Estuaries and Coasts

, Volume 40, Issue 1, pp 187–199 | Cite as

Nonnative Species in British Columbia Eelgrass Beds Spread via Shellfish Aquaculture and Stay for the Mild Climate

  • Megan E. Mach
  • Colin D. Levings
  • Kai M. A. Chan
Article

Abstract

Nonnative species cause economic and ecological impacts in habitats they invade, but there is little information on how they spread and become abundant. This is especially true for nonnative species in native Zostera marina eelgrass beds in coastal British Columbia, Canada, which play a vital role in estuarine ecosystems. We tested how nonnative species richness and abundance were related to both arrival vectors and environmental factors in northeast Pacific eelgrass. Using correlation tests and generalized linear models, we examined how nonnative macroinvertebrates (benthic, epifaunal, and large mobile) and some algae species were related to arrival vectors (shipping and aquaculture) and environmental factors (climate variables, human population density, and native richness and abundance). We found 12 nonnative species, 50 % with known negative impacts within eelgrass habitats. For benthic organisms, both nonnative richness and abundance were strongly correlated with shellfish aquaculture activities, and not with shipping activity. For epifaunal nonnative richness and abundance, neither vector was significantly correlated. Climate (temperature and salinity) helped explain nonnative richness but not abundance; there was no relationship of nonnative richness or abundance to native species richness and abundance or population density. Results suggest that aquaculture activities are responsible for many primary introductions of benthic nonnative species, and that temperature and salinity tolerances are responsible for post-introduction invasion success. While aquaculture and shipping vectors are becoming increasingly regulated to prevent further international spread of nonnative species, it will be important when managing nonnatives to consider secondary spread from intraregional transport through local shellfish aquaculture and shipping.

Keywords

British Columbia Seagrass Nonnative Invasive Exotic Aquaculture Zostera marina 

Introduction

Increased global connectivity has led to the introduction of nonnative marine species (hereafter “nonnatives”) around the world, with harmful economic and ecological effects (Colautti et al. 2006; Lockwood et al. 2005). As nonnative introductions continue to increase (Wonham and Pachepsky 2006), it is important to understand which species have invaded and what factors might limit their success. Despite the role of nonnatives as important drivers of change globally (Didham et al. 2005; Wilcove et al. 1998), there remains a lack of understanding of nonnative marine species and their impacts in coastal marine habitats, such as seagrass beds (Carlton 2009; Williams 2007). High-impact nonnatives have been shown to directly alter seagrass through habitat modification, for example the European green crab, Carcinus maenas, tears up Zostera marina eelgrass when preying on infaunal bivalves (Davis et al. 1998), or by competing with seagrass for resources and preventing reestablishment, as seen in seagrass beds invaded by the Japanese wireweed, Sargassum muticum (den Hartog 1997). In seagrass habitats, which play a vital role in estuarine ecosystems worldwide (Hemminga and Duarte 2000), nonnatives affect the health of seagrass plants and the communities that live within these habitats (Williams 2007).

Z. marina (hereafter eelgrass) is the dominant soft-sediment seagrass in the northeast Pacific, including tidal flats in coastal British Columbia (hereafter, B.C.), Canada (Berry et al. 2003; Levings et al. 1983). Coastal development, habitat modification, and other anthropogenic impacts have resulted in a net loss of eelgrass habitats in B.C. and many other coastal regions (Orth et al. 2006; Short and Wyllie-Echeverria 1996). The additional threat of nonnative introductions to these sensitive and productive systems is of great environmental concern (Waycott et al. 2009). Specific information is lacking on nonnative communities in the eelgrass beds found along the 25,300 km of B.C.’s coastline.

Many factors play a role in the increase of nonnatives in coastal marine systems, but the transport and introduction of organisms via shipping and shellfish aquaculture vectors are widely recognized as important (Carlton 1987; Cohen and Zabin 2009; Haupt et al. 2010; Hewitt and Campbell 2010; Mineur et al. 2007; Ruiz et al. 2011; Sylvester et al. 2011). Transport via international shipping to ports is thought to be one of the greatest influences on the global dispersal of nonnatives (Ruiz et al. 2000). For example, the majority of nonnative species described in San Francisco Bay, one of the most invaded bays in the world (Cohen and Carlton 1998), have arrived with ballast water or hull fouling on ships arriving from overseas (Ruiz et al. 2011). Similarly, trading into Port Metro Vancouver (Vancouver, B.C.), the fourth largest port in North America, results in major volumes of international shipping traffic, shown to transport nonnative species to B.C. (DiBacco et al. 2011; Lo et al. 2012; Sylvester et al. 2011). Shellfish aquaculture also results in the transport of intentionally introduced and unintentional hitchhiking nonnatives (Cohen and Zabin 2009; Levings et al. 2002). In B.C., shellfish aquaculture accounts for over 3700 ha of coastline (Environment Canada 2011). Review studies of nonnatives in coastal ecosystems suggested that Pacific oyster (Crassostrea gigas) aquaculture is the dominant vector for introductions relative to shipping (44 and 16 %, respectively; Gillespie 2007; Levings et al. 2002).

After arrival in a habitat outside their native range, establishment of nonnatives is dependent on favorable environmental conditions and resource availability (Incera et al. 2009; MacLeod et al. 2009). Abiotic factors, such as salinity (Mann and Harding 2003; Miller et al. 2007; Powers et al. 2006) and temperature (Clark and Johnston 2005; Dafforn et al. 2009; Stachowicz et al. 2002), may limit the success of nonnatives in newly invaded habitats. In the northeast Pacific, species richness of nonnatives may be related to temperature, with fewer nonnatives found in colder waters at higher latitudes (deRivera et al. 2005). However, some B.C. fjords and bays are warmer and less saline than the exposed coast at similar latitudes (Thompson 1981). The warm low salinity waters may allow more nonnatives from warm and temperate regions to establish than would otherwise be expected in colder conditions (deRivera et al. 2011).

Nonnatives must also tolerate changes in the biotic and abiotic environment, requiring sufficient habitat and resources (Maron and Marler 2007; Moyle and Light 1996; Shea and Chesson 2002). Human population densities and native species diversity may act as proxies for some aspects of resource availability that affect richness and abundance of nonnative species in eelgrass. Human disturbance causes release of resources such as space and nutrients through altering of natural habitats, nutrient-rich runoff, and other disruptions (Glasby et al. 2007; Martone and Wasson 2008; Piola and Johnston 2007). In coastal B.C., the highest human population densities are found in the southern region of the Strait of Georgia (BCStats 2006). This region may support a greater number of nonnatives than less populated regions to the north (Altman and Whitlatch 2007). Alternatively, in regions of high native species diversity, biotic resistance may prevent the establishment of nonnatives by limiting resource availability (Stachowicz et al. 1999; Tilman 1997).

To study the relative importance of vectors and environment on establishment of nonnatives in eelgrass beds, we conducted biological sampling of benthic, epifaunal, and large mobile invertebrates along the B.C. coast. Due to over 150 years of shipping and aquaculture vectors in this region, we expect that species have had time to migrate making species composition relatively homogenous throughout along the coast.

We addressed the following specific questions:
  1. 1.

    What is the frequency and abundance of nonnative species in B.C. eelgrass?

     
  2. 2.

    Does proximity to international ship traffic or aquaculture facilities explain variation in nonnative richness in eelgrass?

     
  3. 3.

    What is the relationship between the richness of nonnatives and local environmental conditions—climate variables, human population density, and native species richness?

