Habitat loss exacerbates regional extinction risk of the keystone semiarid shrub Ziziphus lotus through collapsing the seed dispersal service by foxes (Vulpes vulpes)

Abstract

Habitat loss and landscape degradation affect animal-mediated seed dispersal, often collapsing the regeneration of endangered plant species and habitats in anthropogenic landscapes. We first compared the role of red fox and other vertebrates as seed disperser for the keystone scrub Ziziphus lotus. Because it turned out that foxes are the major Z. lotus dispersers, we investigated how fox activity and dispersal service relate to habitat loss and landscape alteration in the threatened Ziziphus semiarid scrublands, a priority habitat for conservation in Europe. Considering its opportunistic behavior, we hypothesized that landscape features should affect moderately fox abundance, while influence in a large extent its dispersal service. Accordingly, we predicted that a substantial decline in Ziziphus fruit consumption rather than in disperser activity would be responsible for seed dispersal collapse under severe habitat loss. We evaluated fox activity and dispersal service in 17 populations of Z. lotus spread through the range of its habitat in Spain and found within landscapes with different land-use intensity. We certified the collapse of the dispersal service by fox under severe habitat loss and confirmed that fox activity was less affected by habitat loss or landscape alteration than consumption of Ziziphus fruits. Consequently, the decline of consumption of Ziziphus fruits under severe habitat loss triggers the collapse of its seed dispersal. Results suggest that without increase of the remnant areas other managements may not suffice to achieve seed dispersal and habitat restoring. Dispersal service and natural regeneration in many Ziziphus habitat remnants will possibly cease in the future if habitat loss continues.

Full text

ORIGINAL PAPER
Habitat loss exacerbates regional extinction risk
of the keystone semiarid shrub Ziziphus lotus through
collapsing the seed dispersal service by foxes (Vulpes
vulpes)
Inmaculada Cancio
1
•
Ana Gonza
´
lez-Robles
1
•
Jesu
´
s M. Bastida
1
•
Antonio J. Manzaneda
1
•
Teresa Salido
1
•
Pedro J. Rey
1
Received: 11 January 2016 / Revised: 8 March 2016 / Accepted: 15 March 2016
Ó Springer Science+Business Media Dordrecht 2016
Abstract Habitat loss and landscape degradation affect animal-mediated seed dispersal,
often collapsing the regeneration of endangered plant species and habitats in anthropogenic
landscapes. We first compared the role of red fox and other vertebrates as seed disperser for
the keystone scrub Ziziphus lotus. Because it turned out that foxes are the major Z. lotus
dispersers, we investigated how fox activity and dispersal service relate to habitat loss and
landscape alteration in the threatened Ziziphus semiarid scrublands, a priority habitat for
conservation in Europe. Considering its opportunistic behavior, we hypothesized that
landscape features should affect moderately fox abundance, while influence in a large
extent its dispersal service. Accordingly, we predicted that a substantial decline in Ziziphus
fruit consumption rather than in disperser activity would be responsible for seed dispersal
collapse under severe habitat loss. We evaluated fox activity and dispersal service in 17
populations of Z. lotus spread through the range of its habitat in Spain and found within
landscapes with different land-use intensity. We certified the collapse of the dispersal
service by fox under severe habitat loss and confirmed that fox activity was less affected by
habitat loss or landscape alteration than consumption of Ziziphus fruits. Consequently, the
decline of consumption of Ziziphus fruits under severe habitat loss triggers the collapse of
its seed dispersal. Results suggest that without increase of the remnant areas other man-
agements may not suffice to achieve seed dispersal and habitat restoring. Dispersal service
and natural regeneration in many Ziziphus habitat remnants will possibly cease in the
future if habitat loss continues.
Communicated by Daniel Sanchez Mata.
Electronic supplementary material The online version of this article (doi:10.1007/s10531-016-1085-y)
contains supplementary material, which is available to authorized users.
& Pedro J. Rey
prey@ujaen.es
1
Departamento Biologı
´
a Animal, Biologı
´
a Vegetal y Ecologı
´
a, Universidad de Jae
´
n, 23071 Jae
´
n,
Spain
123
Biodivers Conserv
DOI 10.1007/s10531-016-1085-y

Keywords Carnivore activity  Drylands  Habitat restoring  Landscape degradation 
Red fox diet  Seed dispersal collapse
Introduction
Alteration of natural landscapes by human land use and concomitant loss of natural
habitats (e.g., through farming, urbanization and infrastructures) are among the main
drivers of Global Change and the biodiversity crisis (Sala et al. 2000; Fahrig 2003; Lin-
denmayer and Fischer 2006). Beyond producing direct loss of species, alteration of
landscapes and habitat fragmentation also affect critically important ecosystem services
(Bianchi et al. 2006; Kremen et al. 2007). One of the typically affected ecosystem services
is seed dispersal through mobile organisms (Magrach et al. 2014; Fontu
´
rbel et al. 2015).
Many vertebrates act as vehicles for seed dispersal of many plants species through natural
and anthropogenic landscapes, but they are sensitive in a variable level to landscape
alteration and habitat fragmentation (Cordeiro and Howe 2003; Rodrı
´
guez-Cabal et al.
2007; Garcı
´
a et al. 2010; Rey and Alca
´
ntara 2014). There is growing evidence that present-
day landscape features determine not only animal diversity (Fahrig et al. 2011), but also the
extent and efficiency of ecological services provided by mobile organisms, such as seed
dispersal and pollination (Kremen et al. 2007; Fontu
´
rbel et al. 2015). Moreover, animal-
seed dispersal is fundamental for the natural regeneration of plant populations because it
often has cascading effects along the regeneration cycle (Jordano and Herrera 1995; Rey
and Alca
´
ntara 2000). Thus, to assess the potential for natural regeneration and habitat
recovery in anthropogenic landscapes it is crucial to evaluate whether landscape trans-
formation might limit or even collapse the seed dispersal service provided by vertebrates
(Traveset et al. 2012; Rey and Alca
´
ntara 2014), and the extent to which vertebrate seed-
dispersers play a role as ‘restorers’ in anthropogenic landscapes (Swift et al. 2004; Matı
´
as
et al. 2010; Sua
´
rez-Esteban et al. 2013; Escribano-A
´
vila et al. 2014).
Mediterranean ecosystems have a long history of human disturbance which has resulted
in very diverse stages of landscape and habitat degradation (Valladares et al. 2004). The
effects of anthropogenic landscape disturbance and habitat fragmentation on seed dispersal
have been studied in mesic Mediterranean woodlands (for example, Santos et al. 1999) and
termo-mediterranean scrublands (Traveset et al. 2012; Gonza
´
lez-Varo et al. 2012; Rey and
Alca
´
ntara 2014). However, there is no information on the disruption by human alteration of
the vertebrate-seed dispersal service in Mediterranean semiarid habitats, despite many
semiarid areas have been heavily impacted by changes in human land-use in the last
decades (green-house agriculture intensification and urbanization, e.g., Mota et al. 1996
).
This is the case of the ‘‘Mediterranean arborescent matorral with Ziziphus lotus’’(Ziziphus
habitat hereafter), a singular habitat type from southeastern Iberian Peninsula, which is a
priority habitat protected under European Directive 92/43/EEC (Benito et al. 2009;
Mendoza-Ferna
´
ndez et al. 2015). This habitat has a high ecological and biogeographical
value due to its rich endemic flora and its links with North African ecosystems. Ziziphus
lotus (Rhamnaceae) is the keystone species in this habitat where it facilitates the growth of
many other plant species and serves as refuge for many birds and mammals (Tirado 2009).
Seed dispersal under field conditions has not been previously quantified in Z. lotus
although its seeds are easily visible in scats of some carnivores (authors, personal obser-
vation), especially in Red fox’s (Vulpes vulpes), but also in European badger’s (Meles
Biodivers Conserv
123

