EARTH SURFACE PROCESSES AND LANDFORMS
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
Copyright © 2009 John Wiley & Sons, Ltd.
Published online 16 June 2009 in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/esp.1826
Methodological framework to select plant
species for controlling rill and gully erosion:
application to a Mediterranean ecosystem
7John Wiley & Sons, Ltd.
Methodological framework to select plant species for controlling rill and gully erosion
S. De Baets,1* J. Poesen,1 B. Reubens,2 B. Muys,2 J. De Baerdemaeker3 and J. Meersmans4
Department of Earth and Environmental Sciences, Physical and Regional Geography Research Group, K.U. Leuven, Leuven,
Belgium
2
Department of Earth and Environmental Sciences, Division Forest, Nature and Landscape, K.U. Leuven, Leuven, Belgium
3
Department of Biosystems, Division of Mechatronics, Biostatistics and Sensors, K.U. Leuven, Leuven, Belgium
4
Department of Geography, Université Catholique de Louvain, Louvain-la-Neuve, Belgium
1
Received 22 September 2008; Revised 25 February 2009; Accepted 23 March 2009
* Correspondence to: S. De Baets, Department of Earth and Environmental Sciences, Physical and Regional Geography Research Group, K.U. Leuven, Celestijnenlaan
200E, B-3001 Leuven, Belgium. E-mail address: [email protected]
ABSTRACT: Many studies attribute the effects of vegetation in reducing soil erosion rates to the effects of the above-ground
biomass. The effects of roots on topsoil resistance against concentrated flow erosion are much less studied. However, in a
Mediterranean context, where the above-ground biomass can temporarily disappear because of fire, drought or overgrazing, and
when concentrated flow erosion occurs, roots can play an important role in controlling soil erosion rates. Unfortunately,
information on Mediterranean plant characteristics, especially root characteristics, growing on semi-natural lands, and knowledge
of their suitability for gully erosion control is often lacking. A methodological framework to evaluate plant traits for this purpose is
absent as well. This paper presents a methodology to assess the suitability of plants for rill and gully erosion control and its
application to 25 plant species, representative for a semi-arid Mediterranean landscape in southeast Spain. In this analysis
determination of suitable plants for controlling concentrated flow erosion is based on a multi-criteria analysis. First, four main
criteria were determined, i.e. (1) the potential of plants to prevent incision by concentrated flow erosion, (2) the potential of plants
to improve slope stability, (3) the resistance of plants to bending by water flow and (4) the ability of plants to trap sediments and
organic debris. Then, an indicator or a combination of two indicators was used to assess the scores for the four criteria. In total,
five indicators were selected, i.e. additional root cohesion, plant stiffness, stem density, the erosion-reducing potential during
concentrated flow and the sediment and organic debris obstruction potential. Both above- and below-ground plant traits were
taken into account and measured to assess the scores for the five indicators, i.e. stem density, sediment and organic debris
obstruction potential, modulus of elasticity of the stems, moment of inertia of the stems, root density, root diameter distribution,
root area ratio and root tensile strength. The scores for the indicators were represented on amoeba diagrams, indicating the
beneficial and the weak plant traits, regarding to erosion control. The grasses Stipa tenacissima L. and Lygeum spartum L. and the
shrub Salsola genistoides Juss. Ex Poir. amongst others, were selected as very suitable plant species for rill and gully erosion
control. Stipa tenacissima can be used to re-vegetate abandoned terraces as this species is adapted to drought and offers a good
protection to concentrated flow erosion and shallow mass movements. Lygeum spartum can be used to vegetate concentrated
flow zones or to obstruct sediment inflow to channels at gully outlets. Stipa tenacissima and Salsola genistoides can be used to
stabilize steep south-facing slopes. The methodology developed in this study can be applied to other plant species in areas
suffering from rill and gully erosion. Copyright © 2009 John Wiley & Sons, Ltd.
KEYWORDS:
concentrated flow erosion; soil conservation; plant traits; desertification remediation; root
Introduction
Soil erosion by water is considered a dominant erosion process
in Mediterranean environments, leading to land degradation
and desertification (Poesen et al., 2003). Several studies have
shown that gully erosion results in large soil losses (e.g. Poesen
et al., 2003). To initiate rills by runoff, concentrated surface
or subsurface flow is needed. Once a channel with a crosssection of 929 cm2 is formed, one can term it a gully. Gully
erosion is defined as the erosion process whereby runoff
water accumulates and often recurs in narrow channels, and
over short periods, removes the soil from the narrow area to
considerable depths. Various processes, such as hydraulic
incision, piping, headcut migration and mass movements, are
involved in gully development (Poesen et al., 2003).
The role of vegetation in protecting the soil from erosion
has long been recognized (Morgan, 2005). Vegetation reduces
water-caused erosion by intercepting rainfall, increasing water
METHODOLOGICAL FRAMEWORK TO SELECT PLANT SPECIES FOR CONTROLLING RILL AND GULLY EROSION
infiltration on associated soil-fertility islands, intercepting runoff
at soil surface level, stabilizing the soil by roots, resulting in
larger energy needed to detach soil particles (Bochet and
García-Fayos, 2004) and finally by acting as a roughness
element, causing flow retardance (Styczen and Morgan,
1995). Studies (e.g. Rey, 2003) have also demonstrated the
effectiveness of vegetation growing on gully bottoms for
trapping sediments.
The effectiveness of plants for erosion control depends on
the plant architecture and mechanical properties (Bochet et al.,
2006, Morgan, 2005). Some plants will be more suitable
for erosion protection than others, but how can one evaluate
the effectiveness of various plants for erosion control? The
selection of suitable plant species to control soil erosion and
more importantly a complementary mixture of species requires
a careful balance of considerations. For each field site and for
each set of objectives, the factors to be taken into account
may be different.
The following architectural and mechanical plant properties
will influence the interaction between vegetation and erosive
forces: (1) structural characteristics of the plant individuals,
such as the size and shape of plant stems and roots, (2) the
spatial distribution of the plant stems and roots within a plant
stand and the spatial pattern of plants along or at a site sensitive
to gully erosion and (3) the behaviour of plants in concentrated
overland flows or during soil shearing, expressed by the tensile
strength of plant roots and the flexibility of both individual
plant stems and the whole plant stand (Styczen and Morgan,
1995). All these properties are species specific.
Many studies measured single plant specimen characteristics
to assess the effects of vegetation on soil erosion processes
(e.g. the effects of vegetation cover on water erosion rates, the
effects of roots on soil shear strength). Moreover, the suitability
of plants for water erosion control is often only attributed to
the effects of the above-ground biomass on reducing erosion
rates (e.g. Bochet et al., 2006), whereas the role of the belowground biomass is often neglected (Gyssels et al., 2005). A
standard methodology to evaluate entire plants, including roots,
for erosion control strategies is lacking. A theoretical attempt
to select species for rehabilitation of degraded soils by water
erosion in semi-arid environments has been made by Albaladejo
et al. (1996). In their study, the capacity of plant species to
germinate and establish, the degree of suitability to satisfy
the rehabilitation objectives (e.g. improving soil stability and
fertility), the ecological conditions and the considerations
concerning landscape aesthetics are proposed to evaluate and
select suitable plant species. Other studies focus on environmental conditions of the rehabilitation site and its effects on
successful germination and growth rate (e.g. Bochet et al.,
2007, García-Fayos et al., 2000). Quinton et al. (2002), however,
made a listing of advantageous ecological and bioengineering
properties of potentially useful Mediterranean plant species
for re-vegetation of abandoned lands, but all information
presented is descriptive.
In short, no attempt was made so far to measure all plant
properties important for rill and gully erosion control for a
series of plant species in order to select the most suitable
species.
The main objectives of this study are therefore (1) to develop
a methodological framework to evaluate the suitability of
plants for rill and gully erosion control based on above-ground
and below-ground plant traits, (2) to discuss the methodology
to assess the plant properties required to select suitable species
and (3) to present the results of such selection procedure for
a range of Mediterranean plants growing in three different
habitats, typical for semi-arid degraded areas in southeast Spain
[i.e. (a) ephemeral channels and gully bottoms, (b) steep badland
Copyright © 2009 John Wiley & Sons, Ltd.
1375
slopes and (c) abandoned croplands]. It is hypothesized that
plants growing on steep badland slopes will score better for
improving slope stability and that plant species growing in
ephemeral channels and gully bottoms will be good in trapping sediments and organic debris as plant species will be
somehow adapted to the conditions of their habitat. This study
will focus on collecting and analysing above-ground as well
as below-ground mechanical and architectural properties of
individual Mediterranean plant species in order to make a
selection and to compare the effectiveness of species growing
in different habitats for rill and gully erosion control. Although
this methodology only accounts for the engineering properties
of plants and their potential to control rill and gully erosion,
the studied species are all well adapted to and very common
in the studied semi-arid environment. In addition a table with
all ecological information available for the studied plant
species will be presented in order to help with the selection
of species for application in similar environments.
Materials and Methods
Study area
The measurements of stem and root properties of 25 Mediterranean plant species were conducted in the Cárcavo catchment
(1°34′–1°27′ W, 38°09′–38°14′ N), located about 40 km
northwest of the city of Murcia in southeast Spain, near the
town of Cieza (Figure 1). This catchment was selected as a
target research area for the RECONDES (Conditions for Restoration and Mitigation of Desertified Areas Using Vegetation)
project, because it has very erodible soils (i.e. marls or Quaternary loam deposits), different land-use types are present, prior
knowledge and data were available and the area is reasonably
accessible. The catchment area is 30 km2 and altitudes range
between 220 and 850 m above sea level (a.s.l.) (Lesschen et al.,
2007). The climate is semi-arid, being one of the driest areas
of Europe, with a high inter-annual variability in rainfall
(Lopez-Bermudez et al., 2002). Mean annual air temperature
equals 17 °C and mean annual rainfall is 290 mm (FAO, 2005).
The geology consists of steep Jurassic limestone and dolomite
mountains with calcareous piedmonts, basin deposits of
Cretaceous and Miocene marls, and Keuper gypsum deposits.
Most soils are, following the FAO classification, thin Leptosols,
weakly developed Regosols, Calcisols and Gypsisols. The
current land use consists of farmland, orchards, abandoned
land and reforested land. In the 1970s large parts of degraded
land were reforested with pine (Pinus halepensis Mill.) within
the framework of reforestation and soil conservation programmes. Some almond and olive groves in the central part
are irrigated, while low-yielding cereals on marls are rainfed.
The last few decades, parts of the non-irrigated agricultural
land have been abandoned and are under different stages of
secondary succession. The steeper and higher areas are under
semi-natural vegetation and on north-facing slopes under
forest (Lesschen et al., 2007).
Preselection of plant species
A set of 25 native Mediterranean plant species, growing on
marls and Quaternary loamy deposits, was preselected based
on assessment of abundance. To assess the abundance of these
species 45 transects of 25 m long were randomly selected in
the basin. The abundance of a given species was then defined
as the number of sampling transects along which the species
occurs relative to the total number of sampling transects. To
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
1376
EARTH SURFACE PROCESSES AND LANDFORMS
Figure 1.
