Plant Soil (2009) 324:31–42

DOI 10.1007/s11104-009-9920-5

REGULAR ARTICLE

Linking plant morphological traits to uprooting resistance

in eroded marly lands (Southern Alps, France)
M. Burylo & F. Rey & C. Roumet & E. Buisson &
T. Dutoit

Received: 31 August 2008 / Accepted: 3 February 2009 / Published online: 31 March 2009
# Springer Science + Business Media B.V. 2009

Abstract In marly catchments of the French South- proportion of fine lateral roots and root topology were
ern Alps, soils are subjected to harsh water erosion the best predictors of anchorage strength.
that can result in concentrated flows uprooting small
plants. Evaluating and predicting plant resistance to Keywords Response traits . Uprooting . Erosion .
uprooting from simple plant traits is therefore highly Marls
important so that the most efficient plant strategy for
future restoration of eroded slopes can be defined. Abbreviations
Twelve species growing on marly land were studied. D Stem basal diameter at the root-shoot junction
For each species, in-situ lateral uprooting tests were H Plant height
conducted and morphological plant traits were mea- DBH Diameter at Breast Height
sured on small plants at the early stages of their DWR Total root dry weight
development. The results show that maximum uproot- DWR1 Tap root dry weight
ing force was most positively correlated with stem L Total root length measured using WinR-
basal diameter. Resistance to uprooting depends on a HIZO
combination of several traits. Tap root length, the LR1 Length of the tap root measured using
WinRHIZO
V Total root system volume measured using
Responsible Editor: Alexia Stokes. WinRHIZO
Surf External root surface area calculated with
M. Burylo (*) : F. Rey
WinRHIZO
Cemagref, Unité de recherche Écosystèmes Montagnards,
2 rue de la Papeterie, BP 76, 38402 μ Magnitude: number of external links
St Martin d’Hères cedex, France a Altitude: number of links in the longest single
e-mail: melanie.burylo@cemagref.fr path from an external link to the base link
E(a) Expected altitude calculated from μ on the
C. Roumet
CNRS, Centre d’Écologie Fonctionnelle et Évolutive, assumption of random patterns of root
UMR 5175, 1919 Route de Mende, branching using the algorithm taken from
34293 Montpellier Cedex 5, France Werner and Smart (1973)
E. Buisson : T. Dutoit
Fmax Maximum force reached before plant uprooting
IMEP, UMR-CNRS, IRD, IUT d’Avignon, Agroparc, σ Critical stress calculated as the ratio of Fmax
BP 1207, F-84911 Avignon Cedex 9, France to plant basal cross-sectional area
32 Plant Soil (2009) 324:31–42

Introduction Table 1 List of morphological traits positively correlated with

anchorage strength according to the type of study

Extensive areas of the French Southern Alps are Type of Factors Key references
covered with Black Marl formations. In a Mediterra- material affecting
nean and mountainous climate, with wetting–drying anchorage
cycles, frost in winter and high-intensity rainfalls in Live material
summer, these marly lands are subjected to intense
Temperate DBH, Height, Crook and Ennos
hydric erosion resulting in torrential floods heavily mature Stem mass, Root (1996), Cucchi
loaded with sediment (Oostwoud Wijdenes and trees system depth, et al. (2004),
Ergenzinger 1998; Descroix and Mathys 2003; Root area, Root Mickovski and
bending Ennos (2002),
Mathys et al. 2005). This erosion is responsible for
resistance Nicoll et al.
the sedimentation of river systems and the silting of (2006)
regional reservoirs. Land rehabilitation using vegeta- Temperate Root volume, Karrenberg et al.
tion, through reforestation projects with Austrian juvenile Stem biomass, (2003), Khuder
Black pine dating from the late 19th century (Vallauri trees Tap root length et al. (2007)
et al. 2002), or more recent ecological engineering Tropical DBH Crook and Ennos
methods using bioengineering works (Rey 2005), has trees (1998)
proved effective in reducing erosion and restoring a Herbaceous Stem basal Ennos et al.
dicots diameter, Root (1993a),
certain ecological functioning. In particular, the new biomass, Root Goodman et al.
methods using vegetation barriers allow significant bending (2001), Toukura
quantities of sediment to be retained, by installing resistance et al. (2006)
bioengineering works specifically in gully floors. Grasses Height, Lateral Bailey et al.
These methods are low-cost solutions for ecological root spread, (2002), Crook
Biomass, Lateral and Ennos
rehabilitation, because they guarantee efficient sedi- root number, (1993), Ennos
ment control with as few interventions as possible. Lateral root et al. (1993b),
These solutions use living material (cuttings and volume, Root Mickovski et al.
plants) and favor immediate natural vegetation colo- bending (2005), Stokes
resistance et al. (2007)
nization on the sediment trapped (Rey et al. 2005;
Aquatic macrophytes Root number,
Burylo et al. 2007). Root area
Nevertheless, erosive constraints still limit the Schutten et
sustainability of natural vegetation colonization al. (2005)
(Cohen and Rey 2005). Concentrated runoff loaded Physical model Root system Mickovski et al.
with sediment can uproot plants that have established depth, Root (2007), Stokes
length, Root et al. (1996)
after restoration and that could have otherwise branching, root
stabilized the slopes. Understanding and predicting stiffness
plant resistance to uprooting at the early stages of Numerical model Root system type, Dupuy et al.
their development is therefore a major objective for Root system (2005a, b)
depth, Lateral
both scientists and practitioners. Explaining plant root spread,
resistance in terms of plant traits is essential so that Root number,
taxon-free generalizations of the results can be made Root diameter,
and species compared beyond the local study site. Root branching
Over the past 20 years, many studies have been
conducted on the mechanisms of plant anchorage
(Table 1) contributing a great deal of information on
the factors influencing anchorage strength. To with- properties (Ennos 1990), and type of constraint, i.e.
stand external constraints, plants must transfer the vertical uprooting or lateral loading (Ennos 1993).
force into the soil via their roots. Plant anchorage Several types of anchorage systems (plate, tap,
depends on a combination of root system type and coronal), growth forms (trees, shrubs, herbs), and
root system morphology (Dupuy et al. 2005b), soil type of soil or type of constraint (grazing, wind
Plant Soil (2009) 324:31–42 33

