Pino et al 2006
Hydrobiologia (2006) 570:257–263 Ó Springer 2006
J.M. Caffrey, A. Dutartre, J. Haury, K.J. Murphy & P.M. Wade (eds), Macrophytes in Aquatic Ecosystems: From Biology to Management
DOI 10.1007/s10750-006-0189-x
Invasibility of four plant communities in the Llobregat delta (Catalonia, NE
of Spain) in relation to their historical stability
Joan Pino1,2,*, Josep Maria Seguı´ 3 & Nora Alvarez4
1
Center for Ecological Research and Forestry Applications (CREAF), Universitat Auto`noma de Barcelona, E-08193
Bellaterra, Spain
2
Departament de Biologia Vegetal, Universitat de Barcelona, Diagonal 645, E-08028 Barcelona, Spain
3
´n
Me´ndez Nu ˜ez 1, E-08003 Barcelona, Spain
4
Department of Biology, University of Puerto Rico, PO Box 23360, San Jua, PR 00931-3360 USA
(*Author for correspondence: E-mail: joan.pino@uab.es)
Key words: landscape history, alien species, coastal habitats, Llobregat delta
Abstract
Presence and cover of alien plants were analysed in relation to recent naturalness changes (1956–1999) in
the Llobregat delta by means of GIS techniques and field surveys. Two land cover maps of 1956 and
1999 were generated by photo-interpretation of orthoimages and they were then reclassified into natu-
ralness classes, defined as the degree of preservation of the pristine state. The resulting naturalness maps
were combined in order to obtain a naturalness change map, which was used to design field sampling in
four pristine communities: reedbeds, rushbeds, halophilous scrubs and fixed dune communities. Two
study areas were selected for each community and three stability regimes (stable, semi-stable and non-
stable) obtained from the naturalness change map. Five vegetation inventories were performed on
average in each of these areas using the classical sigmatist method. Results showed a negative rela-
tionship between stability and invasibility, with several variations between communities. No alien species
were found in stable areas of all communities. Alien species number, species percentage and relative cover
increased from semi-stable to non-stable regimes in reedbeds and dune communities, indicating that
reversion towards the climax reduces opportunities for alien establishment in these communities. In
contrast, halophilous habitats such as rushbeds and scrubs did not exhibit significant differences between
semi-stable and non-stable plots, probably because saline stress makes their invasion by alien plants
difficult, even under disturbance.
Introduction Huenneke, 1992; Olenin & Leppanoski, 1999).
¨
Aquatic ecosystems are particularly susceptible to
Invasion by alien species is a worldwide phenom- invasion because of their intermediate to high
enon with recognised negative effects on the con- disturbance regimes (di Castri, 1990; Rauchich &
servation of native biodiversity (Lodge, 1993). The Reader, 1999), with a number of studies indicating
close relationship between habitat disturbance and the concentration of alien species in ponds and
invasibility is widely accepted (di Castri, 1990; rivers (Pysˇ ek & Prach, 1993; Alpert et al., 2000).
Vitousek et al., 1997; Hobbs, 2000). Disturbance Land use changes driven by man are a main
reduces competition and increases the availability source of habitat disturbance, and consequently of
of safe sites, providing more opportunities for alien introduction, in terrestrial ecosystems
alien colonization and spreading (Hobbs & (Hobbs, 2000) but also in wetlands (Ewel, 1986).
258
Despite the major role of land use changes in map of 1956 was photo-interpreted on an
habitat invasibility, there are few studies aimed at orthophotomap generated ad hoc, by geo-cor-
examining the association between stability and rection and mosaic of black and white aerial
the risk of invasion of natural habitats from an photographs. In contrast, the 1999 map was di-
historical perspective. The present study analyses rectly photo-interpreted on colour orthopho-
long-term (1956–1999) changes (including persis- tomaps produced by the Cartographic Institute
tence, recuperation and degradation) and their of Catalonia. Contrasting colour and quality of
possible association with invasibility in several images, and the availability of ancillary infor-
communities in the Llobregat delta (central coast mation on vegetation categories for 1999 but not
of Catalonia, NE of Spain), in the metropolitan for 1956 determined a contrasting thematic res-
area of Barcelona. Because of intense land use olution for both land cover maps. In order to
change and trading, the Llobregat delta has be- homogenise the results of photo-interpretation,
come one of the regions most invaded by alien to facilitate map comparison and to reduce the
plants in Catalonia (Casasayas, 1990). In our number of land cover classes to be analysed, the
study, GIS techniques have been used to find areas land cover classes were reclassified into four
of contrasting habitat stability comparing 1956 naturalness classes. We defined naturalness as
and 1999. A field survey of main wetland and dune the degree of preservation of the pristine state
communities has been conducted in these areas in (Figs. 1a,b): High (natural marshland habitats),
order to assess the relationship between their sta- medium (disturbed natural habitats and formerly
bility and invasibility. abandoned fields), low (crops and recently
mowed areas), and null (urban areas, roads and
railways). A naturalness change map describing
Description of site land stability was then obtained by combining
the naturalness maps of 1956 and 1999 (Fig. 1c).