     

Methods

Field Sampling

We sampled benthic, epifaunal, and mobile species in ten eelgrass beds on the B.C. coast of and one south of B.C., in contiguous Washington State, USA, in June–August 2008 (Fig. 1). For the purposes of our analyses, we include the Washington State data in our B.C. data set. The sampling program spanned approximately 6° latitude (about 666 km).
Fig. 1

Map of British Columbia showing the 11 eelgrass beds (black circles) sampled. Insert map is a close up of sites in southern BC, Canada and Washington State, USA

We collected invertebrates, fish, and algae over 1–2 days, during the same low-tide cycle, using cores, dredge, traps, and visual surveys to sample the full biotic assemblage present in each eelgrass bed (Short and Coles 2001). Each method targeted a different species group: coring—benthic and infaunal invertebrates (e.g., polychaetes, bivalves, gastropods; hereafter, the benthic species group); dredging—epifaunal and sessile invertebrates that utilize eelgrass blades (e.g., amphipods, isopods, tunicates, limpets; hereafter, the epifaunal species group); trapping—large mobile invertebrates (e.g., crabs, sea stars; hereafter, the mobile species group); and visual surveying targeting S. muticum (Asian wireweed) and C. gigas (Pacific Oyster) presence.

Cores and Dredges

At each site, we sampled benthic invertebrates with six sediment cores (10 cm diameter × 17 cm depth; Short and Coles 2001) spaced 25 m apart along two 50-m transects. To sample epifaunal invertebrates, we used a dredge with a semi-circular mouth (30 cm wide × 22 cm high; mesh size 0.5 mm) pulled 50 m through the eelgrass against the water current, parallel to shore (flushing approximately 5.18 m3 of water). Five replicates were obtained. We sieved the samples using a 1-mm sieve and removed plant matter and organisms such as bryozoans by hand. Whole eelgrass plants, bryozoans, and algae found in the sampling were searched for fouling organisms, especially tunicates. All specimens were preserved in 95 % ethanol.

Traps and Visual Surveys

To sample large mobile invertebrates and fish, we used minnow traps with 10-cm-diameter openings. Traps were attached at 10-m intervals, in pairs, along a line and anchored near the middle of the bed, parallel to shore. Traps were left in each bed for 20 to 24 h. Organisms were measured, identified, or photographed (if the species was not identifiable in the field) and returned to the eelgrass bed. We trapped at 9 of the 11 sites (8 traps: Cowichan Bay, Campbell River, Esquimalt Lagoon, Fourth of July, Nanaimo Estuary, Port Alberni; 4 traps: Prince Rupert, East Kaien Island, Tsawwassen). We checked for the presence or absence of S. muticum and C. gigas by walking along a 100 m × 2 m transect parallel to shore. We conducted three visual transects at each site.

We returned preserved collections from core and dredge samples to the lab for identification. Specimens that could not be identified in the lab were sent to Biologica Environmental Services Ltd. (Marine and Freshwater Taxonomy, 634 Humboldt St., Victoria, B.C., V8W 1A4) for identification. More than 50 % of amphipod and polychaete identifications were then re-identified by taxonomic experts to ensure correct identification (Taxonomists: Jeffery R. Cordell, University of Washington; R. Eugene Ruff, Puyallup, WA). We classified organisms as native, cryptogenic, nonnative, or indeterminate. Cryptogenic species are those of unknown origin. “Indeterminate” refers to organisms that we did not identify to a sufficient taxonomic resolution to determine their species identity.

Vector and Environmental Data

Eight of the sampling sites were located in ports with international shipping activity, providing a range of values for the shipping vector. We obtained data on the number of ship arrivals to these ports for November 2006–October 2007 from Lo et al. (2012; their supplementary Table 1).

For the aquaculture vector, we calculated an aquaculture effect score = \( \sum_i^n4.8/{\mathrm{aqdistance}}_i \), where n is the number of aquaculture sites associated with each sample site and aqdistancei is the distance (km) from aquaculture site i to the sample site (developed from Choi 2011). The 4.8 constant is derived from the maximum larval dispersal distance of 480 km; it creates a decay function where the effect score will be ≤0.01 at distances greater than 480 km. This distance was estimated for Nuttalia obscurata, the varnish clam, as modeled from larval duration, sea surface circulation patterns, and stable environmental conditions (Dudas and Dower 2006). Shellfish aquaculture site coordinates were collected from the British Columbia Ministry of Agriculture and Land, and Integrated Land Management Bureau (2005). Larger effect scores represent a greater density of aquaculture located close to the sample site. Distances were measured via waterways, not in overland distances.

Temperature and salinity data were not available for each site, so we used an oceanographic climate model (personal communication with Mike Foreman; Foreman et al. 2008) to generate data on sea surface temperature (SST) and salinity, averaged for the summer and winter. Summer SST was negatively correlated with summer salinity and winter salinity (for all correlations: r > −0.67, p < 0.02), likely a result of river runoff decreasing salinity levels in the warmer Strait of Georgia (Thompson 1981). We therefore chose summer SST as a proxy for summer and winter salinity. Human population density (number of people per km2) data were collected from the 2006 Canadian Census (BCStats 2006) for the census region in closest proximity to a sample site. Aquaculture effect scores, number of ship arrivals per year, and human population density were log-transformed.

Nonnative Abundance and Distribution

Using species accumulation curves, we estimated species richness (including native, cryptogenic, nonnative, and indeterminate) for core, dredge, and trap samples (see Online Resource 1). The rate of species addition reached an asymptote after only a few samples for most sites, suggesting sufficient sampling replication within each site (Gotelli and Colwell 2001). We therefore used pooled samples from each site for benthos (n = 6 cores) and epifauna (n = 5 dredge pulls) to measure total species richness and mean epifaunal abundance. To ensure the independence between benthic and epifaunal analyses, organisms collected by multiple methods were only analyzed in the faunal group in which they were most commonly sampled (as in Table 1).
Table 1

Nonnative (panel A) and cryptogenic (panel B; those of unknown origin) species sampled in British Columbia Z. marina beds, habitat in which they were sampled (Hab), percent of sites occupied (% sites), sites the species was found (listed from south to north), and likely vector of introduction. Quotations around species are unresolved taxa and the value for each species at each site is mean abundance across samples collected at each site. A “+” represents species that were present but abundance data not collected. Native richness of species (includes indeterminate species) in Annelid, Arthropod, and Mollusca phyla and ship arrivals are the number of ships arriving at each port between Nov 2006 and Oct 2007 (Lo et al. 2012). Site abbreviations: Esquimalt Lagoon (EL), Fourth of July (FJ), Cowichan Bay (CB), Tsawwassen (Ts), Port Alberni (PA), Nanaimo Estuary (NE), north of Stanley Park (SP), Mud Bay (MB), Campbell River (CR), Prince Rupert (PR), East Kaien Island (EKI)

Species

Taxa

Hab

% Sites

EL

FJ

CB

Ts

NE

PA

SP

MB

CR

PR

EKI

Taxonomic authority

Native range

Vector

Panel A Nonnative species

Ampithoe valida

Amphipod

E

64

1

  

0.4

15.4

0.6

4

4.6

0.7

  

Smith, 1873

NWAtl

A/S

Batillaria attramentaria

Gastropod

B

9

    

1.3

      

Sowerby II, 1855

NWPac

A

Clymenella torquata

Polychaete

B

18

   

2.8

    

0.5

  

Leidy, 1855

NWAtl

A

Crassostrea gigas

Bivalve

V

18

    

+

  

+

   

Thunberg, 1793

NWPac

N

Grandidierella japonica

Amphipod

E

9

     

0.2

     

Stephensen, 1938

NWPac

A/S

Melita nitida

Amphipod

E

9

     

1.6

     

Smith, 1873

NWAtl

A/S

Monocorophium acherusicum

Amphipod

E

36

  

0.1

0.2

1

  

0.2

   

Costa, 1857

NEAtl

A/S

Monocorophium insidiosum

Amphipod

E

9

    

0.8

      

Crawford, 1937

NAtl

A/S

Mya arenaria

Bivalve

B

27

  

0.2

  

0.4

 

2.7

   

Linnaeus, 1758

NWAtl

N

Sargassum muticum

Algae

V

27

   

+

  

+

+

   

Fensholt, 1955

NWPac

A

Sinelobus sp.

