–3 did. It was difficult to evaluate a slice of Sq 4–5 after
dropping. This had to be broken apart manually, although
often this was made easy by well-defined crack lines.
Aggregate visible porosity
The appearance of the inner faces of aggregates and the
classes of VP VP1, VP2 and VP3 are shown in Figure 2.
There is a significant positive relationship between Sq and
VP, with R2 between 0.58 and 0.89 (Figure 5). Soils of scores
Sq 2 and 3 contained a wide range of aggregate porosity and
even small aggregates in soils of Sq 2 could be VP = 3
(Figure 5). Small aggregates that have no VP within them are
highly compacted (Roger-Estrade et al., 2004).
Aggregate fragmentation and tensile strength
The relationship between VESS and tensile strength was
positive for both soils with R2 = 0.77 and 0.65 for sites D
and E, respectively (Figure 6). The type of aggregates (core
By hand By dropping
5 productivity. When we were assessing the scores, cereal crop
4.2
3.9
4 3.3
3.3
3.0
VESS (Sq)
3
2 1988). Thus, under favourable growing conditions crop
1 differences in scores within the soil slice particularly within
0
SOIL C SOIL D SOIL E
Figure 4 Comparison of visual evaluation of soil structure (VESS)
after break-up by hand and by drop shatter, with the confidence
interval (P < 0.05). Means with confidence interval bars that
overlap are not statistically different.
ª 2011 The Authors. Journal compilation ª 2011 British Society of Soil Science, Soil Use and Management, 27, 395–403
Visual evaluation of soil structure 399
fragments), but not necessarily the porosity, differed between
Sq scores, and should help with the accurate estimation of
Sq. The breakdown of these core fragments is important
because the ease of rounding or of fragmenting immediately
into a round or into a cubic shape helps the user to score Sq.
The descriptions of the faces of ‘core’ aggregates or
fragments of 1.5–2.0 cm diameter are shown in the last
column of Figure 7.
This column can be used to provide a more accurate
description of the soil structure or to confirm the structural
score when the user doubts their Sq score as may occur for
Sq 3–5. In addition to this new column, we improved the
description of aggregate break-up in the first column and
simplified the information on size and appearance of
aggregates, VP and roots in the second and third columns.
We also added the revised guidelines for slice break-up as
described in the Materials and methods section to the first
page of the chart (data not shown).
Discussion
Change in soil quality with depth and the position of the
poorest structured layer are important diagnostics. The Sq
of the first layer and its thickness are the most important
(Figure 1). For example, Sq 4–5 in the first 10 cm is going
to impose worse conditions on most crops, especially during
early growth stages than Sq 4–5 below 10 cm depth.
Wheaton et al. (2008) highlighted the key soil physical
quality indicator as the depth of non-compacted soil where
soil resistance to penetration is <2 MPa. Giarola et al.
(2010) show that VESS is sensitive enough to reveal changes
in Sq between layers which may influence crop growth. Ball
et al. (2007) established that for layers where Sq > 3, soil
improvements by tillage are needed for sustained crop
growth was satisfactory (>0.8 m height) in soils where the
overall Sq was 3.5 > 3. However, scores differed between
layers. In Figure 3 we present the weighted average score for
3.1 0–10 cm and 10–25 cm depths. Although the overall score
was ‡Sq 3 in soils C, E and G, the 0–10 cm layer had scores
<Sq 3. Most of the roots in temperate, humid conditions
are within this layer (Ball & O’Sullivan, 1987; Tardieu,
growth is unlikely to be seriously restricted by the soil
structure below 10 cm. Hence, it is important to detect
the 0–10 cm layer. This would be helpful in choosing tillage
depth, particularly minimum tillage. Thus, it is important to
report the scores and the order of the individual layers as
well as the overall score.
The results from comparing break-up by hand and by
dropping on VESS show that dropping is a useful option for
compact slices, but depends on soil water content because wet
samples can compact on impact. Use of the drop shatter
400 R. M. L. Guimarães et al.