loading, hydrological forces) have been investigated. naturally on eroded lands. Therefore, the aim of the
Experiments on physical and numerical models have present study was to relate uprooting resistance to
made it easier to isolate the effects of a few root plant traits at the early stages of plant development
parameters on the entire plant’s resistance to uproot- using a comparative approach. The following ques-
ing. Among plant traits, plant size (DBH, height, stem tion was addressed: what is the trait, or combination
basal diameter, biomass), root system size (depth and of traits, that can best predict species resistance to
lateral spread), root number, root length, and root lateral uprooting?
branching are the most important traits related to plant To test whether species differ in terms of resistance
anchorage (Table 1). to uprooting, we performed in situ uprooting tests on
It has also been demonstrated that when subjected 12 species growing in the French Southern Alps
to mechanical stress such as wind loading, some representing different growth forms. Ten traits related
species have higher numbers of roots and greater to plant morphology, the root biomass allocation
lateral root branching (Stokes et al. 1997; Mickovski pattern, root system topology, morphology, and
and Ennos 2003), root length (Tamasi et al. 2005), architecture were examined for each of the 12 species
lateral root basal cross-sectional area and root rigidity studied. These traits were chosen because of their
(Goodman and Ennos 1997). These adaptations assumed or demonstrated role in plant anchorage
should improve the anchorage of plants. Finally, it (inspired and adapted from Table 1).
should also be pointed out that other traits such as Analyses were performed on juvenile individuals
root stiffness (Crook and Ennos 1993; Goodman et al. in order to evaluate species survival when they are the
2001; Mickovski et al. 2007), changes in cell wall most vulnerable to the effects of erosion. We chose to
properties, i.e. lignin (Scippa et al. 2006) and apply lateral traction force to mimic the external
cellulose (Genet et al. 2005), and root system forces generated by concentrated flow and sediment
asymmetry (Nicoll and Ray 1996; Mickovski and loading. Plant morphological traits were then exam-
Ennos 2003) can also play a role in plant anchorage. ined and we tested whether species with contrasting
The objectives of most of the studies on plant resistance to uprooting differed in terms of traits.
anchorage carried out over the last 20 years (Table 1)
were to gain information on the anchorage mechanics
of a single species (Mickovski et al. 2005) or to
investigate the influence of one specific factor such as Materials and methods
branching pattern (Stokes et al. 1996), lateral roots, or
root hairs (Bailey et al. 2001) on anchorage efficien- Study site
cy. To date, few studies have been conducted on large
sets of species (Karrenberg et al. 2003; Nicoll et al. The experiment was conducted in the Forêt Doma-
2006; Toukura et al. 2006). Many of the factors niale du Grand Vallon (site 1) and in the Forêt
related to anchorage, such as diameter, biomass, and Domaniale de la Haute Bléone (site 2) in the French
root number, vary with regard to plant size and age. Southern Alps. These two areas, approximately 50 km
Consequently, the results available in the literature are apart, are situated, respectively, near Sisteron and
often valid for one species and do not always enable Digne in the Alpes-de-Haute-Provence, France. At
relevant interspecific comparison. Moreover, a large both sites, the climate is mountainous sub-
number of studies investigated trees, herbaceous Mediterranean with summer droughts interspersed
dicots, and grasses but very few studies have focused with intense storms. The annual total precipitation is
on shrubby species, which are dominant species in 787 mm and 822 mm at site 1 and 2, respectively, and
eroded marly lands and which are often used for the average annual temperature is 10.2°C at both sites
restoration purposes. with four to five cold months (temperature < 7°C;
To define the most efficient strategy for land Rey 2002). At both sites, vegetation has developed on
management in the French Southern Alps, it is partly eroded black marls and the slopes are covered
important to be able to evaluate and predict the with a loose regolith layer made of disintegrated black
resistance to uprooting of young plants and seedlings marl fragments within a silty matrix (Oostwoud
planted for restoration purposes or that develop Wijdenes and Ergenzinger 1998).
34 Plant Soil (2009) 324:31–42

The dominant tree species on the two sites is Pinus architecture and anchorage mechanics (Di Iorio et al.
nigra Arn. ssp. nigra from massive reforestation 2005).
operations for erosion control dating from the We sampled two trees (P. nigra and Q. pubescens),
beginning of the last century (Vallauri et al. 2002). nine shrub species (B. sempervirens, D. pentaphyllum,
The other dominant tree species are Acer opalus Mill., G. cinerea, J. communis, L. angustifolia, O. fruticosa,
Quercus pubescens Wild., and Robinia pseudoacacia O. natrix, S. dubia and T. serpyllum) and one
L. introduced in the 19th century. Ononis fruticosa L., herbaceous dicotyledonous species (A. vulneraria).
Sorbus aria L., and Genista cinerea Vill. mainly Because of variations in local vegetation composition,
compose the shrub layer and Achnatherum calama- all species were not present or very abundant at both
grostis L. the grass layer. sites. Eight species were tested in the Forêt Doma-
niale du Grand Vallon (site 1) while the other species
Species selection were tested in the Forêt Domaniale de la Haute
Bléone (site 2). Moreover, the species selected
Twelve species from the local vegetation were chosen represented different families and successional status.
(Table 2). These species were selected according to Some species, such as Robinia pseudo acacia and
different criteria: i) they are prevalent on marly lands Hippophae rhamnoides, despite their abundance in
of the French Southern Alps and include different local vegetation, could not be included in the
growth forms and families to obtain contrasted selection because they were not available at the
responses to uprooting and to measure a large range juvenile stage at the sites investigated. Uprooting
of plant trait values; ii) they all have tap-like root tests and trait measurements were carried out between
systems, i.e. with an identifiable main vertical root May and June in 2007 and 2008 at the peak of
and smaller lateral roots growing horizontally and vegetative growth. During the experiment, individuals
vertically, to simplify the analysis and species were selected to represent different stem basal
comparison; iii) species already used by practitioners diameters and thus to represent species anchorage
for land management were favored; iv) species are strength and species morphology throughout the
available at the development stage investigated, i.e. range of diameters studied (0–20 mm).
juvenile individuals (stem basal diameter < 20 mm) in
a relatively isolated position to avoid root system Uprooting tests
interaction between plants and modifications of
anchorage strength; and v) species are available on Soil moisture content and soil mechanical properties
flat areas since slope angle influences root system strongly influence anchorage strength. Since uproot-