Tanaid

E

9

     

0.3

     

Sieg, 1980

Unk

P

Venerupis philippinarum

Bivalve

B

18

    

0.6

  

0.6

   

Adams and Reeve, 1850

NWPac

A

Ship arrivals

   

3

0

14

177

14

22

1328

0

2

159

0

   

Native richness

   

33

35

33

36

24

9

35

45

42

25

32

   

Panel B Cryptogenic species

Ampithoe lacertosa

Amphipod

E

18

 

0.6

 

1

       

Bate, 1858

  

Capitella telata (?)

Polychaete

B

9

1

          

Fabricius, 1780

  

Eumidasanguinea

Polychaete

B

18

        

0.17

  

Oersted, 1843

  

Harmothoeimbricata

Polychaete

B

27

       

9.8

 

0.3

0.5

Linnaeus, 1767

  

Harpacticus uniremis group

Copepod

E

18

  

0.3

     

38.5

  

Holmes, 1900

  

Leptocheliadubia

Tanaid

E

64

0.2

4.5

 

1.6

   

0.2

2.8

2.2

8.3

Krøyer, 1842

  

Prionospiosteenstrupi

Polychaete

B

9

        

0.3

  

Malmgren, 1860

  

Spiophanesbombyx

Polychaete

B

9

 

0.3

         

Claparède, 1970

  

Habitat: B benthic, E epifauna, V visual survey of S. muticum and C. gigas presence, no nonnative mobile macroinvertebrates; Native range: Atl Atlantic, Pac Pacific, Unk unknown; Vector: A aquaculture, S ship hull or ballast water, N natural dispersal from aquaculture introduction, P polyvectic

To test the between-site difference in nonnative richness and mean abundance of benthic and epifaunal species between sites, we used a Kruskal-Wallis nonparametric test (Sokal and Rohlf 1995). Data were plotted with the function sciplot() using R software (Morales 2011; R Development Core Team 2012).

Relating Vectors to Nonnative Richness and Abundance

To test the influence of shipping and aquaculture vectors on nonnative richness and mean abundance, we compared these data to the number of ship arrivals per year and aquaculture effect scores using a nonparametric Spearman rank correlation (rho; Sokal and Rohlf 1995). Epifaunal species sampled in the dredge were attributed to both shellfish aquaculture and shipping vectors (hull fouling or ballast water) while benthic species sampled with the core were attributed only to aquaculture (as reviewed in Gillespie 2007; Levings et al. 2002). We described species attributed to multiple vectors as polyvectic (Carlton and Ruiz 2005).

Nonnative shellfish species introduced intentionally via aquaculture, and sampled for richness and abundance in this study, were tested against the shellfish aquaculture effect score. This included Mya arenaria and Venerupis philippinarum, but not C. gigas as these were only sampled for presence or absence.

Relating Environmental Variables and Nonnative Richness and Abundance

We tested the influence of environmental factors on total richness and mean abundance of nonnatives using generalized linear models (GLM) that assume a normal probability distribution. We used an information-theoretical approach to assess performance from the set of candidate models (Burnham and Anderson 2002). As the number of models should not exceed the number of sites sampled (Anderson 2008), we selected a subset of three variables a priori to compare to benthic and epifaunal nonnative richness (summer temperature, human population density, native richness) and abundance (summer temperature, human population density, native abundance; Table 2). A brief rationale for each of the three environmental explanatory variables is as follows:
  • Summer temperature: Nonnative richness and abundance have been positively correlated with summer temperature and negatively correlated with salinity in the coastal northeast Pacific (deRivera et al. 2011; Dethier and Hacker 2005). Quayle (1964) suggested that temperature was one of the primary selection factors favoring establishment of nonnatives in coastal B.C.

  • Human population density: We assumed that human population density correlates with anthropogenic disturbance, a factor that has been positively correlated to nonnative establishment (Altman and Whitlatch 2007). We predicted that richness and abundance of nonnatives would be highest in the densely populated southeastern Strait of Georgia.

  • Native richness and abundance: Regions with high native species richness (includes native, cryptogenic, and indeterminate species for our model analysis) may have “biotic resistance” to establishment of nonnative species, and native richness has been found to be negatively correlated to nonnative richness (Cohen and Carlton 1998; Tilman 1997). On the other hand, in cases where variation in native species richness is explained by productivity and resource availability, native richness has been found to correlate positively with nonnative richness (Shea and Chesson 2002).

Table 2

Variables used in the analysis of relationships between nonnative total richness and mean abundance and the environmental conditions at each site. Mean, standard deviation (S.D.), and minimum and maximum values of each response and explanatory variable are given

Variable

Mean

S.D.

Min.

Max.

Response variables

 Benthic nonnative richness

0.73

0.79

0

2

 Epifaunal nonnative richness

1.36

1.29

0

4

 Benthic nonnative abundance

0.80

1.20

0

3.2

 Epifaunal nonnative abundance

2.82

5.06

0

17.2

Explanatory variables

Arrival vector

    

 Ln shellfish Aquaculture score

3.59

1.39

1.16

6.44

 Ln ship arrivals

2.59

2.41

0

7.19

Environment

 Summer temperature

13.92

2.29

10.78

17.25

 Ln population density per km

3.21

2.29

0

6.54

 Benthic native richness

17.73

7.66

8

34

 Epifaunal native richness

24.45

7.48

11

39

 Benthic native abundance

4.82

7.17

0

19

 Epifaunal native abundance

14.18

25.26

0

86

Ln log normal transformation to the data before analysis

In the absence of any compelling theory or method to group variables into particular candidate models, we used all possible combinations of the three explanatory variables (seven models). We ranked models using the small-sample-size version of Akaike’s Information Criterion (AICc) by the likelihood of being the best model in the group using two ranking measures, as calculated by the R package “AICcmodavg” (Mazerolle 2012): (1) models with a difference of less than 3 units from the first-ranked model were considered to have substantial support and (2) models with high Akaike weights (i.e., the probability of a model being the best explanation of the given data) (Burnham and Anderson 2002). For each of the model tests with benthic and epifaunal data, we inspected the residuals from the full model (all three explanatory variables) to ensure that the statistical assumptions of homoscedasticity and normality were satisfied (Sokal and Rohlf 1995).

To explore additional relationships, we used post hoc tests to assess the potential of the additional environmental variables (those sampled but not selected for model testing) for explaining variation in nonnative richness and mean abundance (see Online Resource 2). Finally, though the assumptions of normality were not met, we conducted a linear regression analyses of benthic and epifaunal richness on significant environmental and vector variables to test whether these variables together better explain nonnative distributions.

