You are on page 1of 11

Effects of 30-year-old Aleppo pine plantations on runoff, soil erosion

,
and plant diversity in a semi-arid landscape in south eastern Spain
E. Chirino
a,
*
, A. Bonet
b
, J. Bellot
b
, J.R. Sa´nchez
b
a
Fundacio´n Centro de Estudios Ambientales del Mediterra´neo (CEAM), Spain
b
Department of Ecology, University of Alicante, Apdo. 99, 03080, Alicante, Spain
Received 9 March 2004; received in revised form 31 August 2005; accepted 13 September 2005
Abstract
Forest management policies in Mediterranean areas have traditionally encouraged land cover changes, with the establishment of tree cover
(Aleppo pine) in natural or degraded ecosystems for soil conservation purposes: to reduce soil erosion and to increase the vegetation
structure. In order to evaluate the usefulness of these management policies on reduced erosion in semi-arid landscapes, we compared 5
vegetation cover types (bare soil, dry grassland, shrublands, afforested dry grasslands and afforested thorn shrublands), monitored in 15
hydrological plots (8Â2 m), in the Vento´ s catchment (Alicante, SE Spain), over 4 years (1996 to 1999). Each cover type represented a
different dominant patch of the vegetation mosaic on the north-facing slopes of this catchment. The results showed that runoff coefficients of
vegetated plots were less than 1% of the precipitation volume; whereas runoff in denuded areas was nearly 4%. Soil losses in vegetation plots
averaged 0.04 Mg ha
À1
year
À1
and increased 40-fold in open-land plots. The evaluation of these forest management policies, in contrast with
the natural vegetation communities, suggests that: (1) thorn shrublands and dry grassland communities with vegetation cover could control
runoff and sediment yield as effectively as Aleppo pine afforestation in these communities, and (2) afforestation with a pine stratum improved
the stand’s vertical structure resulting in pluri-stratified communities, but reduced the species richness and plant diversity in the understorey
of the plantations.
D 2005 Elsevier B.V. All rights reserved.
Keywords: Runoff; Erosion rate; Afforestation; Semi-arid; Soil erosion
1. Introduction
During the 1950s and 1960s, Aleppo pine (Pinus
halepensis Miller) plantation over low cover formations
was the most widespread forest administration management
policy in most areas of Spain (Valle and Bocio, 1996;
Vallejo, 1997; Cortina et al., 2004). The medium and long-
term purpose of these afforestation policies was to recover
the vegetation cover for soil conservation and to reduce the
effects of soil erosion. With these aims, about 2.5 million
hectares in Spain, have been afforested over the last 50
years; that is approximately 5% of the national territory
(Vilagrosa et al., 1997). The II National Forest Inventory
(IFN) of the Ministerio de Medio Ambiente in Spain (1998)
reported that Aleppo pine forest had increased the woodland
surface over all in Spain by 75%. Aleppo pine forest
occupies practically all of the woodland surface of the
Spanish semi-arid region. Aleppo pine forest constitutes the
dominant forest species in the Alicante province covering
more than 90% of the forested area. As a consequence of
this forest policy the semi-arid lands have become a patch-
mosaic landscape of a mixed formation of grasslands and
shrublands with afforested woodlots, occupying the north-
facing slopes, where water availability is less restricted.
Whereas Alpha grass steppes of Stipa tenacissima L. are
widespread in south-facing slopes and the presence of pine
trees and other shrub species is less significant (Bonet et al.,
2001). Recently, the increase of abandoned land has
changed this patch-mosaic as described by Reynolds et al.
0341-8162/$ - see front matter D 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.catena.2005.09.003
* Corresponding author. CEAM - Department of Ecology, University of
Alicante, Apdo. 99, 03080, Alicante, Spain. Tel.: +34 965903555; fax: +34
965909832.
E-mail address: esteban.chirino@ua.es (E. Chirino).
Catena 65 (2006) 19 – 29
www.elsevier.com/locate/catena
(1999), introducing new patches for woodland connectivity
or for overall landscape fragmentation (Bonet et al., 2004).
The performance of Aleppo pine forest has been strongly
related to water balances and the disturbance regime in
Mediterranean conditions (Zavala et al., 2000). Moreover, it
has also been suggested that Aleppo pine plantations in
extreme water stressed arid conditions could not develop
self-persistent populations, mature and multi-layer commu-
nities, unlike the situation in other wetter Mediterranean
areas (Rivas-Martı ´nez, 1987; De la Torre, 1988; De la Torre
and Alı ´as, 1996; Costa, 1987), in spite of their drought
resistant characteristics in dryer regions (Schiller and
Cohen, 1998; Atzmon et al., 2004). This is probably due
to low growth and unsuccessful regeneration dynamics in
extreme water stressed events (Ravento´ s et al., 2001).
The objective of the work presented here is to evaluate
afforestation using Aleppo pine as the appropriate recovery
strategy in semi-arid regions, in contrast with the natural
recovery of these areas, considering the effects on the
magnitude of runoff, soil losses, plant cover, species
richness and plant diversity. This management strategy is
based on the hypotheses that erosion decreases exponen-
tially with vegetation cover (Elwell and Stocking, 1976),
despite the work of Francis and Thornes, which indicates
that this hypotheses is true but more-or-less independent of
cover type (Francis and Thornes, 1990a,b), and that the
advantages of Aleppo pine afforestations are only a little
above those of natural and anthropogenically degraded
shrubland and grassland (Francis and Thornes, 1990b).
From this perspective it was expected that the effects of an
Aleppo pine stratum planted 30 years ago, would signifi-
cantly reduce erosion, in contrast with the areas without
active management, where the regeneration of the plant
cover only took place slowly through natural vegetation
dynamics (Bonet, 2004). To evaluate the previous hypoth-
esis, we analysed five land-cover types that represented the
composition of the patch-mosaic landscape in north-facing
semi-arid environments (Bonet et al., 2001, 2004). In the
study area, north-facing afforestations with Aleppo pine
tress had been more successful in extent, growth and
survival than in south-facing aspects (Cortina et al., 2004;
Bonet et al., 2004). These cover types included a gradient
from more degraded (open areas without vegetation) to less
degraded (grassland and shrubland) landscape units and the
result of afforestation practices. The aim of this paper is to
evaluate the effectiveness of Aleppo pine afforestation as a
common technique to restore plant cover, facilitate the
establishment of other understorey species, and reduce the
effects of erosion in a semi-arid landscape.
2. Area description
The research was carried out in the Vento´ s-Agost
Catchment Experimental Station (University of Alicante),
located in the Municipality of Agost, Alicante Province, SE
Spain (38-28VN, 0-37VW). The study area covers approxi-
mately 537 ha, with altitudes ranging around 600 m a.s.l.,
and slopes between 23- and 26-. The climate is semi-arid
Mediterranean, with a very high inter-annual variability.
Mean annual temperature is 18.2 -C and annual rainfall is of
291.7 mm (mean of the 1961–1999 period), most of which
falls in autumn (Pe´rez Cuevas, 1994). Lithology is loam and
loamy limestone supporting calcareous Cambisol (FAO-
UNESCO). The soils are characterised as shallow (0–30
cm), with a medium content of organic matter (5.7%),
between 48% and 58% of clay content, soil bulk density of
1246.0 kg m
À3
, and a soil particle density of 2352.0 kg
m
À3
. The hydraulic performance of the soil is determined
by a total porosity of 46.0%, field capacity of 25.0%, a
saturated soil hydraulic conductivity of 3.37 cm s
À1
, and
average infiltration rate of 512.6 L m
À2
h
À1
. Resistance to
the penetration of a cone of 0–30 cm of depth (18%
humidity) varies between 31 and 97 kPa (Derouiche et al.,
1996; Bellot et al., 2001).