Table 2 Ecological characteristics and sampling site of the 12 species studied

Species Family Growth form Successional status Sampling site

Pinus nigra Arn. ssp. nigra Pinaceae T Pioneer 1

Quercus pubescens Wild. Fagaceae T Dryad 1
Buxus sempervirens L. Buxaceae S Post-pioneer 2
Dorycnium pentaphyllum Scop. Fabaceae S Post-pioneer 2
Genista cinerea Vill. Fabaceae S Post-pioneer 1
Juniperus communis L. Cupressaaceae S Pioneer 1
Lavandula angustifolia Mill. Lamiaceae S Post-pioneer 2
Ononis fruticosa L. Fabaceae S Pioneer 1
Ononis natrix L. Fabaceae S Post-pioneer 1
Staehelina dubia L. Asteraceae S Pioneer 2
Thymus serpyllum L. Lamiaceae S Post-pioneer 1
Anthyllis vulneraria L. Fabaceae H Post-pioneer 1

T = Tree; S = Shrub; H = Herbaceous

Plant Soil (2009) 324:31–42 35

Table 3 Soil shear strength (kPa) at 5-and 10-cm depths at site (σ in MPa), i.e. the force per unit area necessary to
1 and 2
induce uprooting, was calculated as Fmax divided by
Depth Site 1 Site 2 plant stem basal cross-sectional area (mm2).

5 cm Mean 58.8 47.1 Trait measurement

SE 3.6 3.6
10 cm Mean 121.4 150.8 On the basis of the existing literature on anchorage
SE 9.2 10.9 mechanics and root traits (Table 1), ten morphological
traits were selected (Table 4). For each species, at
least six juvenile individuals were carefully excavated
with hand tools to avoid damaging the roots and were
ing tests were carried out on different sites and preserved at 5°C. The week following harvest, the
different dates, particular precautions were taken to plants were cleaned using a fine stream of water to
be sure tests were valid. To prevent soil moisture remove soil particles. After cleaning, plant height (H)
content differences, the tests were carried out in the and stem basal diameter (D) were measured.
morning, at least 2 days after an intensive rainfall Root samples were separated into tap root (R1) and
event. Soil shear strength at 5-and 10-cm depths was lateral roots and were conserved in ethanol 50% (v:v)
measured with a shear vane (Geonor H-60, Geonor until analysis. The tap roots and laterals were scanned
Inc., Oslo, Norway) to determine the soil’s mechan- and analyzed using WinRHIZO Pro (version 2003b,
ical properties at sites 1 and 2 (15 tests at each site Regent Instrument, Quebec, Canada), following the
and depth). A repeated measures analysis of variance scanning protocol proposed by Bouma et al. (2000).
showed that soil cohesion increased with soil depth The roots were stained with blue methylene (5 g.l−1)
(Effect of depth: F = 151.4, p = 0.000) but that there in order to increase contrast and then scanned at a
were no significant differences in soil shear strength resolution of 400 dpi, a root density less than 0.5 mm
between sites (Effect of site: F = 1.1, p = 0.29; root/mm2 surface and with the automatic threshold
Interaction Depth x Site: F = 9.2, p = 0.005–Table 3). option. WinRHIZO was used to determine root length
Uprooting tests were performed on six to twelve
individuals per species (Table 5) using the method
Table 4 Morphological traits measured on 12 species growing
described in Mickovski et al. (2005). Before each test,
on marly lands
the superficial litter layer was removed to clear the
stem base; plant height and stem basal diameter were Traits Abbreviations Measurement
then measured. A non-stretch rope was bound to the (units)
stem base at one end and to a portable force gauge
Plant slenderness ratio H/D H/D
(Alluris FMI-300) at the other end. A horizontal (cm.cm−1)
traction force was then applied manually until the Root slenderness ratio LR1/D LR1/D
plant was uprooted. Whenever the sample area was not (cm.cm−1)
Relative root volume V/D V/D
completely flat, the force was applied parallel to the
(cm3.cm−1)
downslope direction. The main drawback of this Relative root dry weight DWR/D DWR/D
method is that speed of pulling, which influences (g.cm−1)
anchorage resistance, cannot be controlled, as it can be Relative external root Surf/D Surf/D
surface area (cm.cm−1)
when using a winch. Consequently, the traction force
Percentage of root system %DWR1 (%) DWR1/DWR
was applied slowly, as regularly as possible, to avoid dry weight accounted for
altering the results. Many uprooting tests failed by the tap root
because of rope or stem breakage or the rope slipping Root Tissue Density RTD (g.cm−3) DWR/V
−1
around the stem. During the valid tests, as the plants Specific Root Length SRL (m.g ) L/DWR
began to fail, roots could be heard breaking until Topological index a/E(a) (−) a/E(a)
complete root breakage and root system dislodgement Proportion of root length %FR (%) Fine root length (<
from the sediment. The maximum force reached before with diameter < 0.5 mm 0.5 mm)/total
root length
uprooting (Fmax in N) was noted and the critical stress
36 Plant Soil (2009) 324:31–42