Map of the Cárcavo basin. Black squares indicate the sites where studied plants were sampled.
be considered as a main species an occurrence >10% had to
be achieved. All selected plant species are widespread in
southeast Spain (Alcaraz Ariza et al., 2002). The preselected
species were present in three habitats that are very prone
to concentrated flow erosion phenomena, i.e. (a) ephemeral
channels and gully bottoms, (b) steep badland slopes and
(c) abandoned fields (Figure 2). Most of the selected species
are habitat-specific for the study area. Full-grown, mediumsized grasses, forbs, shrubs, a reed, a rush and a small tree
were studied. No large trees were selected for potential use
in restoration works for soil conservation in these semi-arid
degraded areas, as the weight of large trees might surcharge
the slope, increasing normal and downhill shear force components (Gray and Sotir, 1996) and hence increasing landslide
risk. Moreover, grasses and shrubs can germinate quickly
when site conditions are favourable (Brindle, 2003) and can
reduce concentrated flow erosion on degraded soils in a
relatively short time span (De Baets et al., 2006). Inevitable,
plant size of the observed specimen differs largely. To allow
comparison on the suitability of plant species for rill and
gully erosion control, all plant properties were measured
relative to the size of the plant specimen. This means that the
vertical orthogonal projection of the above-ground biomass is
used as a reference area to calculate for example stem density
or root density.
a desirable plant trait for soil erosion control in arid and
semi-arid regions. Quinton et al. (2002) recommend species
with a dense ground cover and a root system that promotes
macro-porosity of the soil to re-vegetate degraded soils in
Mediterranean environments. Desirable root characteristics
for predicting the erosion-reducing potential of vegetation will
depend both on the erosion process of interest and the site
conditions. For preventing rill and gully erosion the initiation of concentrated flow erosion has to be prevented. Once
gullies have developed, further gully wall retreat has to be
prevented as well. While for gully wall stabilization deep and
dense rooting patterns are preferred (Simon and Collison,
2001), a shallow but dense lateral-spreading root system
seems to be more effective for preventing water erosion by
concentrated flow (e.g. Gyssels et al., 2005, De Baets et al.,
2006). According to Reubens et al. (2007) a woody species
with a large, deep and strong central part of the root system,
having some rigid vertical roots penetrating deeply into the
soil and anchoring into firm strata, as well as a large number
of finer roots numerously branching from the main lateral
roots would be most effective to increase shallow slope
stability. An asymmetrical distribution of the laterals parallel
to the contours may also be beneficial, as they can possibly
form a barrier for runoff and sediment. Based on a literature
survey, the following major criteria were selected to assess
the suitability of plants for erosion control:
Methodological framework
•
•
•
•
According to Gökbulak (2003), having a dense and relatively
deep root system, especially during the early growth stages, is
Copyright © 2009 John Wiley & Sons, Ltd.
high resistance against concentrated flow erosion;
high potential for slope stabilization;
high threshold for bending by water flow;
good ability to trap sediments and organic debris.
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
METHODOLOGICAL FRAMEWORK TO SELECT PLANT SPECIES FOR CONTROLLING RILL AND GULLY EROSION
1377
Figure 2. Illustration of typical sites where Mediterranean plants were sampled: (a) ephemeral river channels, (b) badland slopes, (c) abandoned
cropland (formerly almond grove). This figure is available in colour online at www.interscience.wiley.com/journal/espl
As both rills and gullies are initiated by concentrated flow
erosion, the same criteria will be used to assess the resistance
of plants to soil erosion by concentrated runoff. For controlling
rills, the criterion ‘potential for slope stabilization’ will not be
accounted for. This criterion becomes important once rills
have developed into gullies for the stabilization of their walls.
High resistance against concentrated flow erosion
In this study the effects of roots on topsoil resistance against
concentrated flow erosion are predicted using root density
(RD, in kg dry matter m–3) and root diameter (D, in metres) as
explanatory variables, as De Baets et al. (2007a) showed that
the erosion-reducing effects of plant roots during concentrated
runoff decrease exponentially with increasing root density
and that this effect is less pronounced with increasing root
diameter. It is also known that the number of roughness
elements per unit area (stem density) influence flow velocity
(Kouwen et al., 1981) and therefore also flow erosivity. As
such, the erosion-reducing potential of the studied plant
species during concentrated flow is determined using RD and
D information of the roots present in the topsoil (0– 0·1 m)
together with stem density (SD, in m2 m–2) information. The
topsoil erosion-reducing potential of plant root systems was
selected because it is in this soil layer that incision by
concentrated flow erosion has to be prevented. In this way,
the effects of the above-ground and the below-ground biomass on the resistance against concentrated flow erosion are
accounted for.
Copyright © 2009 John Wiley & Sons, Ltd.
High potential for slope stabilization
To assess the effects of plants on slope stability (of gully walls
or terrace slopes) root area ratio (RAR, dimensionless) and
root tensile strength (Tr, in MPa) information are combined
to obtain root cohesion values (Cr, in kPa), expressing the root
reinforcement effect on slope stability (Gray and Sotir, 1996).
This root reinforcing effect was assessed for a soil depth of 0·3
to 0·4 m, because this depth can be regarded as a common
depth at which overhanging gully walls become unstable.
Moreover at greater depths, root presence decreases. Hence,
the studied species can only prevent shallow mass movements.
High threshold to bending by water flow
An index of stiffness (MEI, in N) was determined as a measure
for the resistance of plants to bending. It is assumed that the
stiffer the plant stems, the more resistant the plant is to bending by water flow (Mant, 2002), as part of the flow energy is
absorbed by rigid stems and hence cannot contribute to soil
detachment anymore.
Ability to trap sediments and organic debris
A sediment and organic debris obstruction potential (SOP, in
m m–1) was measured in this study to assess the ability of plants
to trap sediments and organic debris. This one-dimensional
characteristic was selected to assess the sediment obstruction
potential of plants, as Isselin-Nondedeu and Bédécarrats (2007)
amongst others report, that plant roundness was inversely
correlated with sediment trapping and mound formation for
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
1378
EARTH SURFACE PROCESSES AND LANDFORMS
Figure 3. Multi-criteria approach for selecting plant species to control rill and gully erosion. Cr (in kPa) is root cohesion at 0·3–0·4 m soil depth,
MEI (in N) is index of stiffness, SD (in m2 m–2) is stem density, RSD (dimensionless) is topsoil erosion-reducing potential of plant roots during
concentrated flow erosion, SOP (in m m–1) is sediment obstruction potential, Tr (in kPa) is mean root tensile strength, RAR (in m2 m–2) is root area
ratio, M (in m–2) is stem density, Emod (in Pa) is modulus of elasticity, I (in m4) is second moment of inertia, RD (in kg m–3) is root density and D (in
metres) is mean root diameter.
species growing in the French alps on steep (25° and 35°)
gypsum-rich slopes.
A representation of this multi-criteria analysis used to select
suitable species is shown in Figure 3. The following indicators were selected to provide information on the four main
criteria: stem density (SD, in m2 m–2), sediment and organic
debris obstruction potential (SOP, in m m–1), index of plant
stiffness (MEI, in N), topsoil (0–0·1 m) erosion-reducing potential of plant roots during concentrated flow (RSD, dimensionless) and root cohesion (Cr, in kPa) in deeper soil layers
(0·3 –0·4 m).
To assess scores for the five indicators data on both aboveground as well as below-ground plant characteristics were
used, i.e. stem density (SD, in m2 m–2), sediment and organic
debris obstruction potential (SOP, in m m–1), the number of
stems under the crown of the plant (M, in m–2), the modulus
of elasticity of the stems (Emod, in Pa), the second moment of
inertia of the stems (I, in m4), root density (RD, in kg m–3), root
diameter (D, in metres) distribution, root area ratio at 0·3–
0·4 m (RAR, fraction) and root tensile strength (Tr, in MPa).
Each plant property was given an equal weight.
To determine the resistance against concentrated flow erosion
a mean value of the scores for the indicators SD and RSD was
calculated. Each indicator was given an equal weight. For the
criteria ‘potential to stabilize slopes’, ‘resistance to bending
by water flow’ and ‘the ability of plants to trap sediments and
organic debris’ the Cr, MEI and SOP scores were respectively
assigned (Figure 3).
Above-ground indicators and related plant traits
determine the resistance of vegetation to flow shear stress. In
this study, stem density was measured (1) as the area occupied
by the stems per unit area as a measure for the resistance to
water erosion and (2) as the number of stems per unit area for
the determination of the resistance to bending by water flow.
For five individuals per plant species, the number of stems,
their diameters at the plant base and the area occupied by
the vertical projection of the above-ground biomass were
measured. The unit area, equalling the soil surface occupied
by the vertical projection of the above-ground biomass, is
thus species dependent. The stems were assumed to have a
circular cross-section. Stem density as a measure for the
resistance to water erosion (SD, in m2 m–2) is calculated as
follows for shrubs:
n
SDshrub =
Copyright © 2009 John Wiley & Sons, Ltd.
(1)
i
Ar
where ds,i (in metres) is the diameter of each stem at the plant
base and Ar (in m2) is the total surface occupied by the
vertical projection of the above-ground biomass.
For grasses not all stem diameters were measured separately.
Only a representative horizontal area (c. 5 cm2) of a section
of grasses (As, in m2), the corresponding number of stems in
this section (ns) and their mean diameter at the plant base
were measured (Ds, in metres). Additionally the fraction of
this area (As, in m2) relative to the area occupied by the stems
(Ab, in m2) and the area occupied by the vertical projection of
the above-ground biomass (Ar, m2) were assessed to calculate
total stem density (SD, in m2 m–2) for grasses (Equation 2).
Stem density
Kouwen et al. (1980) argued that the number of roughness
elements per unit of area (i.e. stem density) is needed to
∑ π (d s,i / 2)2
SDgrass =
Ab
As
⎛
⎛ Ds ⎞ ⎞
* ⎜⎜ ns ⋅ π ⎜ ⎟ ⎟⎟
⎝ 2 ⎠⎠
⎝
Ar
2
(2)
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
METHODOLOGICAL FRAMEWORK TO SELECT PLANT SPECIES FOR CONTROLLING RILL AND GULLY EROSION
where ns is the number of stems in a horizontal section with
area As, Ds (in metres) is mean stem diameter at the plant
base, Ab (in m2) is the area occupied by the stems of a grass
stand, Ar (in m2) is the total surface occupied by the vertical
projection of the above-ground biomass.
Stem density (M, in m–2) used in the formula of Kouwen
et al. (1980) to calculate the resistance of plants to water
flow, is calculated as follows (both for shrubs as for grasses):
ns
(3)
Ar
where ns is the number of stems and Ar (in m2) is the total
surface occupied by the vertical projection of the aboveground biomass.