The plantation period lasted from 1960 to 1983, but was
especially intense during the first decade. The Forest
Service afforested the area with P. halepensis Miller
woodlots over land degraded grasslands and shrublands,
maintaining 30% of the territory managed with no inter-
vention. Afforestation was carried out by planting in holes
made by hand, with a mean plantation density of 5280T692
trees ha
À1
over grasslands and 3733T1097 trees ha
À1
over
shrublands. In these areas forest management was restricted
to a minimum selective pruning, and no thinning was
carried out after the treatments.
The present vegetation landscape in the study area is
composed of different land cover types (patch-mosaic) that
can be situated in a successional and structural vegetation
complexity gradient: degraded open land or bare soil (B),
sometimes occupied by microphytic crusts (Maestre et al.,
2002); dry grassland formations ( G) of Brachypodium
retusum Pers. Beauv., with dwarf scrubs (Anthyllis cyti-
soides L., Helianthemum syriacum (Jacq.) Dum.-Cours. and
Thymus vulgaris L.); and more mature landscape patches
composed of scattered thorn and sclerophyllous shrublands
(S) with Quercus coccifera L., Pistacia lentiscus L., Erica
multiflora L., Rhamnus lyciodes L. and Rosmarinus
officinalis L. (Bonet et al., 2001). Other anthropic compo-
nents of the present landscape mosaic are afforested dry
grasslands (AG) and afforested thorn shrublands (AS).
3. Methods and experimental design
The research was conducted over a 4-year period (1996–
1999) and 30 years after the plantation work. In each land-
cover type (B, G, S, AG and AS) we installed 3 hydrological
plots of 16 m
2
(8Â2 m) randomly distributed in each
landscape patch. After each rain event, the volumes of
runoff and soil loss were collected and measured in the
Gerlach box of 2 m width, installed at the bottom of each
E. Chirino et al. / Catena 65 (2006) 19–29 20
plot. The precipitation and other climatic variables (temper-
ature, radiation, wind, and air humidity) were measured and
registered continuously by means of an automatic meteoro-
logical station (Campbell CR10). The analysis of annual
runoff coefficient and erosion rates were calculated consid-
ering 4 years (1996–1999), and the one-way ANOVA was
used for statistical analyses of vegetation type responses
during 1998 and 1999. This analysis also was carried out in
two groups of rainfall events, those exceeding 10 mm and
those below 10 mm.
In order to determine the afforestation effect on
vegetation structure, the vegetation survey was performed
in two ways to characterise the hydrological plots and the
landscape patches, respectively. First, each hydrological plot
was divided into eight subplots (2 m
2
) to determine the
cover (%) of each plant species, stoniness, litter cover and
bare soil. As a functional indicator of vegetation community
structure, we also recorded the leaf area index (LAI), which
was estimated by means of a destructive analysis in 0.25 m
2
plots close to hydrological plots. The second vegetation
survey was carried out on landscape patches over a wider
scale during the spring of 1998, using 5 stands (as
replicates) of 100 m
2
in each land cover type. On each
stand, 10 squares (1 mÂ1 m) were placed in a random
procedure. A complete species list was recorded and the
species cover (%) was estimated in each through the
projection of each plant species in the soil horizontal plane
by means of a grid (5Â5 cm unit). Species richness and
Shannon’s diversity index (HV), were calculated on each
plot. Species were grouped in life forms (trees, shrubs,
dwarf shrubs, and herbs: perennial grasses, perennial forbs
and annuals). The differences between the vegetation
characteristics in the hydrological plots and the landscape
patches surveys were analysed by using the Kruskall-Wallis
test and the posteriory Wilcoxon test. The statistical
analyses on plant diversity and hydrological variables were
carried out by doing a one-way ANOVA on log transformed
data. The statistical comparison of the vegetation cover
types in the hydrological plots was carried out by Principal
Components Analysis—PCA (Statistic, version 3.0). Vege-
tation composition and cover of hydrological plots was
analysed through means of Detrended Correspondences
Analysis—DCA (Ter Braak and Colin Prentice, 1988) using
CANOCO version 4 (Ter Braak and S
ˇ
milauer, 1998).
4. Results and discussion
4.1. Analysis of the physical characteristics and vegetation
composition of hydrological plots
In order to prove the homogeneity of the characteristics of
the different hydrological plots in the cover types analysed,
we carried out a Principal Components Analysis (PCA)
using several variables that characterised the plots and the
accumulated values observed in runoff and soil loss treat-
ments. A correlation matrix between the 12 variables (runoff,
soil loss, slope, total LAI, total vegetation cover, trees
stratum cover, shrub stratum cover, herbaceous stratum
cover, moss stratum cover, bare soil surface, stones cover,
and litter cover), ratified the existence of significant
correlation ( p <0.05) between the variables runoff and soil
loss with the total vegetation cover and the bare soil
percentages. After this analysis we rejected those variables
that introduced redundant information in the PCA analysis.
The PCA results (Fig. 1), indicate eigenvalues (k) of 2.93,
2.13 and 1.57 for factors 1, 2 and 3, respectively. The first
factor explains 36.6% of the variance and is composed of
vegetation cover, and trees and shrub stratum; factor 2
explains 26.7% and is made up of the LAI and the
herbaceous stratum and factor 3 explains the 19.6%
represented by the slope.
Grassland plots showed more variability in factors 1 and
2, regarding differences in vegetation cover (Fig. 1A). Shrub
plots also showed some variability in relation to factor 3
(Fig. 1B), as consequence of slope, which ratifies the effect
of this variable in the generation of greater runoff in S in
comparison to G. These results showed that the hydrolog-
ical response in the plots is a function of the vegetation
structure (vegetation strata) and physical characteristics,
mainly slope. The PCA ordination is consistent with the plot
selection by structure and vegetation cover types.
In order to verify the homogeneity of the vegetation of
hydrological plots, a DCA analysis on species composition
and cover was performed using an input matrix of 12
samples (plots with vegetation cover) by 52 species with
224 occurrence samples. DCAwas analysed with detrending
Fig. 1. PCA ordination plot of cover type characteristics showing the
position of each hydrological plot in relation to the main ordination factors.
S, filled circle; AS, open circle; G, open square; AG, filled square.
E. Chirino et al. / Catena 65 (2006) 19–29 21
by segments and downweighting of rare species. Ordination
eigenvalues were higher for the first axis (k
1
=0.320) than
for the second (k
2
=0.139). Cumulative variances of species
data improve greatly through addition of the second and
third axis (c
1
=28.4; c
2
=40.3; c
3
=44.0). The DCA results
(Fig. 2) showed that ordination of vegetation composition is
consistent with the selected plot design. Ordination revealed
a close relationship of species cover between hydrological
plots pertaining to each land cover type. Meanwhile
Brachypodium retusum (Pers.) Beauv. was the most
abundant species in all plots, and specially in G, plots
under this land cover type also presented Coronilla juncea
L. and Fumana ericoides (Cav.) Gand. Cistus albidus L.
was recorded in pine understory of AS and AG, but other
dwarf shrubs were absent in forested plots (Globularia
alypum L., Helichrysum stoechas (L) Moench., Lithodora
fruticosa (L.) Griseb, and Thymus vulgaris L.). On the other
hand, Q. coccifera L. showed high cover values in S and
AS. Despite the existence of a P. halepensis Miller layer,
differences in composition and cover of hydrological plots
between S and AS were due mainly to the abundance of E.
multiflora L. in S and the abundance of P. lentiscus L. and
Rhamnus lycioides L. and the absence of dwarf shrubs in
AS. The first DCA axis is negatively correlated with LAI
(Kendall’s correlation coef. =0.879, p <0.000, n =12),
meanwhile the second axis is negatively correlated with
tree cover (Kendall’s correlation coef. =À0.638, p =0.006,
n =12), and positively correlated with sediment yield
(Kendall’s correlation coef. =0.545, p =0.014, n =12), and
slope (Kendall’s correlation coef. =À0.450, p =0.045,
n =12).