(L), surface (Surf), volume (V), and the percentage of diameter (D) from 1.8 to 14.8 mm, and its maximum
fine root (%FR). After scanning, samples were dried uprooting force (Fmax) from 29 to 430 N (Table 5).
at 60°C for 72 h and weighed. This large variability in Fmax is mainly explained by
Root Tissue Density (RTD), Specific Root Length the variation in D. Indeed, we found a linear positive
(SRL), and the percentage of tap root biomass to total relationship between Fmax and D (Fmax = 25.1D + 12;
root biomass (%DWR1) were calculated. The branch- r = 0.66, p<0.001). Root breakage generally occurred
ing intensity of root systems was analyzed using between 10 and 20 cm below the soil surface and the
Fitter’s terminology (Fitter, 1985). The magnitude (µ) root system section dislodged from the soil consisted
and altitude (a) were determined manually from the of the tap root, main lateral roots, and a few fine roots
scanned images. The expected altitude [E(a)] was without soil particles. However, for L. angustifolia,
calculated and was then used to determine a root ‘root balls’, i.e. root-soil aggregates containing many
topological index [a/E(a)]. Values of a/E(a) < 1 fine roots were observed around the main root
indicate a largely dichotomous root system, while branches.
values > 1 indicate a herringbone-like root system When Fmax was normalized by stem basal cross-
where branching is largely confined to the tap root. sectional area (σ: critical stress), the intraspecific
Values of 1 indicate a random branching structure. variability was lower (Fig. 1).
Finally, we calculated the plant slenderness ratio (H/ The ANCOVA analysis showed that σ differed
D) and the root slenderness ratio (LR1/D). To compare between species (Fig. 1). The highest values were
differences between plant traits, all traits that were found in O. natrix (9.08 MPa), and the lowest in L.
plant size-dependent (DWR, V, Surf) were standard- angustifolia (2.77 MPa). Species were classified into
ized by stem basal diameter (DWR/D, V/D, Surf/D). four resistance groups using Duncan’s multiple range
test:
Statistical analysis
– Group 1: O. natrix, G. cinerea, O. fruticosa and
B. sempervirens
An analysis of covariance (ANCOVA) with D as a
– Group 2: A. vulneraria and Q. pubescens
covariate and an analysis of variance (ANOVA) were
– Group 3: J. communis, T. serpyllum and D.
used to test for differences between species in uprooting
pentaphyllum
resistance and traits (Tukey HSD tests). The assumption
– Group 4: P. nigra, S. dubia and L. angustifolia
of normal distribution was checked before analysis
(Shapiro-Wilk’s test). Duncan’s multiple range test was Species classification according to resistance to
used to classify species into four resistance groups; a uprooting did not highlight a particular influence of
discriminant analysis was then performed to determine growth form or successional status. The four most
which traits best discriminate between the groups. Trait resistant species and the least resistant one, were shrubby
differences between groups were investigated with species. Tree species were found in Groups 2 and 4.
ANOVA and Tukey HSD tests. We used correlation
analysis to investigate relationships between resistance Plant traits and relationship with resistance
to uprooting and plant traits. All the analyses were to uprooting
carried out with STATISTICA (version 7.1 for Win-
dows, Statsoft 1984). All traits differed significantly between species and
between resistance groups (Table 6). For example, the
root slenderness ratio (LR1/D) ranged from 54.2 (S.
Results dubia) to 190.9 (D. pentaphyllum) and the percentage
of root biomass allocated to the tap root (%DWR1)
Uprooting resistance from 0.45 (B. sempervirens) to 0.96 (Q. pubescens).
Group 1, with high critical stress, is characterized by
Plant height, stem basal diameter and maximum high LR1/D and plant slenderness ratio (H/D), a low
uprooting force showed large inter-and intra-specific topological index (a/E(a)), and a low relative root dry
variability. For example, the plant height (H) of Q. weight (DWR/D). Species from group 4, which were
pubescens ranged form 6 to 24 cm, its stem basal less resistant, invested less length and biomass in the
Plant Soil (2009) 324:31–42 37

Table 5 Results of aboveground traits and uprooting tests

Species Plant Height (H, cm) Stem basal diameter (D, mm) Maximum uprooting force Fmax (N) Number of valid tests

Mean Min Max Mean Min Max Mean Min Max

P. nigra 39 23 53 6.5 4.9 8.1 126.4 32 245 11

Q. pubescens 15.9 6 24 4.8 1.8 14.8 108.7 29 430 10
B. sempervirens 21.2 10 32 4.9 3.2 5.8 158.5 66 294 10
D. pentaphyllum 4.8 11 27 6.5 3.9 9.9 164.9 79 387 12
G. cinerea 35.1 17 44 6.7 4.8 12.2 266.6 134 358 10
J. communis 29.8 17 43 5.7 3.7 7.7 158.5 63 266 12
L. angustifolia 29.6 23 45 10.1 4.4 17.5 203.8 74 379 11
O. fruticosa 28.8 19 42 7.1 4.9 8.1 333.3 182 402 6
O. natrix 19.4 11 25 5.1 3.2 8.4 174 65 347 9
S. dubia 21.9 16 30 6.6 4.3 9.9 126.3 37 279 10
T. serpyllum 5 8 24 7.8 2.9 12.4 211.4 71 369 11
A. vulneraria 6.6 5 10 4.2 1.7 7.2 102.7 33 185 10

tap root (low LR1/D and %DWR1) and were charac- (a)=0.56), while the opposite traits characterized the
terized by coarse roots (low Specific Root Length least resistant species from group 4 (e.g. S. dubia with
(SRL) and percentage of fine roots (%FR) and high %FR=54.6, LR1/D=54.2, and a/E(a)=1.47). Species
relative root volume (V/D)). with a low %FR or LR1/D value, or a high a/E(a)
Discriminant analysis showed that among traits, % value were found in the intermediate groups (group 2
FR, LR1/D and a/E(a) best discriminated the groups or 3). Despite having the highest LR1/D value (190.9),
(Table 7). To a lesser extent, %DWR1 also significantly D. pentaphyllum was one of the least resistant species
influenced plant anchorage as well as H/D, Root Tissue (σ = 5.2 MPa) because of a high a/E(a) value (1.62)
Density (RTD), and relative root surface (Surf/D). and a low %FR value (46.4). Similarly, the rank of Q.
Species with high %FR and LR1/D values and low a/E pubescens (group 2) can be explained by its high
(a) values were the most resistant (e.g. B. sempervirens topological index (a/E(a)=1.93). Despite equivalent or
from group 1 with %FR=88, LR1/D=113.2, and a/E even better values than species from group 1, L.