M=
Sediment and organic debris obstruction potential
The obstruction potential for sediment and organic debris
(SOP, in m m–1) was calculated as a measure for the obstruction capacity assuming that the flow is unidirectional. SOP
for shrubs is the ratio of the sum of the diameters at the plant
base (ds,i, in metres) of the horizontally projected stems on a
line perpendicular to the dominant flow direction and the
maximum length of this perpendicular line (Ltot, in metres)
defined by the vertical projection of the above-ground
biomass, i.e.
SOPshrub =
∑ d s,i
L tot
(4)
where SOP (in m m–1) is sediment obstruction potential,
ds,i (in metres) is stem diameter at the plant base and Ltot
(in metres) is the length determined by the projection of
the above-ground biomass, in a direction perpendicular to
the dominant flow direction (assessed topographically).
For grasses the number of stems and the mean stem diameter
were assessed for a unit length (e.g. 0·01 m). The sediment
obstruction potential (SOP, in m m–1) was then calculated as
follows:
SOPgrass =
Lb * n1cm * Ds
Ltot
(5)
where n1cm is the number of stems along 1 cm unit length, Lb
(in metres) is the length occupied by the stems projected to a
horizontal line perpendicular to the dominant flow direction
at the plant base, Ltot (in metres) is total length determined by
1379
the vertical projection of the above-ground biomass, in a
direction perpendicular to the dominant flow direction and Ds
(m) is mean stem diameter.
Plant stiffness
To assess the resistance of individual plants to bending in this
study a MEI index was calculated following a formula proposed
by Kouwen et al. (1980). In this formula mean bending
modulus values (Emod, in Pa) are multiplied by the number of
stems within the area occupied by the vertical orthogonal
projection of the above-ground biomass (M, in m–2) and by
the second moment of inertia (I, in m4) of the stems.
In the laboratory, three-point bending tests (Figure 4) were
performed on stems of 0·15 m long, stored in an alcohol
solution, with the UTS test system (GmBh., Ulm, Germany).
Modulus of elasticity (Emod) is a material characteristic and
can be calculated for a three-point bending test as follows
(Goodman et al., 2001):
Emod =
L3(dF / dY )
12π r 4
(6)
where L (in metres) is the length between the fixed points,
F (in N) is force, Y (in metres) is displacement in the vertical
direction and r (in metres) is the radius of the stem. For each
species, 10 samples of 0·15 m long stems were tested. Different
stem diameters were tested. The length between the fixed
points (L, in metres) was changed according to stem diameter
(L = 10*2r).
Stem density (M, in m–2) used in the formula of Kouwen et al.
(1980) to calculate resistance of plants to water flow, is calculated using Equation 3. For stems with a circular cross-section,
which is not hollow, I equals π d s4 /64 , where ds (in metres) is
mean stem diameter.
The product, MEI (in N), can be used as an index of stiffness
(Dunn and Dabney, 1996) and is used to determine the overall
resistance of a grass stand to bending under a shear force of
flowing water. This stiffness index will be used in this study to
rank different Mediterranean plant species according to their
resistance to bending under flow shear forces.
The formula proposed by Kouwen et al. (1980) to assess
stiffness of entire plant stand to flow shear forces for grasses
had to be adapted for shrubs as shrubs do not have a uniform
stem diameter distribution.
For grasses the following formula is used to assess the stiffness of the plant:
Figure 4. Three-point bending test. F (in N) is force, Y (in metres) is displacement, r (in metres) is the radius of the stem and L (in metres) is the
length between the fixed points of the apparatus. This figure is available in colour online at www.interscience.wiley.com/journal/espl
Copyright © 2009 John Wiley & Sons, Ltd.
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
1380
EARTH SURFACE PROCESSES AND LANDFORMS
Table I.
Transformation of absolute values into scores
SD (m2 m−2)
Score
SOP (m m−1)
Score
MEI (N)
Score
RSD
Score
Cr (kPa)
Score
0–0·001
0·001–0·002
0·002–0·003
0·003–0·007
>0·007
0
1
2
3
4
0–0·05
0·05–0·075
0·075–0·10
0·10–0·15
>0·15
0
1
2
3
4
0–0·25
0·25–1
1–10
10–25
>25
0
1
2
3
4
>0·50
0·50–0·50
0·10–0·25
0·01–0·10
0–0·01
0
1
2
3
4
0–0·1
0·1–1
1–5
5–10
>10
0
1
2
3
4
Note: SOP (in m m−1) is sediment obstruction potential, SD (in m2 m−2) is stem density, MEI (in N) is index of stiffness, RSD (dimensionless) is topsoil
erosion-reducing potential of plant roots during concentrated flow erosion and Cr (in kPa) is root cohesion.
MEIgrass = MEmodI (from Dunn and Dabney, 1996)
(7)
where M (in m–2) is the number of stems per unit area (i.e. area
occupied by the vertical projection of the above-ground vegetation elements), Emod (in Pa) is the mean bending modulus of
the stems and I (in m4) is the second moment of inertia of the
stems.
For shrubs this formula was adapted as follows:
n
MEIshrub =
∑ IiEmod
i
(8)
Ar
where Ii (in m4) is the second moment of inertia of each
individual stem, Emod (in Pa) is the mean bending modulus of
the stems and Ar (in m2) is the area occupied by the vertical
projection of the above-ground vegetation elements.
For each plant species mean MEI values of five mediumsized, representative individuals were established. It must be
highlighted that these calculations predict the hydraulic
resistance of individual shrubs and do not account for the
spatial configuration of plants.
Below-ground plant indicators and related plant traits
The erosion-reducing potential of plant roots during
concentrated flow
Empirically-based relationships were used to assess the erosionreducing potential for the 25 selected Mediterranean plant
species, of which root density (for different root diameter
classes) distribution with depth was measured. To assess the
erosion-reducing potential of plant roots, concentrated flow
experiments with rootless and root-permeated silt loam topsoils
were conducted in a laboratory flume. Empirical relationships
were established for silt loam topsoils permeated with grass
roots (representing fine-branched root systems), silt loam
topsoils permeated with carrot roots (representing tap root
systems) and silt loam topsoils permeated with both grass
roots and tap roots. Soil detachment rate was then related to
root density and root diameter information. Root density and
root diameter information for the Mediterranean plants was then
filled in into these equations to assess the erosion-reducing
potential of Mediterranean plants during concentrated flow.
For more information reference is made to De Baets et al.
(2007a,b).
Root reinforcement in deeper soil layers to prevent
shallow mass movements
To calculate the potential of root systems to prevent shallow
mass movements the model of Wu et al. (1979) was used.
For more information reference is made to De Baets et al.
(2008).
Copyright © 2009 John Wiley & Sons, Ltd.
Scoring and graphical representation
In order to compare species, the numerical data were rescaled
to scores ranging from 0 to 4 according to Table I. The transition from one score to another was fixed when there was a
gap in the absolute data range (visual cluster analysis). The
minimum and maximum interval values were determined by
the minimum and maximum values measured. In order to
have a good distribution of data over the classes, a non-linear
scale was adapted for assigning the scores. For each set of
continuous data (i.e. for each indicator) five groups were
made, i.e. 0 = very low value, 1 = low value, 2 = medium value,
3 = high value, 4 = very high value. The higher the score, the
more effective the plant is for controlling rill and gully erosion.
The score for each indicator is then plotted on amoeba diagrams,
with the five indicators forming the axes. The amoeba diagrams
are constructed for visual interpretation of the suitability of a
plant species for rill and gully erosion control.
Statistical analysis
In order to group the species according to their scores for the
five different indicators, a cluster analysis is performed in the
statistical program SAS Enterprise Guide 4.1 using the average
linkage method. Cluster analysis groups population elements
(e.g. plant species) based on their values for certain variables
(e.g. scores for the five indicators) into groups whereby the
similarity in one group is as good as possible and whereby
the differences between groups are as large as possible. To
assign an observation to a cluster, the mean distance between
this observation and the members of the cluster is calculated
and should be minimal (Buijs, 2000). To determine the number
of clusters, visual interpretation of a tree chart is performed,
whereby the average distance between the cluster elements is
not too high. The one-way ANOVA test was used to compare
mean values for the criteria between different habitat groups.
Results
Scores for the above-ground and below-ground
indicators
Table II presents the results of measured above-ground and
below-ground architectural or mechanical plant properties.
SD ranges from 0·06 to 12·61%. The two forbs Plantago
albicans L. and Limonium supinum (Girard) Pignatti have the
highest stem density. These are small plant species with a
good soil cover in spring. Some grasses like Piptatherum
miliaceum (L.) Coss and Helictotrichon filifolium (Lag.) Henrard,
the studied rush Juncus acutus L. and the reed Phragmites
australis Cav. also show high stem densities. Other grasses
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
METHODOLOGICAL FRAMEWORK TO SELECT PLANT SPECIES FOR CONTROLLING RILL AND GULLY EROSION
1381
Table II. Measured values and scores for the five above-ground and below-ground indicators, selected to evaluate the suitability of plants for rill
and gully erosion control
Plant name
Limonium supinum
Plantago albicans
Brachypodium retusum
Stipa tenacissima
Lygeum spartum
Avenula bromoides
Piptatherum miliaceum
Helictotrichon filifolium
Phragmites australis
Juncus acutus
Fumana thymifolia
Teucrium capitatum
Thymelaea hirsuta
Artemisia barrelieri
Dittrichia viscosa
Retama sphaerocarpa
Atriplex halimus
Salsola genistoides
Anthyllis cytisoides
Thymus zygis
Dorycnium pentaphyllum
Ononis tridentata
Nerium oleander
Rosmarinus officinallis
Tamarix canariensis
Vegetation
type
SD
(m2 m−2)
SD
score
SOP
(m1 m−1)
SOP
score
MEI
(N)
MEI
score
RSD
(−)
RSD
score
Cr
(kPa)
Cr
score
forb
forb
grass
grass
grass
grass
grass
grass
reed
rush
shrub
shrub
shrub
shrub
shrub
shrub
shrub
shrub
shrub
shrub
shrub
shrub
shrub
shrub
tree
0·0401
0·1261
0·0006
0·0008
0·0019
0·0021
0·0028
0·0055
0·0051
0·0070
0·0007
0·0009
0·0011
0·0014
0·0014
0·0018
0·0018
0·0019
0·0021
0·0023
0·0026
0·0029
0·0042
0·0045
0·0011
4
4
0
0
1
2
2
3
3
4
0
0
1
1
1
1
1
1
2
2
2
2
3
3
1
0·1840
0·3538
0·1408
0·0498
0·0612
0·0981
0·0736
0·2161
0·2101
0·1478
0·0326
0·0456
0·0350
0·0413
0·0499
0·0705
0·0451
0·0514
0·0652
0·0684
0·0669
0·0577
0·0787
0·1179
0·0554
4
4
3
0
1
2
1
4
4
3
0
0
0
0
0
1
0
1
1
1
1
1
2
3
1
0·25
n.a.
0·05
0·32
0·41
0·45
2·09
3·62
5·54
1·77
0·64
0·62
17·04
2·70
3·58
27·51
54·16
42·17
8·56
11·96
22·27
97·54
25·17
159·80
48·48
0
n.a.