4.2. Effects of afforestation on runoff and erosion
Many authors have studied the relationships between
vegetation cover, runoff and sediment yield (Elwell and
Stocking, 1976; Morgan, 1986; Trimble, 1988; Francis and
Thornes, 1990b; Teixeira and Misra, 1997; Albadalejo et al.,
1998). This relationship is a function of several factors:
rainfall characteristics (volume, intensity, duration and
frequency); soil parameters (physics and hydrophysics);
and also vegetation structure. All the research carried out in
the study zone (Derouiche et al., 1996; Bellot et al., 1998,
2001; Chirino et al., 2001) has indicated that the protective
effect of the vegetation cover controls runoff and soil
erosion. Vegetation cover diminishes the quantity of water
that reaches the soil through the effect of interception
(Bellot and Escarre´, 1998), and also increases the hetero-
geneity of the spatial distribution of the rainwater toward the
soil, through the stemflow and throughfall (Lo´ pez, 1989;
Bellot and Escarre´, 1998).
According to the structure of vegetation cover, some
differences were expected in runoff and soil erosion between
AG, AS, G and S. Through the analysis of the entire research
period we observe the lower values of the annual runoff
coefficients in the vegetated plots (AS, AG, G and S), which
do not exceed 1.0% of precipitation (Table 1), and approx-
imately increased 8-fold in open land plots (B). In conse-
quence the calculated annual soil loss rates (through runoff)
are also low (Table 1). The average values of the annual soil
loss rates were calculated in all vegetated plots (less than 0.05
Mg ha
À1
year
À1
). The open lands (B) obtained higher values
(1.90 Mg ha
À1
year
À1
). Table 1 also expresses the sediment
yield as percentage of bare soil annual values. The result
appears to indicate that afforested communities (AG and AS)
show small mean values with respect to the non-afforested
ones (the half of the G and S, respectively), despite the high
inter annual variability of each cover type.
The statistical analysis of the runoff values and soil
losses (Table 2), detected significant differences only
between the accumulated runoff ( F
Cover type
=8.105,
p =0.004, n =15), and soil loss ( F
Cover type
=16.554,
p =0.000, n =15), of the vegetated plots (AS, AG, S and
G) in comparison to the open lands (B). These results
indicate that in the aforementioned conditions, a 30-year-old
afforestation with Aleppo pine (AS and AG) does not
significantly reduce the runoff and erosion in comparison to
natural vegetation (S and G), which is very important for
management. In summary, these results are entirely consis-
tent with early expectations and we are relieved that there is
no significant difference between cover types.
These results are due to the high percentage of vegetation
cover in non-afforested plots. These values rise to 70–95%
in all (Table 1), which are enough to reduce the runoff and
soil loss in every type of vegetation cover analysed. If we
take into account the estimated values of net precipitation
using the hydrological model VENTOS (Bellot et al., 2001),
31–40% of the rainfall is intercepted by vegetation cover in
plots, reducing both the runoff by 85%, and the soil loss by
95%, with respect to the open lands. Several authors
(Francis and Thornes, 1990a; Cerda`, 1998; Andreu et al.,
1998) have reported similar figures. In our study, shrublands
and grasslands with more than 70% of vegetation cover
protect the soil in the same way as the afforested
communities, however, the AS plots reveal higher protec-
Fig. 2. DCA ordination plot of the plant species cover showing the position
of each hydrological plots. S, filled circle; AS, open circle; G, open square;
AG, filled square.
E. Chirino et al. / Catena 65 (2006) 19–29 22
tion (0.40% of runoff and 0.019 Mg ha
À1
year
À1
of
erosion), but the difference was not statistically significant
from the other vegetated plots. Vacca et al. (2000) estimate
runoff coefficients of 0.65–1.59%, and erosion rates
between 0.03 and 0.05 Mg ha
À1
in plots of 20 m
2
covered
by herbaceous plants and shrubs, while in Eucalyptus sp.
plots (15 years old and 25% vegetation cover) the estimated
rates were 2.01% and 0.19 Mg ha
À1
, respectively. The
registered rates of soil erosion in our vegetated plots are
lower than those reported in other research works carried out
in semi-arid environments of Spain. Romero-Dı ´az et al.
(1988) calculated annual soil losses of 0.08–2.55 Mg ha
À1
year
À1
in a catchment with 35% of vegetation cover with
marls as substrate. In a microcatchment with 60% of
vegetation cover, Albadalejo and Stocking (1989) deter-
mined rates between 0.5 and 1.2 Mg ha
À1
year
À1
, and
Lo´ pez Bermu´ dez et al. (1991) reported annual losses of 0.1
Mg ha
À1
year
À1
in plots with 80% shrub cover. Areas with
reduced vegetation cover (lower than 50%) caused by
human interference or affected by wildfires can increase soil
loss in the first years after disturbance (Sa´nchez, 1997; Soto
and Dı ´az-Fierros, 1997; Bautista, 1999). In our experimental
area, the bare soil (B) registered higher values of runoff
coefficient and soil loss (4.42% and 1.90 Mg ha
À1
year
À1
,
respectively). Cerda` (1998) obtained similar results when
comparing discontinuous Stipa tenacı ´ssima steppes, herb
cover and dwarf shrub communities by means of rainfall
simulation experiments.
4.3. Effects of precipitation characteristics on runoff and
erosion
Besides the vegetation cover (%), rainfall characteristics
determine the runoff and erosion magnitude (Francis and
Thornes, 1990a; Cerda`, 1997). Runoff and loss of soil
caused by rainfall events reflect small values in both
variables, as responses to the vegetation cover and to the
characteristics of the rain. 75% of events in vegetated plots
Table 2
Accumulated runoff and soil loss rates in hydrological plots by vegetation cover type during 1998–1999 years (ANOVA—one way; averageTstandard error.
DMS post hoc)
Vegetation type cover
B G S AG AS
Runoff (L m
À2
) 20.85T15.46 a 2.56T0.36 b 3.36T0.75 b 4.01T3.12 b 2.20T0.15 b
Soil loss (Mg h
À1
) 3.64T2.23 a 0.05T0.00 b 0.07T0.01 b 0.02T0.01b 0.02T0.00 b
Values of runoff and soil loss followed by same letter are not significantly different at p 0.05.