Fig. 1 Uprooting resistance 12

(σ) of the 12 species studied E E
(means ± SE). Letters indi- DE E E
10
cate significant differences DE
Critical stress σ (MPa)

between species (ANCOVA CD

8
D: F = 41.3, p<0.001; spe- BC
cies: F = 6.97, p<0.001– A BC
Duncan’s multiple range 6
AB AB
test, α = 0.05) A
4

0
sempervirens
pentaphyllum

T. serpyllum
P. nigra

A. vulneraria

O. frutisosa

G. cinerea
Q. pubescens
J. communis

O. natrix
S. dubia
L. angustifolia

D.

B.
 38
Table 6 Plant traits of the 12 species studied, individually and when classed into four groups depending on resistance to lateral uprooting, and results of the ANOVA. Species are
classified according to their resistance to uprooting. When letters differ, differences are significant between groups using a Tukey HSD test (p<0.05). Significance levels: ns
nonsignificant, *p<0.05, **p<0.001, ***p<0.0001

Species/ H/D LR1/D V/D DWR/D Surf/D %DWR1 RTD SRL a/E(a) %FR
resistance group (cm/cm) (cm/cm) (cm3/D) (g/D) (cm/D) (g.cm−3) (m.g−1) (<0.5 mm)

ANOVA between species F=11.7*** F=6.8*** F=10.5*** F=11.4*** F=9.6 *** F=12.9*** F=31.7*** F=12*** F=30.1*** F=35.1***
ANOVA between groups F=11.1*** F=4.8** F=3.2* F=3.6* F=2.9* F=4.9*** F=7.8*** F=1.2 ns F=11.2*** F=14.6***
Ononis natrix 51.9 158.1 0.21 0.08 84.1 0.88 0.36 6.37 0.75 88.6
Genista cinerea 65.7 103.1 0.32 0.21 170.9 0.57 0.63 5.61 0.71 73.2
Ononis fruticosa 49.2 100.8 0.43 0.16 147.3 0.7 0.37 6.87 0.69 77.6
Buxus sempervirens 53.1 113.2 0.5 0.23 329.5 0.45 0.46 11.7 0.56 88
Group 1 54.8 A 115.9 A 0.38 AB 0.18 A 189.6 A 0.63 AB 0.46 A 7.76 ABC 0.67 A 81.4 A
Anthyllis vulneraria 31.7 99 0.22 0.07 124.9 0.51 0.29 15.1 0.82 80.9
Quercus pubescens 28.1 100.7 0.33 0.26 82.3 0.96 0.79 1.4 1.93 91.5
Group 2 29.8 B 99.9 AB 0.28 A 0.17 AB 102.2 B 0.74 A 0.56 AB 7.77 ABC 1.41 C 86.5 A
Juniperus communis 65.6 77 0.21 0.13 103.1 0.59 0.6 5.55 0.87 79.6
Thymus serpyllum 20.2 94.9 0.20 0.14 136.3 0.57 0.70 11.2 0.59 88.8
Dorycnium pentaphyllum 34.4 190.9 0.79 0.44 344.9 0.47 0.56 3.83 1.62 46.4
Group 3 43.6 B 110.1 A 0.35 AB 0.21 AB 171.8 AB 0.56 B 0.62 B 6.95 ABC 0.96 B 74.6 A
Pinus nigra 39.2 77 0.56 0.32 176.9 0.57 0.57 2.25 0.77 42.2
Staehelina dubia 34 54.2 0.38 0.21 180.2 0.5 0.57 4.8 1.47 54.6
Lavandula angustifolia 38.7 91.6 0.44 0.24 263.3 0.56 0.55 11.3 0.37 88.6
Group 4 37.5 B 73.9 B 0.47 B 0.27 B 201.2 A 0.55 B 0.56 B 5.46 ABC 0.88 AB 58.5 B
Plant Soil (2009) 324:31–42
Plant Soil (2009) 324:31–42 39