The study was done in the southern hemidelta of Only three stability regimes were considered for
the Llobregat, adjacent to the city of Barcelona the present study: stable (high naturalness in
and still dominated by natural and agricultural 1956 and 1999), semi-stable (medium or low
habitats (Fig. 1). At present, marshlands and naturalness in 1956, and high in 1999), and non-
fixed dunes correspond to 13% of the total land stable (medium naturalness in 1999).
cover, with reedbeds (Phragmition australis The relationship between invasibility and sta-
W. Koch), rushbeds (Juncion maritimi Br. Bl.), bility was assessed in four plant communities
and secondary pine (Pinus pinea L.) forests as representing several of the pristine habitats in the
dominant plant communities. Fragments of hal- ` `
Llobregat delta (Bolos & Bolos, 1950) and
ophilous scrublands (Arthrocnemion fruticosi Br. responding to contrasting conditions of water
Bl.) and fixed dune communities (Crucianellion availability and conductivity: reedbeds, rushbeds,
maritimae Rivas Goday and Rivas Mart.) with halophilous scrublands, and fixed dune commu-
variable conservation status still persist. Despite nities. Two study areas for each plant community
some of the remaining natural areas having been and stability regime were selected using the nat-
declared Natura 2000 sites, their conservation is uralness change map and additional information
threatened by intense land-use change, water on current vegetation. In each area, a mean of 5
pollution and degradation. About 15% of the 10Â10 m plots were marked out and vegetation
areas occupied by natural habitats exhibit mod- inventories were performed therein following the
erate to intense degradation. classical sigmatist method. The number and per-
centage of alien species and their relative cover
were compared among communities, stability re-
Material and methods gimes and areas, by means of ANOVA tests after
normalisation by arcsin transformation. Pairwise
Two land cover maps of 1956 and 1999 of the comparisons between semi stable and non-stable
study area were generated by photo-interpreta- regimes were carried out a posteriori for each
tion of orthophotomaps at 1:5 000 scale. The community using Tukey’s test.
259
Results communities studied. Cortaderia selloana (Schultes
ex Schultes) Asch. et Graetbn. was found in reedbeds
A total of 17 alien species were found in the study and secondarily in fixed dune communities, whereas
(Table 1). Fixed dune communities were invaded by Cuscuta campestris Yuncker was mainly located in
most species (11), followed by reedbeds (10), rush- fixed dunes and occasionally in rushbeds. Oenothera
beds (4) and halophilous scrubs (2). Aster squamatus glazioviana Minchx. and Carpobrotus edulis (L.)
(Spreng.) Hieron. was the most frequent alien, N.E. Br.were exclussively found in fixed dune com-
growing in 25% of samples and colonizing all the munities. Seven species were found only once.
Figure 1. Habitat naturalness in (a) 1956, and (b) 1999 in the southern hemidelta of Llobregat, and (c) the associated naturalness
change map showing the stability regimes selected for the study (see text for more details).