meles), Stone marten‘s (Martes foina), and some ungulates (Wild boar, Sus scrofa). Other
carnivores, primates and cattle are known to disperse other Ziziphus species (Zhang and
Wang 1995; Grice 1996; Varela and Bucher 2006).
Large-sized carnivores are top predators commonly considered to be sensitive to habitat
loss and fragmentation because of their large spatial requirements, low reproductive rate,
and human persecution (Woodroffe 2000; Gittleman et al. 2001). However, there is ample
evidence that some small and medium-sized carnivores (for example, many canid and
mustelids), like the red fox, are able to adapt to human-modified and fragmented habitats
(Contesse et al. 2004; Baker and Harris 2007; Salek et al. 2014), being able to live in
deforested agricultural environments with small remnants of natural habitats (Fedriani
et al. 2001; Virgo
´
s et al. 2002). In particular, the adaptability of red fox to anthropogenic
landscapes is based on its high plasticity and opportunistic behavior (Lovari et al. 1996;
Contesse et al. 2004). Red fox frequently feeds also on cultivated and wild fruits and has
been described as seed disperser of many fruit species in the Mediterranean region
(Rosalino and Santos-Reis 2009; Fedriani and Wiegand 2014).
As part of ongoing research on the potential for natural regeneration and recovery of the
Ziziphus habitat, we show that red fox is the major seed disperser for Z. louts and inves-
tigate how red fox’s activity and dispersal service to Z. lotus relate to habitat loss and
landscape alteration throughout the plant’s distribution range in the Iberian Peninsula. Our
working hypothesis is that provided the red fox opportunism, landscape alteration and
habitat loss (that is, present-day landscape features) should affect only moderately its
abundance and activity, whereas its seed dispersal service for Z. lotus should be affected in
a higher extent by landscape features. Under such hypothesis we predict that the role of red
fox as Z. lotus seed-disperser would be proportional to the Ziziphus natural habitat cover
remaining in the landscape, collapsing where natural habitat became relict. Accordingly, a
substantial decline in Ziziphus fruit consumption by red fox rather than in disperser
abundance would be triggering Z. lotus seed dispersal collapse under severe habitat loss
conditions.
Materials and methods
Species and study system
The focus of our study is the ‘‘Mediterranean arborescent matorral with Z. lotus’’ (habitat
5220 in Annex 1 of the European Union’s Habitats Directive 92/43/EEC). This habitat is
dominated by Z. lotus, accompanied by open pre-desertic scrubs like Periploca laevigata,
Lycium intrincatum, Asparagus stipularis, A. albus, Maytenus senegalensis and Whitania
frutescens. In Europe, this habitat is restricted to semiarid areas in the southeast of the
Iberian Peninsula, where it should occupy most of the plains and seasonal streams (namely
‘ramblas’) in the coastal belt of the region. However, this original area of distribution has
been reduced and fragmented by intensive greenhouse-based agriculture and urban
expansion in the last 55 years (Mota et al. 1996). In many sites less than 5 % of the area
occupied in the 50 s is left, which has led to a dramatic situation for conservation of this
type of habitat.
Ziziphus lotus is a North African taxon distributed across Northern Africa, Sahara desert
and Arabian Peninsula. In Europe, it only occurs in Spain and Sicily. Ziziphus lotus is a
keystone and ecosystem engineering species because provides shelter for vertebrates,
Biodivers Conserv
123

facilitates the growth of many plant species, and its potent radicular system creates islands
of soil fertility and influences the hydraulic properties of the habitat (Tirado 2009). Cur-
rently Z. lotus populations suffer a collapse in natural regeneration, with low density and
establishment of seedlings (PJ Rey et al. unpublished), persisting mainly thanks to its
sprouting ability. Ziziphus lotus produces globular brown-reddish drupes of 9–12 mm
diameter, each containing two seeds enclosed within a hard endocarp (stone). Hereafter we
will use the term seed in reference to the whole dispersal unit formed by the two seeds and
the stone. Fructification occurs between July and early August when fruits are ripe and
available for consuming. To our knowledge there is no previous published information on
seed dispersal of Z. lotus, but based on our surveys of vertebrate scats (carnivores, wild
ungulates and lizards) in Ziziphus habitat remnants, red fox acts as major seed disperser in
the region (see ‘‘Results’’). Furthermore, although we cannot rule out a role of medium-
sized birds (thrushes, pigeons or starlings) in seed dispersal of Z. lotus, we have not
detected seeds offside cover of conspecifics that could be attributable to birds (PJ Rey et al.
unpublished).
Study area
This study was conducted in 17 Z. lotus populations localized in Ziziphus habitat remnants
covering much of the original range of the habitat in the Iberian Peninsula. Most popu-
lations were more than 1 km apart from each other, spanning a total of 140 km (Fig. 1).
Further details of each population are shown in Table 1.
Fig. 1 Study populations in southeastern Spain
Biodivers Conserv
123