Table 7 Results of the discriminant analysis on ten plant traits mature plants for which root system size, morpholo-
(Wilks lambda = 0.19, p<0.001). Wilks lambda values reflect
gy, and biomechanical properties may show greater
the discriminant power: the lower the Wilks lambda, the higher
the discriminant power differences between trees, shrubs, and grasses.
The correlation analyses showed that anchorage
Trait Wilks lambda p strength cannot be related to one particular plant trait.
%FR 0.75 <0.001
As expected, it depends on a combination of several
traits. In this study, we found that resistance to lateral
LR1/D 0.77 <0.001
uprooting is mainly determined by three traits: the
a/E(a) 0.79 <0.001
root slenderness ratio (LR1/D), the percentage of fine
%DWR1 0.85 0.0013
roots (%FR), and the topological index (a/E(a)). LR1/
Surf/D 0.88 0.006
D is related to anchorage in depth. High LR1/D values
RTD 0.88 0.007
correspond to a greater growth of the tap root in
H/D 0.88 0.008
length than in width and thus result in a long and thin
SRL 0.90 0.018
tap root. The major mechanical role of the tap root in
DWR 0.94 0.11
providing resistance to lateral uprooting by external
V 0.94 0.13
forces has already been highlighted (Ennos 1993;
Ennos 2000). In juvenile individuals, this main rigid
axis acts as a stake in the ground with lateral roots
angustifolia was found to be the least resistant species acting as guy ropes to withstand lateral loading
(σ = 2.77 MPa; LR1/D=91.6; a/E(a)=0.37; %FR=88.6). (Ennos 1994; Fourcaud et al. 2008). The two other
The correlation analysis showed no significant traits, a/E(a) and %FR, are related to horizontal
correlations between critical stress and any one plant anchorage and lateral roots. In particular, a low
trait (Table 8). topological index indicates a highly branched root
system per unit volume of soil with branches on
second-and third-order lateral roots, while high values
Discussion indicate a herringbone-like root system with branches
on the tap root only. Low a/E(a) values and high %FR
In all species, the maximum uprooting force increased values result in more root-soil contact and in a quicker
linearly with stem basal diameter, a surrogate for plant transfer of the external forces into the soil. The results
size: the larger the plant, the higher its anchorage for L. angustifolia suggest that a combination of high
strength. This result is not surprising and confirms, LR1/D and %FR values and a low a/E(a) value does
for small plants and seedlings, the relationship not always result in a strong anchorage. In this case,
between plant size and resistance to uprooting already there were too many roots. Soil failure, rather than
found for other types of plants in previous studies root failure, occurred around the root balls which
(Mickovski et al. 2005; Schutten et al. 2005). were then removed from the soil at very low forces
Moreover, Fmax values, which ranged from 29 to
430 N, are similar to those reported in other studies Table 8 Correlations be-
tween critical stress σ and Trait r p
(Karrenberg et al. 2003; Mickovski et al. 2005).
ten plant traits. The param-
Standardized by the basal cross-sectional area (σ), the eters presented are correla- %FR 0.48 0.11
values differed significantly between species, suggest- tion coefficients (r) and the H/D 0.47 0.12
ing that anchorage strength may be dependent on significance of the relation- DWR/D −0.43 0.16
other plant traits. ship (p)
%DWR1 0.40 0.19
The results did not show any particular influence LR1/D 0.37 0.24
of growth forms or successional status on uprooting V/D −0.33 0.3
resistance of species at the early stages of develop- RTD −0.33 0.3
ment. Indeed, tests were carried out on small plants Surf/D −0.28 0.38
with comparable tap-like root systems whose rooting
SRL 0.1 0.76
depth did not exceed 50 cm and did not reach the
a/E(a) −0.08 0.81
bedrock. This conclusion might be different on
40 Plant Soil (2009) 324:31–42

(Ennos 2000). Consequently, there is an optimal small plants that colonize eroded lands after restoration.
number of roots to efficiently anchor the plant to the These findings can have important applications in land
soil, which could be described by threshold a/E(a) and management of eroded marly lands of the French
%FR values. Southern Alps. Firstly, these results can be used to
Two other traits, %DWR1 and H/D, may have an evaluate, the resistance of species used by practitioners
influence on anchorage strength. %DWR1, like LR1/D, for erosion control or the efficiency of the existing
refers to rooting depth and vertical anchorage. H/D vegetation cover to withstand erosive pressures, based
can be considered the aboveground analog of LR1/D. on plant traits. Secondly, the results can be used for
Insofar as anchorage is most often related to below- prediction purposes. However, they have to be dis-
ground traits, the measurement of which is destruc- cussed with regard to species ecology (Barrouillet 1982;
tive, a routine and immediate application of results by Rameau et al. 1993). Parameters such as germination
practitioners is not easy. Therefore, the efficiency of and growth rates, vegetative multiplication ability, and
this trait (H/D), simple and easy to measure, in invasion risk can balance the conclusions on species
evaluating uprooting resistance should be further efficiency in resisting uprooting. B. sempervirens and
investigated in future studies. G. cinerea were found to be among the most resistant
For juvenile plants growing on marly soils, our species but their slow growth rates do not make them
results suggest that the best root system design to the best species for rapid colonization. On the other
resist lateral loading on marls is a long and thin tap hand, O. fruticosa and O. natrix are pioneer species
root with fine lateral ramifications. This conclusion is that have much higher seed production, germination,
in agreement with Khuder et al. (2007), who and growth rates and can colonize large surface areas
highlighted the role of a deep tap root in young trees in a few years. In addition, O. fruticosa, already used
in preventing toppling, and with previous results on for land rehabilitation, is known to spread widely by
lateral roots (Stokes et al. 1996; Bailey et al. 2001). vegetative multiplication. J. communis and S. dubia are
The present study focused on morphological traits pioneer species and therefore would be suitable for the
inspired from Table 1. Of course, traits related to root colonization of the most degraded soils. Nevertheless,
biomechanical properties, such as root tensile strength S. dubia would be more vulnerable to intense erosion
or root stiffness also influence plant anchorage. than J. communis since it was found to be one of the
Several authors have studied root tensile strength least resistant species. The other shrubby species, T.
and have found significant differences between serpyllum, D. pentaphyllum, and L. angustifolia, all
species (Bischetti et al. 2005; Mattia et al. 2005). post-pioneer woody species, had quite low anchorage
Mickovski et al. (2007) showed that more rigid root strength and present slow growth in the field compared
models had greater resistance to uprooting than more to A. vulneraria, a herbaceous post-pioneer species,
flexible root models. Neither our experiment, nor the which anchored well in the soil. This result suggests
data available in the literature allow us to specify the that self-supporting herbaceous species (forbs and
role of root strength in uprooting resistance, but legumes) could resist erosive constraints and could
further investigations on that area would provide allow faster soil restoration (humus layer, soil structur-
useful information. ing, depth). Among tree species, P. nigra, a pioneer
Given the restrictions of field research, the number species, is very efficient for degraded soil colonization,
of samples per species was sometimes low in our as evidenced by old reforestation operations and the
study, especially for the two Ononis species for which large surface areas it covers. However, its anchorage
only six and nine valid uprooting tests could be strength is low and adults have a high mortality due to
retained. Moreover, our results only concern Mediter- mistletoe infestation (Viscum album L.). Forest regen-
ranean vegetation growing on black marls. Further eration with late succession species such as Q.
investigations are needed to determine whether root pubescens, which is much more resistant to uprooting
traits involved in juvenile plant anchorage are the than P. nigra, would therefore be useful.
same for different soil types and different plant To carry out a global diagnosis, results on species
species. efficiency for soil protection, e.g. soil fixation by
The results of our study provide new elements that roots and sediment trapping, should also be taken into
can help explain and predict uprooting resistance of account. Flume experiments on species efficiency in
Plant Soil (2009) 324:31–42 41