Influence of Nonnative Species Composition and Environment on Site Similarity

To interpret differences in nonnative species composition between sites, we created a single matrix of site-by-species data of nonnative richness estimates at 8 sites for 12 nonnatives (4 benthic, 6 epifaunal, S. muticum, and C. gigas). We excluded the three sites where we did not find nonnative species. Using the Kulczynski coefficient, which is appropriate for presence/absence species estimates, we then transformed the site-by-species data matrix into a distance matrix of the dissimilarities in nonnative composition between each pair of sites using the envfit() function in the R package vegan (Legendre and Legendre 1998; Oksanen et al. 2012). Based on the site dissimilarity matrix, we then visualized nonnative composition using a cluster analysis and nonmetric multidimensional scaling (nMDS), an ordination technique that simplifies multidimensional relationships among sites to create a smaller number of dimensions that are easier to visualize and interpret (Kruskal and Wish 1978). We plotted nMDS scores for nonnative composition against site. Each point represents the nonnative community for a given site, and the distance between points represents the similarity of sites based on the composition of nonnative species.

To cross-check the nMDS site groupings, we used a cluster analysis to perform hierarchic clustering of weighted averages calculated from average neighbor resemblances of site similarity. These groupings were then overlaid on the nMDS plot for sites with 50 and 70 % similarity. To test if site differences in species composition can be explained by environmental variables, we compared the site dissimilarity matrix to average summer SST (Summer SST), human population density (Ln Population), and native species richness (Native Richness) using the envfit() function. Correlation significance was calculated using a permutation test (n = 11). The number of sites sampled limited the number of permutations possible. All statistical analyses were performed using R software (R Development Core Team 2012).

Results

Nonnative Frequency. Abundance, and Distribution

We sampled a total of 32,401 individuals from 181 different species; 12 were classified as nonnative (6.6 % of total), 8 as cryptogenic (4.4 % of total), 29 as indeterminate (16 % of total), (Table 1), and the remaining 132 were classified as native (Table 1; species recorded in Mach 2012). We found nonnative species in 8 of 11 eelgrass beds. The maximum number of nonnatives (n = 6) was found at Mud Bay and Nanaimo estuary. We did not find nonnatives at East Kaien Island, Prince Rupert, or Fourth of July Beach. Maximum abundance of benthic nonnatives was found at Mud Bay, and the greatest abundance of epifaunal nonnatives at Nanaimo estuary (Table 1).

Richness of benthic and epifaunal nonnatives was significantly different across sites (Fig. 2; benthic, chi210 = 29.28, P = 0.001; epifaunal: chi210 = 43.96, P < 0 .0001) as was mean abundance (benthic: chi210 = 29.77, P = 0.0009; epifaunal: chi210 = 44.686, P < 0.0001). For benthic nonnatives, M. arenaria had the greatest abundance at Mud Bay while at Tsawwassen, Clymenella torquata was the most abundant species. Epifaunal nonnatives were dominant at Nanaimo estuary, driven by the abundance of the most common nonnative, the amphipod Ampithoe valida (15.4 per 5.18 m3 dredge). A. valida was found at 7 of the 11 sites. All other nonnatives were found at one to four sites and at low abundances relative to A. valida. S. muticum and C. gigas were present at less than half of the sites (Table 1).
Fig. 2

Bar plots (with standard error) of total nonnative benthic richness (a total species per 6 × 79 m3) and epifaunal richness (b total species per 5 × 5.18 m3), mean nonnative benthic abundance (c number of individuals per 79 m3, over six cores), and epifaunal abundance (d number of individuals per 5.18 m3, over five dredge pulls) in each eelgrass bed. Black circles represent total richness. Sites are in order of latitude from south to north. Kruskal-Wallis nonparametric tests of site differences are significant for nonnative richness and abundance: benthic richness (chi2 = 29.28, P = 0.001) and abundance (chi2 = 29.77, P = 0.0009), epifaunal richness (chi2 = 43.96, P < 0.0001), and abundance (chi2 = 44.686, P < 0.0001)

Aquaculture Versus Shipping Vectors

Sites with high aquaculture effect scores correlated closely with benthic nonnative richness and abundance in Spearman’s rank test (Table 3; Fig. 3a, b). However, sites with high aquaculture effect scores did not correlate with epifaunal nonnative richness or abundance. Abundance of nonnative aquaculture species, V. philippinarum, was correlated to a high aquaculture effect score (S = 86.51, rho = 0.61, P = 0.048), while M. arenaria was not (S = 184.39, rho = 0.162, P = 0.634). Neither benthic nor epifaunal nonnative richness or abundance was significantly related to ship arrivals.
Table 3

Spearman rank correlation (rho) and probability-value (P) for benthic and epifaunal nonnative richness and abundance with distance to and frequency of shellfish aquaculture sites (aquaculture) and the number of ship arrivals at each port (ship arrivals)

Vector

Richness

Abundance

 

rho

P

rho

P

Benthic nonnatives

 Aquaculture

0.63

0.037

0.61

0.046

 Ship arrivals

−0.06

0.86

−0.01

0.97

Epifaunal nonnatives

 Aquaculture

0.31

0.36

0.54

0.085

 Ship arrivals

0.29

0.36

0.18

0.59

Significant P values in italic, N = 11

Fig. 3

Correlation of nonnative richness (total species per 6 × 79 m3) and abundance and transformed aquaculture effect score. Relationships significant in the Spearman rank test plotted as dashed line

Model Results for Environment Effect on Nonnative Richness and Abundance

Nonnative richness of both benthic and epifaunal communities was best explained by summer SST alone (Table 4). Models that included the other two variables (human population density and native richness or abundance) were non-significant. When summer SST was compared to benthic and epifaunal nonnative richness in a Pearson correlation test, both were significant (P < 0.05) with r2 values of 0.73 and 0.84, respectively, suggesting summer SST, or variables correlated with summer SST such as salinity, predicted 73 % of the variance in benthic nonnative richness and 84 % of epifaunal nonnative richness (Fig. 4). While these data were non-normal, the high statistical significance of the relationship between nonnative richness and SST alone supports the accuracy of this model-test. The best model of mean nonnative abundance contained only the y-intercept and no explanatory variables, suggesting that the variables selected for this analysis do not significantly explain trends (Burnham and Anderson 2002).
Table 4

Linear models of nonnative benthic and epifaunal total richness and mean abundance in the 11 eelgrass beds sampled. The three best-ranked candidate models of the full factorial of possible models (full model: y ~ SummerSST + LnPopulation + Native; 7 models tested) are listed with the number of parameters (K), corrected AIC (AICc), the difference in AIC between the candidate model and the best model (deltai), Akaike weights (wi), and log-likelihood (LL)

Model

K

AICc

Deltai

wi

LL

Benthic ANS richness

SummerSST

3

26.03

0

0.68

−8.30

 SummerSST + Benthic.Native

4

29.23

3.20

0.14

−7.28

 Intercept

2

30.38

4.34

0.08

−12.44

Epifaunal ANS richness

SummerSST

3

31.49

0

0.79

−11.03

 SummerSST + Epifauna.Native

4

34.97

3.48

0.14

−10.15

 SummerSST + LnPopulation

4

36.58

5.07

0.06

−10.96

Benthic ANS abundance

Intercept

2

39.61

0

0.41

−17.06

 SummerSST

3

39.72

0.10

0.39

−15.14

 LnPopulation

3

43.31

3.70

0.06

−16.94

Epifaunal ANS abundance

Intercept

2

71.34

0

0.43

−32.92

 SummerSST

3

72.11

0.77

0.30

−31.34

 Epifauna.Native

3

73.57

2.23

0.14

−32.07

The “best” models are in italic. “Intercept” are those models without any of the three variables

Fig. 4

Nonnative benthic (a) and epifaunal (b) richness as predicted by SummerSST (y ~ SumTemp, the “best” AICc model). Relationships significant in the Pearson correlation test plotted as dashed line

The linear regression analysis of summer SST and aquaculture effects together revealed that both contribute to explaining the variation of benthic and epifaunal nonnative richness (benthic: r2 = 0.62, F8 = 9.13, P = 0.009; epifauna: r2 = 0.63, F8 = 9.86, P = 0.007). However, the relationship of these two variables together to nonnative richness was less than summer SST and aquaculture effects alone.