Table 1 Study populations and landscape characterization
Population Coordinates
UTM (X)
Coordinates
UTM (Y)
Landscape
type
Population size
(No. of adults)
% of natural or
seminatural habitat
Total number of red
fox scats collected
Track
length (km)
No. of years
of sampling
Be
´
jar 608753 4162689 Relict 89 5.72 29 1.479 3
Campohermoso 573207 4089516 Fragmented 20 25.69 9 0.694 2
Ctro. de Visitantes 566902 4074470 Variegated 1367 92.57 31 1.600 3
Dalı
´
as 516503 4068393 Relict 15 7.29 0 0.664 3
El Ejido 517838 4068287 Relict 64 8.74 9 1.185 3
Ferna
´
n-Pe
´
rez 584412 4086871 Fragmented 80 15.90 24 2.473 1
Guazamara 608941 4134043 Fragmented 81 31.89 40 1.999 3
Nı
´
jar 570227 4088796 Fragmented 255 40.37 60 2.728 3
Playazo 587885 4079328 Fragmented 101 10.93 87 4.248 3
Puerto Lumbreras 605781 4154249 Fragmented 102 27.10 117 0.838 3
Rambla Retamar 562378 4077811 Fragmented 68 56.91 2 3.829 1
Retamar 564303 4079496 Variegated 80 83.36 20 1.272 2
S. Alhamilla 552531 4089557 Variegated 133 80.62 101 1.817 3
Sta. Marı
´
a Aguila 520643 4071120 Relict 109 6.42 0 1.212 1
Torre Garcı
´
a 563288 4075416 Variegated 4598 64.76 45 2.132 3
Toyo-Alquia
´
n 558505 4078249 Fragmented 1015 41.71 13 1.183 3
Vega Can
˜
ada 557620 4082307 Fragmented 9 22.06 2 0.645 2
Sampling effort in each population and percentage cover of natural or seminatural habitat is also given. See methods for details on landscape characterization (following
McIntire and Hobbs 1999) and quantification of percentage cover of natural and seminatural habitat
Biodivers Conserv
123

Collection and analyses of feces and the standardized index of red fox activity
During three consecutive years (2012–2014) we surveyed transects in 17 populations of Z.
lotus for red fox and other vertebrates scats collection. Transects were surveyed monthly
along the period of Ziziphus ripe fruit availability on the ground (August–March). The
transect at each population was conducted by two observers, its length in each population
varied according to the area occupied by the population (Table 1) and the scats were
recorded in a 6 m-width band. Each scat was identified according to size, shape, color and
smell combination (see Fedriani et al. 2010;Lo
´
pez-Bao and Gonza
´
lez-Varo 2011). From
these data we obtained an index of red fox and other vertebrates activity in each popu-
lation. Specifically, we calculated the number feces collected in each population each year
per km of transect (IKA hereafter) and averaged it across years. In the particular case of red
fox, this value is finally used here as a Standardized Index of red fox activity in each
population (e.g. Cavallini 1994). Although not free of problems, scats density or abundance
have been reputedly considered as an efficient and inexpensive method to evaluate the
relative abundance and activity of canids, and is commonly used with red fox (for example,
Cavallini 1994; Webbon et al. 2004).
For analyses of scats content, samples were broken up in the lab to disaggregate all
components and remove all seeds. We counted all seeds found in feces. Because our focus
here is on how landscape transformation and habitat loss affect the seed dispersal service
for Ziziphus we paid attention to Z. lotus seeds occurrence and number of seeds in each
scat. Variation in the whole diet and its relationship to landscape features will be con-
sidered elsewhere.
Frequency of Ziziphus seed occurrence and seed mobility index (SMI)
We obtained for each population the frequency of Z. lotus seeds occurrence in red fox scats
(OF) and the average number of Z. lotus seeds in feces with seeds (NS). The first was used
as an estimator of the relative frequency of red fox fruit consumption in the population. NS
assesses both intensity of consumption and seed clumping. An especially important
composite variable that integrates the contribution of red fox to Z. lotus seed dispersal is
the seed mobility index (SMI): SMI = IKA 9 OF 9 NS.
SMI evaluates the amount of Ziziphus seeds mobilized from the source plants by red fox
in a population as a consequence of fox abundance or activity (IKA), and frequency and
intensity of Ziziphus fruit consumption. We should note that the soundness of using this
approach relies in the assumption that fox scat detectability is similar across study sites
To weight the role of red fox as seed disperser for Z. lotus in relation to other verte-
brates, we also estimated the product of IKA 9 OF for other vertebrate scats (Stone
marten, European badger, Common genet, Wild boars and Lacerta lepida) collected in our
surveys. The product of these two components of the dispersal service comparatively
assesses the mobilization of Z. lotus seeds by different vertebrates to different points within
the habitat remnants.
Seed viability and germination tests
We assessed the viability of seeds defecated by red fox through flotation. We collected 356
ripe Z. lotus fruits from at least five different adults in 9 populations and 95 seeds from red
fox feces. Floating seeds were assumed to be non-viable. To corroborate the reliability of
Biodivers Conserv
123

the floating test we also tested 10 non-floating and 10 floating seeds for viability using
tetrazolium chloride (ISTA 1999), which is considered a more reliable viability test. Seeds
were cut in half along the longitudinal axis to expose the embryo, and then submerged
during 4 h in a 0.5 % tetrazolium chloride solution in a 0.1 M phosphate buffer. We also
subjected soy and pea seeds to the staining solution as a positive control. The embryo was
classified as either viable (pink/red) or dead (white embryo). All non-floating seeds became
stained, denoting viability, while no floating seed yielded positive results in this test. All
remaining undispersed Z. lotus seeds (322 seeds; range 18–65 depending on population)
and 21 defecated seeds were sown in pots under common garden conditions and checked
for germination during one year.
Landscape type characterization
We characterized landscape type following the classification of McIntire and Hobbs (1999)
for fragmented landscapes. This classification is based on the representation of natural or
seminatural habitats in the landscape and defines 4 types of landscapes according to their
continuity and level of alteration: (1) relict (natural or seminatural habitat repre-
sents \10 % in the landscape), (2) fragmented (10–60 %), (3) variegated (60–90 %) and
(4) continuous or intact ([90 %). Our ascription of each population to a particular land-
scape type was based on the cover currently represented by natural or seminatural states of
this habitat relative to other land uses surrounding the focal population. More specifically,
we (geo)located with GPS all adult individuals in each population and then defined the
population area as the convex hull encapsulating all the individuals of each population. We
used a 1.5 km radius circle centered on the centroid of this hull as the unit to evaluate
habitat and land use cover in the landscape. For this evaluation we used the land-use and
vegetation cover cartography 2006 (1:10.000) of Junta de Andalucı
´
a(http://www.
juntadeandalucia.es/medioambiente/site/rediam) for 15 Andalusian populations and
CORINE Land Cover 2006 (http://www.eea.europa.eu/data-and-maps/data/corine-land-
cover-2006-raster#tab-additional-information) (1:100.000, 100 meters resolution) for 2
Murcian populations. In order to classify land-use cover within our 1.5 km radius circles,
we considered natural or seminatural states of the habitat those land-uses corresponding to
‘‘ramblas’’ and natural scrublands (dense or disperse) in relatively flat areas or smooth
hillsides, which are typical habitat requirements for Z. lotus. Land-use cartography was
cross-checked based on our field knowledge of each population and its surroundings as
well as on a close examination of land-use polygons using orto-photographs from year
2013. The percentage cover of the natural and seminatural habitat together with its cor-
responding habitat use is given in Table 1. Land-use may be found in Online Resource 1,
Table A1. We further obtained an index of diversity of cover use (Shannon index) as a
measure of landscape complexity or compositional heterogeneity (Fahrig et al. 2011) based
on proportion cover of each use. Finally, we also considered the population size (total
number of adults Z. louts in the area covered by the population).
Statistical analyses
Variation in SMI and its components between landscape types was analyzed using General
Linear Models. Prior to analyses, we log transformed SMI, IKA, and mean NS, and applied
angular transformation to OF. Likewise, we analyzed the relationships between SMI
components and landscape and population features (population size log transformed for
accounting for large between population differences). Separate linear and quadratic
Biodivers Conserv
123