preventing marly topsoil erosion during concentrated pinaster Ait.) growing in different podzolic soil condi-
flow are in preparation and should provide useful tions. Trees (Berl) 18:460–466 doi:10.1007/s00468-004-
0330-2
additional information. Long-term predictions could Descroix L, Mathys N (2003) Processes, spatio-temporal
be made by combining the results of the present study factors and measurements of current erosion in the French
with knowledge of vegetation colonization and Soutern Alps: a review. Earth Surf Process Landf 28:993–
succession on marly lands (Vallauri 1997; Rey et al. 1011 doi:10.1002/esp.514
Di Iorio A, Lasserre B, Scippa GS, Chiatante D (2005) Root
2005; Burylo et al. 2007). This knowledge would system architecture of Quercus pubescens trees growing
improve existing ecological tools so that the evolution on different sloping conditions. Ann Bot (Lond) 95:351–
of vegetation cover after rehabilitation could be 361
modeled. Therefore, plant community efficiency for Dupuy L, Fourcaud T, Stokes A (2005a) A numerical
investigation into factors affecting the anchorage of roots
erosion control could be better predicted when in tension. Eur J Soil Sci 56:319–327 doi:10.1111/j.1365-
diagnosing how vulnerable marly lands are to erosion. 2389.2004.00666.x
Dupuy L, Fourcaud T, Stokes A (2005b) A numerical
investigation into the influence of soil type and root
Acknowledgements We thank Sophie Labonne and Soizig architecture on tree anchorage. Plant Soil 278:119–134
Le Stradic for their help in collecting and analyzing root doi:10.1007/s11104-005-7577-2
samples and Fabrice Grassein for statistical advice and valuable Ennos AR (1990) The anchorage of leek seedlings: the effect of
comments and suggestions. root length and soil strength. Ann Bot (Lond) 65:409–416
Ennos AR (1994) The biomechanics of root anchorage.
Biomimetics 2:129–137
Ennos AR (1993) The scaling of root anchorage. J Theor Biol
References
161:61–75 doi:10.1006/jtbi.1993.1040
Ennos AR (2000) The mechanics of root anchorage. Adv Bot
Bailey PHJ, Currey JD, Fitter AH (2002) The role of root Res 33:133–157 doi:10.1016/S0065-2296(00)33042-7
system architecture and root hairs in promoting anchorage Ennos AR, Crook MJ, Grimshaw C (1993a) A comparative
against uprooting forces in Allium cepa and root mutants study of the anchorage systems of himalayan balsam
of Arabidopsis thaliana. J Exp Bot 53:333–340 Impatiens glandulifera and mature sunfower Helianthus
doi:10.1093/jexbot/53.367.333 annuus. J Exp Bot 44:133–146 doi:10.1093/jxb/44.1.133
Barrouillet J (1982) La revégétalisation dans les Alpes du Sud: Ennos AR, Crook MJ, Grimshaw C (1993b) The anchorage
choix et emploi des espèces herbacées et arbustives. mechanics of maize Zea mays. J Exp Bot 44:147–153
Cemagref, Grenoble, p 143 doi:10.1093/jxb/44.1.147
Bischetti GB, Chiaradia EA, Simonato T, Speziali B, Vitali B, Fitter AH (1985) Functional significance of root morphology
Vullo P, Zocco A (2005) Root strength and root area ratio and root system architecture. In: Fitter AH, Atkinson D,
of forest species in Lombardy (Northern Italy). Plant Soil Read DJ, Usher MB (eds) Ecological interactions in soil.
278:11–22 doi:10.1007/s11104-005-0605-4 Blackwell Scientific, Oxford, UK, pp 87–106
Bouma TJ, Nielsen KL, Koutstaal K (2000) Sample preparation Fourcaud T, Ji JN, Zhang ZQ, Stokes A (2008) Understanding
and scanning protocol for computerised analysis of root the impact of root morphology on overturning mecha-
length and diameter. Plant Soil 218:185–196 doi:10.1023/ nisms: a modelling approach. Ann Bot (Lond) 101:1267–
A:1014905104017 1280 doi:10.1093/aob/mcm245
Burylo M, Rey F, Delcros P (2007) Abiotic and biotic factors Genet M, Stokes A, Mickovski SB et al (2005) The influence
influencing the early stages of vegetation colonization in of cellulose content on tensile strength in tree roots. Plant
restored marly gullies (Southern Alps, France). Ecol Eng Soil 278:1–9 doi:10.1007/s11104-005-8768-6
30:231–239 doi:10.1016/j.ecoleng.2007.01.004 Goodman AM, Ennos AR (1997) The responses of field-grown
Cohen M, Rey F (2005) Dynamiques végétales et érosion sunfower and maize to mechanical support. Ann Bot
hydrique sur les marnes dans les Alpes françaises du Sud. (Lond) 79:703–711 doi:10.1006/anbo.1997.0409
Gémorphologie : relief, processus, environnement 1:31–44 Goodman AM, Crook MJ, Ennos AR (2001) Anchorage
Crook MJ, Ennos AR (1993) The mechanics of root lodging in mechanics of the tap root system of winter-sown oilseed
winter wheat Triticum aestivum L. J Exp Bot 44:1219– rape (Brassica napus L.). Ann Bot (Lond) 87:397–404
1224 doi:10.1093/jxb/44.7.1219 doi:10.1006/anbo.2000.1347
Crook MJ, Ennos AR (1996) The anchorage mechanics of Karrenberg S, Blaser S, Edwards PJ et al (2003) Root
deeprooted larch, Larix europea x L. japonica. J Exp Bot anchorage of saplings and cuttings of woody pioneer
47:1509–1517 doi:10.1093/jxb/47.10.1509 species in a riparian environment. Funct Ecol 17:170–177
Crook MJ, Ennos AR (1998) The Increase in Anchorage with doi:10.1046/j.1365-2435.2003.00709.x
Tree Size of the Tropical Tap Rooted Tree Mallotus wrayi, Khuder H, Stokes A, Danjon F, Gouskou K, Lagane F (2007) Is
King (Euphorbiaceae). Ann Bot (Lond) 82:291–296 it possible to manipulate root anchorage in young trees?
doi:10.1006/anbo.1998.0678 Plant Soil 294:87–102 doi:10.1007/s11104-007-9232-6
Cucchi V, Meredieu C, Stokes A et al (2004) Root anchorage of Mathys N, Klotz S, Esteves M, Descroix L, Lapetite JM (2005)
inner and edge trees in stands of Maritime pine (Pinus Runoff and erosion in the Black marls of the French Alps:
42 Plant Soil (2009) 324:31–42