Table 1
Annual runoff coefficient and soil loss in hydrological plots by vegetation type cover
Characteristics Slope and vegetation cover of hydrological plots
B G S AG AS
Slope (-) 22 26 29 23 21
Vegetation cover (%) 0 70 90.2 95.8 95.2
Years Acum. rainfall. (mm) Annual runoff coefficient (%)
B G S AG AS
1996 308.9 nd
a
0.60 0.46 0.45 0.29
1997 483.3 5.36 0.62 0.63 0.62 0.45
1998 288.3 3.80 0.45 0.52 0.80 0.39
1999 241.1 4.11 0.52 0.77 0.71 0.45
Mean 330.4 4.42 0.55 0.60 0.65 0.40
30/09/97
b
134.6 12.54 0.59 0.86 1.02 0.56
Annual soil loss (Mg ha
À1
year
À1
)
1997 483.3 2.070 0.100 0.055 0.088 0.036
1998 288.3 0.873 0.003 0.004 0.003 0.003
1999 241.1 2.761 0.004 0.067 0.014 0.019
Mean 337.6 1.901 0.049 0.042 0.035 0.019
30/09/97 134.6 1.70 0.07 0.05 0.10 0.03
Annual soil loss
(Mg ha
À1
year
À1
)
Annual soil loss (% of bare soil)
1997 483.3 2.070 4.83 2.65 4.25 1.74
1998 288.3 0.873 0.34 0.46 0.34 0.34
1999 241.1 2.761 0.14 2.42 0.50 0.68
Mean 337.6 1.901 2.57 2.21 1.84 1.00
30/09/97 134.6 1.70 4.11 2.94 5.88 1.76
a
Treatment not available.
b
Extraordinary event.
E. Chirino et al. / Catena 65 (2006) 19–29 23
and 53% in bare soils showed runoff values of 0.1 L
m
À2
. During the whole study period we registered 94
rainfall events, with an annual mean of 330.4 mm. 90% of
these events present volumes lower than 10 mm (Fig. 3),
with an average intensity lower than 5 mm h
À1
and short
duration (less than 2 h). These characteristics of precipitation
and the vegetation cover (higher of 70%) may justify the
lower measured runoff and erosion rates. The temporal
distribution of rainfall events is also relevant in this region.
These are concentrated in a few days of very heavy rain. The
rainfall events with more than 10.1 mm (10% of the events)
represent 57% of accumulated annual precipitation volume,
with a mean maximum intensity (I
15
) of 18.9 mm h
À1
and a
mean duration of 3 h. Also storm events are frequent in the
semi-arid area, called ‘‘flood’’, ‘‘cold drop’’ or ‘‘cut-off
low’’, similar to that which occurred on 30/09/97, with 134.6
mm in 7 h 30 min, which represented 24% of the annual
rainfall.
Considering that 90% of the precipitation events do not
exceed 10 mm, and that these events generated between
81% and 89% of the runoff, and between 92% and 96% of
soil loss in the different vegetation cover type; we carried
out a second statistical analysis considering two rainfall sets
(events until 10 mm and rainfall over 10 mm). The results
(not shown) reinforce the previous interpretations detecting
significant statistical differences between the means of the
accumulated runoff and soil loss of the vegetated plots (AS,
AG, S and G) and open land (B), but not between vegetation
communities (Table 2). The effects of the extraordinary
event of 30/09/97 only just increased the values of runoff
and soil loss (Table 1), but not the relationships between
different vegetation cover types.
Another aspect evaluated in this work has been the
contribution of rainfall intensity on runoff and soil losses.
Using multiple regressions to estimate runoff, through
rainfall amount, maximum intensity at 15 min, mean
weighted intensity, and rainfall duration, as independent
variables, we obtain significant correlation coefficients in all
vegetation cover types (Table 3). However, maximum
intensity seems to have made the major contribution,
together with the rainfall amount to explain the daily runoff
and soil loss variability.
4.4. Afforestation effects on vegetation structure, plant life
forms, species richness and diversity
Despite the observed changes in species composition and
cover (Fig. 2), the statistical analysis of the vegetation cover
structure of the hydrology plots (Table 4) indicates that
afforestation with Aleppo pine only modifies the previous
structure introducing the tree strata which was not present in
natural vegetation communities ( G and S). The surface
occupied by the shrub stratum in the AS plots does not
significantly differ from the surface in the S plots, being
significantly smaller in AG. Nor does the surface occupied
by herbaceous stratum in AG differ from the plots with
natural vegetation (S and G), but the differences were
significant between these and AS. On the other hand, we did
not find differences in mosses and bare soil between the
different hydrology plots; but the difference exists with
regard to the surface covered by litter, being similar in AS,
AG and S, and different from G (Table 4).
Table 3
Multiple regression by vegetation cover type
R
2
F b
0
Coefficient of the independent variable
Rainfall Int
15
mw
_
Int D
AS 0.926 268,7838** À1.59eÀ02 5.404eÀ03*** 6.045eÀ03*** À6.07eÀ3*** À1.47eÀ03***
AG 0.896 185,250*** 8.925eÀ05 1.014eÀ02*** 7.192eÀ03** À1.01eÀ02** À2.28eÀ03***
S 0.942 353,265*** À1.20eÀ02 8.281eÀ03*** 7.647eÀ03*** À8.91eÀ03*** À2.09eÀ03***
G 0.884 220,442*** À3.90eÀ02 4.224eÀ03*** 1.149eÀ02*** À1.149e02***
B 0.968 395,198*** 6.089eÀ02 0.148*** À0.111** 9.812eÀ02** À4.69eÀ02***
y =runoff, b
0
=slope. x
1
=rainfall, x
2
=maximum intensity 15 min (Int
15
), x
3
=mean weigh intensity (mw_Int), and x
4
=rainfall duration (D); gl =86.
***p <0.001, **p <0.01.
Fig. 3. Rainfall characteristics. Relationship between rainfall volume and
mean intensity, and between rainfall volume and rainfall duration for each
precipitation event (Vento´ s Meteorological Station, years 1996–1999).
E. Chirino et al. / Catena 65 (2006) 19–29 24
The conflict between pine plantations and conservation
of pre-existing thorn shrublands is a relevant topic in the
management of the Mediterranean semi-arid ecosystem
(Esteve et al., 1990). From the medium to long-term,
Aleppo pine afforestation produces a positive effect in-
creasing the overall vegetation cover with respect to non
afforested cover; which implies similar results regarding
LAI (Table 4). On the other hand, a negative effect of the
afforestation is located in species richness values, being
higher in natural vegetation communities where the estab-
lishment of dwarf shrubs is facilitated, in contrast to the
afforested plots where the tree stratum reduces the presence
of other species in the understory. This effect is contrary to
one of the priorities of the EU Environmental Policy, which
aims at the conservation of biodiversity of these Mediter-
ranean ecosystems (grassland and shrubland), through their
high rates of richness and species diversity (Peco et al.,
1983).
To evaluate the representativeness of these results on
structure, richness and diversity of species obtained in the
hydrology plots, we carried out another complementary
study in landscape patches on the same community types.