Table 2 Abundance of scats (number of scats per km and study year) of red fox and other vertebrates detected in our surveys averaged across Z. lotus habitat remnants (range
between populations is shown within parentheses)
Vulpes vulpes Martes foina Meles meles Genetta genetta Sus scrofa Lacerta lepida
Abundance of scats per km of track (IKA) 8.91 (0.00–46.54) 0.25 (0.00–1.48) 0.03 (0.00–0.21) 0.0035 (0.00–0.06) 0.86 (0.00–5.07) 0.87 (0.00–7.59)
Z. lotus seed occurrence in scats (OF) 57/583 2/26 2/4 0/1 9/83 1/70
IKA x OF 0.871 0.023 0.015 0.000 0.093 0.012
It is also shown Z. lotus seed occurrence (referred to all scats found across populations) in scats of each vertebrate species and the product of these two components of the
dispersal service. Such a product comparatively represents the weight of each disperser in the mobilization of Z. lotus seeds to distinct points in the habitat remnants.
Normalized to sum 1 across vertebrate species, such product shows that a single species (V. vulpes) was responsible for the mobilization of 87 % of seeds to distinct points in
the remnants
Biodivers Conserv
123

-1,0
-0,5
0,0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
4,0
Log IKA
-0,2
-0,1
0,0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
occurrence of Ziziphus
-4
-2
0
2
4
6
8
10
12
nº seeds of Ziziphus/scat with seeds
RELIC T
FRAGMENTED
VARIEGATED
-1,0
-0,5
0,0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
Log SMI
(a)
(b)
(c)
(d)
0,5 0,6 0,7 0,8 0,9 1,0 1,1
Landscape use diversity (H')
-0,5
0,0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
4,0
4,5
Log IKA Red fox scats
0 102030405060708090100
Natural habitat cover in the landscape
-0,1
0,0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
Occurrence of Ziziphus
(angular transformed)
0 102030405060708090100
Natural habitat cover in the landscape
-0,5
0,0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
Log SMI
(e)
(f)
(g)
a
b
a,b
a
a
a
b
b
b
Fig. 2 Left panels: variation in SMI by red fox (d) and its components IKA (a), frequency of occurrence of
Ziziphus seeds in feces (b), and number of Ziziphus seeds in samples with Ziziphus (c), according to
landscape types. Predicted least-squares means ± 95 % confidence intervals are shown, identifying Tukey’s
post hoc differences with different letters. Right-hand panels (e–g) shows the detected significant
relationships of SMI and its components to landscape features and rabbit abundance
Biodivers Conserv
123

regression models were conducted for each predictor and dependent variable. We did not
try fitting multiple regressions due to sample size limitations.
We compared the viability and germination rates of defected and intact seeds between
populations. We included a factor identifying defecated/intact seeds nested within popu-
lation and used contrast tests to explore whether germination rates were higher for defe-
cated compared to intact seeds. Generalized Linear Models with binomial error and logit
link function were considered for these analyses. Otherwise stated, least-square adjusted
means ± 1 SE are shown in results.
Results
We collected a total of 590 scats of red fox during 3 study years. In two populations, Dalı
´
as
and Sta. Marı
´
a del A
´
guila, no sample was ever collected in the surveys, thereby these
populations were only used for estimation of IKA and SMI (i.e., 0 in both cases). Ziziphus
fruit remains and seeds occurred in 57 out of 583 scats (range among populations 0–34.8 %
of scats), lacking in scats from 5 populations.
We further collected scats of other vertebrates in the same surveys, but they were at least
tenfold less abundant than red fox’s scats (Table 2) in the habitats remnants through the
whole study region. In particular, European badger and Common genet were extremely rare
in the habitat remnants, while Z. lotus seed dispersal by Lacerta lepida was very infrequent
(Table 2). The product of IKA 9 OF for each vertebrate indicates that red fox accounts for
87 % of Z. lotus seed dispersal to distinct points within the habitat remnants (Table 2).
Variation in seed dispersal service by red fox and its components
Red fox activity in Ziziphus habitats estimated as standardized feces abundance (IKA),
varied significantly according to landscape type (F
2.14
= 3.69, P = 0.05) with mean IKA
tending to be lower in relict than in variegated landscapes (Fig. 2a). In particular, we did
not find any red fox fecal samples in tracks conducted in two relict populations located on
relict landscapes (Dalı
´
as and Sta. Marı
´
a del A
´
guila). However, red fox IKA did not relate
significantly to Ziziphus habitat cover in the landscape (F
1.15
= 2.30, P = 0.15 in linear
regression; F
2.14
= 1.20, P = 0.33, in the test of the whole quadratic model). In contrast,
red fox IKA was related to land-use diversity, describing a non-linear relationship
(quadratic regression: F
2.14
= 7.15, P = 0.007, R
2
= 0.51, Fig. 2e), peaking at landscapes
with intermediate values of land-use diversity. Red fox IKA did not relate significantly to
Z. lotus population size (F
1.15
= 1.44, P = 0.25, in linear regression; F
2.14
= 1.17,
P = 0.34, in the test of the whole quadratic model).
Occurrence of Ziziphus (OF) in excrements varied significantly among landscape types
(F
2.12
= 7.50, P = 0.008), with fragmented landscape showing significantly lower OF in
the diet than both variegated and relict landscapes, which did not differ from each other
(Fig. 2b). Accordingly, OF did not relate linearly (F
1.13
= 1.31, P = 0.27) but kept an
U-shaped relationship with natural habitat cover (Fig. 2f) in the landscape (whole model
test of the quadratic regression: F
2.12
= 4.39, P = 0.037, R
2
= 0.42), showing a minimum
at intermediate values. It tended also to decrease linearly with landscape land-use diversity
(b =-0.93 ± 0.45, t =-2.41, P = 0.06). OF did not relate to Z. lotus population size
(F
1.13
= 0.78, P = 0.20, in linear regression; F
2.12
= 0.93, P = 0.42, in the whole quad-
ratic model).
Biodivers Conserv
123