260
Table 1. Percentage of inventories of each community and stability regime with presence of each alien species. Stability regimes: SS,
semi-stable; NS, non-stable. There were no alien species in the inventories of the stable regime
Reedbeds Fixed dunes Rushbeds Halophilous
scrubs
SS NS SS NS SS NS SS NS
Amaranthus retroflexus L. 7.7
Ambrosia coronopifolia Torrey et A. Gray 7.7
Arundo donax L. 23.1 7.7
Asparagus officinalis L. 7.7
Aster squamatus (Spreng.) Hieron. 18.2 23.1 9.1 38.5 75.0 83.3 45.5 50.0
Carpobrotus edulis (L.) N.E. Br. 9.1 30.8
Chenopodium ambrosioides L. 7.7 7.7
Conyza bonariensis ( L.) Cronq. 7.7
Conyza sumatrensis (Retz.) E. Walter 30.8 18.2 30.8 8.3
Cortaderia selloana (Schultes ex Schultes) Asch. et Graetbn. 30.8 7.7
Cuscuta campestris Yuncker 36.4 30.8 8.3
Ipomoea indica (Burm.) Merr. 7.7
Lonicera japonica Thunb. in Murray 7.7
Oenothera glazioviana Minchx. 45.5 23.1
Phoenix canariensis Chabaud 9.1
Rumex palustris Sm 23.1 40.0
Xanthium echinatum Murray 38.5 33.3
Non-stable plots concentrated the majority of significant (p = 0.057) for the relative number of
alien citations in reedbeds and dune communities. aliens. This was the only significantly different
A.squamatus, C. edulis, Conyza sumatrensis (Retz.) factor between areas. The interactions between
E. Walker, and Xanthium echinatum Murray were stability and community, and between area and
found in more than 30% of non-stable plots, but community were significant for both the number
in less than 20% of semi stable ones in fixed dune and the percentage of aliens, but not for the rela-
communities. C. sumatrensis and C. selloana were tive cover of aliens. The interactions between sta-
found in more than 30% of non-stable plots in bility and area were never significant, whereas
reedbeds, whereas Arundo donax L., A. squamatus, third-order interactions between all the parameters
and Rumex palustris Sm colonised more than 20% studied were always significant. Semi-stable and
of these plots. Semi stable plots were only colon- non-stable regimes exhibited contrasting invasi-
ised by A. squamatus, which was also commonly bility patterns among communities (Fig. 2). Non-
found in rushbeds and in halophilous scrubs, stable reedbeds and fixed dune communities
either in semi-stable or non-stable plots. R. palus- showed significantly higher number, percentage
tris and X. echinatum were, respectively, frequent and relative cover of alien plants than semi-stable
in non-stable plots in halophilous scrubs and ones. In contrast, there were no significant differ-
rushbeds, but absent in semi-stable plots. There ences between non-stable and semi-stable plots in
were no records of alien species in stable plots, rushbeds and halophilous scrubs.
despite A. squamatus being observed at times in
areas nearby these plots.
Stability was significantly related to the number Discussion
and the proportion of alien species, and also to
their relative cover (Table 2). Differences between Most of the Catalan alien species occur in heavily
communities were significant for the number and human-disturbed habitats, with few of them able
the relative cover of alien species, and marginally to succeed in natural communities (Casasayas,
261
Table 2. Summary of ANOVA aimed at comparing the effects of stability regime, community, and area on the number and the
percentage of alien species, and on the percentage of species cover corresponding to aliens
df MS F p
Number of alien species
Stability 2 169.216 25.567 <0.001
Community 3 481.798 72.795 <0.001
Area 1 4.796 0.725 0.396
Stability  community 6 107.464 16.237 <0.001
Stability  area 2 3.512 0.531 0.590
Community  area 3 86.410 13.056 <0.001
Stability  community  area 6 34.821 5.261 0.001
Percentage of alien species
Stability 2 8.383 63.063 <0.001
Community 3 0.343 2.584 0.057
Area 1 1.010 7.597 0.007
Stability  community 6 0.461 3.471 0.003
Stability  area 2 0.323 2.429 0.093
Community  area 3 0.584 4.397 0.006
Stability  community  area 6 1.288 9.688 <0.001
Percentage of alien species cover
Stability 2 3.695 36.210 <0.001
Community 3 0.290 2.840 0.041
Area 1 0.251 2.462 0.119
Stability  community 6 0.196 1.923 0.083
Stability  area 2 0.063 0.619 0.540
Community  area 3 0.186 1.825 0.146
Stability  community  area 6 0.745 7.296 <0.001
1990). We have found that natural, coastal plant invasion. Disturbance might provide gaps that
communities in the Llobregat delta are potentially would enhance the establishment of invaders, as
invaded by a number of alien plants, which appear Jones & Doren (1997) reported for the exotic tree
to be relatively non-specific with the exception of Schinus terebinthifolius in the Everglades.
several characteristic aliens of dune communities Invasibility of reedbeds and fixed dune com-
(C. edulis, O. erythrosepala, and to a minor extent munities is related to historical stability, since
C. campestris). In addition, habitat invasibility was alien species number and proportion and relative
related negatively to stability. Communities en- alien cover increased significantly from semi-
closed in areas that have persisted unaltered since stable to non-stable regimes. Assuming that
1956 exhibited a very low invasibility, with only a historical change indicates how long ago main
single species, A. squamatus, growing at extremely disturbances occurred, the decrease in invasibility
low densities. It is well established that disturbed from non-stable to semi-stable regimes would
habitats are regarded to be more vulnerable to indicate a reversion towards the climax. During
invasion than unaltered ones (di Castri, 1990; this process, the establishment of non-natives
Vitousek et al., 1997; Hobbs, 2000). This is par- may be less opportune, basically as a result of
ticularly true in perennial communities dominated decreasing resources or increasing colonization
by one or a few species, especially in reedbeds, but of clonal, dominant species that would reduce
also in several rushbeds and even dune communi- safe sites for alien germination and establishment
ties, whose vegetative regeneration mechanisms (Hobbs & Huenneke, 1992). However, this
are responsible for a dense canopy that prevents pattern was not valid for rushbeds nor for
262
Figure 2. Number and percentage of alien species, and relative alien cover in four coastal communities in the Llobregat delta, in
relation to the stability regimes selected for the study (see text for more details). Means and standard errors are shown. Significance of
pairwise comparisons between semi-stable and non-stable regimes is shown (*: p < 0.05; **: p < 0.01; n: no significant).