Table 5 shows the statistical results of the comparative
analysis between the natural vegetation communities with
their corresponding pairs of Aleppo pine afforestation ( G–
AG and S–AS). Overall richness and diversity present lower
values in afforested than in natural communities. Some
shrubs like R. officinalis were present in S, but not in AS,
Table 5
Vegetation characteristics of natural and afforested communities at landscape patches in the north-facing slopes in the Ventos catchment: Plant cover (%), Mean
species richness (S) and Shannon diversity index (HV) for all cover type and each plant life form
G AG S AS
Mean (s.d.) Mean (s.d.) F p Mean (s.d.) Mean (s.d.) F p
Overall community
Species richness (S) 37.00 (6.16) 19.80 (4.09) 27.04 ** 34.20 (5.36) 26.20 (3.03) 8.44 *
Diversity (HV) 3.00 (0.15) 2.28 (0.18) 48.73 *** 3.35 (0.22) 2.88 (0.17) 13.83 **
Plant cover (%) 71.00 (9.62) 72.60 (9.42) 0.90 ns 85.60 (3.20) 99.20 (0.45) 120.71 ***
Life forms
Species richness (S)
Shrubs 2.20 (1.64) 0.40 (0.55) 5.40 * 6.20 (2.17) 2.60 (1.14) 10.80 *
Dwarf shrubs 17.80 (1.30) 11.00 (1.41) 62.48 *** 15.40 (0.89) 15.20 (1.92) 0.04 ns
Grasses 5.20 (0.84) 1.00 (0.00) 126.00 *** 3.60 (0.55) 1.80 (0.45) 32.40 ***
Forbs 10.00 (3.46) 6.40 (3.05) 3.04 ns 8.40 (4.77) 6.20 (2.59) 0.82 ns
Annuals 1.80 (0.84) 1.00 (0.00) 4.57 ns 0.60 (0.55) 0.00 (0.00) 6.00 ns
Diversity (HV)
Shrubs 0.86 (0.96) 0.00 (0.00) 4.08 ns 1.51 (0.33) 0.77 (0.54) 6.81 *
Dwarf shrubs 3.44 (0.10) 2.86 (0.45) 8.08 * 3.70 (0.05) 3.55 (0.21) 2.54 ns
Grasses 0.73 (0.24) 0.00 (0.00) 45.84 *** 0.97 (0.20) 0.10 (0.20) 45.92 ***
Forbs 2.72 (0.58) 2.36 (0.63) 0.88 ns 2.68 (0.90) 2.32 (0.47) 0.67 ns
Annuals 0.60 (0.60) 0.00 (0.00) 5.01 ns 0.00 (0.00) 0.00 (0.00) – –
Means and standard deviations are calculated in 5-replicated 100 m
2
plots. ANOVA (*p <0.05, **p <0.01, ***p <0.001).
Table 4
Statistical significance of the differences between vegetation cover (%) of the hydrological plots
Mean cover (%)
G S AG AS X
2
p-value
Layers
Trees 0.00 (0.00) 0.00 (0.00) 54.50 a (15.55) 44.05 a (14.40) 13.152 0.011*
Shrubs 9.71 b (8.13) 58.58 a (12.66) 5.10 b (3.82) 72.25 a (0.33) 12.300 0.015*
Herbs 60.92 a (13.72) 72.71 a (13.14) 69.54 a (15.21) 30.46 b (9.43) 10.267 0.036*
Mosses 0.42 (0.45) 0.00 (0.00) 0.27 (0.24) 0.13 (0.22) 5.777 0.216 ns
Bare 7.83 (7.49) 1.00 (1.31) 1.63 (2.06) 0.29 (0.40) 8.102 0.088 ns
Stones 44.29 a (19.62) 12.79 b (9.60) 2.54 c (0.95) 4.50 bc (3.15) 11.867 0.018*
Litter 16.11 b (6.43) 83.92 a (15.32) 95.61 a (3.15) 97.06 a (2.21) 11.583 0.021*
Total plant cover 70.02 c (3.22) 90.17 b (2.10) 95.83 a (2.40) 95.21 a (2.81) 9.359 0.025*
Vegetation characteristics
Species richness 24.33 a (1.15) 25.00 a (2.00) 19.00 b (6.00) 19.00 b (3.46) 9.862 0.043*
LAI 1.197 a (0.112) 2.303 a (0.135) 1.732 ac (0.038) 3.012 b (0.239) 13.233 0.010**
Mean values (s.d.) results of the Kruskall-Wallis test (X
2
) and post hoc Wilcoxon pairs comparison (values of mean cover followed by same letter are not
significantly different at p 0.05; the * in p-values indicates the significance of the Wilcoxon test: *p <0.05, **p <0.01).
E. Chirino et al. / Catena 65 (2006) 19–29 25
and the dominant Q. coccifera showed lower cover values
in afforested plots (Table 6). These effects were especially
evident in G, where mean richness values decreased near
50% in afforested plots AG (Table 5). Moreover, afforesta-
tion with pines increased plant cover values in AS,
increasing practically all the total available soil surface
whereas there was not such an effect on AG. However, plant
cover alone is not considered to be so important. There is a
threshold cover value above which there is no significant
impact on run-off and erosion.
The results indicate that afforested communities had a
different species composition when compared with non-
managed communities. Differences are clearer if we take into
account the plant life forms as crude functional groups (Table
5). We did not find variation in forbs and annuals richness
nor diversity in any of the cases that are compared. Grasses
demonstrate lower richness and diversity values with pine
afforestation in both cases, AG and AS. However, the
perennial grass B. retusum seems to be more constant than
other grasses (Table 6). Moreover, dwarf shrubs presented
lower richness and diversity values with afforestation only in
G. Shrub richness was lower in afforested communities than
in non-afforested ones, whereas we obtained statistically
significant differences in shrub diversity only between S and
AS due to the low shrub abundance in G (Table 4). Most
shrub species had fewer cover values in afforested plots,
except P. lentiscus (a bird-dispersed shrub), which could
have been facilitated through the presence of P. halepensis.
Verdu´ and Garcı ´a-Fayos (1996) described several examples
of facilitation pathways in colonisation dynamics of this
species by means of trees that can act as perches for
frugivorous birds disseminating seeds.
In summary, the grassland communities presented greater
shrub, dwarf shrub and grass species richness than in
grassland afforestation. On the other hand, shrubland
communities presented greater shrub and grass species
richness and overall species diversity, with no differences
in relation to dwarf shrubs. Several authors indicate similar
Table 6
Mean cover (%) of dominant species in analysed vegetation type cover in landscape patches
S AS G AG
Mean (s.d.) Mean (s.d.) Mean (s.d.) Mean (s.d.)