NS, the last component of the SMI, did not vary among landscape types (F
2.7
= 2.19,
P = 0.18, seven populations lacking fecal samples with Ziziphus were excluded from these
analyses) because there was a notable variation within each landscape type (Fig. 2c). No
significant linear (F
1.8
= 3.17, P = 0.11) or quadratic relationship (F
2.7
= 1.60, P = 0.27)
was detected between NS and the percentage cover of natural habitat in the landscape, the
land-use diversity (linear: F
1.8
= 1.26, P = 0.29; quadratic: F
2.7
= 1.42 P = 0.30), or the
population size of Z. lotus (linear: F
1.8
= 1.98, P = 0.20; quadratic: F
2.7
= 0.87
P = 0.46).
Finally, SMI varied largely between landscape types (F
2.14
= 8.91, P = 0.003). We did
not detect seed dispersal service for Z. lotus (SMI = 0) in 7 populations, due to either lack
of red fox activity (2 populations) or to absence of Z. lotus seeds in the fecal samples (5
populations). Post-hoc tests clearly showed that SMI was higher in variegated than both in
fragmented and relict landscapes, which did not differ from each other. In fact, predicted
mean SMI was four to five-fold higher in variegated than in fragmented or relict landscapes
(Fig. 2d). While SMI was null in 5 out of 9 populations from fragmented landscapes (in all
cases by lack of Z. lotus in the fecal samples) and 2 out of 4 populations of relict land-
scapes (both lacking red fox fecal samples in the tracks), all 4 populations from variegated
landscapes have some dispersal service by red fox. This trend was confirmed by the highly
significant linear positive relationship of SMI to percentage cover of natural or seminatural
habitat (F
1.15
= 7.27, P = 0.006, R
2
= 0.41, Fig. 2g). However, land-use diversity did not
affect dispersal service for Z. lotus (linear relationship: F
1.15
= 0.52, P = 0.58; quadratic:
F
2.14
= 3.02, P = 0.08). In contrast to previously found for its components, SMI did relate
significantly to Z. lotus population size, increasing linearly (F
1.15
= 6.38; P = 0.023,
R
2
= 0.25; b = 0.37 ± 0.14, in linear regression; no significant pattern found in quadratic
models, F
2.14
= 2.98, P = 0.08), although such a relationship was weak compared to other
predictors of SMI (see above).
Viability and germination rates of seeds dispersed by red fox
Seed viability was clearly affected by seed source (tree collected vs. fecal samples; Wald
Chi square = 21.12, df = 1, P \ 0.0001). Almost all seeds dispersed by red fox were
viable (LS-mean viability = 0.98, N = 95), whereas 20 % of seeds from the trees were
non-viable (LS-mean viability = 0.82, N = 356). Germination rate was variable among
populations (Wald Chi square = 50.1, df = 9, P \ 0.0001). Mean seed germination for
undispersed seeds ranged between 68.0 % ± 0.44 in Puerto Lumbreras and 12.9 % ± 0.38
in Toyo, whereas it was 66.6 % ± 0.47 for dispersed seeds. A contrast test between
dispersed and undispersed seeds shows significant difference in seed germination rate
(Likelihood ratio Chi square = 4.23, df = 1, P = 0.0396), indicating that seed pass
through the red fox gut affected positively germination.
Discussion
Landscape effects on the red fox seed dispersal service for Ziziphus lotus
Studies on the effects of landscape degradation and habitat fragmentation on animal-
mediated seed dispersal have rendered disparate results. While many case studies and a
meta-analytical review (Magrach et al. 2014) showed negative effects on animal-mediated
Biodivers Conserv
123

seed dispersal (Santos and Tellerı
´
a 1994; Rodrı
´
guez-Cabal et al. 2007, Cordeiro et al.
2009) other meta-analytical approaches (Markl et al. 2012) did not show effect of frag-
mentation but effects of other disturbances as logging and hunting. Other recent meta-
analytical review by Fontu
´
rbel et al. (2015) found that habitat degradation has overall a
negative effect on seed-disperser animal diversity, whereas habitat fragmentation nega-
tively affects interaction rates. They specifically conclude that while mammal diversity
does not seem to be affected by landscape degradation, seed dispersal rates were negatively
affected. We do not explore here how the diversity of vertebrate seed dispersers was
affected by landscape degradation and habitat loss, though our data on activity of other
vertebrates and Z. lotus seeds OF in their scats clearly reflect the over-dominance of red
fox in Z. lotus seed dispersal. In fact, comparison of the product of IKA 9 OF for each
vertebrate (Table 2) renders that red fox is responsible for the mobilization of ca. 90 % of
seeds that are dispersed to distinct points in the Z. lotus habitat remnants. Thus Z. lotus
seed dispersal would be clearly compromised by those factors affecting the interaction rate
with red fox.
We asked how landscape degradation and habitat loss could affect Ziziphus seed dis-
persal service by red fox, and predicted that dispersal service would be proportional to
Ziziphus habitat cover remaining in the landscape, collapsing where this habitat became
relict. We confirmed this prediction since we detected a highly significant and positive
linear relationship between the Ziziphus seed mobility index (SMI) and natural habitat
cover in the landscape, and a collapse of seed dispersal by red fox in some relict land-
scapes, achieving its regional maximum in quasi-continuous Ziziphus habitat landscapes
(over 90 % cover). Collapse of seed dispersal has been found under strong habitat loss and
fragmentation in other species both for seeds dispersed by mammals and by birds (Santos
et al. 1999; Cordeiro and Howe 2003; Rodrı
´
guez-Cabal et al. 2007). In fact, only this
variable explained more than 40 % of variation in SMI in our study. This, along with the
finding that SMI did not relate to landscape heterogeneity (i.e. to land-use diversity) and
only related weakly to Z. lotus population size, suggests that habitat loss and fragmenta-
tion, rather than changes in other attributes of the landscape associated to degradation,
affect Ziziphus seed dispersal by red fox. In any case, the complex relationships between
landscape features, habitat fragmentation, fruit production and abundance of alternative
food resources on the dispersal service of red fox and other vertebrate merit further
research.
Studies on seed dispersal effectiveness and dispersal service by frugivores should
examine quantitative and qualitative components of dispersal (Schupp et al. 2010), how-
ever these components have rarely been examined under scopes of landscape and habitat
degradation and/or fragmentation (but see Santos et al. 1999; Rey and Alca
´
ntara 2014).
Even more scarce are studies examining dispersal effectiveness with carnivore seed dis-
persal under habitat degradation and loss (Santos et al. 1999;Lo
´
pez-Bao and Gonza
´
lez-
Varo 2011). Because of the opportunism and dietary plasticity of red fox (Dell’Arte et al.
2007; Delibes-Mateos et al. 2008; Hartova
´
-Nentvichova
´
a et al. 2010;Dı
´
az-Ruiz et al.
2013) and its adaptation to rural and urban habitats (Lovari et al. 1996
; Goldyn et al., 2003;
Contesse et al. 2004), we expected that the component of the Z. lotus dispersal service
related to activity (i.e., abundance) of red fox were less affected by severe habitat loss and
anthropogenic landscape alteration than components related to the use of Ziziphus as food
(Ziziphus OF in de diet and the NS in feces with seeds). Our results confirm such
expectation substantially.
On the one hand, Ziziphus habitat cover in the landscape (an inverse measure of
habitat loss) did not relate significantly to activity of red fox, while landscape use
Biodivers Conserv
123