Top Layer Middle Layer Botton Layer

Soil C Soil D
3 3

Visible porosity (VP)

Visible porosity (VP)

2 2

1 1
y = 0.3815*x + 0.7648* y = 0.452*x + 0.8572*
R2 = 0.59 R2 = 0.89
0 0
0 1 2 3 4 5 0 1 2 3 4 5
VESS (Sq) VESS (Sq)

Soil E Soil F
3 3

Visible porosity (VP)

Visible porosity (VP)

2 2

1 1
y = 0.4282*x + 0.8294* y = 0.4961*x + 0.7374*
R2 = 0.78 R2 = 0.58
0 0
0 1 2 3 4 5 0 1 2 3 4 5
VESS (Sq) VESS (Sq)

Soil G
3
Visible porosity (VP)

1
y = 0.4056*x + 0.6648*
R2 = 0.73
0
0 1 2 3 4 5
VESS (Sq)

Figure 5 Correlation between visual evaluation of soil structure (VESS) structural quality and aggregate eye-visible porosity (VP), soils C–G.
Each graph has a minimum of 24 points and each point is the mean of five aggregates. *Significant at P < 0.001. The average thicknesses of the
layers are 0–5, 5–15 and 15–25 cm for top, middle and bottom layers, respectively.

technique to break-up soil slices may be more subjective than compressive force during the tensile strength test, commonly
it appears at first sight. On impact, the irregular shape of soil used to measure aggregate strength. This compressive force
aggregates causes the area of contact with the hard surface to by one hand is ca. 100 N. Dropping the slice causes soil
vary. With small contact areas, there is higher pressure inside layers to mix, especially in loose or sandy soils when the fine
the aggregate and the resultant shock wave increases material spreads out making it hard to distinguish and
aggregate break-up. The VESS method of applying force with measure the layer from which aggregates originate. The
one hand to break-up aggregates simulates break-up by a dropping technique is also useful for breaking individual

ª 2011 The Authors. Journal compilation ª 2011 British Society of Soil Science, Soil Use and Management, 27, 395–403
 Visual evaluation of soil structure 401

350 Soil E
1200 Soil D
300

Tensile strength (kPa)

Tensile strength (kPa) 1000 y = 194.48**x – 12.353* y = 69.451*x – 64.613*
R2 = 0.77 250 R2 = 0.65
800
200
600
150
400 100
200 50
0 0
0 1 2 3 4 5 0 1 2 3 4 5
VESS (Sq) VESS (Sq)

Figure 6 Relationship between tensile strength and visual evaluation of soil structure (VESS) score. **Significant at P < 0.001; *significant at
P < 0.05.

Structure Size and Visible porosity Appearance after Appearance after Distinguishing Appearance and description of
quality appearance of and Roots break-up: various break-up: same soil feature natural or reduced fragment
aggregates soils different tillage of ~ 1.5 cm diameter

Sq1 Mostly < 6 mm after Highly porous 1 cm The action of breaking

Friable crumbling the block is enough to
Roots throughout reveal them. Large
Aggregates the soil aggregates are
readily composed of smaller
crumble with ones, held by roots.
fingers
Fine aggregates

Sq2 A mixture of porous, Most aggregates 1 cm Aggregates when

Intact rounded aggregates are porous obtained are rounded,
from 2 mm –7 cm. very fragile, crumble very
Aggregates No clods present Roots throughout easily and are highly
easy to break the soil porous.
with one hand
High aggregate
porosity

Sq3 A mixture of porous Macropores and 1 cm Aggregate fragments are

Firm aggregates from cracks present. fairly easy to obtain. They
2 mm –10 cm; less have few visible pores
Most than 30% are <1 cm. Porosity and roots and are rounded. Roots
aggregates Some angular, non- both within usually grow through the
break with one porous aggregates aggregates. aggregates.
hand (clods) may be Low aggregate
present porosity

Sq4 Mostly large > 10 cm Few macropores Aggregate fragments are

1 cm
Compact and sub-angular and cracks easy to obtain when soil
non-porous; is wet, in cube shapes
Requires horizontal/platy also All roots are which are very sharp-
considerable possible; less than clustered in edged and show cracks
effort to break 30% are <7 cm macropores and internally.
aggregates around aggregates Distinct
with one hand macropores

Sq5 Mostly large > 10 Very low porosity. 1 cm Aggregate fragments are
Very compact cm, very few < 7 cm, Macropores may easy to obtain when soil
angular and non- be present. May is wet, although
Difficult to porous contain anaerobic considerable force may
break up zones. be needed. No pores or
Few roots, if any, cracks are visible usually.
and restricted to
Grey-blue colour
cracks

Figure 7 Improved chart. This chart can be downloaded from http://www.sac.ac.uk/vess.

large lumps. Slices of Sq 3 shattered, but slices of Sq 4–5 did The VP results provide evidence for previous compaction
not. Overall, slice break-up by hand is quicker and easier, in all soils as a result of earlier management such as wheeling
and shows the same Sq as dropping. at harvest. A lack of VP suggests persistence of dense