Trees
Pinus halepensis 0.00 (0.00) 50.80 (11.78) 0.00 (0.00) 52.40 (7.60)
Shrubs
Erica multiflora 4.00 (2.92) 0.20 (0.45) 0.00 (0.00) 0.00 (0.00)
Pistacia lentiscus 2.20 (3.49) 9.60 (15.65) 0.20 (0.45) 0.00 (0.00)
Quercus coccifera 47.60 (16.74) 28.20 (6.53) 0.00 (0.00) 0.00 (0.00)
Rhamnus lycioides 1.40 (1.149) 1.20 (2.17) 0.60 (0.55) 0.40 (0.55)
Rosmarinus officinalis 9.00 (4.95) 0.00 (0.00) 0.60 (0.55) 0.00 (0.00)
Dwarf shrubs
Anthyllis cytisoides 1.00 (1.73) 1.40 (1.14) 10.80 (3.03) 1.80 (1.30)
Asparagus horridus 1.00 (0.71) 0.40 (0.42) 1.60 (0.89) 0.30 (0.45)
Atractylis humilis 0.70 (0.45) 1.00 (0.71) 1.40 (2.07) 2.40 (1.34)
Coronilla minima subsp lotoides 2.00 (1.58) 0.30 (0.45) 1.00 (1.22) 0.00 (0.00)
Fumana ericoides 0.80 (0.84) 1.80 (0.84) 0.80 (0.84) 0.00 (0.00)
Globularia alypum 2.00 (2.12) 1.40 (1.52) 0.60 (0.55) 0.40 (0.55)
Helianthemum syriacum 0.60 (0.55) 1.00 (0.71) 4.80 (3.11) 1.40 (1.52)
Helianthemum violaceum 0.60 (0.55) 0.80 (0.76) 1.20 (0.84) 1.40 (1.52)
Helichrysum decumbens 0.30 (0.45) 3.20 (1.64) 1.40 (0.55) 9.40 (7.64)
Phagnalon rupestre 0.00 (0.00) 0.20 (0.45) 1.00 (0.00) 1.80 (0.84)
Phagnalon saxatile 0.00 (0.00) 0.80 (1.30) 3.20 (0.84) 1.20 (1.10)
Teucrium homotrichum 1.00 (0.71) 2.60 (2.41) 0.24 (0.43) 0.40 (0.55)
Thymus vulgaris 1.20 (1.10) 3.40 (1.52) 3.00 (1.87) 1.60 (1.52)
Grasses
Brachypodium retusum 50.60 (18.94) 27.40 (16.15) 65.60 (12.82) 54.80 (9.42)
Lygeum spartum 0.60 (0.55) 0.00 (0.00) 2.40 (1.52) 0.00 (0.00)
Stipa parviflora 0.00 (0.00) 0.00 (0.00) 3.40 (1.95) 0.00 (0.00)
Forbs
Eryngium campestre 0.00 (0.00) 0.30 (0.45) 1.40 (0.89) 0.40 (0.55)
Pallenis spinosa 0.00 (0.00) 0.20 (0.27) 0.60 (0.55) 1.40 (1.67)
Teucrium pseudochamaepitys 0.70 (0.45) 0.60 (0.55) 0.50 (0.50) 0.80 (0.45)
Annuals
Reichardia tingitana 0.80 (0.84) 0.30 (0.45) 0.04 (0.09) 0.00 (0.00)
E. Chirino et al. / Catena 65 (2006) 19–29 26
effects associated with afforestation management. Hartley
(2002) and Watson et al. (2000) detected the decrease of plant
diversity of pre-existing native species after the monospecific
afforestations with pines in non-degraded lands. Sometimes,
current management techniques like thinning or pruning, can
affect the composition of forest species (Freedman et al.,
1996; Wagner et al., 1998). In spite of this, it is recognised
that plantations can be of benefit to landscape composition by
increasing the ecotone density and landscape diversity
(Estades and Temple, 1999; Bonet et al., 2001).
4.5. Relationships between vegetation cover, runoff and soil
loss
In order to contrast these results with the so called 30%
rule (Francis and Thornes, 1990b), we carried out a
regression analysis between runoff and soil loss with
vegetation cover. The best equation obtained was the
negative exponential model (Fig. 4), as proposed by Elwell
and Stocking (1976), with R
2
>0.91. This function indicated
that in the presence of high vegetation cover percentages,
the responses in runoff and soil loss in the different
vegetation patches analysed tended to have little differen-
tiation. Obviously, small values of vegetation cover increase
soil losses, the most significant loss being for covers under
30% of surface. Further studies on Alpha grass steppe
(37.0% vegetation cover) reveal runoff and soil loss values
similar to the estimated ones using these equations (Fig. 4).
The response of the dry grassland is related to the
presence of B. retusum cover, that occupies 65.6% of the
surface of the plots (Table 6); being capable of intercepting
between 38% and 57% of the precipitation, according to
results of the rainfall simulations in laboratory (Derouiche et
al., 1996). Cerda` (1997) reported runoff coefficients of
0.03–0.06% in herbaceous micro-plots (with B. retusum as
dominant species) and soil loss of 0.014 Mg ha
À1
. Gutierrez
and Herna´ndez (1996) in experiments with rainfall simulator
(137 mm h
À1
) and different herbaceous covers showed that
a cover of 50% of herbaceous (in growth phase) and 70%
(in dormancy phase) was sufficient to significantly reduce
the runoff.
Based on our results (Table 2), the runoff generated by
vegetation cover type should present the following order:
AS <AG<S <G<B, but the measured annual runoff coef-
ficients indicated a different order: AS <G<S <AG<B.
Moreover it was found to be significant that G formation
generated less runoff than S and AG. In this case the
difference between G and S may be because the mean slope
of the shrub plots (29.2-) exceeds the slope in the G plots
(26-), and despite the importance of soil surface roughness
(Lavee et al., 1995). The difference between G and AG, can
be related to the characteristics of the upper soil layer on the
surface. Both vegetation cover types present soil upon
loamy limestone substrate, but the AG plots are found
located in the medium third of the hillside, the zone of
greater erosion, for which the surface presents loamy
limestone horizon, with a certain degree of crusting that
reduces the infiltration and favours the generation of runoff.
Similar explanations are reported by Cerda` (1999), which
obtained greater values of runoff and erosion in soil
supported by loamy limestone.
Most authors indicate that the soil loss tolerance rates
may differ according to several characteristics of the soils,
topography and vegetation cover. In our study area, the
values measured in all the vegetation cover types (afforested
and non-afforested vegetation communities) and bare soil
are lower than the 5.0 Mg ha
À1
year
À1
of annual rate of soil
loss indicated by Smith and Staney (1965) and Mitchell and
Bubenzer (1980). Only bare soil (in years 1997 and 1999)
presented annual soil losses that exceeded the rate of 2.0 Mg
ha
À1
year
À1
established by Arnoldus (1977) for soils whose
roots reach 25 cm depth, as in our case.
5. Conclusions
Our study provides basic information on runoff, sediment
yield and plant diversity in semi-arid natural and managed
communities and can be used to design effective recovery
strategies for degraded semi-arid ecosystems. Although 4
Fig. 4. Exponential functions and regression coefficients for runoff and soil
loss as a function of vegetation type cover during the study period. Together
with hydrological plots studied in this paper, the data of Alfa grass plots
with 37% of soil cover is included to illustrate the relevance of the so called
‘‘30% rule’’.
E. Chirino et al. / Catena 65 (2006) 19–29 27
years is insufficient time in which to evaluate erosion
processes in semi-arid Mediterranean lands, due to high
climatic variability, this study reveals that dry grassland
formations ( G) retained soil and runoff as efficiently as thorn
shrublands (S) and afforested communities (AG and AS).
The average rainfall event in the semi-arid study area had
low volume, low intensity and short duration, and conse-
quently low erosivity risk for the vegetation communities
analysed (more than 70% of land covered). When compar-
ing afforested communities (AS and AG) with natural
vegetation communities ( G and S), the results reveal that
there were no differences in the runoff generation and soil
loss. Therefore, in the conditions under which the research
was conducted (semi-arid climate), afforestation with an
Aleppo pine stratum; does not significantly reduce erosion
on a long-term scale (30 years), in comparison to the natural
vegetation without trees. Consequently, this kind of
afforestation is not the most efficient procedure either in
the long-term or in the short-term to improve land cover in
non-degraded landscapes with the aim of soil and plant
conservation. Alternatively, the conservation of natural
communities, or the restoration of semi-arid ecosystems
with native shrub species can be an adequate method for
reducing the effects of erosion and improving plant diversity
and ecosystem resilience, especially when re-sprouting
species can be used.
In conclusion, afforestation, as a management practice
has improved the structure of the natural communities
through the addition of pine stratum in order to establish a
pluri-stratified forest; however species richness was reduced
as well as plant diversity in these afforested semi-arid areas.