0
1
1
1
2
2
2
2
1
1
3
2
2
4
4
4
2
3
3
4
4
4
4
0·37
1·10−5
8·10−5
0·03
2·41·10−7
0·3·10−12
0·01
1·61·10−6
0·60
2·72·10−8
0·25
0·32
0·50
0·07
0·19
0·03
0·18
0·03
2·29×10−3
0·32
0·11
0·45
0·19
0·15
0·01
1
4
4
3
4
4
3
4
0
4
1
1
0
3
2
3
2
3
4
1
2
1
2
2
3
0
0
0
2·53
0
0
0
0
35·18
5·21
0
0
2·81
0
4·31
38·56
9·62
10·27
16·22
0
3·46
4·03
0·23
0·17
36·74
0
0
0
2
0
0
0
0
4
3
0
0
2
0
2
4
3
4
4
0
2
2
1
1
4
Note: SOP (in m m−1) is sediment obstruction potential, SD (in m2 m−2) is stem density, MEI (in N) is index of stiffness, RSD (dimensionless) is topsoil
(0–0.1m) erosion-reducing potential of plant roots during concentrated flow erosion and Cr (in kPa) is root cohesion at 0·3–0·4 m soil depth.
like Stipa tenacissima L. and Lygeum spartum L. do not have
such high stem densities as expected. This can be explained
by the angle of the stems (c. 45° with the vertical), whereby
the total surface under the crown is higher as compared to
other grasses.
SOP ranges between 3·3 and 35·4%. The two forbs Plantago albicans and Limonium supinum, as well as most of the
grasses (except Lygeum spartum), the rush Juncus acutus and
the reed Phragmites australis have a high potential to obstruct
sediment and organic residues. But also some shrubs like
Rosmarinus officinalis L. have a high obstruction potential.
These findings are supported by Rey (2003), who reports
that the percentage of low vegetation cover, that is to say
herbaceous and under-shrub layers, is positively related with
gully-inactivity.
The results of the MEI calculations show that the shrubs
and trees are the most resistant and the grasses and some
small shrubs the least resistant to bending through simulated
flow shear forces. The reed and rush show intermediate
values.
Grass roots have the highest erosion-reducing potential (i.e.
very low RSD value) and highly increase the resistance of
topsoils to concentrated flow erosion, but also some shrubs
such as Anthyllis cytisoides L. or Salsola genistoides Juss. Ex
Poir. have a high erosion-reducing potential through their
roots during concentrated flow. This can be attributed to the
high density of fine roots in the topsoil for these species.
The shrubs Anthyllis cytisoides, Retama sphaerocarpa (L.)
Boiss., Salsola genistoides, the tree Tamarix canariensis Willd.
and the reed Phragmites australis strongly reinforce the soil at
greater depth (0·3– 0·4 m).
Copyright © 2009 John Wiley & Sons, Ltd.
Amoeba diagrams
The amoeba diagrams corresponding to each studied species
are presented in Figure 5. It can be observed from Figure 5
that plant species growing on abandoned fields are much less
suitable for rill and gully erosion control as compared to the
species growing in ephemeral channels and on steep badland
slopes, as the areas occupied by the amoeba diagrams are
much smaller for species growing on abandoned fields.
Generally, from the species growing on abandoned fields, the
shrubs Rosmarinus officinalis and Dorycnium pentaphyllum
Scop. are most suitable for controlling rill and gully erosion.
The forb Plantago albicans and the grasses are effective to
prevent erosion by concentrated flow, but are not able to
improve slope stability and can be easily bent.
Juncus acutus seems to be the best scoring species of all
species growing in ephemeral channels, although Tamarix
canariensis and Retama sphaerocarpa also have high scores,
except for stem density and sediment obstruction potential.
On steep slopes, the grass Helictotrichon filifolium performs
well, but can not be used to prevent shallow mass movements, because no roots are present in deeper soil layers. The
shrubs Anthyllis cytisoides and Salsola genistoides also have a
high potential to resist erosion and bending by water flow.
Cluster analysis
In order to summarize the results represented with amoeba
diagrams and with scores for each indicator, a cluster analysis
was performed. In doing so, species are grouped in such a
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
1382
EARTH SURFACE PROCESSES AND LANDFORMS
Figure 5. Amoeba diagrams indicating the suitability of 25 Mediterranean plants for rill and gully erosion control. (a) plant species growing in
ephemeral river channels, (b) plant species growing on steep badland slopes, (c) plant species sampled on abandoned croplands (formerly
almond grove).
Copyright © 2009 John Wiley & Sons, Ltd.
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
METHODOLOGICAL FRAMEWORK TO SELECT PLANT SPECIES FOR CONTROLLING RILL AND GULLY EROSION
Figure 5.
1383
(Continued )
way that plants belonging to the same cluster have similar
scores for all criteria.
By visual interpretation of the cluster tree chart, eight clusters
were selected. A maximum mean distance of 0·50 (units are
score values) was used as a selection tool to assign an observation to a cluster or to group clusters and the plants species
belonging to each cluster are presented in Table III.
The shrubs Fumana thymifolia (L.) Spach and Teucrium
capitatum (L.) ssp. gracillimum (Rouy) (Cluster 1) have low
scores for the four criteria. Clusters 2 and 5 represent a mix of
shrub and grass species with medium or low scores for the
criteria, except for species belonging to Cluster 2 having high
scores for their resistance to bending by water flow. Cluster 3
groups very resistant shrubs that have a high potential for
improving slope stability and are very resistant to bending by
water flow. Cluster 4, however groups species, mainly grasses
or small shrubs, with a medium to high potential to prevent
soil erosion by concentrated flow, but a low resistance to bending by water flow and a low sediment obstruction potential,
whereas Cluster 6 also groups species (forbs and grasses) with
Table III. Plant species grouped in eight clusters according to their scoring for the four main requirements, i.e. (1) the potential to prevent incision by concentrated flow erosion, (2) the potential to improve slope stability, (3) the potential to resist bending by water flow and (4) the ability
to trap sediments and organic debris
Cluster
Plant species name
Cluster description
1
Low resistance against concentrated flow ersosion, low sediment obstruction potential, not
resistant to bending by water flow, no potential to improve slope stability
Low potential for slope stabilization, medium potential to prevent erosion by concentrated
runoff, high resistance to bending by water flow, medium sediment obstruction potential
High potential for slope stabilization, very resistant to bending by water flow, low sediment
obstruction potential and medium to high potential to prevent concentrated flow erosion
7
Fumana thymifolia
Teucrium capitatum
Nerium oleander
Rosmarinus officinalis
Anthyllis cytisoides
Retama sphaerocarpa
Salsola genistoides
Tamarix canariensis
Atriplex halimus
Thymus zygis
Artemisia barrelieri
Lygeum spartum
Avenula bromoides
Piptatherum miliaceum
Stipa tenacissima
Thymelaea hirsuta
Dittrichia viscosa
Ononis tridentata
Dorycnum pentaphyllum
Plantago albicans
Limonium supinum
Helictotrichon filifolium
Brachypodium retusum
Juncus acutus
8
Phragmites australis
2
3
4
5
6
Copyright © 2009 John Wiley & Sons, Ltd.
Medium to high potential to prevent erosion by concentrated flow, low potential for slope
stabilization, medium sediment obstruction potential and low resistance to bending by water
flow
Medium potential for slope stabilization, low potential to prevent incision by water flow,
medium to high resistance to bending by water flow, low sediment obstruction potential
Medium to high potential to prevent concentrated flow erosion, easily bended, low potential
for slope stabilization and high sediment obstruction potential
High potential to prevent concentrated flow erosion, high sediment obstruction potential,
medium resistant to bending by water flow
Medium resistant to bending by water flow, high sediment obstruction potential, low potential
to prevent concentrated flow erosion
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
1384
EARTH SURFACE PROCESSES AND LANDFORMS
a high potential to prevent erosion, but with in addition a
high sediment obstruction potential. The rush Juncus acutus
and the reed Phragmites australis form two separate clusters.
Juncus acutus appears to be a very suitable plant species for
application in erosion control strategies. Whereas Phragmites
australis has a low potential to prevent erosion by concentrated
flow, but has high scores for the other criteria, Juncus acutus
scores medium to high for all criteria. It has to be noticed that
both plant species can only grow in moist environments. A
remark has to be made on the potential to improve slope
stability for Phragmites australis and Juncus acutus: whereas
they have a large potential to reinforce the soil to greater
depths, these plants prefer gentle sloping areas to grow, so
the applicability for slope stabilization will be very limited.
Classification according to erosion control potential
Table IV reports the scores for the four main criteria, i.e. (1) the
potential to prevent incision by concentrated flow erosion,
(2) the potential to improve slope stability, (3) the potential to
resist bending or pull-out caused by floods and (4) the ability
to trap sediments and organic debris.
Table IV shows that most studied grasses are very effective
for preventing topsoils from being eroded by concentrated
flow. Their high stem density, resulting in a decrease of mean
flow velocity and a decrease in erosivity combined with their
dense network of fine roots in the topsoil offers a great protection to concentrated flow erosion. But also some shrubs, such
as Anthyllis cytisoides, the forb Plantago albicans and the rush
Juncus acutus are suitable to prevent erosion by concentrated
flow. However, when flow depth exceeds plant height, these
species (e.g. Plantago albicans) will not reduce flow erosivity
any longer.
Field observations and statistical analysis clearly indicate
that shrubs on earthen terrace walls increased terrace failure
risk, whereas almost no failure was observed on earthen terraces
covered with grasses (Lesschen et al., 2008). It is therefore
recommended to vegetate terraces with the mentioned grasses,
as these species offer a good protection to incision by
concentrated flow when runoff flows over a terrace.
To prevent shallow mass movements on gully walls or steep
slopes, dense and deep rooting species, such as the shrubs
Retama sphaerocarpa, and Salsola genistoides or the tree
Tamarix canariensis are very effective and offer good protection (Table IV).
Most shrubs such as Nerium oleander or Salsola genistoides
are very resistant to bending by water flow, but also the tree
Tamarix canariensis offer a large resistance to bending forces
exerted by concentrated flows (Table IV).
As mentioned earlier, the reed, rush, forb and grass species,
but also the shrub Rosmarinus officinalis have a high obstruction
potential for sediments and organic debris (Table IV).
ANOVA tests show that the studied plant species growing
on steep badland slopes and in ephemeral channels have a
significantly (p = 0·03) higher potential to improve slope
stability as compared to plant species growing on abandoned
croplands. This can be attributed to the presence of favourable environmental conditions, encouraging plant growth
and strength, for species growing in ephemeral channels and
on steep slopes. Plant species growing in ephemeral channels
are moistened by seasonal flows and water availability is
known to be crucial for seed establishment, growth and function (e.g. García-Fayos et al., 2000, Pugnaire et al., 2006).
Plant species growing on steeps slopes need to anchor themselves and have to resist gravitational forces, which results
in stronger roots (Greenway, 1987). However, most plants
growing in ephemeral channels prefer gentle sloping areas.