ª 2011 The Authors. Journal compilation ª 2011 British Society of Soil Science, Soil Use and Management, 27, 395–403
402 R. M. L. Guimarães et al.
aggregates even in adequate seedbeds created by tillage. In
such cases, tillage operations appear only to have reduced the
size of the aggregates, but not to have increased their
porosity. In layers containing roots, VP can be between 1 and
2, mainly because the roots can be seen by the naked eye and
can guide the operator deciding on the score. This is
important in sandy soils. A lack of roots makes estimation of
VP more difficult. Soils with an average VP >2 as shown in
Figure 5 contain some aggregates with no porosity which can
restrict plant growth, particularly if they occur in the top
layer.
The VP appeared unhelpful for scoring Sq because of the
wide range of VPs found in each category. It could be useful
to confirm a score of Sq 4–5. In terms of management, if
most aggregates are VP3 in a Sq 3 layer, improvement by
tillage, weathering and biological activity is required (Roger-
Estrade et al., 2004). In such cases, porosity can be improved
through cropping with extensive and aggressive root systems
or by better timed cultivations. Long-term no-tillage can also
increase porosity inside aggregates via organic matter
incorporation (Álvaro-Fuentes et al., 2008) and conservation
of earthworm activities in the soil (Johnson-Maynard et al.,
2007).
The aggregate reduction method improved VESS,
irrespective of texture, made scoring less subjective and
helped with the assessment of Sq. The rounded core
fragments from Sq 1 to 3 have lower tensile strength than the
angular or cubic fragments from Sq 4 and 5 which are more
dense (Figure 6). This supports our choice of core aggregate
shape as used in the new last column of the chart (Figure 7).
Soils with good structure have friable, fine, porous, sub-
angular and sub-rounded aggregates whereas poorly
structured soils have dense, very firm, angular or blocky clods
with high tensile strength (Shepherd, 2009).
The positive correlation between tensile strength and VESS
emphasizes the importance of aggregate strength for VESS to
underline the importance of making subjective evaluations of
the tensile strength of aggregates as part of the test. Tensile
strength was measured on dry aggregates and was high for
the clay soil (Figure 6) because of greater cohesion than in
the lighter textured soil. Although these improvements are
useful in making the method more objective, we stress that
successful application, particularly for scientific purposes
where layer diagnosis is important, is still best with at least
two well-trained operators. They should regularly cross-check
slice evaluations.
Conclusion
Identification of the location and quality of layers of non-
compacted soil is possible with VESS. For example for a
given score, sites with non-compacted soil below a compact
layer may impose more restrictions on the crop than when
the non-compacted layer is near the surface. Breaking up the
ª 2011 The Authors. Journal compilation ª 2011 British Society of Soil Science, Soil Use and Management, 27, 395–403
slice of topsoil by dropping resulted in the same Sq score as
breaking by hand and the latter was easier and quicker.
Evaluation of VP of aggregates was not helpful as an
additional diagnostic to VESS. However, if a considerable
proportion of a layer of Sq 3 is VP3, this indicates the likely
need for improved soil management. Progressive reduction in
aggregates and evaluation of the shape of the created core
fragments improved VESS and made it less subjective.
Acknowledgements
We are grateful to Mr J. Miller, Ferrygate, North Berwick,
Mr J. Grant-Suttie, Sheriffhall, North Berwick, Mr K. Gray,
Preston Mains, East Linton, Mr R. Aitken, Traprain, East
Linton, Agr. Eng. Moacir Ferro, Campos Verdes and Dr R.
Wheatley, James Hutton Institute, Dundee, for access to their
sites. We also thank our colleagues Messrs Colin Crawford,
Davide Tarsitano, Craig Rogers, Edner Betioli Júnior,
Everton Blainski and Wagner Moreira for help in field
assessments, and Paul Hallett, James Hutton Institute, South,
Dundee, for allowing access to his equipment for measuring
tensile strength and for his advice on its use. This work was
sponsored by Coordenação de Aperfeiçoamento de Pessoal
de Nı́vel Superior – CAPES and by the Scottish Government
Environment and Rural Affairs Department, Sustainable
Crop Systems Work Package.
References
Álvaro-Fuentes, J., Arrúe, J.L., Gracia, R. & López, M.V. 2008.
Tillage and cropping intensification effects on soil aggregation:
temporal dynamics and controlling factors under semiarid
conditions. Geoderma, 45, 390–396.
Ball, B.C. & O’Sullivan, M.F. 1987. Cultivation and nitrogen
requirements for drilled and broadcast winter barley on a surface
water gley (Gleysol). Soil & Tillage Research, 9, 103–122.
Ball, B.C., Batey, T. & Munkholm, L.J. 2007. Field assessment of
soil structural quality – a development of the Peerlkamp test. Soil
Use and Management, 23, 329–337.
Dexter, A.R. & Kroesbergen, B. 1985. Methodology for
determination of tensile strength of soil aggregates. Journal of
Agricultural Engineering Research, 31, 139–147.
Gabriel, K.A. 1987. Simple method of multiple comparisons of
means. Journal of the American Statistical Association, 73, 724–
729.
Gautronneau, Y. & Manichon, H. 1987. Guide me´thodique du profil
cultural. CEREF-GEARA, ISARA, Lyon.
Giarola, N.F.B., Tormena, C.A., Silva, A.P. & Ball, B. 2009.
Método de avaliação visual da qualidade da estrutura aplicado a
Latossolo Vermelho Distroférrico sob diferentes sistemas de uso e
manejo. Cieˆncia Rural, 39, 2531–2534.
Giarola, N.F.B., Silva, A.P., Tormena, C.A., Ball, B. & Rosa, J.A.
2010. Visual soil structure quality assessment on oxisols under no-
tillage system. Scientia Agricola, 67, 479–482.
Guimarães, R.M.L., Tormena, C.A., Alves, S.J., Fidalski, J. &
Blainski, E. 2009. Tensile strength, friability and organic carbon in