Arthrocnemum scrubs in the Llobregat delta. even native plants is extremely difficult, being
Invasibility of a given habitat also depends on restricted to vegetation patches and nearby areas
traits related to their adequacy for plant estab- with lower conductivity and higher water and
lishment, such as resource supply, non-biotic nutrient contents (Rubio-Casal et al., 2001).
and biotic conditions (Alpert et al., 2000; Heger,
2001). Saline stress probably prevents these hal-
ophilous communities from invasion, as Alpert Acknowledgements
et al. (2000) reported for other harsh habitats
such as xeric grasslands and desert vegetation in This work has been funded by the CICYT pro-
relation to drought and nutrient stress. Although jects REN2000-0361 GLO and SEC2000-0836-
bare soil might be abundant in halophil- C04-04 of the Spanish Ministry of Science and
ous communities, its colonisation by alien or Technology.
263
References Hobbs, R. J. & L. F Huenneke, 1992. Disturbance, diversity
and invasion: implications for conservation. Conservation
Alpert, P., E. Bone & C. Holzapfel, 2000. Invasiveness, inva- Biology 6: 324–337.
sibility and the role of environmental stress in the spread of Jones, D. T. & R. F. Doren, 1997. The distribution, biology and
non-native plants. Perspectives in Plant Ecology, Evolution control of Schinus terebinthifolius in southern Florida. With
and Systematics 3: 52–66. special reference to Everglades National Park. In Brock, J.
` ` ´
Bolos, A. & O. Bolos, 1950. La vegetacion de las comarcas H., M. Wade, P. Pysˇ ek & D. Green (eds), Plant Invasions:
´
barcelonesas. Instituto Espanol de Estudios Mediterraneos,
˜ Studies from North America and Europe. Blackhuys pub-
Barcelona, 350 pp. lishers, Leiden, 81–93.
Casasayas, T., 1990. Widespread adventive plants in Catalonia. Lodge, D. M., 1993. Species invasions and deletions. In Kare-
In di Castri, F., A. J. Hansen & M. Debussche (eds), Bio- iva, P. M., J. G. Kingsolver & R. B. Honey (eds), Biotic
logical Invasions in Europe and the Mediterranean Basin. Interactions and Global Change. Sunderland, Massachus-
Kluwer academic Publishers, Dodretch, 85–104. sets, 367–387.
di Castri, F., 1990. On invading species and invaded ecosys- Olenin, S. & E. Leppanoski, 1999. Non-native animals in the
¨
tems: the interplay of historical chance and biological Baltic Sea: alteration of benthic habitats in coastal inlets and
necessity. In di Castri, F., A. J. Hansen & M. Debussche lagoons. Hydrobiologia 393: 233–243.
(eds), Biological Invasions in Europe and the Mediterranean Pysˇ ek, P. & K. Prach, 1993. Plant invasion and the role of
Basin. Kluwer academic Publishers, Dodretch, 3–16. riparian habitats: a comparison of four species alien to
Ewel, J. J., 1986. Invasibility: lessons from South Florida. In central Europe. Journal of Biogeography 20: 413–420.
Mooney, H. A. & J. A. Drake (eds), Ecology of biological Rauchich, J. & R. J. Reader, 1999. An experimental study of
invasions of North America and Hawaii. Springer-Verlag, wetland invisibility by purple loosestrife (Lythrum salicaria).
New York, 214–230. Canadian Journal of Botany 77: 1499–1503.
Heger, T., 2001. A model for interpreting the process of inva- Rubio-Casal, A. E., J. M. Castillo, C. J. Luque & M. E. Fig-
sion: crucial situations favouring special characteristics of uero, 2001. Nucleation and facilitation in salt pans in
invasive spacies. In Brundu, G., J. Brock, I. Camarda, L. Mediterranean salt marshes. Journal of Vegetation Science
Child & M. Wade (eds), Plant Invasions: Species Ecology and 12: 761–770.
Ecosystem Management. Backhuys Publishers, Leiden, 3–9. Vitousek, P. M., H. A. Mooney, J. Lubchenco & J. M. Melillo,
Hobbs, R. J., 2000. Land-use changes and invasions. In Moo- 1997. Human domination of Earth’s ecosystems. Science
ney H. A. & R. J. Hobbs (eds), Invasive Species in a 277: 494–499.