diversity did it in a non-linear way, i.e., activity peaking at intermediate values but
steady decreasing in continuous or in very heterogeneous landscapes. Several studies
have found that the activities of red fox and other wild canids are affected by landscape
diversity, increasing linearly (Oehler and Litvaitis 1996) or peaking at intermediate
levels of agricultural land occurrence (Kurki et al. 1998; Fedriani et al. 1999). The multi-
functionality for red fox of intermediate levels of land use diversity probably underlies
such a relationship but the exploration of the exact causes shaping the observed activity
pattern is beyond the scope of this paper and is dealt elsewhere (I. Cancio et al.
unpublished). On the other hand, Ziziphus OF in red fox scats varied significantly among
landscape types showing a minimum sited through the range of the natural habitat cover
at intermediate to low values (significant U-shaped relationship), with red fox consuming
frequently Ziziphus in more continuous natural landscapes, especially from 60 % cover
onwards (Fig. 2f.). Congruently, OF further tended to decrease linearly with land use
diversity, since very high diversity emerges from fragmentation and degradation of
continuous natural landscapes.
Finally, the last component of the dispersal service here considered, number of seeds
deposited in the same scat (NS) is indicative of seed clumping and may be interpreted as
inversely related with seed dispersal quality, since just one seedling could eventually
become a reproductive adult from such scat. This was the least predictable component of
the SMI. Such unpredictability may be caused by the scarce power of our tests since only
10 populations had feces with Ziziphus seeds. In any case it may be concluded that this
component of the seed dispersal service did not shape the detected relationship of SMI to
habitat cover since it neither varied among landscape types nor was related to habitat cover
or land-use diversity.
Our results suggests thus that the decline in the frequency of use of Ziziphus fruits rather
than the abundance and activity of the red fox in remnants of this habitat is responsible of a
marked decline in dispersal service of Ziziphus associated to habitat loss.
Viability and germination rates of Z. lotus seeds processed in red fox gut
A better understanding of the quality of seed dispersal service by vertebrates requires
knowledge on the effects of seed processing in vertebrate gut. The benefits of seed pro-
cessing in the guts of foxes have been investigated in terms of viability and germination of
seeds and subsequent seedling survival. Results were inconsistent among seed species
(Varela and Bucher 2006; Silva et al. 2005; Fedriani and Delibes 2009; Rosalino et al.
2010) with some species enhancing their germination and/or survival of seedlings while
other experiencing neutral or prejudicial effects. We found that Ziziphus seeds processed in
the red fox gut had enhanced viability and germination compared to seeds taken directly
from trees, something also found for other Mediterranean species like Celtis australis
(Traba et al. 2006)orPyrus bourgaena (Rosalino et al. 2010). Likewise, enhanced ger-
mination after processing in the gut of carnivores has been reported in other species of
Ziziphus (Varela and Bucher 2006; Maraghini et al. 2010), which is probably related to the
need of scarification of their thick stony endocarp. The majority of red fox scats with
Ziziphus seeds were found in open interspaces (81.6 %). Since Ziziphus species need direct
solar exposition for seedling establishment (Varela 2004), red fox acts as an efficient
disperser for this species because it moves seeds away from parent plants to potentially
safe microsites favorable for establishment, although further seedling recruitment studies
are needed to confirm this suggestion.
Biodivers Conserv
123

Implications for regeneration and conservation of Ziziphus scrubland
habitats
Knowledge about seed dispersal service is important for evaluating the state of the plant
regeneration and has potential for management, conservation and restoration (Matı
´
as et al.
2010; Rey and Alca
´
ntara 2014; Escribano-A
´
vila et al. 2014). In the case of Z. lotus there
was virtually no information about seed dispersal. We have shown in this paper that red fox
is the major fruit consumer and seed disperser of this species. Much of the Ziziphus seed
dispersal in the semiarid Ziziphus scrublands habitats of southeastern Iberian Peninsula
seems to depend on red fox, because it is largely the most abundant of its dispersers,
consume its fruits frequently and is an efficient seed disperser that enhances germination of
the seeds.
Some findings of this study provide insights on how habitat loss and landscape alter-
ation are heavily affecting the potential as ‘restorer’ of red fox through disruption of its
seed dispersal service. We have shown that drastic habitat alteration, fragmentation and
loss of the Ziziphus scrubland habitats dramatically diminish the seed dispersal service
(estimated as SMI) and some of its components, frequently causing collapse. Our results
suggest that the decay of the use of Ziziphus as food under severe habitat loss (a measure of
interaction rate) rather than abundance of red fox would be triggering the collapse of seed
dispersal service in this species under severe habitat loss. These results have important
implications for understanding the potential for natural regeneration and recovery of
Ziziphus scrubland habitat. Despite red fox is considered a generalist and opportunistic
species able to inhabit anthropogenic landscapes, we found that in 7 out of the 17 studied
Ziziphus populations the dispersal service collapse (i.e. it was null or undetectable through
three study years), either by no activity of red fox in the habitat remnants (2 populations in
relict landscapes) or by lack of Ziziphus fruit consumption (5 populations in fragmented
landscapes). Moreover, the fact that a single variable, Ziziphus habitat cover in the land-
scape, explained more than 40 % of the dispersal service should alert us on the exacerbated
risk of collapse of the dispersal service and natural regeneration of the species in other
remnants in the future, provided that the loss of this habitat is still ongoing in the region.
The risk of extinction of the seed dispersal interaction, and of its ecological function,
clearly exists and could be preceding the regional species loss (Valiente-Banuet et al.
2015). Our results further warn about the difficulty of natural recovery of this habitat, even
when active management is dedicated to improve the state of the habitat remnants, if we do
not increase the cover of the habitat in the landscape to attract dispersers and increase its
interaction with Ziziphus. Z. lotus is considered a keystone species in this ecosystem,
providing food and shelter for many animals, constituting fertility islands for the estab-
lishment of many other species and regulating its hydraulic function (Tirado 2009).
Consequently, the collapse of its seed dispersal and recruitment may have disproportionate
cascading effects on species diversity maintenance and community dynamics, as well as in
the function and services (Maestre et al. 2012) of this singular semiarid ecosystem.
Acknowledgments We thank to Jorge Isla, Francisco Valera and Javier Rodrı
´
guez for field assistance.
Julio M. Alca
´
ntara offered continuous support and discussion on multiple aspects of this MS and kindly
reviewed it. Jose
´
M. Fedriani and one anonymous reviewer contributed to improve this paper. This work was
funded by projects CGL-2009-08130 of MINECO and RNM-766 of Junta de Andalucı
´
a to PJR and by
European FEDER. Inma Cancio was supported by FPI grant BES-2010-035999 of MINECO.
Biodivers Conserv
123