Hence their applicability to improve slope stability will be
limited. For the other criteria, no significant differences could
be detected between different habitat groups. Plants growing in
channels did not score significantly better on their resistance
to bending by water flow or their potential to trap sediments
and organic debris.
Table IV. Scores indicating (1) the potential of plants to prevent incision by concentrated flow erosion, (2) the potential to improve slope stability, (3) the potential to resist bending by water flow and (4) the ability to trap sediments and organic debris
Plant species name
Limonium supinum
Plantago albicans
Avenula bromoides
Brachypodium retusum
Helictotrichon filifolium
Lygeum spartum
Piptatherum miliaceum
Stipa tenacissima
Phragmites australis
Juncus acutus
Anthyllis cytisoides
Artemisia barrelieri
Atriplex halimus
Dittrichia viscosa
Dorycnium entaphyllum
Fumana thymifolia
Nerium oleander
Ononis tridentata
Retama sphaerocarpa
Rosmarinus officinalis
Salsola genistoides
Teucrium capitatum
Thymelaea hirsuta
Thymus zygis
Tamarix canariensis
Vegetation type
Resistance to erosion
Slope stabilization
Resistance to bending
Ability to trap sediments
forb
forb
grass
grass
grass
grass
grass
grass
reed
rush
shrub
shrub
shrub
shrub
shrub
shrub
shrub
shrub
shrub
shrub
shrub
shrub
shrub
shrub
tree
2·5
4·0
3·0
2·0
3·5
2·5
2·5
1·5
1·5
4·0
3·0
2·0
1·5
1·5
2·0
0·5
2·5
1·5
2·0
2·5
2·0
0·5
0·5
1·5
2·0
0
0
0
0
0
0
0
2
4
3
4
0
3
2
2
0
1
2
4
1
4
0
2
0
4
0
0
1
0
2
1
2
1
2
2
2
2
4
2
3
1
4
4
4
4
4
1
3
3
4
4
4
2
3
4
1
1
0
4
3
1
0
0
0
1
0
2
1
1
3
1
0
0
1
1
Copyright © 2009 John Wiley & Sons, Ltd.
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
METHODOLOGICAL FRAMEWORK TO SELECT PLANT SPECIES FOR CONTROLLING RILL AND GULLY EROSION
Classification according to habitat
All studied plant species are adapted to a Mediterranean
environment, as only common, representative, native species
were selected. Still, each species searches its optimal habitat
within this Mediterranean environment. It is therefore
necessary to gain information on the optimal environmental
site conditions required for plant growth and survival and
on the interactions between plants and these environmental
conditions in order to select species for potential use in
erosion control structures in a given environment. Information on Mediterranean plant species is rather scarce and for
some species no information exists at all. Nevertheless an
attempt was made to summarize for each plant species all
ecological and additional important information for plant
species selection. The results are presented in Table V.
Identifying the ecological site properties allowed us to sort
suitable plant species by habitat and erosion prevention
requirements (Table VI). All species with scores >2 (Table IV)
are presented in Table VI. It has to be recognized that the
plant species are only reported for the habitat in which
they were sampled, whereas some species may occur in several
habitats [e.g. Rosmarinus officinalis, Thymelaea hirsuta (L.)
Endl.].
Discussion
Methodological framework
The strength of the framework presented and applied in this
study is the use of relatively few and easily measured data.
For the first time, above-ground and below-ground plant traits
were combined and related to their erosion-reducing potential.
This methodological framework can be applied to other woody
shrubs and stiff grasses, growing in other environments.
Caution should be made with more flexible plants and trees,
as the methodology to measure the plant properties selected
in this study, shall be more complex and probably requires
adaptation.
Ecological constraints
Inevitably it is not possible to have an ideal mixture of suitable
plants for every situation and combination of circumstances.
A variety of factors can influence the choice of a certain plant
species for rehabilitation of degraded sites.
It was mentioned in the previous paragraphs that Tamarix
canariensis, Retama sphaerocarpa and the reed Juncus acutus
are preferred for restoration works to combat rill and gully
erosion in channels, with a preference for Juncus acutus, as
its trapping capacity and its potential to prevent erosion by
concentrated flow is higher due to the higher stem density
and density of fine roots. In addition, these species are well
adapted to this habitat. Juncus acutus and Tamarix canariensis
both prefer fine substrates and need moist environments for
seed germination (Table V). Consequently, they are mostly
found in sedimentation zones and near pools within the
channel where they can prevent concentrated flow erosion of
the channel bottom. Also Retama sphaerocarpa needs moist
environments, but once established and once the roots have
reached deeper water resources, this species is adapted to
seasonal drought (Table V). As Retama sphaerocarpa reinforces
the soil to a larger depth (i.e. at least to 0·4 m depth, De
Baets et al., 2008), it can be used to stabilize riverbanks. Of
the grasses, Lygeum spartum is identified as having a good
Copyright © 2009 John Wiley & Sons, Ltd.
1385
potential for reducing water erosion in channels (Table IV).
While Lygeum spartum has been found to establish in a range
of substrates, even on places with a high salt concentration
(Pugnaire and Haase, 1996), a preference for fines appears to
exist. Plant density even increases as the proportion of fines
increases. Extended monitoring indicates that (1) this species
is resistant to medium erosive floods (Sandercock et al., submitted for publication b), (2) has the ability to trap sediments
when it forms dense stands (Navarro-Cano, 2004) and (3) contributes to the net aggradation of the channel bed (Sandercock
et al., 2007). The sediment obstruction potential measured in
this study for Lygeum spartum was rather low (Table II),
because this property was measured for an individual plant.
Due to the umbrella morphology of the stems, the area
occupied by the stems relative to the area occupied by the
vertical orthogonal projection of the above-ground biomass
was rather small. However, this plant can form dense stands
(Navarro-Cano, 2004), and has the potential to act as a barrier
for sediments. It is suggested to plant Lygeum spartum grasses
in channels in close proximity to gullies, supplying fines, with
the aim of trapping these incoming sediments (Hooke et al.,
submitted for publication).
The shrub Rosmarinus officinalis is preferred for rill and
gully erosion control on abandoned croplands (Figure 5). This
shrub is preferred over Dorycnium pentaphyllum, because
Rosmarinus officinalis has a better potential to trap sediments
and organic debris and is also effective in preventing splash
(González-Hildago et al., 1997) and interrill erosion (Bochet
et al., 2006). Moreover this species is often found in concavities, where it can prevent gully development. Plantago
albicans can also resist erosion by concentrated flow to a
large extent in this habitat (Table IV) and has a good
ground cover (Table V), but its resistance to bending by
water flow is very low (Table IV). Nevertheless, this plant can
be very import-ant for the colonization of bare or abandoned
areas. The rapid establishment of a good vegetation cover is
a key step for effective mitigation of erosion and requires
first short-lived species that are resistant to removal and
grow quickly so that later the ‘late successional’ shrubs can
establish.
Considering plant reproduction, plants that are able to
resprout, such as Plantago albicans, Dorycnium pentaphyllum
and Piptatherum miliaceum (Table V), may possess an advantageous strategy over seeders as it allows species to persist
after disturbance. But plants that sprout tend to be poor
recruiters and have generally less seed production, smaller
seedbanks, slower growth rates and less seedling survival
than non-sprouters. This is due to the characteristic allocation
of resources to storage in sprouting species which has an
important cost on growth or reproduction (Bochet et al., 2007).
Piptatherum miliaceum can be established on abandoned
fields or along roads to prevent concentrated flow erosion
(Table V). The rhizomatous species Brachypodium retusum (pers.)
Beauv. can send out roots and shoots from the horizontal,
usually underground, stem and is able to resprout after fire.
The soil under Brachypodium retusum plants has a higher soil
aggregate stability as compared to the soil under Rosmarinus
officinalis plants (Cerda, 1998). This is probably linked with
the higher root density under Brachypodium retusum (De Baets
et al., 2007b). Cerda (1998) recommends that immediately
after fire, Brachypodium retusum grassland seems to be the
best option for soil protection. It must be highlighted that
Brachypodium retusum prefers to grow in shadow-rich places
and is mostly found on north-facing slopes (Table VI).
Avenula bromoides also has a favourable root system to resist
concentrated flow erosion (Table IV), but only grows along
road tracks (Table V).
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
Copyright © 2009 John Wiley & Sons, Ltd.
–
–
*channel bottom11
*salty soils5
*borders of roads11
*borders of roads11
Atriplex halimus
Avenula bromoides
*sunny sites19
Fumama thymifolia
*rhizomatous2
*seeds: needs constant moisture to
germinate21
*seed germination2
*rhizomatous2
*north slops
*forest communities2
*marls, limestone2
*low gradient16
*wadis and depressions21
*thalweg and ponding sites16
*fine substrates16
*salty soils2
*marly ephemeral streams and gully
bottoms2
*fine substrates16
*salt tolerant22
*ephemeral streams, gully bottoms2
Helictotrichon filifolium
Juncus acutus
Limonium supinum
Lygeum spartum
*seed germination
*seed germination19
*abandoned grasslands2
*marly gully bottoms2
*ephemeral streams2
*able to grow in acid and alkaline, low
fertility soils17
Dorycnium pentaphyllum
2
*seed germination2
*resprouter2 (after fire, cut, grazing)
*marly ephemeral streams and gully
bottoms2
*gypsum rocky gullies2
*gravel bars16
Dittrichia viscosa
20
*seed germination2
*resprouter2 (after fire, cut, grazing)
*north slopes
*natural forest or reforested areas9
*fine substrates15
Brachypodium retusum
*rhizomatous
*seed germination2
*fine substrates10
*abandoned fields9
*low slopes9
Artemisia barrelieri
13
*seed germination2 (dispersal by
wind and secondly by water)7
*different soil types1
*poor soils1
Anthyllis cytisoides
2
Reproduction
Habitat
Plant species name
13
*leaves vulnerable for grazing42
*can form dense stands2
*>3–5 years for a clump41
*<1 year for adult plant41
*fast recovery after flood damage16
*temperature is more important in
regulating seedling growth than
moisture20
*3–5 years for a clump41
*2 years for adult plant41
*3–5 years for adult plant41
*can form dense swards2
*<1 year for adult plant41
*faster growing than seed species
*able to resprout after fire13
*<1 year for adult plant41
*<1 year for adult plant41
*slow growth rate12
*>3–5 years for adult plant41
*linked to seasonal percipitation8
*2 years for full-grown plant41
*linked with rainfall3
*able to resprout after perturbation6
*better growth on north slopes7
*≥3 years for adult plant41
Growth
*tolerant to drought16
–
*adapted to water
stress21
–
–
*adapted to water
stress18
–
–
–
*drought resistant12
–
*drought resistant4
Resistance to droughts
*easily flattened16
*high soil aggregate stability18
*architecture of the leaves is negative
for splash erosion (drip effects)23
*ability to trap sediments16
*easily washed away by floods16
*easily flattened16
–
–
–
*ability to trap sediment in dense
stand16
*low to moderate resistant16
*high soil aggregate stability under
Brachypodium clumps14
–
*high roughness16
*high rainfall interception capacity5
*good cover12
–
*sedimentation when enough
stems6
*good cover and litter production5
Resistance to floods/erosion
Table V. Additional information on preferential habitat and substrate requirements, reproduction strategy, growth requirements, growth speed, adaptation or vulnerability to drought and information on the resistance to floods or other erosion processes for the 25 studied Mediterranean matorral species
1386
EARTH SURFACE PROCESSES AND LANDFORMS
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
Copyright © 2009 John Wiley & Sons, Ltd.