Changing World. Island Press, Washington, DC, 55–64.
J.M. Caffrey, A. Dutartre, J. Haury, K.J. Murphy & P.M. Wade (eds), Macrophytes in Aquatic Ecosystems: From Biology to Management
DOI 10.1007/s10750-006-0189-x
Invasibility of four plant communities in the Llobregat delta (Catalonia, NE
of Spain) in relation to their historical stability
Joan Pino1,2,*, Josep Maria Seguı´ 3 & Nora Alvarez4
1
Center for Ecological Research and Forestry Applications (CREAF), Universitat Auto`noma de Barcelona, E-08193
Bellaterra, Spain
2
Departament de Biologia Vegetal, Universitat de Barcelona, Diagonal 645, E-08028 Barcelona, Spain
3
´n
Me´ndez Nu ˜ez 1, E-08003 Barcelona, Spain
4
Department of Biology, University of Puerto Rico, PO Box 23360, San Jua, PR 00931-3360 USA
(*Author for correspondence: E-mail: joan.pino@uab.es)
Key words: landscape history, alien species, coastal habitats, Llobregat delta
Abstract
Presence and cover of alien plants were analysed in relation to recent naturalness changes (1956–1999) in
the Llobregat delta by means of GIS techniques and field surveys. Two land cover maps of 1956 and
1999 were generated by photo-interpretation of orthoimages and they were then reclassified into natu-
ralness classes, defined as the degree of preservation of the pristine state. The resulting naturalness maps
were combined in order to obtain a naturalness change map, which was used to design field sampling in
four pristine communities: reedbeds, rushbeds, halophilous scrubs and fixed dune communities. Two
study areas were selected for each community and three stability regimes (stable, semi-stable and non-
stable) obtained from the naturalness change map. Five vegetation inventories were performed on
average in each of these areas using the classical sigmatist method. Results showed a negative rela-
tionship between stability and invasibility, with several variations between communities. No alien species
were found in stable areas of all communities. Alien species number, species percentage and relative cover
increased from semi-stable to non-stable regimes in reedbeds and dune communities, indicating that
reversion towards the climax reduces opportunities for alien establishment in these communities. In
contrast, halophilous habitats such as rushbeds and scrubs did not exhibit significant differences between
semi-stable and non-stable plots, probably because saline stress makes their invasion by alien plants
difficult, even under disturbance.
Introduction Huenneke, 1992; Olenin & Leppanoski, 1999).
¨
Aquatic ecosystems are particularly susceptible to
Invasion by alien species is a worldwide phenom- invasion because of their intermediate to high
enon with recognised negative effects on the con- disturbance regimes (di Castri, 1990; Rauchich &
servation of native biodiversity (Lodge, 1993). The Reader, 1999), with a number of studies indicating
close relationship between habitat disturbance and the concentration of alien species in ponds and
invasibility is widely accepted (di Castri, 1990; rivers (Pysˇ ek & Prach, 1993; Alpert et al., 2000).
Vitousek et al., 1997; Hobbs, 2000). Disturbance Land use changes driven by man are a main
reduces competition and increases the availability source of habitat disturbance, and consequently of
of safe sites, providing more opportunities for alien introduction, in terrestrial ecosystems
alien colonization and spreading (Hobbs & (Hobbs, 2000) but also in wetlands (Ewel, 1986).
258
Despite the major role of land use changes in map of 1956 was photo-interpreted on an
habitat invasibility, there are few studies aimed at orthophotomap generated ad hoc, by geo-cor-
examining the association between stability and rection and mosaic of black and white aerial
the risk of invasion of natural habitats from an photographs. In contrast, the 1999 map was di-
historical perspective. The present study analyses rectly photo-interpreted on colour orthopho-
long-term (1956–1999) changes (including persis- tomaps produced by the Cartographic Institute
tence, recuperation and degradation) and their of Catalonia. Contrasting colour and quality of
possible association with invasibility in several images, and the availability of ancillary infor-
communities in the Llobregat delta (central coast mation on vegetation categories for 1999 but not
of Catalonia, NE of Spain), in the metropolitan for 1956 determined a contrasting thematic res-
area of Barcelona. Because of intense land use olution for both land cover maps. In order to
change and trading, the Llobregat delta has be- homogenise the results of photo-interpretation,
come one of the regions most invaded by alien to facilitate map comparison and to reduce the
plants in Catalonia (Casasayas, 1990). In our number of land cover classes to be analysed, the
study, GIS techniques have been used to find areas land cover classes were reclassified into four
of contrasting habitat stability comparing 1956 naturalness classes. We defined naturalness as
and 1999. A field survey of main wetland and dune the degree of preservation of the pristine state
communities has been conducted in these areas in (Figs. 1a,b): High (natural marshland habitats),
order to assess the relationship between their sta- medium (disturbed natural habitats and formerly
bility and invasibility. abandoned fields), low (crops and recently
mowed areas), and null (urban areas, roads and
railways). A naturalness change map describing
Description of site land stability was then obtained by combining
the naturalness maps of 1956 and 1999 (Fig. 1c).