References
Baker PJ, Harris SH (2007) Urban mammals: what does the future hold? An analysis of the factors affecting
patterns of use of residential gardens in Great Britain. Mamm Rev 37:297–315
Benito BM, Martı
´
nez-Ortega MM, Mun
˜
oz LM, Lorite J, Pen
˜
as J (2009) Assessing extinction—risk of
endangered plants using species distribution models as: a case study of habitat depletion caused by the
spread greenhouse. Biodivers Conserv 18:2509–2520
Bianchi FJJA et al (2006) Sustainable pest regulation in agricultural landscapes: a review on landscape
composition, biodiversity and natural pest control. Proc R Soc B 273:1715–1727
Cavallini P (1994) Faeces count as an index of fox abundance. Acta Theriol 39:417–424
Contesse P, Hegglinb D, Gloorc S, Bontadinad F, Deplazes P (2004) The diet of urban foxes (Vulpes vulpes)
and the availability of anthropogenic food in the city of Zurich, Switzerland. Mamm Biol 69:81–95
Cordeiro NJ, Howe HF (2003) Forest fragmentation severs mutualism between seed dispersers and an
endemic African tree. Proc Natl Acad Sci USA 100:14052–14056
Cordeiro NJ, Ndangalasi HJ, Mcentee JP, Howe HF (2009) Disperser limitation and recruitment of an
endemic African tree in a fragmented landscape. Ecology 90:1030–1041
Delibes-Mateos M, Fernandez de Simon J, Villafuerte R, Ferreras P (2008) Feeding responses of the red fox
(Vulpes vulpes) to different wild rabbit (Oryctolagus cuniculus) densities: a regional approach. Eur J
Wildl Res 54:71–78
Dell’ Arte GL, Laaksonen T, Norrdahl K, Korpima
¨
ki E (2007) Variation in the diet composition of a
generalist predator, the red fox, in relation to season and density of main prey. Acta Oecol 31:276–281
Dı
´
az-Ruiz F, Delibes-Mateos M, Garcı
´
a-Moreno JL, Lo
´
pez-Martı
´
n JM, Ferreira C, Farreras P (2013)
Biogeographical patterns in the diet of an opportunistic predator: the red fox Vulpes vulpes in the
Iberian Peninsula. Mamm Rev 43:59–70
Escribano-A
´
vila G, Calvino-Cancela M, Pı
´
as B, Virgo
´
s E, Valladares F, Escudero A (2014) Diverse guilds
provide complementary dispersal services in a woodland expansion process after land abandonment.
J Appl Ecol 51:1701–1714
Fahrig L (2003) Effect of habitat fragmentation on biodiversity. Annu Rev Ecol Evol Syst 34:487–515
Fahrig L, Baudry J, Brotons L, Burel FG, Crist TO, Fuller RJ et al (2011) Functional landscape hetero-
geneity and animal biodiversity in agricultural landscapes. Ecol Lett 14:101–112
Fedriani JM, Delibes M (2009) Functional diversity in fruit-frugivore interactions: a field experiment with
Mediterranean mammals. Ecography 32:983–992
Fedriani JM, Wiegand T (2014) Hierarchical mechanisms of spatially contagious seed dispersal in complex
seed-disperser networks. Ecology 95:514–526
Fedriani JM, Palomares F, Delibes M (1999) Niche relations among three sympatric Mediterranean car-
nivores. Oecologia 121:138–148
Fedriani JM, Fuller TK, Sauvajot R (2001) Does availability of anthropogenic foods enhance densities of
omnivorous mammals? An example with coyotes in southern California. Ecography 24:325–331
Fedriani JM, Wiegand T, Delibes M (2010) Spatial pattern of adult trees and the mammal-generated seed
rain in the Iberian pear. Ecography 33:545–555
Fontu
´
rbel FE, Candia AB, Malebra
´
n J, Salazar DA, Gonza
´
lez-Browne C, Medel R (2015) Meta-analysis of
anthropogenic habitat disturbance effects on animal-mediated seed dispersal. Glob Change Biol
21:3951–3960
Garcı
´
a D, Zamora R, Amico GC (2010) Birds as suppliers of seed dispersal in temperate ecosystems:
conservation guidelines from real world landscapes. Conserv Biol 24:1070–1079
Gittleman JL, Funks SM, Macdonald D, Wayne RK (2001) Carnivore conservation. Cambridge University
Press, Cambridge
Goldyn B, Hromada M, Surmacki A, Tryjanowski P (2003) Habitat use and diet of the red fox Vulpes vulpes
in an agricultural landscape in Poland. Z Jagdwiss 49:191–200
Gonza
´
lez-Varo JP, Nora S, Aparicio A (2012) Bottlenecks for plant recruitment in woodland remnants: an
ornithochorous shrub in a Mediterranean ‘relictual’ landscape. Perspect Plant Ecol Evol Syst
14:111–122
Grice AC (1996) Seed production, dispersal and germination in Crytostegia grandiflora and Ziziphus
mauritiana, two invasive shrubs in tropical woodlands of northern Australia. Aust J Ecol 21:324–331
Hartova
´
-Nentvichova
´
aM,S
ˇ
a
´
leka M, C
ˇ
erveny
´
J, Koubek P (2010) Variation in the diet of the red fox
(Vulpes vulpes) in mountain habitats: effects of altitude and season. Mamm Biol 75:334–340
ISTA (International Seed Testing Association) (1999) Proceedings of the international seed testing asso-
ciation. Seed Science and Technology 27 (Suppl), 333 pp
Jordano P, Herrera CM (1995) Shuffling the offpring: uncoupling and spatial discordance of multiple stages
in vertebrate seed dispersal. E
´
coscience 2:230–237
Biodivers Conserv
123