Alternatively, these positive effects can be achieved through
recovery of the natural vegetation, which can be managed
with these aims.
Acknowledgements
This research was partially funded by the Spanish
Government, through the CICYT programmes (HID97-
1047 and REN2000-0592-HID), the LUCDEME pro-
gramme of MMA (RESEL network), and Generalitat
Valenciana Regional Program /GV97-RN-14-4.
References
Albadalejo, J., Stocking, M.A., 1989. Comparative evaluation of two
models in predicting storm soil loss from erosion plots in semiarid
Spain. Catena 16, 227–236.
Albadalejo, J., Martı ´nez-Mena, M., Rolda´n, A., Castillo, V., 1998. Soil
degradation and desertification induced by vegetation removal in
semiarid environment. Soil Use Manage. 14, 1–5.
Andreu, V., Rubio, J.L., Gimeno-Garcı ´a, E., Llinares, J.V., 1998. Testing
three Mediterranean shrub species in runoff reduction and sediment
transport. Soil Tillage Res. 45, 441–454.
Arnoldus, H.M.J., 1977. Predicting soil losses due to sheet and rill erosion.
FAO Conserv. Guide 1, 99–124.
Atzmon, N., Moshe, Y., Schiller, G., 2004. Ecophysiological response to
severe drought in Pinus halepensis Mill. Trees of two provenances.
Plant Ecol. 171, 15–22.
Bautista, S., 1999. Regeneracio´ n post-incendio de un pinar (Pinus
halepensis, Miller) en ambiente semia´rido. Erosio´ n del suelo y medidas
de conservacio´ n a corto plazo. Ph.D. Thesis, Alicante University, Spain.
Bellot, J., Escarre´, A., 1998. Stemflow and throughfall determination in a
resprouted Mediterranean holm-oak forest, and changes by precipitation
trends. Ann. Sci. For. Paris 55 (7), 847–865.
Bellot, J., Sa´nchez, J.R., Bonet, A., Chirino, E., Abdelli, F., Herna´ndez, N.,
Martı ´nez, J.M., 1998. Effect of different vegetation type cover on the
soil water balance in semi-arid areas of south eastern Spain. Phys.
Chem. Earth (b) 24 (4), 353–357.
Bellot, J., Bonet, A., Sa´nchez, J.R., Chirino, E., 2001. Likely effects of land
use changes on the runoff and aquifer recharge in a semi-arid landscape
using a hydrological model. Landsc. Urban Plan. 778, 1–13.
Bonet, A., 2004. Secondary succession on semi-arid Mediterranean old-
fields in South-eastern Spain: insights for conservation and restoration
of degraded lands. J. Arid Environ. 56, 213–233.
Bonet, A., Pen˜a, J., Bellot, J., Cremades, M., Sa´nchez, J.R., 2001. Changing
vegetation structure and landscape patterns in semi-arid Spain. In:
Villacampa, Y., Brebbia, C.A., Uso´ , J.L. (Eds.), Ecosystems and
Sustainable Development III, pp. 377–386.
Bonet, A., Bellot, J., Pen˜a, J., 2004. Landscape dynamics in a semiarid
mediterranean catchment (SE Spain). In: Mazzoleni, S., Di Pasquale,
G., de Martino, P., Rego, F. (Eds.), Recent Dynamics of Mediterranean
Vegetation Landscape. Wiley, U.K., pp. 47–56.
Cerda`, A., 1997. Rainfall drop size distribution in the Western Mediterra-
nean basin, Valencia, Spain. Catena 30, 169–182.
Cerda`, A., 1998. The effect of patchy distribution of Stipa tenacissima L.
on runoff and erosion. J. Arid Environ. 36, 37–51.
Cerda`, A., 1999. Parent material and vegetation affect soil erosion in eastern
Spain. Soil Sci. Soc. Am. J. 63 (2), 362–368.
Chirino, E., Sa´nchez, J.R., Bonet, A., Bellot, J., 2001. Effect of
afforestation and vegetation dynamics on soil erosion in semi-arid
environment (SE Spain). In: Villacampa, Y., Brebbia, C.A., Uso´ , J.L.
(Eds.), Ecosystems and Sustainable Development III, pp. 239–248.
Cortina, J., Bellot, J., Vilagrosa, A., Caturla, R.N., Maestre, F., Rubio, E.,
Martı ´nez, J.M., Bonet, A., 2004. Restauracio´ n en semia´rido. In: Vallejo,
R., Alloza, J.A. (Eds.), Avances en el Estudio de la Gestio´ n del Monte
Mediterra´neo. Fundacio´ n CEAM. Generalitat Valenciana, Valencia,
pp. 339–400.
Costa, M., 1987. El Paı ´s Valenciano. In: Peinado, M., Rivas-Martı ´nez, S.
(Eds.), La Vegetacio´ n de Espan˜a. Aula Abierta. Universidad de Alcala´
de Henares, pp. 281–308.
De la Torre, A., 1988. Flora, Vegetacio´ n y Suelos de la Sierra del
Maigmo´ (Alicante). Caja de Ahorros Provincial de Alicante, Alicante,
Spain. 248 pp.
De la Torre, A., Alı ´as, L.J., 1996. Suelos y vegetacio´ n en el Alto Vinalopo´ .
Universidad de Alicante, Alicante, Spain. 250 pp.
Derouiche, A., Bellot, J., Cartagena, D., 1996. Effect du Covert Ve´ge´tal Sur
le Comportement Hydrique et le Transport Solide Dans une Pine`de a´
Sous-Bois Arbustif et Herbace´. Re´seau E
´
rosion 13. 10 pp.
Elwell, H.A, Stocking, M.A., 1976. Vegetal cover to estimate soil erosion
hazard in Rhodesia. Geoderma 15, 61–70.
Estades, C.F., Temple, S.A., 1999. Deciduous-forest bird communities in
plantations with broad-leaved and coniferous trees in a fragmented
landscape dominated by exotic pine plantations. Ecol. Appl. 9, 573–585.
Esteve, M.A., Ferrer, D., Ramı ´rez-Dı ´az, L., Calvo, J.F., Sua´rez-Alonso,
M.L., Vidal Abarca, M.R., 1990. Restauracio´ n de la vegetacio´ n en
ecosistemas a´ridos y semia´ridos: algunas reflexiones ecolo´ gicas.
Ecologı ´a. Fuera Ser. 1, 497–510.
Francis, C.F., Thornes, J.B., 1990a. Runoff hydrographs from three
Mediterranean vegetation cover types. In: Thornes, J.B. (Ed.), Vegeta-
tion and Erosion. John Wiley & Sons Ltd., pp. 363–384.
Francis, C.F., Thornes, J.B., 1990b. Matorral: erosion and reclamation. In:
Albaladejo, J., Stocking, M.A., Dı ´az, E. (Eds.), Soil Degradation and
E. Chirino et al. / Catena 65 (2006) 19–29 28
Rehabilitation in Mediterranean Environmental Conditions. CSIC,
Murcia, Spain, pp. 87–115.
Freedman, B., Zelazny, V., Beaudette, D., Flemming, T., Flemming, S.,
Forbes, G., Gerrow, J.S., Johnson, G., Woodley, S., 1996. Biodiversity
implications of changes in the quantity of dead organic matter in
managed forests. Environ. Rev. 4, 238–265.