The study was done in the southern hemidelta of Only three stability regimes were considered for
the Llobregat, adjacent to the city of Barcelona the present study: stable (high naturalness in
and still dominated by natural and agricultural 1956 and 1999), semi-stable (medium or low
habitats (Fig. 1). At present, marshlands and naturalness in 1956, and high in 1999), and non-
fixed dunes correspond to 13% of the total land stable (medium naturalness in 1999).
cover, with reedbeds (Phragmition australis The relationship between invasibility and sta-
W. Koch), rushbeds (Juncion maritimi Br. Bl.), bility was assessed in four plant communities
and secondary pine (Pinus pinea L.) forests as representing several of the pristine habitats in the
dominant plant communities. Fragments of hal- ` `
Llobregat delta (Bolos & Bolos, 1950) and
ophilous scrublands (Arthrocnemion fruticosi Br. responding to contrasting conditions of water
Bl.) and fixed dune communities (Crucianellion availability and conductivity: reedbeds, rushbeds,
maritimae Rivas Goday and Rivas Mart.) with halophilous scrublands, and fixed dune commu-
variable conservation status still persist. Despite nities. Two study areas for each plant community
some of the remaining natural areas having been and stability regime were selected using the nat-
declared Natura 2000 sites, their conservation is uralness change map and additional information
threatened by intense land-use change, water on current vegetation. In each area, a mean of 5
pollution and degradation. About 15% of the 10Â10 m plots were marked out and vegetation
areas occupied by natural habitats exhibit mod- inventories were performed therein following the
erate to intense degradation. classical sigmatist method. The number and per-
centage of alien species and their relative cover
were compared among communities, stability re-
Material and methods gimes and areas, by means of ANOVA tests after
normalisation by arcsin transformation. Pairwise
Two land cover maps of 1956 and 1999 of the comparisons between semi stable and non-stable
study area were generated by photo-interpreta- regimes were carried out a posteriori for each
tion of orthophotomaps at 1:5 000 scale. The community using Tukey’s test.
259
Results communities studied. Cortaderia selloana (Schultes
ex Schultes) Asch. et Graetbn. was found in reedbeds
A total of 17 alien species were found in the study and secondarily in fixed dune communities, whereas
(Table 1). Fixed dune communities were invaded by Cuscuta campestris Yuncker was mainly located in
most species (11), followed by reedbeds (10), rush- fixed dunes and occasionally in rushbeds. Oenothera
beds (4) and halophilous scrubs (2). Aster squamatus glazioviana Minchx. and Carpobrotus edulis (L.)
(Spreng.) Hieron. was the most frequent alien, N.E. Br.were exclussively found in fixed dune com-
growing in 25% of samples and colonizing all the munities. Seven species were found only once.
Figure 1. Habitat naturalness in (a) 1956, and (b) 1999 in the southern hemidelta of Llobregat, and (c) the associated naturalness
change map showing the stability regimes selected for the study (see text for more details).
260
Table 1. Percentage of inventories of each community and stability regime with presence of each alien species. Stability regimes: SS,
semi-stable; NS, non-stable. There were no alien species in the inventories of the stable regime
Reedbeds Fixed dunes Rushbeds Halophilous
scrubs
SS NS SS NS SS NS SS NS
Amaranthus retroflexus L. 7.7
Ambrosia coronopifolia Torrey et A. Gray 7.7
Arundo donax L. 23.1 7.7
Asparagus officinalis L. 7.7
Aster squamatus (Spreng.) Hieron. 18.2 23.1 9.1 38.5 75.0 83.3 45.5 50.0
Carpobrotus edulis (L.) N.E. Br. 9.1 30.8
Chenopodium ambrosioides L. 7.7 7.7
Conyza bonariensis ( L.) Cronq. 7.7
Conyza sumatrensis (Retz.) E. Walter 30.8 18.2 30.8 8.3
Cortaderia selloana (Schultes ex Schultes) Asch. et Graetbn. 30.8 7.7
Cuscuta campestris Yuncker 36.4 30.8 8.3
Ipomoea indica (Burm.) Merr. 7.7
Lonicera japonica Thunb. in Murray 7.7
Oenothera glazioviana Minchx. 45.5 23.1
Phoenix canariensis Chabaud 9.1
Rumex palustris Sm 23.1 40.0
Xanthium echinatum Murray 38.5 33.3
Non-stable plots concentrated the majority of significant (p = 0.057) for the relative number of
alien citations in reedbeds and dune communities. aliens. This was the only significantly different
A.squamatus, C. edulis, Conyza sumatrensis (Retz.) factor between areas. The interactions between
E. Walker, and Xanthium echinatum Murray were stability and community, and between area and
found in more than 30% of non-stable plots, but community were significant for both the number
in less than 20% of semi stable ones in fixed dune and the percentage of aliens, but not for the rela-
communities. C. sumatrensis and C. selloana were tive cover of aliens. The interactions between sta-