Differences in Nonnative Community Composition

The differences in nonnative assemblage at each site were interpreted using nMDS ordination space as shown in Fig. 5. Port Alberni was separated from other sites (negative nMDS values along the y-axis) as a result of three arthropod species not found at other sites: Sinelobus sp. (Tanaidacea), Melita nitidia, and Grandidierella japonica (Amphipoda). Nanaimo, Mud Bay, and Cowichan Bay are located on mid-Vancouver Island on the Strait of Georgia and were characterized by three nonnative molluscs: Ruditapes philippinarum, C. gigas, Batillaria attramentaria, and Monocorophium spp. (Amphipoda) separating them to the right of the nMDS (positive nMDS values along x-axis). These sites had species compositions that were more than 50 % similar to one another in cluster analyses. Other sites with similar species composition (Tsawwassen, Stanley Park, Esquimalt, and Campbell River) were separated by the Strait of Georgia from east to west and spread between the northern and southern ends of the Strait. These four sites also have species compositions that were 50 % similar and group together in the negative nMDS values along the x-axis.
Fig. 5

Nonmetric multidimensional scaling (nMDS) plot visualizing site dissimilarity of nonnative community composition with polygons from cluster analysis, (50 % similarity = gray, 70 % similarity = white). Location of species names (in italics) in relation to site names on the nMDS figure demonstrates site similarities in species composition. Includes only those sites with nonnatives. Relationships of environmental variables to the site dissimilarities are depicted as vectors in the bottom right of the plot. The length of vectors represents the difference in r2 values between each variable: SummerSST (r2 = 0.25), Ln Population (r2 = 0.20), and Native Richness (r2 = 0.28)

Native richness and human population density were positively correlated with sites positioned with negative values on the x-axis and positive values on the y-axis (Campbell River, Tsawwassen, Stanley Park, and Esquimalt Lagoon) (Fig. 5). Native richness and human population density explained 28 and 20 % of site differences, respectively, while the summer SST vector explained 25 % of variance between sites and was negatively related to these same sites, pointing to the opposite corner of the nMDS plot.

Discussion

There are few nonnative species in B.C. eelgrass beds when compared to the more than 62 nonnative species recorded by past reviews that summarized all coastal B.C. habitats, at least up to 2007 (Gillespie 2007; Levings et al. 2002). We found only 12 species, 3 of which were introduced intentionally for aquaculture. Nonnative species composition between sites differed significantly and no two sites had the same set of nonnatives. The rarity of nonnatives suggests that secondary spread is a major concern for the future, as regional dispersal will increase introductions to uninvaded sites via intraregional transport of goods, recreational boating, and local aquaculture trade (Clarke Murray et al. 2011; Cohen and Carlton 1998; Davidson et al. 2008).

Relationship to Shipping and Aquaculture Vectors

In B.C., the majority of nonnatives have been hypothesized to arrive via ballast and hull fouling on shipping and through import of aquaculture species (Gillespie 2007; Levings et al. 2002; Quayle 1964). As such, we expected a strong positive relationship between these vectors and nonnative richness in eelgrass beds. However, if secondary spread and natural dispersal of nonnatives had continued since initial establishment, there would be no significant relationship with either vector (Mineur et al. 2010). Our data showed a significant positive relationship between richness and abundance of benthic nonnatives and the distance and quantity of shellfish aquaculture, but no relationship with nonnative epifauna. Richness and abundance of nonnatives of both faunal groups was not significantly related to the number of ships arriving at the nearest port.

The relationship of benthic nonnatives to aquaculture but not shipping suggests introductions from aquaculture may be the most important vector explaining the current distribution of benthic nonnative species in B.C. eelgrass beds. However, epifaunal species did not relate to either vector, and this suggests they have likely dispersed widely since their original introduction. At a local scale, aquaculture is likely to remain an important vector through secondary transport of nonnatives via the movement of shellfish and equipment, until preventative management action reduces further spread (Clarke Murray et al. 2011; Cohen and Zabin 2009). Reduced dispersion rates are expected as aquaculture arrivals are already slowing from international sources as a result of decreased shellfish imports and improved regulations (ICES, Canadian Food Inspection Agency 2011, 2005). Our results do not support findings other analyses of invasions that have shown shipping to be an important vector in the introduction of molluscs, arthropods, and annelids to seagrass beds elsewhere in the northeast Pacific (Williams 2007). The reasons for this difference are not clear.

While shipping remains a common source of introduced species globally, this vector may be reduced somewhat in B.C. now that recent ballast flushing regulations require mid-ocean exchange of ballast water before ships arrive at ports (Transport Canada 2006), though introductions still occur (DiBacco et al. 2011). It is also possible that ballast-waterborne species (e.g., copepods) have colonized eelgrass beds but were not sampled in our study because of methodological differences. A common limitation in studies of marine nonnatives, including this study, is the sampling scale (Carlton 2009). Sampling usually captures macrofauna (Dafforn et al. 2009; Vermonden et al. 2010; Wyatt et al. 2005), but less commonly meio- or microfauna that can only be caught on smaller sieve sizes than we used (but see Cordell et al. 2007; Kask et al. 1982).

Climate Influences Nonnative Abundance and Distribution

Our results suggest that temperature and/or salinity are more important selection factors than the availability of habitat and resources or biotic resistance by native species for nonnative establishment in B.C.’s eelgrass habitats. Species richness of benthic and epifaunal nonnatives was significantly correlated with summer SST variation across sites, but abundance was not. Perhaps this was because species establishment was constrained by the local marine climate but invertebrate population growth may be limited by other factors such as available resources (Elahi and Sebens 2012). For example, in a study of subtidal fouling nonnative species on hard substrates in B.C., the richness of nonnatives was related to temperature; however, human population density was also important in explaining distribution patterns of these species (Clarke Murray 2012). Environmental variables also did not significantly relate to species assemblage, though this may result from the limited number of sites and species richness across those sites, and does not necessarily indicate whether environmental variables tested are important for species at these sites.

The Strait of Georgia receives fresh water input from the Fraser River leading to lower surface salinities while increased water mass residence time and lack of coastal upwelling results in higher temperatures than found elsewhere on the B.C. coast (Thompson 1981) creating a biogeographic break between habitats inside and outside the Straight. Increased chance of nonnative arrivals owing to high propagule pressure from the many aquaculture sites in the Strait and the above-mentioned unique environmental conditions may have influenced the greater richness of nonnatives for seagrass beds in this region.

However, Port Alberni, on the west coast of Vancouver Island, had the lowest summer salinity (less than six parts per thousand) and warmest summer temperatures. This site had extremely low native species richness (nine natives) and was one of the most invaded sites (five nonnatives) with three nonnatives found only at this site. Nonnatives have been demonstrated to fare better under more stressful conditions than their native counterparts (Sortie et al. 2013).

The lack of nonnatives in the two northernmost sites may well be the result of cold temperatures limiting their establishment (however, see Sloan and Bartier 2004). Additionally, because of geographic distance and because trading routes are different for ships arriving to northern B.C., it is possible the nonnatives found further south have yet to arrive in Prince Rupert. However, transport of ballast waterborne species between the Strait of Georgia and the north has been demonstrated (DiBacco et al. 2011; Piercey et al. 2000). While some nonnatives may be limited from northward spread, others such as M. arenaria are clearly not. This species was found on the north coast of B.C. by Quayle (1964), though we did not record it there.

Missing Nonnatives?