*seed germination (dispersal by wind
and secondly by water)7
*flood plains and low terraces8
*dry valleys and river banks8
*prefers acid soils5
*occurs on thin soils over
non-calcareous bedrock8
Retama sphaerocarpa
*marly ephemeral streams and gully
bottoms2
*dry beds, stream banks2
*very steep degraded slopes2
*all slope orientations2
*seed germination2
*marly ephemeral streams and gully
bottoms2
*dry beds, stream banks2, flat positions,
abandoned land9
*quaternary colluvium9
Plantago albicans
Salsola genisoides
*seed germination2
*resprouter2 (after fire, cut, grazing)
*dry beds30
*open grassy places, pasture30
Piptatherum miliaceum
*marly ephemeral streams and gully
bottoms2
*dry beds, stream banks2
*concave hillslopes2
*calcareous soils31
*able to resprout40
*<1 year for adult plant41
*seed germination, clonal
reproduction, asexual multiplication,
rhizomatous25
*salt tolerant areas28
*downstream of checkdams16
*prefers fine substrates27
*high water level sites27
Phragmites australis
Rosmarinus officinalis
*dependent on surface soil moisture
for full top development16
*<1 year for adult plant41
*seed germination2
*marly and gypsium rich soils2
*gullies2
*reforestation areas9
Ononis tridentata
*competition of grasses may
suppress succession towards shrub
cover31
*≥3–5 years for adult plant41
*≥3–5 years for adult plant41
*seed germination2
*seed germination2
*resprouter2 (after fire, cut, grazing)
*provides shade for growth of other
grasses, shrubs or summer annuals7,
N fixing, improves soil fertility5
*<5 years for adult plant41
*few seasons are needed for
complete development26
*invasive!26
*<1 year for adult plant41
*3–5 years for adult plant41
*able to resprout after damage24
*<5 years for adult plant41
*seed germination2
*seeds require moist conditions and
are sensitive to drought24
*marly ephemeral streams and gully
bottoms2
*coarse substrate16
*steeper channel sections16
*zones of erosion16
Nerium oleander
Growth
Reproduction
Habitat
Plant species name
Table V. (Continued )
–
*drought tolerant
(rapid closing of
stomae)31
*phreatophyte8
*uses deep water
resources3
*drought avoider by
habitat4
*high temperature
avoider by leave
morphology4
*not resistant to drought
>12 m7
–
–
*adapted to water
stress18
–
*once deep roots are
established: able to
withstand long periods
of drought16
Resistance to droughts
–
*sedimentation as water flows
through the branches6
*high protection against spash
erosion6
*soil loss due to interrill erosion is
much reduced as compared to bare
soil32
*good infiltration5
*soil binding5
*excellent ground cover5
*good infiltration5
–
*easily swept over16
*young seedlings may be killed27
*reduced flow velocity and
enhaced sedimentation29
*good cover and surface
protection5
*survives extensive flood damage24
Resistance to floods/erosion
METHODOLOGICAL FRAMEWORK TO SELECT PLANT SPECIES FOR CONTROLLING RILL AND GULLY EROSION
1387
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
Copyright © 2009 John Wiley & Sons, Ltd.
*requires moist, high-light
environments for germination and
establishment39
*fast growth38
*germinates and propagates easily16
*invasive!, dominant16
*<5 years for adult plant41
*seed germination2
*resprouter2 (after fire, cut, grazing)
*seed dispersal by wind or water35
*seed germination2
*seed germination2
*seed germination2
*marly ephemeral streams and gully
bottoms2
*dry beds, swamped beds2
*fine substates16
*salt tolerant areas35
*thalwegs and bars in rivers16
*close to checkdams, ponds16
*borders of roads2
*abandoned fields2
*reforestation areas9
*abondoned fields19
*fine substrate9
*marly ephemeral streams and gully
bottoms2
*dry beds, stream banks2
*gypsium rocky gullies2
Tamarix canariensis
Teucrium capitatum
Thymelaea hirsuta
Thymus zygis
–
–
–
*phreatophyte35
*drops of its leaves to
withstand long periods
of drought35
*drougth tolerant5
(curles its leaves)
*high temperature
endurer4
Resistance to droughts
*very effective as plant-cover strip37
–
–
*can withstand extreme flows16
*able to trap sediments16
*offers a high resistance to flows16
*when removed: formation of large
woody debris16
*high protection against spash
erosion6
*soil loss due to interrill erosion is
much reduced as compared to bare
soil32
*good infiltration34
Resistance to floods/erosion
Note: This table does not pretend to present a complete overview.
1
Robledo et al., 1990 –1991; 2Navarro-Cano, 2004; 3Archer et al., 2002; 4Puigdefabregas et al., 1996; 5Quinton et al., 2002; 6Bochet et al., 1998; 7Pugnaire et al., 2006; 8Haase et al., 1996; 9Sandercock and Hooke,
submitted for publication a; 10Frietag, 1971; 11personal observations; 12Mattia et al., 2005; 13De Luis et al., 2004; 14Cerda, 1998; 15Maestre and Cortina, 2002; 16Sandercock and Hooke, submitted for publication b;
17
Alegre et al., 1998; 18Caravaca et al., 2004; 19Alcaraz Ariza et al., 2002; 20Tong, 1990; 21Snogerup, 1993; 22Pugnaire and Haase, 1996; 23Hidalgo et al., 1997; 24Herrera, 1991; 25Clevering et al., 2001; 26Hara et al.,
1993; 27Haslam, 1972; 28Matsushita and Matoh, 1991; 29Brix, 1997; 30Garcia et al., 2004; 31Clary et al., 2004; 32Bochet et al., 2006; 33Cerda, 1997; 34Gasque and García-Fayos, 2004; 35DiTomaso, 1998; 36El-Keblawy
et al., 1997; 37Martinez-Raya et al., 2006; 38Friederici, 1995; 39Cohn, 2005; 40Bochet et al., 2007; 41personal observation by Dr G. Barbera (CEBAS-CSIC, Murcia, Spain); 42 Béchet et al., 1982.
*>3 years for adult plant41
*requires a number of consecutive
days with percipitation for
germination36
*low germination and growth rate36
*1–2 years for adult plant41
*<1 year for adult plant41
*>5 years for adult plant41
*rhizomatous2
*resprouter2 (after fire, cut, grazing)
*bare grounds9
*all slopes (can colonize very steep
slopes)34
*stony soils15
Stipa tenacissima
Growth
Reproduction
Habitat
Plant species name
Table V. (Continued )
1388
EARTH SURFACE PROCESSES AND LANDFORMS
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
Brachypodium retusum
Thymelaea hirsuta Dorycnium
pentaphyllum Rosmarinus
officinalis
Moist
Rosmarinus officinalis
Plantago albicans
Dry
Abandonned fields
Stipa tenacissima
South
North
Steep badland slopes
Ephemeral channels
Copyright © 2009 John Wiley & Sons, Ltd.
Piptatherum miliaceum
Plantago albicans
Avenula bromoides
Artemisia barrelieri
Rosmarinus officinalis
Dorycnium pentaphyllum
Brachypodium retusum
Salsola genistoides
Stipa tenacissima
Ononis tridentata
Dorycnium pentaphyllum
Dittrichia viscosa
Thymelaea hirsuta
Ononis tridentata
Salsola genistoides
Rosmarinus officinalis
Plantago albicans
Avenula bromoides
Helictotrichon filifolium
Anthyllis cytisoides
Juncus acutus
Limonium supinum
Phragmites australis
Nerium oleander
Lygeum spartum
Nerium oleander
Atriplex halimus
Retama sphaerocarpa
Tamarix canariensis
Atriplex halimus
Juncus acutus
Retama sphaerocarpa
Tamarix canariensis
Juncus acutus
Lygeum spartum
Limonium supinum
Nerium oleander
Retama sphaerocarpa
Tamarix canariensis
Helictotrichon filifolium
Anthyllis cytisoides
Salsola genistoides
Resistant to bending
Erosion prevention requirements
Suitable for preventing shallow
mass movements (0·3–0·4 m)
Resistant to concentrated flow erosion
Resistant to splash and
interrill erosion
Slope orientation/
moisture status
Habitiat
Table VI.
Suitable species for erosion control ranked according to habitat and erosion prevention requirements
High sediment obstruction
potential
METHODOLOGICAL FRAMEWORK TO SELECT PLANT SPECIES FOR CONTROLLING RILL AND GULLY EROSION
1389
Salsola genistoides and Anthyllis cytisoides are the most suitable species to control gully erosion on steep slopes (Table IV,
Figure 5), but also the grasses Helictotrichon filifolium and
Stipa tenacissima can be planted to increase the resistance to
erosion. The shrubs Salsola genistoides and Anthyllis cytisoides
are resistant to concentrated flow erosion and bending by water
flow, and can improve slope stability (Table IV). Whereas
Anthyllis cytisoides is often found on north-facing slopes,
Salsola genistoides is observed on all slopes (Table VI). When
Anthyllis cytisoides can develop enough stems, sedimentation
can also occur. Whereas Anthyllis cytisoides can only reproduce
by seeds, Salsola genistoides is able to resprout after fire, grazing or cutting (Table V). In highly eroded slopes under severe
water stress, such as south-facing roadcuts or badland slopes,
sprouting may be an advantageous plant strategy. Yet, seedling
bending by water flow occurs only occasionally (GarcíaFayos et al., 2000). Seedling establishment is not limited by
seed availability, but by microsite conditions. Soil water availability and salinity are the main factors responsible for differences in seed emergence and seedling survival. The main
cause of mortality of seedlings on roadcut slopes in semi-arid
Spain is drought (García-Fayos et al., 2000). Of the grasses,
Stipa tenacissima and Helictotrichon filifolium can colonize
steep slopes (Table VI). Helictotrichon filifolium prefers north
slopes, because its seeds can not germinate when temperatures are too high (Tong, 1990). Hence, this species has a
small applicability. Stipa tenacissima is drought tolerant and
mostly grows in bare areas. Moreover, this species resprouts
after grazing, fire or cutting (Table V). The sediment obstruction potential reported in this study for Stipa tenacissima is
rather low (Table II). This can be attributed to the umbrella
shape morphology of the stems, resulting in a small area
occupied by the base of the stems relative to the area covered
by the vertical orthogonal projection of the stems. Stipa
tenacissima can also be used to prevent failure of abandoned
terraces as it can reinforce the soil to a large extent (De Baets
et al., 2008) and as it has a good network of fine roots in the
topsoil to prevent concentrated flow erosion (De Baets et al.,
2007b).