found in more than 30% of non-stable plots in bility and area were never significant, whereas
reedbeds, whereas Arundo donax L., A. squamatus, third-order interactions between all the parameters
and Rumex palustris Sm colonised more than 20% studied were always significant. Semi-stable and
of these plots. Semi stable plots were only colon- non-stable regimes exhibited contrasting invasi-
ised by A. squamatus, which was also commonly bility patterns among communities (Fig. 2). Non-
found in rushbeds and in halophilous scrubs, stable reedbeds and fixed dune communities
either in semi-stable or non-stable plots. R. palus- showed significantly higher number, percentage
tris and X. echinatum were, respectively, frequent and relative cover of alien plants than semi-stable
in non-stable plots in halophilous scrubs and ones. In contrast, there were no significant differ-
rushbeds, but absent in semi-stable plots. There ences between non-stable and semi-stable plots in
were no records of alien species in stable plots, rushbeds and halophilous scrubs.
despite A. squamatus being observed at times in
areas nearby these plots.
Stability was significantly related to the number Discussion
and the proportion of alien species, and also to
their relative cover (Table 2). Differences between Most of the Catalan alien species occur in heavily
communities were significant for the number and human-disturbed habitats, with few of them able
the relative cover of alien species, and marginally to succeed in natural communities (Casasayas,
261
Table 2. Summary of ANOVA aimed at comparing the effects of stability regime, community, and area on the number and the
percentage of alien species, and on the percentage of species cover corresponding to aliens
df MS F p
Number of alien species
Stability 2 169.216 25.567 <0.001
Community 3 481.798 72.795 <0.001
Area 1 4.796 0.725 0.396
Stability  community 6 107.464 16.237 <0.001
Stability  area 2 3.512 0.531 0.590
Community  area 3 86.410 13.056 <0.001
Stability  community  area 6 34.821 5.261 0.001
Percentage of alien species
Stability 2 8.383 63.063 <0.001
Community 3 0.343 2.584 0.057
Area 1 1.010 7.597 0.007
Stability  community 6 0.461 3.471 0.003
Stability  area 2 0.323 2.429 0.093
Community  area 3 0.584 4.397 0.006
Stability  community  area 6 1.288 9.688 <0.001
Percentage of alien species cover
Stability 2 3.695 36.210 <0.001
Community 3 0.290 2.840 0.041
Area 1 0.251 2.462 0.119
Stability  community 6 0.196 1.923 0.083
Stability  area 2 0.063 0.619 0.540
Community  area 3 0.186 1.825 0.146
Stability  community  area 6 0.745 7.296 <0.001
1990). We have found that natural, coastal plant invasion. Disturbance might provide gaps that
communities in the Llobregat delta are potentially would enhance the establishment of invaders, as
invaded by a number of alien plants, which appear Jones & Doren (1997) reported for the exotic tree
to be relatively non-specific with the exception of Schinus terebinthifolius in the Everglades.
several characteristic aliens of dune communities Invasibility of reedbeds and fixed dune com-
(C. edulis, O. erythrosepala, and to a minor extent munities is related to historical stability, since
C. campestris). In addition, habitat invasibility was alien species number and proportion and relative
related negatively to stability. Communities en- alien cover increased significantly from semi-
closed in areas that have persisted unaltered since stable to non-stable regimes. Assuming that
1956 exhibited a very low invasibility, with only a historical change indicates how long ago main
single species, A. squamatus, growing at extremely disturbances occurred, the decrease in invasibility
low densities. It is well established that disturbed from non-stable to semi-stable regimes would
habitats are regarded to be more vulnerable to indicate a reversion towards the climax. During
invasion than unaltered ones (di Castri, 1990; this process, the establishment of non-natives
Vitousek et al., 1997; Hobbs, 2000). This is par- may be less opportune, basically as a result of
ticularly true in perennial communities dominated decreasing resources or increasing colonization
by one or a few species, especially in reedbeds, but of clonal, dominant species that would reduce
also in several rushbeds and even dune communi- safe sites for alien germination and establishment
ties, whose vegetative regeneration mechanisms (Hobbs & Huenneke, 1992). However, this
are responsible for a dense canopy that prevents pattern was not valid for rushbeds nor for
262
Figure 2. Number and percentage of alien species, and relative alien cover in four coastal communities in the Llobregat delta, in
relation to the stability regimes selected for the study (see text for more details). Means and standard errors are shown. Significance of
pairwise comparisons between semi-stable and non-stable regimes is shown (*: p < 0.05; **: p < 0.01; n: no significant).