Most nonnatives sampled in this study were well-established species found in other B.C. coastal ecosystems with known impacts to seagrass habitats (Online Resource 3; Quayle 1964; Levings et al. 2002; Gillespie 2007). However, there were important nonnatives we anticipated finding, but did not. In particular, C. maenas (European green crab), a potential ballast water invader, is currently found on the outer coast of Vancouver Island from Barkley Sound to Kyuquot Sound (Yamada and Gillespie 2008). It has not been found to date in the Strait of Georgia (personal observation by Thomas Therriault; Gillespie et al. 2007). This species is of concern for eelgrass in B.C. because C. maenas is a destructive consumer that rips up eelgrass when it feeds at high densities (Davis et al. 1998).

Interestingly, we did not find any nonnative tunicates in our study. Nonnative tunicates are more frequently found near shipping and marina docks (Carman et al. 2009) and several species are already present on the B.C. coast (Clarke Murray et al. 2011). We did find native fouling organisms (i.e., Membranipora membranacea, Smithora naiadum). We are fairly confident our sampling techniques would have detected solitary or colonial tunicates had they been present. Limiting factors specific to tunicates may be restricting their use of eelgrass beds.

Conclusions

There is growing evidence that seagrass meadows are experiencing worldwide decline primarily as a result of human disturbances, such as direct physical damage and deterioration of water quality (Hemminga and Duarte 2000; Short and Wyllie-Echeverria 1996). With more than half of nonnatives found in B.C. eelgrass having known negative impacts on eelgrass habitats, the threats of these species in conjunction with anthropogenic pressures has the potential to impact the health of B.C. eelgrass beds (Ban and Alder 2008). As we found temperature to be a key variable explaining nonnative distribution, global warming may lead to further dispersal of nonnatives on the B.C. coast. Further studies and monitoring of nonnatives in eelgrass should aim to investigate species characteristics, dispersal factors, and environmental limitations to clarify why, currently, nonnatives are relatively rare in these habitats.

Notes

Acknowledgments

The authors would like to thank T. Goodman for help with collection and identification of species, taxonomists R.E. Ruff and J.R. Cordell for identification support, and C.C. Murray and R. Naidoo for constructive comments on writing and analysis. Funding for this research came from the Canadian Aquatic Invasive Species Network. Sampling equipment was provided by the Department of Fisheries and Oceans, Canada.

Supplementary material

12237_2016_124_MOESM1_ESM.docx (401 kb)
ESM 1(DOCX 400 kb)
12237_2016_124_MOESM2_ESM.docx (158 kb)
ESM 2(DOCX 158 kb)
12237_2016_124_MOESM3_ESM.docx (172 kb)
ESM 3(DOCX 171 kb)