Kremen C et al (2007) Pollination and other ecosystem services produced by mobile organisms: a con-
ceptual framework for the effects of land-use change. Ecol Lett 10:299–314
Kurki S, Nikula A, Helle P, Linden H (1998) Abundances of red fox and pine marten in relation to the
composition of boreal forest landscapes. J Anim Ecol 67:874–886
Lindenmayer D, Fischer J (2006) Habitat fragmentation and landscape change: an ecological and conser-
vation synthesis. Island Press, Whashington
Lo
´
pez-Bao JV, Gonza
´
lez-Varo JP (2011) Frugivory and spatial patterns of seed deposition by carnivorous
mammals in anthropogenic landscapes: a multi-scale approach. PloS One 6:e14569
Lovari S, Lucherini M, Crema G (1996) Individual variations in diet, activity and habitat use of red foxes in
a mediterranean rural areas. J Wild Res 1:24–31
Maestre FT, Quero JL, Gotelli NJ, Escudero A, Ochoa V, Delgado-Baquerizo M et al (2012) Plant species
richness and ecosystem multifunctionality in global drylands. Science 335:214–218
Magrach A, Laurance WF, Larrinaga AR, Santamarı
´
a L (2014) Meta-analysis of the effects of forest
fragmentation on interspecific interactions. Conserv Biol 28:1342–1348
Maraghini M, Gorai M, Neffati M (2010) Seed germination at different temperatures and water stress levels,
and seedling emergence from different depths of Ziziphus lotus. S Afr J Bot 76:453–459
Markl J, Schleuning M, Forget PM, Jordano P, Lambert JE, Traveset A, Wright SJ, Bo
¨
hning-Gaese K (2012)
Meta-analysis of the effects of human disturbance on seed dispersal by animals. Conserv Biol
26:1072–1081
Matı
´
as L, Zamora R, Mendoza I, Ho
´
dar JA (2010) Seed dispersal patterns by large frugivorous mammals in
a degraded mosaic landscape. Restor Ecol 18:619–627
McIntire S, Hobbs R (1999) A framework for conceptualizing human effects on landscapes and its relevance
to management and research models. Conserv Biol 13:1282–1292
Mendoza-Ferna
´
ndez AJ, Martı
´
nez-Herna
´
ndez F, Pe
´
rez-Garcı
´
a FJ, Garrido-Becerra JA, Benito BM, Sal-
mero
´
n-Sa
´
nchez E, Guirado J, Merlo ME, Mota JF (2015) Extreme habitat loss in a Mediterranean
habitat: Maytenus senegalensis subsp. europaea. Plant Biosyst 149:503–511
Mota JF, Pen
˜
as J, Castro H, Cabello J, Guirado JS (1996) Agricultural development vs biodiversity con-
servation: the Mediterranean semiarid vegetation in El Ejido (Almeria, southeastern Spain). Biodivers
Conserv 5:1597–1617
Oehler JD, Litvaitis JA (1996) The role of spatial scale in understanding responses of medium-sized
carnivores to forest fragmentation. Can J Zool 74:2070–2079
Rey PJ, Alca
´
ntara JM (2000) Recruitment dynamics of a fleshy-fruited plant (Olea europaea): connecting
patterns of seed dispersal to seedling establishment. J Ecol 88:622–633
Rey PJ, Alca
´
ntara JM (2014) Effects of habitat alteration on the effectiveness of plant-avian seed dispersal
mutualisms: consequences for plant regeneration. Perspect Plant Ecol Evol Syst 16:21–31
Rodrı
´
guez-Cabal MA, Aizen MA, Novaro AJ (2007) Habitat fragmentation disrupts a plant-disperser
mutualism in the temperate forest of South America. Biol Conserv 139:195–202
Rosalino LM, Santos-Reis M (2009) Fruit consumption by carnivores in Mediterranean Europe. Mamm Rev
39:67–78
Rosalino LM, Rosa S, Santos-Reis M (2010) The role of carnivores as Mediterranean seed dispersers. Ann
Zool Fenn 47:195–205
Sala OE, Chapin FS, Armesto JJ, Berlow E, Bloomfield J, Dirzo R et al (2000) Global biodiversity scenarios
for the year 2100. Science 287:1770–1774
Salek M, Drahnı
´
kova
´
L, Tkadlec E (2014) Changes in home range sizes and population densities of
carnivore species along the natural to urban habitat gradient. Mamm Rev 45:1–14
Santos T, Tellerı
´
a JL (1994) Influence of forest fragmentation on seed consumption and dispersal of Spanish
Juniper Juniperus thurifera. Biol Conserv 70:129–134
Santos T, Tellerı
´
a JL, Virgo
´
s E (1999) Dispersal of Spanish juniper Juniperus thurifera
by birds and
mammals in a fragmented landscape. Ecography 22:193–204
Schupp EW, Jordano P, Go
´
mez JM (2010) Seed dispersal effectiveness revisited: a conceptual review. New
Phytol 188:333–353
Silva SI, Bozinovic F, Jaksic FM (2005) Frugivory and seed dispersal by foxes in relation to mammalian
prey abundance in a semiarid thornscrub. Austral Ecol 30:739–746
Sua
´
rez-Esteban A, Delibes M, Fedriani JM (2013) Barriers or corridors? The overlooked role of unpaved
roads in endozoochorous seed dispersal. J Appl Ecol 50:767–774
Swift MJ, Izac AMN, van Noordwijk M (2004) Biodiversity and ecosystem services in agricultural land-
scapes—are we asking the right questions? Agric. Ecosyst Environ 104:113–134
Tirado R (2009) 5220 Matorrales arborescentes con Ziziphus. In: VV.AA. Bases ecolo
´
gicas preliminares
para la conservacio
´
n de los tipos de ha
´
bitat de intere
´
s comunitario en Espan
˜
a. Ministerio de Medio
Ambiente, Medio Rural y Marino. Madrid, p 68
Biodivers Conserv
123

Traba J, Arrieta S, Herranz J, Clamagirand MC (2006) Red fox (Vulpes vulpes L.) favour seed dispersal,
germination and seedling survival of Mediterranean Hackberry (Celtis australis). Acta Oecol 30:39–45
Traveset A, Gonza
´
lez-Varo JP, Valido A (2012) Long-term demographic consequences of a seed dispersal
disruption. Proc R Soc Lond B Biol Sci 279:3298–3303
Valiente-Banuet A et al (2015) Beyond the species loss: the extinction of ecological interactions in a
changing world. Funct Ecol 29:299–307
Valladares F, Camarero JJ, Pulido F, Gil-Pelegrı
´
n E (2004) El bosquemediterra
´
neo, un sistema humanizado
y dina
´
mico. In: Valladares F (ed) Ecologı
´
a del bosque mediterra
´
neo en un mundo cambiante. Min-
isterio de Medio Ambiente, EGRAF, Madrid, pp 13–25
Varela O (2004) Frugivorı
´
a y dispersion de semillas de 13especies de vertebtrados en el Charco salten
˜
o,
Argentina. Tesis doctoral. Universidad de Co
´
rdoba, Argentina
Virgo
´
s E, Telleria JL, Santos T (2002) A comparison on the response to forest fragmentation by medium-
sized Iberian carnivores in central Spain. Biodivers Conserv 11:1063–1079
Varela O, Bucher EH (2006) Passage time, viability, and germination of seeds ingested by foxes. J Arid
Environ 67:566–578
Webbon CC, Baker PJ, Harris S (2004) Faecal density counts for monitoring changes in red fox numbers in
rural Britain. J Appl Ecol 41:768–779
Woodroffe R (2000) Predators and people: using human density to interpret declines of large carnivores.
Anim Conserv 3:165–173
Zhang SY, Wang LX (1995) Fruit consumption and seed dispersal of Ziziphus cinnamomum (Rhamnaceae)
by two sympatric primates (Cebiis apella and Ateles paniscus) in French Guiana. Biotropica
27:397–401
Biodivers Conserv
123