Arthrocnemum scrubs in the Llobregat delta. even native plants is extremely difficult, being
Invasibility of a given habitat also depends on restricted to vegetation patches and nearby areas
traits related to their adequacy for plant estab- with lower conductivity and higher water and
lishment, such as resource supply, non-biotic nutrient contents (Rubio-Casal et al., 2001).
and biotic conditions (Alpert et al., 2000; Heger,
2001). Saline stress probably prevents these hal-
ophilous communities from invasion, as Alpert Acknowledgements
et al. (2000) reported for other harsh habitats
such as xeric grasslands and desert vegetation in This work has been funded by the CICYT pro-
relation to drought and nutrient stress. Although jects REN2000-0361 GLO and SEC2000-0836-
bare soil might be abundant in halophil- C04-04 of the Spanish Ministry of Science and
ous communities, its colonisation by alien or Technology.
263
References Hobbs, R. J. & L. F Huenneke, 1992. Disturbance, diversity
and invasion: implications for conservation. Conservation
Alpert, P., E. Bone & C. Holzapfel, 2000. Invasiveness, inva- Biology 6: 324–337.
sibility and the role of environmental stress in the spread of Jones, D. T. & R. F. Doren, 1997. The distribution, biology and
non-native plants. Perspectives in Plant Ecology, Evolution control of Schinus terebinthifolius in southern Florida. With
and Systematics 3: 52–66. special reference to Everglades National Park. In Brock, J.
` ` ´
Bolos, A. & O. Bolos, 1950. La vegetacion de las comarcas H., M. Wade, P. Pysˇ ek & D. Green (eds), Plant Invasions:
´
barcelonesas. Instituto Espanol de Estudios Mediterraneos,
˜ Studies from North America and Europe. Blackhuys pub-
Barcelona, 350 pp. lishers, Leiden, 81–93.
Casasayas, T., 1990. Widespread adventive plants in Catalonia. Lodge, D. M., 1993. Species invasions and deletions. In Kare-
In di Castri, F., A. J. Hansen & M. Debussche (eds), Bio- iva, P. M., J. G. Kingsolver & R. B. Honey (eds), Biotic
logical Invasions in Europe and the Mediterranean Basin. Interactions and Global Change. Sunderland, Massachus-
Kluwer academic Publishers, Dodretch, 85–104. sets, 367–387.
di Castri, F., 1990. On invading species and invaded ecosys- Olenin, S. & E. Leppanoski, 1999. Non-native animals in the
¨
tems: the interplay of historical chance and biological Baltic Sea: alteration of benthic habitats in coastal inlets and
necessity. In di Castri, F., A. J. Hansen & M. Debussche lagoons. Hydrobiologia 393: 233–243.
(eds), Biological Invasions in Europe and the Mediterranean Pysˇ ek, P. & K. Prach, 1993. Plant invasion and the role of
Basin. Kluwer academic Publishers, Dodretch, 3–16. riparian habitats: a comparison of four species alien to
Ewel, J. J., 1986. Invasibility: lessons from South Florida. In central Europe. Journal of Biogeography 20: 413–420.
Mooney, H. A. & J. A. Drake (eds), Ecology of biological Rauchich, J. & R. J. Reader, 1999. An experimental study of
invasions of North America and Hawaii. Springer-Verlag, wetland invisibility by purple loosestrife (Lythrum salicaria).
New York, 214–230. Canadian Journal of Botany 77: 1499–1503.
Heger, T., 2001. A model for interpreting the process of inva- Rubio-Casal, A. E., J. M. Castillo, C. J. Luque & M. E. Fig-
sion: crucial situations favouring special characteristics of uero, 2001. Nucleation and facilitation in salt pans in
invasive spacies. In Brundu, G., J. Brock, I. Camarda, L. Mediterranean salt marshes. Journal of Vegetation Science
Child & M. Wade (eds), Plant Invasions: Species Ecology and 12: 761–770.
Ecosystem Management. Backhuys Publishers, Leiden, 3–9. Vitousek, P. M., H. A. Mooney, J. Lubchenco & J. M. Melillo,
Hobbs, R. J., 2000. Land-use changes and invasions. In Moo- 1997. Human domination of Earth’s ecosystems. Science
ney H. A. & R. J. Hobbs (eds), Invasive Species in a 277: 494–499.
Changing World. Island Press, Washington, DC, 55–64.