References

  1. Altman, S., and R.B. Whitlatch. 2007. Effects of small-scale disturbance on invasion success in marine communities. Journal of Experimental Marine Biology and Ecology 342: 15–29.CrossRefGoogle Scholar
  2. Anderson, D.R. 2008. Model based inference in the life sciences: primer on evidence. New York: Springer.CrossRefGoogle Scholar
  3. Ban, N., and J. Alder. 2008. How wild is the ocean? Assessing the intensity of anthropogenic marine activities in British Columbia, Canada. Aquatic Conservation: Marine and Freshwater Ecosystems 18: 55–85.CrossRefGoogle Scholar
  4. BCStats. 2006. 2006 Census, Statistics Canada. In. http://www12.statcan.gc.ca/census-recensement/2006/index-eng.cfm/. Accessed Jan 2010.
  5. Berry, H.D., A.T. Sewell, S. Wyllie-Echeverria, et al. 2003. Puget sound submerged vegetation monitoring project: 2000–2002 monitoring report, 60. Olympia: Nearshore Habitat Program, Washington State Department of Natural Resources.Google Scholar
  6. Burnham, K.P., and D.R. Anderson. 2002. Model selection and multimodel inference: a practical information-theoretic approach, 2nd edn. New York: Springer.Google Scholar
  7. Canadian Food Inspection Agency. 2011. Aquatic animal imports. In. http://www.inspection.gc.ca/animals/aquatic-animals/imports/eng/1299156741470/1320599337624/. Accessed May 2012.
  8. Carlton, J.T. 1987. Patterns of transoceanic marine biological invasions in the Pacific Ocean. Bulletin of Marine Science 41: 452–465.Google Scholar
  9. Carlton, J.T. 2009. Chapter 2. Deep invasion ecology and the assembly of communities in historical time. In Biological invasions in marine ecosystems, eds. G. Rilov, and J.A. Crooks, 13–56. Berlin: Springer-Verlag.CrossRefGoogle Scholar
  10. Carlton, J.T., and G.M. Ruiz. 2005. Vector science and integrated vector management in bioinvasion ecology: conceptual frameworks. In Invasive alien speices: a new synthesis, eds. H.A. Mooney, R.N. Mack, J.A. Mc Neely, L.E. Neville, P.J. Schei, and J.K. Waage, 36–58. Covelo: Island Press.Google Scholar
  11. Carman, M.R., K.E. Hoagland, E. Green-Beach, et al. 2009. Tunicate faunas of two North Atlantic-New England islands: Martha’s vineyard, Massachusetts and Block Island, Rhode Island. Aquatic Invasions 4: 65–70.CrossRefGoogle Scholar
  12. Choi, F.M. 2011. Assessing intertidal marine non-indigenous species in Canadian ports. Earth and Ocean Sciences. University of British Columbia, Vancouver, Masters Thesis, pp. 118.Google Scholar
  13. Clark, G.F., and E.L. Johnston. 2005. Manipulating larval supply in the field: a controlled study of marine invasibility. Marine Ecology-Progress Series 298: 9–19.CrossRefGoogle Scholar
  14. Clarke Murray, C. 2012. The role of recreatinal boating in the introduction and spread of marine invasive species. Earth and Ocean Sciences. University of British Columbia, Vancouver, Ph.D. Thesis, pp. 191.Google Scholar
  15. Clarke Murray, C., E.A. Pakhomov, and T.W. Therriault. 2011. Recreational boating: a large unregulated vector transporting marine invasive species. Diversity and Distributions 17: 1161–1172.CrossRefGoogle Scholar
  16. Cohen, A.N., and J.T. Carlton. 1998. Accelerating invasion rate in a highly invaded estuary. Science 279: 555–558.CrossRefGoogle Scholar
  17. Cohen, A.N., and C.J. Zabin. 2009. Oyster shells as vectors for exotic organisms. Journal of Shellfish Research 28: 163–167.CrossRefGoogle Scholar
  18. Colautti, R.I., S.A. Bailey, C.D.A. van Overdijk, et al. 2006. Characterised and projected costs of nonindigenous species in Canada. Biological Invasions 8: 45–59.CrossRefGoogle Scholar
  19. Cordell, J.R., R. Draheim, and M. Sytsma. 2007. First record of the harpacticoid genus Harpacticella in the Pacific northwest, USA: another probable introduction. Aquatic Biology 1: 17–20.CrossRefGoogle Scholar
  20. Dafforn, K.A., T.M. Glasby, and E.L. Johnston. 2009. Links between estuarine condition and spatial distributions of marine invaders. Diversity and Distributions 15: 807–821.CrossRefGoogle Scholar
  21. Davidson, I.C., L.D. McCann, P.W. Fofonoff, et al. 2008. The potential for hull-mediated species transfers by obsolete ships on their final voyages. Diversity and Distributions 14: 518–529.CrossRefGoogle Scholar
  22. Davis, R.C., F.T. Short, and D.M. Burdick. 1998. Quantifying the effects of green crab damage to eelgrass transplants. Restoration Ecology 6: 297–302.CrossRefGoogle Scholar
  23. denHartog, C. 1997. Is Sargassum muticum a threat to eelgrass beds? Aquatic Botany 58: 37–41.CrossRefGoogle Scholar
  24. deRivera, C., Ruiz, G., Crooks, J., et al. 2005. Broad-scale nonindigenous species monitoring along the west coast in National Marine Sanctuaries and National Estuarine Research Reserves. Report to the National Fish and Wildlife Foundation, pp. 126.Google Scholar
  25. deRivera, C.E., B.P. Steves, P.W. Fofonoff, et al. 2011. Potential for high-latitude marine invasions along western North America. Diversity and Distributions 17: 1198–1209.CrossRefGoogle Scholar
  26. Dethier, M.N., and S.D. Hacker. 2005. Physical factors vs. biotic resistance in controlling the invasion of an estuarine marsh grass. Ecological Applications 15: 1273–1283.CrossRefGoogle Scholar
  27. DiBacco, C., D.B. Humphrey, L.E. Nasmith, et al. 2011. Ballast water transport of non-indigenous zooplankton to Canadian ports. ICES Journal of Marine Science. doi:10.1093/icesjms/fsr133.Google Scholar
  28. Didham, R.K., J.M. Tylianakis, M.A. Hutchison, et al. 2005. Are invasive species the drivers of ecological change? Trends in Ecology & Evolution 20: 470–474.CrossRefGoogle Scholar
  29. Dudas, S.E., and J.F. Dower. 2006. Reproductive ecology and dispersal potential of the varnish clam Nuttallia obscurata, a recent invader in the Northeast Pacific. Marine Ecology-Progress Series 320: 195–205.CrossRefGoogle Scholar
  30. Elahi, R., and K. Sebens. 2012. Consumers mediate natural variation between prey richness and resource use in a benthic marine community. Marine Ecology-Progress Series 452: 131–143.CrossRefGoogle Scholar
  31. Environment Canada. 2011. Shellfish aquaculture statistics for British Columbia. In: Oceans and Marine Fisheries Branch, Ministry of the Environment http://www.env.gov.bc.ca/omfd/fishstats/aqua/shellfish.html/. Accessed March 2012.
  32. Foreman, M.G.G., W.R. Crawford, J.Y. Cherniawsky, et al. 2008. Dynamic Ocean topography for the Northeast Pacific and its continental margins. Geophysical Research Letters 35: 1–5.CrossRefGoogle Scholar
  33. Gillespie, G., Phillips, A., Paltzat, D., et al. 2007. Status of the European green crab, Carcinus maenas, in British Columbia - 2006. Department of Fisheries and Oceans, Canadian Technical Report of Fisheries and Aquatic Sciences, 2700, pp. 49.Google Scholar
  34. Gillespie, G.E. 2007. Distribution of non-indigenous intertidal species on the Pacific coast of Canada. Bulletin of the Japanese Society of Scientific Fisheries (Nippon Suisan Gakkaishi) 73: 1133–1137.CrossRefGoogle Scholar
  35. Glasby, T.M., S.D. Connell, M.G. Holloway, et al. 2007. Nonindigenous biota on artificial structures: could habitat creation facilitate biological invasions? Marine Biology 151: 887–895.CrossRefGoogle Scholar
  36. Gotelli, N.J., and R.K. Colwell. 2001. Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecology Letters 4: 379–391.CrossRefGoogle Scholar
  37. Haupt, T.M., C.L. Griffiths, T.B. Robinson, et al. 2010. Oysters as vectors of marine aliens, with notes on four introduced species associated with oyster farming in South Africa. African Zoology 45: 52–62.CrossRefGoogle Scholar
  38. Hemminga, M.A., and C.M. Duarte. 2000. Seagrass ecology. Cambridge: Cambridge University Press.CrossRefGoogle Scholar
  39. Hewitt, C.L., and M.L. Campbell. 2010. Relative contribution of vectors to the introduction and translocation of marine invasive species. Canberra City: Australian Government Department of Agriculture, Fisheries and Forestry.Google Scholar
  40. Incera, M., C. Olbarria, J.S. Troncoso, et al. 2009. Response of the invader Sargassum muticum to variability in nutrients supply. Marine Ecology-Progress Series 377: 91–101.CrossRefGoogle Scholar
  41. International Council for the Exploration of the Sea (ICES). 2005. ICES code of practice on the introductions and transfers of marine organisms. Copenhagen: ICES.Google Scholar
  42. Kask, B.A., J.R. Sibert, and B. Windecker. 1982. A check list of marine and brackish water harpacticoid copepods from the Nanaimo estuary, southwestern British Columbia. Syesis 15: 25–38.Google Scholar
  43. Kruskal, J.B., and M. Wish. 1978. Multidimensional scaling. Beverly Hills: Sage Publications, Inc..CrossRefGoogle Scholar
  44. Legendre, P., and L. Legendre. 1998. Numerical ecology, 2nd edn. Amsterdam: Elsevier Science B.V.Google Scholar
  45. Levings, C.D., R.E. Foreman, and V.J. Tunnicliffe. 1983. A review of the benthos of the strait of Georgia and contiguous fjords. Canadian Journal of Fisheries and Aquatic Sciences 40: 1120–1141.CrossRefGoogle Scholar