an oxisol under a crop-livestock system. Scientia Agricola, 66,
499–505.
Imhoff, S., Silva, A.P. & Dexter, A. 2002. Factors contributing to
the tensile strength and friability of oxisols. Soil Science Society of
America Journal, 66, 1656–1661.
Johnson-Maynard, J.L., Umiker, K.J. & Guy, S.O. 2007.
Earthworm dynamics and soil physical properties in the first three
years of no-till management. Soil & Tillage Research, 94, 338–345.
Mueller, L., Kay, B.D., Hu, C., Li, Y., Schindler, U., Behrendt, A.,
Shepherd, T.G. & Ball, B.C. 2009. Visual assessment of soil
structure: evaluation of methodologies on sites in Canada, China
and Germany Part I: comparing visual methods and linking them
with soil physical data and grain yield of cereals. Soil & Tillage
Research, 103, 178–187.
Payne, R.W., Harding, S.A., Murray, D.A., Soutar, D.M., Baird,
D.B., Glaser, A.I., Channing, I.C., Welham, S.J., Gilmour, A.R.,
Thompson, R. 2008. GenStat release 11 reference manual. Part 2.
Directives. VSN International, Hemel Hempstead, UK.
Peerlkamp, P.K. 1959. A visual method of soil structure evaluation.
Meded. V.d. Landbouwhogeschool en Opzoekingsstations van de
Staat te Gent, XXIV, 216–221.
Peigné, J., Cannavaciuolo, M., Gautronneau, Y., Aveline, A.,
Giteau, J.L. & Cluzeau, D. 2009. Earthworm populations under
different tillage systems in organic farming. Soil & Tillage
Research, 104, 207–214.
Richard, G., Boizard, H., Roger-Estrade, J., Boiffin, J. &
Guérif, J. 1999. Field study of soil compaction due to traffic:
ª 2011 The Authors. Journal compilation ª 2011 British Society of Soil Science, Soil Use and Management, 27, 395–403
Visual evaluation of soil structure 403
pore space and morphological analysis. Soil & Tillage Research,
51, 151–160.
Roger-Estrade, J., Richard, G., Caneill, J., Boizard, H., Coquet, Y.,
Defossez, P. & Manichon, H. 2004. Morphological
characterization of soil structure in tilled fields: from a diagnosis
method to the modelling of structural changes over time. Soil &
Tillage Research, 79, 33–49.
Shepherd, T.G. 2000. Visual soil assessment. Volume 1. Field guide
for cropping and pastoral grazing on flat to rolling country.
Horizons.mw & Landcare Research, Palmerston North, New
Zealand.
Shepherd, T.G. 2009. Visual soil assessment. Volume 1. Field guide
for cropping and pastoral grazing on flat to rolling country, 2nd
edn. Horizons Regional Council, Palmerston North, New Zealand.
Skiba, U. & Ball, B. 2002. The effect of soil texture and soil
drainage on emissions of nitric oxide and nitrous oxide. Soil Use
and Management, 18, 56–60.
Tardieu, F. 1988. Analysis of the spatial variability of maize root
density. Plant and Soil, 107, 259–266.
Watts, C.W. & Dexter, A.R. 1998. Soil friability: theory,
measurement and the effects of management and organic carbon
content. European Journal of Soil Science, 49, 73–84.
Wheaton, A.D., McKenzie, B.M. & Tisdall, J.M. 2008. Management
to increase the depth of soft soil improves soil condition and
grapevines performance in an irrigated vineyard. Soil & Tillage
Research, 98, 68–80.