observations and measurements at the plot scale. Catena bioengineering works in mountainous marly gullies
63:261–281 doi:10.1016/j.catena.2005.06.010 in a Mediterranean climate (Southern Alps, France).
Mattia C, Bischetti GB, Gentile F (2005) Biotechnical character- Plant Soil 278:149–158 doi:10.1007/s11104-005-8422-
istics of root systems of typical Mediterranean species. Plant 3
Soil 278:23–32 doi:10.1007/s11104-005-7930-5 Scippa GS, Di Michele M, Di Iorio A et al (2006) The
Mickovski SB, Ennos AR (2002) A morphological and Response of Spartium junceum roots to slope: anchorage
mechanical study of the root systems of suppressed crown and gene factors. Ann Bot (Lond) 97:857–866
Scots pine Pinus sylvestris. Trees (Berl) 16:274–280 doi:10.1093/aob/mcj603
doi:10.1007/s00468-002-0177-3 Schutten J, Dainty J, Davy AJ (2005) Root anchorage and its
Mickovski SB, Ennos AR (2003) Anchorage and asymmetry in significance for submerged plants in shallow lakes. J Ecol
the root system of Pinus peuce. Silva Fenn 37:161–173 93:556–571 doi:10.1111/j.1365-2745.2005.00980.x
Mickovski SB, van Beek LPH, Salin F (2005) Uprooting Stokes A, Ball J, Fitter AH et al (1996) An experimental
resistance of vetiver grass (Vetiveria zizanioides). Plant investigation of the resistance of model root systems to
Soil 278:33–41 doi:10.1007/s11104-005-2379-0 uprooting. Ann Bot (Lond) 78:415–421 doi:10.1006/
Mickovski SB, Bengough AG, Bransby MF, Davies MCR, anbo.1996.0137
Haallett PD (2007) Material stiffness, branching pattern Stokes A, Nicoll B, Coutts MP, Fitter AH (1997) Responses of
and soil matric potential affect the pullout resistance of young Sitka spruce clones to mechanical perturbation and
model root systems. Eur J Soil Sci 58:1471–1481 nutrition: Effects on biomass allocation, root development,
doi:10.1111/j.1365-2389.2007.00953.x and resistance to bending. Can J Res 27:1049–1057
Nicoll BC, Ray D (1996) Adaptive growth of tree root systems doi:10.1139/cjfr-27-7-1049
in response to wind action and site conditions. Tree Stokes A, Lucas A, Jouneau L (2007) Plant biomechanical
Physiol 16:891–898 strategies in response to frequent disturbance: uprooting of
Nicoll BC, Gardiner BA, Rayner B, Pearce AJ (2006) Phyllostachys nidularia (Poaceae) growing on landslide
Anchorage of coniferous trees in relation to species, soil prone slopes in Sichuan, China. Am J Bot 94:1129–1136
type, and rooting depth. Can J Res 36:1871–1883 doi:10.3732/ajb.94.7.1129
doi:10.1139/X06-072 Tamasi E, Stokes A, Lasserre B et al (2005) Influence of wind
Oostwoud Wijdenes DJ, Ergenzinger P (1998) Erosion and loading on root system development and architecture in
sediment transport on steep marly hillslopes, Draix, Haute- oak (Quercus robur L.) seedlings. Trees (Berl) 19:374–
Provence, France: an experimental field study. Catena 384 doi:10.1007/s00468-004-0396-x
33:179–200 doi:10.1016/S0341-8162(98)00076-9 Toukura Y, Devee E, Hongo A (2006) Uprooting and shearing
Rameau J C, Mansion D and Dumé G (1993) Flore forestière resistances in the seedlings of four weedy species. Weed Biol
française, Tome 2 Montagnes. Institut pour le développe- Manage 6:35–43 doi:10.1111/j.1445-6664.2006.00192.x
ment forestier Vallauri D (1997) Dynamique de la restauration forestière des
Rey F (2002) Influence de la distribution spatiale de la substrats marneux avec Pinus nigra J.F. Arnold ssp. nigra
végétation sur la production sédimentaire de ravines dans le secteur haut-provençal. PhD thesis, Cemagref,
marneuses dans les Alpes du Sud. PhD thesis, Cemagref, Grenoble, France
Grenoble, France Vallauri DR, Aronson J, Barbero M (2002) An analysis of
Rey F (2005) Efficacité des ouvrages de génie biologique pour forest restoration 120 years after reforestation on badlands
le piégeage des sédiments dans des ravines incisées dans in the Southwestern Alps. Restor Ecol 10:16–26
des marnes (Alpes du Sud, France). Géomorphologie: doi:10.1046/j.1526-100X.2002.10102.x
relief, processus, environnement 1 Werner C, Smart JS (1973) Some new methods of topological
Rey F, Isselin-Nondedeu F, Bédécarrats A (2005) Vegeta- classification of channel networks. Geogr Anal 5:271–
tion dynamics on sediment deposits upstream of 295