  46. Levings, C.D., D. Kieser, G.S. Jamieson, et al. 2002. Marine and estuarine alien species in the strait of Georgia, British Columbia. In Alien invaders in Canada’s waters, wetlands, and forests, eds. R. Claudi, P. Nantel, and E. Muckle-Jeffs, 113–131. Canada: Canadian Forest Service, Natural Resoures.Google Scholar
  47. Lo, V.B., C.D. Levings, and K.M.A. Chan. 2012. Quantifying potential propagule pressure of aquatic invasive species from the commercial shipping industry in Canada. Marine Pollution Bulletin 64: 295–302.CrossRefGoogle Scholar
  48. Lockwood, J.L., P. Cassey, and T. Blackburn. 2005. The role of propagule pressure in explaining species invasions. Trends in Ecology & Evolution 20: 223–228.CrossRefGoogle Scholar
  49. Mach, M.E. 2012. Research on marine coastal impacts to promote ecosystem-based management: Nonnative species in northeast Pacific estuaries. Resource Management and Environmental Studies. University of British Columbia, Vancouver, Ph.D. Thesis, pp. 189.Google Scholar
  50. MacLeod, C.J., S.E. Newson, G. Blackwell, et al. 2009. Enhanced niche opportunities: can they explain the success of New Zealand’s introduced bird species? Diversity and Distributions 15: 41–49.CrossRefGoogle Scholar
  51. Mann, R., and J.M. Harding. 2003. Salinity tolerance of larval Rapana venosa: implications for dispersal and establishment of an invading predatory gastropod on the north American Atlantic coast. Biological Bulletin 204: 96–103.CrossRefGoogle Scholar
  52. Maron, J.L., and M. Marler. 2007. Native plant diversity resists invasion at both low and high resource levels. Ecology 88: 2651–2661.CrossRefGoogle Scholar
  53. Martone, R.G., and K. Wasson. 2008. Impacts and interactions of multiple human perturbations in a California salt marsh. Oecologia 158: 151–163.CrossRefGoogle Scholar
  54. Mazerolle, M.J. 2012. AICcmodavg: model selection and multimodel inference based on (Q)AIC(c). R package version 1.24 Available at: http://CRAN.R-project.org/package=AICcmodavg/. Accessed Feb 2012.
  55. Miller, A.W., G.M. Ruiz, M.S. Minton, et al. 2007. Differentiating successful and failed mollusan invaders in estuarine ecosystems. Marine Ecology-Progress Series 332: 41–51.CrossRefGoogle Scholar
  56. Mineur, F., T. Belsher, M.P. Johnson, et al. 2007. Experimental assessment of oyster transfers as a vector for macroalgal introductions. Biological Conservation 137: 237–247.CrossRefGoogle Scholar
  57. Mineur, F., A.J. Davies, C.A. Maggs, et al. 2010. Fronts, jumps and secondary introducitons suggested as different invasion patterns in marine species, with an increase in spread rates over time. Proceedings of the Royal Society B 277: 2693–2701.CrossRefGoogle Scholar
  58. Morales, M. 2011. Sciplot: scientific graphing functions for factorial designs. R package version 1.0–9, Available at: http://CRAN.R-project.org/package=sciplot/. Accessed Apr 2012.
  59. Moyle, P.B., and T. Light. 1996. Fish invasions in California: do abiotic factors determine success? Ecology 77: 1666–1670.CrossRefGoogle Scholar
  60. Oksanen, J., Blanchet, F.G., Kindt, R., et al. 2012. Vegan: community ecology package. R package version 2.0–3. Available at: http://CRAN.R-project.org/package=vegan/. Accessed Feb 2012.
  61. Orth, R.J., T.J.B. Carruthers, W.C. Dennison, et al. 2006. A global crisis for seagrass ecosystems. Bioscience 56: 987–996.CrossRefGoogle Scholar
  62. Piercey, G.E., C.D. Levings, M. Elfert, et al. 2000. Invertebrate fauna in ballast water collected in vessels arriving in British Columbia ports, especially those from the western North Pacific. Canadian Data Report of Fisheries and Aquatic Sciences 1060: 50.Google Scholar
  63. Piola, R.F., and E.L. Johnston. 2007. Pollution reduces native diversity and increases invader dominance in marine hard-substrate communities. Diversity and Distributions 14: 329–342.CrossRefGoogle Scholar
  64. Powers, S.P., M.A. Bishop, J.H. Grabowski, et al. 2006. Distribution of the invasive bivalve Mya arenaria L. On intertidal flats of southcentral Alaska. Journal of Sea Research 55: 207–216.CrossRefGoogle Scholar
  65. Quayle, D. 1964. Distribution of introduced marine mollusca in British Columbia waters. Journal of the Fisheries Research Board of Canada 21: 1155–1181.CrossRefGoogle Scholar
  66. R Development Core Team. 2012. A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing ISBN 3-900051-07-0, http://www.R-project.org/.Google Scholar
  67. Ruiz, G.M., P.W. Fofonoff, J.T. Carlton, et al. 2000. Invasion of coastal marine communities in North America: apparent patterns, processes, and biases. Annual Review of Ecology and Systematics 31: 481–531.CrossRefGoogle Scholar
  68. Ruiz, G.M., P.W. Fofonoff, B. Steves, et al. 2011. Marine invasion history and vector analysis of California: a hotspot for western North America. Diversity and Distributions 17: 362–373.CrossRefGoogle Scholar
  69. Shea, K., and P. Chesson. 2002. Community ecology theory as a framework for biological invasions. Trends in Ecology & Evolution 17: 170–176.CrossRefGoogle Scholar
  70. Short, F.T., and R.G. Coles. 2001. Global seagrass research methods. Amsterdam: Elsevier Science.Google Scholar
  71. Short, F.T., and S. Wyllie-Echeverria. 1996. Natural and human-induced disturbance of seagrasses. Environmental Conservation 23: 17–27.CrossRefGoogle Scholar
  72. Sloan, N.A., and P.M. Bartier. 2004. Introduced marine species in the Haida Gwaii (queen Charlotte Islands) region, British Columbia. Canadian Field-Naturalist 118: 77–84.Google Scholar
  73. Sokal, R.R., and F.J. Rohlf. 1995. Biometry: the principals and practice of statistics in biological research, 3rd edn. New York: W.H. Freeman and Company.Google Scholar
  74. Sortie, C.J.B., I. Ibanez, D.M. Blumenthal, et al. 2013. Poised to prosper? A cross-system pomparison of climate change effects on native and non-native species performance. Ecology Letters 16: 261–270.CrossRefGoogle Scholar
  75. Stachowicz, J., R. Whitlatch, and R. Osman. 1999. Species diversity and invasion resistance in a marine ecosystem. Science 286: 1577.CrossRefGoogle Scholar
  76. Stachowicz, J.J., J.R. Terwin, R.B. Whitlatch, et al. 2002. Linking climate change and biological invasions: ocean warming facilitates nonindigenous species invasions. Proceedings of the National Academy of Sciences of the United States of America 99: 15497–15500.CrossRefGoogle Scholar
  77. Sylvester, F., O. Kalaci, B. Leung, et al. 2011. Hull fouling as an invasion vector: can simple models explain a complex problem? Journal of Applied Ecology 48: 415–423.CrossRefGoogle Scholar
  78. Thompson, R.E. 1981. Oceanography of the British Columbia coast. Canadian Special Publication of Fisheries and Aquatic Sciences 56: 291.Google Scholar
  79. Tilman, D. 1997. Community invasibility, recruitment limitation, and grassland biodiversity. Ecology 78: 81–92.CrossRefGoogle Scholar
  80. Transport Canada. 2006. Ballast water control and management regulations. In. http://laws.justice.gc.ca/en/SOR-2006-129/. Accessed May 2012.
  81. Vermonden, K., R.S.E.W. Leuven, and G. van der Velde. 2010. Environmental factors determining invasibility of urban waters for exotic macroinvertebrates. Diversity and Distributions 16: 1009–1021.CrossRefGoogle Scholar
  82. Waycott, M., C.M. Duarte, T.J.B. Carruthers, et al. 2009. Accelerating loss of seagrasses across the globe threatens coastal ecosystems. Proceedings of the National Academy of Sciences 106: 12377–12381.CrossRefGoogle Scholar
  83. Wilcove, D.S., D. Rothstein, J. Dubow, et al. 1998. Quantifying threats to imperiled species in the United States. Bioscience 48: 607–615.CrossRefGoogle Scholar
  84. Williams, S.L. 2007. Introduced species in seagrass ecosystems: status and concerns. Journal of Experimental Marine Biology and Ecology 350: 89–110.CrossRefGoogle Scholar
  85. Wonham, M.J., and E. Pachepsky. 2006. A null model of temporal trends in biological invasion records. Ecology Letters 9: 663–672.CrossRefGoogle Scholar
  86. Wyatt, A.S.J., C.L. Hewitt, D.I. Walker, et al. 2005. Marine introductions in the Shark Bay world heritage property, Western Australia: a preliminary assessment. Diversity and Distributions 11: 33–44.CrossRefGoogle Scholar
  87. Yamada, S.B., and G.E. Gillespie. 2008. Will the European green crab (Carcinus maenas) persist in the Pacific northwest? ICES Journal of Marine Science 65: 725–729.CrossRefGoogle Scholar

Copyright information

© Coastal and Estuarine Research Federation 2016

Authors and Affiliations

  • Megan E. Mach
    • 1
    • 2
  • Colin D. Levings
    • 3
  • Kai M. A. Chan
    • 1
  1. 1.Institute for Resources, Environment and Sustainability, Resource Management and Environmental SustainabilityUniversity of British ColumbiaVancouverCanada
  2. 2.Center for Ocean SolutionsMontereyUSA
  3. 3.Department of Fisheries and Oceans CanadaCentre for Aquaculture and Environment ResearchWest VancouverCanada

Personalised recommendations