Gutierrez, J., Herna´ndez, I.I., 1996. Runoff and interrill erosion as affected
by grass cover in a semiarid rangeland of northern Mexico. J. Arid
Environ. (34), 287–295.
Hartley, M.J., 2002. Rationale and methods for conserving biodiversity in
plantation forests. For. Ecol. Manag. 155, 81–95.
Lavee, H., Kutiel, P., Segev, M., Benyamini, Y., 1995. Effect of surface
roughness on runoff and erosion in a mediterranean ecosystem: the role
of fire. Geomorphology 11, 227–234.
Lo´ pez, M.V., 1989. Estudio de flujos hı ´dricos y de la contribucio´ n de la
deposicio´ n seca y lixiviacio´ n al aporte de nutrientes a un suelo forestal
bajo encina (Q. ilex) y pino (P. sylvestris) Tesis Master of Sciences.
CIHEAM.
Lo´ pez Bermu´ dez, F., Alias, L.J., Marı ´nez, J., Romero, M.A., Marı ´n, P.,
1991. Escorrentı ´a y pe´rdida de suelo en calcisol pe´trico bajo ambiente
mediterra´neo semia´rido. Cuatern. Geomorfol. 5, 7–89.
Maestre, F.T, Huesca, M., Zaady, E., Bautista, S., Cortina, J., 2002.
Infiltration, penetration resistance, and microphytic crust composition in
contrasted microsites within a Mediterranean semi-arid steppe. Soil
Biol. Biochem., 895–898.
Ministerio de Medio Ambiente, 1998. II Inventario Forestal Nacional
(1986–1996). Secretarı ´a General de Medio Ambiente. Direccio´ n
General de Conservacio´ n de la Naturaleza. Publicaciones del Organ-
ismo Auto´ nomo de Parques Nacionales.
Mitchell, J.K., Bubenzer, G.D., 1980. Soil loss estimation. In: Kirby,
M.J., Morgan, R.P.C. (Eds.), Soil Erosion. John Wiley & Sons Ltd.,
pp. 35–88.
Morgan, R.P.C., 1986. Soil Erosion and Conservation. Longman Scientific
& Technical, Essex.
Peco, B., Levassor, C., Pineda, F.D., 1983. Diversite´ et structure spatiale de
paˆturages me´diterrane´ens en cours de succession. Ecol. Mediter. 9,
223–234.
Pe´rez Cuevas, A.J., 1994. Atlas Clima´tico de la Comunidad Valenciana.
Consellerı ´a d’ Obres Publiques, Urbanisme i Transport. Generalitat
Valenciana, pp. 168–183.
Ravento´ s, J., De Luis, M., Gras, M.J., C
ˇ
ufar, K., Gonzalez-Hidalgo, C.,
Bonet, A., Sa´nchez, J.R., 2001. Growth of Pinus pinea and Pinus
halepensis as affected by dryness, marine spray and land use changes
in a Mediterranean semiarid ecosystem. Dendrochronologia 19 (2),
211–220.
Reynolds, J.F., Virginia, R.A., Kemp, P.R., de Soyza, A.G., Tremmel, D.C.,
1999. Impact of drought on desert shrubs: effects of seasonality and
degree of resource island development. Ecol. Monogr. 69, 69–106.
Rivas-Martı ´nez, S., 1987. Memoria del mapa de series de vegetacio´ n de
Espan˜a. ICONA, Madrid, Spain. 268 pp.
Romero-Dı ´az, M.A., Lo´ pez-Bermudez, F., Thornes, J.B., Francis, C.F.,
Fisher, G.C., 1988. Variability of overland flow erosion rates in
semiarid Mediterranean environment under matorral cover. Murcia
Spain. In: Harvey, A.M., Salas, M. (Eds.), Geomorphic Systems. Catena
Suplement, vol. 13, pp. 1–11.
Sa´nchez, J.R., 1997. Estimacio´ n de las pe´rdidas erosivas inducidas por
las te´cnicas de preparacio´ n del suelo previa a la reforestacio´ n en el
Sur de la Comunidad Valenciana. Ph. D. Thesis, Alicante University,
Spain.
Schiller, G., Cohen, Y., 1998. Water balance of Pinus halepensis Mill.
afforestations in an arid region. For. Ecol. Manag. 105, 121–128.
Smith, R.M., Staney, M.L., 1965. Determining the range of tolerance
erosion. Soil Sci. 100, 414–424.
Soto, B., Dı ´az-Fierros, F., 1997. Runoff and soil erosion from areas of burnt
scrub: comparison of experimental results with those predicted by the
WEPP model. Catena 31, 257–270.
Ter Braak, C.J.F., Colin Prentice, I., 1988. A theory of gradient analysis.
Adv. Ecol. Res. 18, 271–313.
Ter Braak, C.J.F., S
ˇ
milauer, P., 1998. CANOCO Reference Manual and
User’s Guide to Canoco for Windows. Software for Canonical
Community Ordination (version 4). Centre for Biometry Wageninger.
351 pp.
Teixeira, P.C., Misra, R.P., 1997. Erosion and sediment characteristics of
cultivated forest soil as effected by the mechanical stability of
aggregates. Catena 30, 119–134.
Trimble, S.W., 1988. The impact of organisms on overall erosion rates
within catchment in temperature regions. In: Viles, H.A. (Ed.),
Biogeomorphology. Blackwell, Oxford, pp. 83–142.
Vacca, A., Loddo, S., Ollesch, G., Puddu, R., Serra, G., Tomasi, D., Aru,
A., 2000. Measurement of runoff and soil erosion in three areas under
different land use in Sardinia (Italy). Catena 40, 69–92.
Valle, F., Bocio, I., 1996. Restauracio´ n de la vegetacio´ n en el sureste de la
Peninsula Ibe´rica. Cuad. SECF 3, 109–122.
Vallejo, R., 1997. Presentacio´ n. In: Vallejo, V.R. (Ed.), La restauracio´ n de la
cubierta vegetal en la Comunidad Valenciana. Fundacio´ n CEAM, GV.,
Valencia, pp. 1–7.
Verdu´ , M., Garcı ´a-Fayos, P., 1996. Nucleation processes in a Mediterranean
bird-dispersed plant. Funct. Ecol. 10, 275–280.
Vilagrosa, A., Seva, J.P., Valdecantos, A., Cortina, J., Alloza, J.A.,
Serrasolsas, I., Diego, V., Abril, M., Ferran, A., Bellot, J., Vallejo,
V.R., 1997. Plantaciones para la restauracio´ n forestal en la
Comunidad Valenciana. In: Vallejo, V.R. (Ed.), La restauracio´ n de
la cubierta vegetal en la Comunidad Valenciana. CEAM, GV, Valencia,
pp. 435–546.
Wagner, R.G., Flynn, J., Gregory, R., Mertz, C.K., Slovic, P., 1998.
Aceptable practices in Ontario’s forests: differences between the public
and forestry professionals. New For. 16, 139–154.
Watson, R.T., Noble, I.R., Bolin, B., Ranindranath, N.H., Verardo, D.J.,
Dokken, D.J., 2000. Land Use, Land-Use Change and Forestry. Special
Report of the Intergovernmental Panel on Climate Change. Cambridge
University Press, UK.
Zavala, M.A., Espelta, J.M., Retana, J., 2000. Constraints and trade-offs in
Mediterranean plant communities: the case of holm oak–Aleppo pine
forests. Bot. Rev. 66 (1), 119–149.
E. Chirino et al. / Catena 65 (2006) 19–29 29