Recommendations
Figure 6 illustrates the position in the landscape were suitable
plants for rill and gully erosion control grow naturally. Fast
establishing plant species that provide a good cover, such as
Plantago albicans, are recommended for immediate soil protection after abandonment or destruction by fire. Brachypodium
retusum also provides a good cover, but prefers to grow in
shadow-rich places. Later on, late-successional shrubs or grasses,
such as Rosmarinus officinalis or Piptatherum miliaceum can
colonize abandoned fields or their growth can be encouraged
by planting, hydroseeding or irrigation. Plant growth speed
should also be considered when selecting species for soil
restoration works. However, quantitative information on growth
speed of Mediterranean plants is limited, but some indications based on personal observations are given in Table V.
Rosmarinus officinalis for example needs more than 3–5 years
to become an adult plant and prefers to grow in concavities
at the foot of steep slopes, whereas Piptatherum miliaceum
grows faster (<1 year for standard size) and can be planted in
strips between crops or trees, at downslope borders of fields
or in natural drainage lines to prevent erosion by concentrated
flow. Especially the root system of this grass species is effective for this purpose. Avenula bromoides also grows fast
(<1 year for standard size) and can be planted along road
tracks to prevent concentrated flow erosion (Table V).
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
1390
EARTH SURFACE PROCESSES AND LANDFORMS
Figure 6. A landscape cross-section with indication of the natural occurence of plants suitable for erosion rill and gully control: (1) Brachypodium
retusum on shadow-rich north-facing slopes, (2) Salsola genistoides and Stipa tenacissima on steep north or south-facing slopes, (3) Juncus acutus
and Tamarix canariensis on ephemeral channel bottoms, (4) Lygeum spartum at gully bottoms and gully outlets, (5) Avenula bromoides along
unpaved roads, (6) Pipthaterum miliaceum and Plantago albicans on abandoned cropland and (7) Retama sphaerocarpa on river banks.
Salsola genistoides and Stipa tenacissima have a high potential
to withstand drought and destruction by fire or grazing and
are able to grow on steep slopes. It is therefore recommended
to use these species to re-vegetate steep slopes. However, it can
take more than five years to become a full-grown standardsize Stipa or Salsola plant (Table V). Planting or sowing species
perpendicular to the dominant flow direction provides an
optimal resistance and increases the capability of plants to
trap sediment and organic debris.
To protect terraces or banks and decrease their failure risk,
vegetating them with grass species, such as Stipa tenacissima
or Lygeum spartum, is recommended. To stabilize gully walls
or steep slopes and prevent shallow mass movements a
combination of a grass species reinforcing the topsoil to a
great extent and a deep-rooted species providing a high
reinforcement at greater depth, such as Salsola genistoides or
Anthyllis cytisoides is recommended. To stabilize river banks,
Retama sphaerocarpa is put forward.
To prevent incision by concentrated flow in zones where
runoff concentrates (i.e. natural drainage lines) establishing
grass buffer strips of grassed waterways with Lygeum spartum
will offer a good protection against erosion by concentrated
flow. It is a drought and salt tolerant species. If plant density
is large enough, this species has a high ability to trap sediments and organic debris. Lygeum spartum grasses can also
be planted in channels at gully inlets, with the aim of trapping the incoming fine sediments. It has to be acknowledged
that it can take more than 3–5 years for a standard-size
Lygeum clump to establish (Table V).
Tamarix canariensis or Juncus acutus are suitable species to
vegetate ephemeral channel bottoms. Tamarix canariensis can
be grown in less than five years (Table V), has the potential to
resist extreme floods and provides the soil with extra cohesion,
while Juncus acutus will be most effective for preventing
erosion by concentrated flow and has a high potential to trap
sediments. Both species need moist environments to grow
and survive.
From individual plant level to landscape level
It has to be remarked that plants were selected on an individual basis in this study and not on their performance or
spatial distribution within the landscape. However, the spatial
Copyright © 2009 John Wiley & Sons, Ltd.
organization of plants is important for predicting the effects of
plants on a hillslope or catchment scale and needs more
attention in future research.
Conclusions
The selection of suitable plant species depends on the process
of interest. Four main criteria were put forward in this study
to meet the requirements for controlling rill and gully erosion,
i.e. plants having (1) a high potential to prevent incision by
concentrated flow erosion, (2) the potential to improve slope
stability, (3) the potential to resist bending by water flow and
(4) the ability to trap sediments and organic debris. The scoring of plants on these criteria was based on a multi-criteria
analysis and made after measuring stem density, sediment
obstruction potential, modulus of elasticity of the stems, moment
of inertia of the stems, topsoil root density, root diameter distribution, root area ratio of deeper soil layers and root tensile
strength. All information was graphically summarized in amoeba
diagrams (Figure 5). The major advantage of the methodological framework developed in this study is the use of
relatively few and easily measured plant traits. A combination
of species (e.g. on the one hand a grass having a high potential
to resist concentrated flow erosion and a high ability to trap
sediments and on the other hand a shrub with a high resistance
to bending by water flow and a high potential to improve slope
stability) or the allocation of species to specific target areas
(e.g. grasses in concentrated flow zones and on terrace walls,
deep-rooted species to stabilize gully walls) is recommended.
The reed Juncus acutus, the grasses Stipa tenacissima and
Lygeum spartum and the shrub Salsola genistoides amongst
others, were selected as very suitable plant species for gully
erosion control. Stipa tenacissima can be used to re-vegetate
abandoned terraces as this species is adapted to drought and
offers a good protection to concentrated flow erosion and shallow
mass movements. On north-facing terraces, Brachypodium
retusum can be established as well. Lygeum spartum can be
used to vegetate concentrated flow zones or to obstruct sediment inflow to channels at gully outlets. Whereas Retama
sphaerocarpa can be used to stabilize riverbanks, Stipa
tenacissima and Salsola genistoides can be used to stabilize
steep south-facing slopes. Helictotrichon filifolium and Anthyllis
cytisoides can be established to improve slope stability on
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
METHODOLOGICAL FRAMEWORK TO SELECT PLANT SPECIES FOR CONTROLLING RILL AND GULLY EROSION
north-facing slopes. The rush Juncus acutus and the tree
Tamarix canariensis can be established to prevent erosion of
ephemeral channel bottoms.
Acknowledgements—This research was part of the RECONDES
(Conditions for Restoration and Mitigation of Desertified Areas Using
Vegetation) project funded by the European Commission, DirectorateGeneral of Research, Global Change and Desertification Programme,
Project No. GOCE-CT-2003-505361 (2004–2007). Thanks goes to
the Research Fund of the K.U. Leuven for providing a post-doctorate
fellowship and to Gonzalo Gonzáles Bárbera and José Antonio Navarro
Cano for providing background information on the selected plant
species. Special thanks go to Flip Bamelis and Kristof Mertens of the
Laboratory for Agricultural Machinery and Processing (K.U. Leuven)
for their assistance in making the UTS apparatus operational for
bending plant stems and pulling plant roots.
References
Albaladejo J, Castillo V, Roldan A. 1996. Rehabilitation of degraded
soils by water erosion in semiarid environments. In Soil Degradation
and Desertification on Mediterranean Environments, Rubio JL,
Calvo A (eds). Geoforma Ediciones: Logroño; 265–278.
Alcaraz Ariza F, Botias Pelegrin M, Garcia Ruiz R, Rios Ruiz S, Rivera
Nunez D, Robledo Miras A. 2002. Flora Basica de la region de
Murcia. Novograf: Murcia.
Alegre J, Toledo JL, Martinez A, Mora O, De Andres EF. 1998. Rooting
ability of Dorycnium spp. under different conditions. Scientia
Horticulturae 76: 123–129.
Archer N, Hess T, Quinton J. 2002. Belowground relationships of soil
texture, roots and hydraulic conductivity in two-phase mosaic vegetation in south-east Spain. Journal of Arid Environments 52: 535–
553.
Béchet G, Nedjraoui D, Djebaili S. 1982. Valeur énergétique des
principales espéces des hautes plaines steppiques de la Wilaya de
Saida. Biocénoses, Bulleting d'écologie terrestre 1: 79–94.
Bochet E, García-Fayos P. 2004. Factors controlling vegetation establishment and water erosion on motorway slopes in Valencia, Spain.
Restoration Ecology 12: 166–174.
Bochet E, Poesen J, Rubio JL. 1998. Relative efficiency of tree
representative matorral species in reducing water erosion at the
microscale in a semi-arid climate (Valencia, Spain). Geomorphology
23: 139–150.
Bochet E, Poesen J, Rubio JL. 2006. Runoff and soil loss under
individual plants of a semi-arid Mediterranean shrubland: influence
of plant morphology and rainfall intensity. Earth Surface Processes
and Landforms 31: 536–549.
Bochet E, García-Fayos P, Alborch B, Tormo J. 2007. Soil water availability effects on seed germinations account for species segregation
in semiarid roadslopes. Plant and Soil 295: 179–191.
Brindle FA. 2003. Use of native vegetation and biostimulants for
controlling soil erosion on steep terrain. Journal of the Transportation
Research Board, Eighth International Conference on Low-volume
Roads 1: 203–209.
Brix H. 1997. Do macrophytes play a role in constructed treatment
wetlands. Water Science Technolology 35: 11–17.
Buijs A. 2000. Statistiek om Mee Verder te Werken. Stenfert Kroese:
Groningen.
Caravaca F, Alguacil MM, Vassileva M, Diaz G, Roldán A. 2004. AM
fungi inoculation and addition of microbially-treated dry olive cakeenhanced afforestation of a desertified Mediterranean site. Land
Degradation & Development 15: 153–161.
Cerda A. 1997. The effect of patchy distribution of Stipa tenacissima
L. on runoff and erosion. Journal of Arid Environments 36: 37–51.
Cerda A. 1998. Soil aggregate stability under different Mediterranean
vegetation types. Catena 32: 73–86.
Clary J, Save R, Biel C, De Herralde F. 2004. Water relations in
competitive interactions of Mediterranean grasses and shrubs. Annals
of Applied Botany 144: 149–155.
Clevering OA, Brix H, Lukavska J. 2001. Geographic variation in growth
responses in Phragmites australis. Aquatic-Botany 69: 89–108.
Copyright © 2009 John Wiley & Sons, Ltd.
1391
Cohn JP. 2005. Tiff over Tamarisk: can a nuisance be nice, too?
Bioscience 55: 648–654.
De Baets S, Poesen J, Gyssels G, Knapen A. 2006. Effects of grass
roots on the erodibility of topsoils during concentrated flow.
Geomorphology 76: 54–67.
De Baets S, Poesen J, Knapen A, Galindo Morales P. 2007a. Impact
of root architecture, soil characteristics and flow shear stress on the
erosion-reducing potential of roots during concentrated flow. Earth
Surface Processes and Landforms 32: 1323–1345.
De Baets S, Poesen J, Knapen A, Gonzáles Barberá G, Navarro JA.
2007b. Root characteristics of representative Mediterranean plant
species and their erosion-reducing potential during concentrated
runoff. Plant and Soil 294: 169–183.
De Baets S, Poesen J, Reubens B, Muys B, De Baerdemaecker J,
Wemans K. 2008. Root tensile strength and root distribution of
typical Mediterranean plant species and their implications for
increasing soil shear strength. Earth Surface Processes and Landforms
305: 207–226.
De Luis M, Raventos J, Cortina J, Gonzales-Hidalgo JC, Sanchez JR.
2004. Fire and torrential rainfall: effects on the perennial grass
Brachypodium retusum. Plant Ecology 173: 225–232.
DiTomaso JM. 1998. Impact, biology, and ecology of saltcedar
(Tamarix spp.) in the southwestern United States. Weed-Technology
12: 326–336.
Dunn GH, Dabney SM. 1996. Modulus of elasticity and moment of
inertia of grass hedge stems. Transactions of the American Society
of Agricultural Engineers 39: 947–952.
El-Keblawy A, Shaltout KH, Lovett DJ, Ramadan A. 1997. Population
dynamics of an Egyptian desert shrub Thymelaea hirsuta. Canadian
Journal of Botany 75: 2027–2037.
FAO. 2005. New_LocClim software – local climate estimator,
Environment and Natural Resources Working Paper No. 20. Environment and Natural Resources Service, FAO: Rome (CD-ROM).
Friederici P. 1995. The alien saltcedar. American Forests 101: 45–47.
Frietag H. 1971. Die naturlcihe vegetation des sudost-spanischen
Trockengebietes. Botanik Jahrbuch 91: 11–310.
Garcia G, Faz A, Cunha M. 2004. Performance of Piptatherum
miliaceum (Smilo grass) in edaphic Pb and Zn phytoremediation
over a short growth period. International Biodeterioration & Biodegradation 54: 245–250.
García-Fayos P, García-Ventoso B, Cerdà A. 2000. Limitations to
plant establishment on eroded slopes in southeastern Spain. Journal
of Vegetation Science 11: 77–86.
Gasque M, García-Fayos P. 2004. Interaction between Stipa tenacissima
and Pinus halepensis: consequences for reforestation and the
dynamics of grass steppes in semi-arid Mediterranean areas. Forest
Ecology and Manangement 189: 251–261.
Gökbulak F. 2003. Comparision of growth performance of Lolium
perenne L., Dactylis glomerata L. and Agropyron elongatum (Host.)
P. Beauv. for erosion control in Turkey. Journal of Environmental
Biology 24: 45–53.
González-Hildago JC, Raventos J, Echevarria MT. 1997. Comparison
of sediment ratio curves for plants with different architectures.
Catena 29: 333–340.
Goodman AM, Crook MJ, Ennos AR. 2001. Anchorage mechanics of
the tap root system of winter-sown oilseed rape (Brassica napus L.).
Annals of Botany 87: 397–404.
Gray DH, Sotir RB. 1996. Biotechnical and Soil Bioengineering Slope
Stabilization: A Practical Guide for Erosion Control. John Wiley
and Sons: Toronto.
Greenway DR. 1987. Vegetation and slope stability. In Slope Stability:
Geotechnical Engineering and Geomorphology, Anderson MG,
Richards KS (eds). John Wiley and Sons: Chichester; 187–230.
Gyssels G, Poesen J, Bochet E, Li Y. 2005. Impact of plant roots on
the resistance of soils to erosion by water: a review. Progress in
Physical Geography 29: 189–217.
Haase P, Pugnaire FI, Fernandez EM, Puigdefabregas J, Clark SC,
Incoll LD. 1996. An investigation of rooting depth of the semiarid
shrub Retama sphaerocarpa (L.) Boiss. By labelling of ground water
with a chemical tracer. Journal of Hydrology 177: 23–31.
Hara T, Van der Toorn J, Hook JH. 1993. Growth dynamics and size
structure of shoots of Phragmites australis, a clonal plant. Journal of
Ecology 81: 47–60.
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp
1392
EARTH SURFACE PROCESSES AND LANDFORMS
Haslam SM. 1972. Phragmites communis Trin. The Journal of Ecology
60: 585–610.
Herrera J. 1991. The reproductive-biology of a riparian Mediterranean
shrub, Nerium oleander L. (Apocynaceae). Botanical Journal of the
Linnean Society 106: 147–172.
Hidalgo JCG, Raventos J, Echevarria MT. 1997. Comparison of
sediment ratio curves for plants with different architectures. Catena
29: 333–340.
Hooke JM, Sandercock PJ, van Wesemael B, Meerkerk A, Torri D,
Borselli L, Salvador MP, Castillo V, Barberá GG, Navarro-Cano JA,
Cammeraat LH, Lesschen LP, Poesen J, De Baets S, Marchamalo M.
The connectivity approach to sustainable land management and
soil erosion reduction in Mediterranean lands. Landscape Ecology.
Submitted for publication.
Isselin-Nondedeu F, Bédécarrats A. 2007. Influence of alpine plants
growing on steep slopes on sediment trapping and transport by
runoff. Catena 71: 330–339.
Kouwen N, Li R, Members A. 1980. Biomechanics of vegetative
channel linings. Journal of the Hydraulics Division 106: 1085–1103.
Kouwen N, Li RM, Simons DB. 1981. Flow resistance in vegetated
waterways. Transactions of the American Society of Agricultural
Engineers 24: 684–698.
Lesschen JP, Kok K, Verburg PH, Cammeraat LH. 2007. Identification
of vulnerable areas for gully erosion under different scenarios of
land abandonment in southeast Spain. Catena 71: 110–121.
Lesschen JP, Cammeraat LH, Nieman T. 2008. Erosion and terrace
failure due to agricultural land abandonment in a semi-arid environment. Earth Surface Processes and Landforms 33: 1574–1584.
Lopez-Bermudez F, Barbera GG, Alonso-Sarria F, Belmonte Serrato F.
2002. Natural resources in the Guadalentin basin (south-east
Spain): water as a key factor. In Mediterranean Desertification, A
Mosaic of Processes and Responses, Brandt CJ, Geeson NA,
Thornes JB (eds). John Wiley and Sons: Chichester; 233–245.
Maestre FT, Cortina J. 2002. Spatial patterns of surface soil properties
and vegetation in a Mediterranean semi-arid steppe. Plant and Soil
241: 279–291.
Mant J. 2002. Vegetation in Ephemeral Channels of Southeast
Spain: Its Impact on and Response to Morphological Change,
PhD Thesis, Department of Geography, University of Portsmouth;
259–281.
Martinez-Raya A, Duran Zuazo VH, Francia Martinez JR. 2006. Soil
erosion and runoff response to plant-cover strips on semiarid
slopes (SE Spain). Land Degradation and Development 17: 1–11.
Matsushita N, Matoh T. 1991. Characterization of Na+ exclusion
mechanisms of salt tolerant reed plants in comparison with saltsensitive rice plants. Plant Physiology 83: 170–176.
Mattia C, Bischetti GB, Gentile F. 2005. Biotechnical characteristics
of root systems of typical Mediterranean species. Plant and Soil
278: 23–32.
Morgan RPC. 2005. Soil Erosion and Conservation, 3rd edition.
Blackwell Science: Oxford.
Navarro-Cano JA. 2004. Plant Species and Vegetation in Channels
and Gullies in Cárcavo Basin, Internal Report for RECONDES
Project. University of Portsmouth: Portsmouth.
Copyright © 2009 John Wiley & Sons, Ltd.
Poesen J, Nachtegaele J, Verstraeten G, Valentin C. 2003. Gully erosion
and environmental change: importance and research needs. Catena
50: 91–133.
Pugnaire F, Haase P. 1996. Comparative physiology and growth of
two perennial tussock grass species in a semi-arid environment.
Annals of Botany 77: 81–86.
Pugnaire FI, Luque MT, Armas C, Gutiérrez L. 2006. Colonization
processes in semi-arid Mediterranean old-fields. Journal of Arid
Environments 65: 591–603.
Puigdefabregas J, Alonso JM, Delagro L, Domingo F, Cueto M,
Gutiírrez L, Lázaro R, Nicolau JM, Sùanchez G, Solé A, Vidal S.
1996. The Rambla Honda field site; interactions of soil and
vegetation along a catena in semi-arid southeast Spain. In Mediterranean Desertification and Land Use, Brandt J, Thornes JB (eds).
John Wiley and Sons: Chichester.
Quinton JN, Morgan RPC, Archer NA, Hall GM, Green A. 2002.
Bioengineering principles and desertification mitigation. In Mediterranean Desertification, A Mosaic of Processes and Responses,
Brandt CJ, Geeson NA, Thornes JB (eds). John Wiley and Sons:
Chichester; 93–105.
Reubens B, Poesen J, Danjon F, Geudens G, Muys B. 2007. The role
of fine and coarse roots in shallow slope stability and soil erosion
control with a focus on root system architecture: a review. Trees –
Structure and Function 21: 385–402.
Rey F. 2003. Influence of vegetation distribution on sediment yield in
forested marly gullies. Catena 50: 549–562.
Robledo A, Rios S, Correal E. 1990–1991. Estimación de biomasa en
los matorrales de albaida (Anthyllis cytisoides) del Suresta de
Espana. Pastos 20–21: 107–129 (in Spanish).
Sandercock PJ, Hooke JM. Vegetation zonation and conditions for
growth of plant species in ephemeral stream channels in SE Spain.
River Research and Applications. Submitted for publication. a.
Sandercock PJ, Hooke JM. Assessment of vegetation effects on
hydraulics and of feedbacks on plant survival and zonation in
ephemeral channels. Hydrological Processes. Submitted for
publication. b.
Sandercock PJ, Hooke JM, Mant JM. 2007. Vegetation in dryland
river channels and its interaction with fluvial processes. Progress in
Physical Geography 31: 107–129.
Simon A, Collison A. 2001. Scientific Basis for Streambank Stabilization
using Riparian Vegetation. Proceedings of the Seventh Federal
Interagency Sedimentation Conference, 25–29 March, Reno, NV;
47–54.
Snogerup S. 1993. A revision of Juncus subgen. Juncus (Juncaceae).
Willdenowia 23: 23–73.
Styczen ME, Morgan RPC. 1995. Engineering properties of vegetation.
In Slope Stabilization and Erosion Control: A Bioengineering Approach,
Morgan RPC, Rickson RJ (eds). E&FN Spon: London; 5–58.
Tong STY. 1990. Growth analysis of Aristida coerulescens and
Helictotrichon filifolium seedlings under simulated temperature and
nutrient conditions. Journal of Arid Environments 19: 161–167.
Wu TH, McKinnell III WP, Swanston DN. 1979. Strength of tree roots
and landslides on Prince of Wales Island, Alaska. Canadian
Geotechnical Journal 16: 19–33.
Earth Surf. Process. Landforms 34, 1374–1392 (2009)
DOI: 10.1002/esp