Limits...
Rate of biological invasions is lower in coastal marine protected areas

View Article: PubMed Central - PubMed

ABSTRACT

Marine biological invasions threaten biodiversity worldwide. Here we explore how Marine Protected areas, by reducing human use of the coast, confer resilience against the introduction of non-indigenous species (NIS), using two very different Pacific islands as case studies for developing and testing mathematical models. We quantified NIS vectors and promoters on Vancouver (Canada) and Moorea (French Polynesia) islands, sampled and barcoded NIS, and tested models at different spatial scales with different types of interaction among vectors and between marine protection and NIS frequency. In our results NIS were negatively correlated with the dimension of the protected areas and the intensity of the protection. Small to medium geographical scale protection seemed to be efficient against NIS introductions. The likely benefit of MPAs was by exclusion of aquaculture, principally in Canada. These results emphasize the importance of marine protected areas for biodiversity conservation, and suggest that small or medium protected zones would confer efficient protection against NIS introduction.

No MeSH data available.


Scatter plot obtained from Principal Component Analysis of the sites and variables considered in this study.The diagonals (in green) represent the variables analyzed (red squares) and are proportional in length to the relative contribution of each variable to the total variance. Site acronyms (M for Moorea, green ellipses; V for Vancouver, orange ellipses): PP, Pao-Pao; Pa, Papetoai; Th, Tiahura; Fa, Faarehau; Vn, Vai’ane; Vr, Vai’are; Ti, Tiki; At, Atiha; Hu, Hauru; Mh, Maharepa, Te, Temae; Ma, Maatea; EB, Entre-2-Baies; Op, Opunohu; Af, Afareitou; CI, Cortes Island; Cr, Crofton; FB, Fanny Bay; NB, Nanoose Bay; PI, Portland Island; Ld, Ladysmith; Sd, Sydney; Vi, Victoria; Sk, Sooke; CB, China Beach; Bf, Bamfield; PA, Port Alberni; SB, Salmon Beach; LB, Long Beach.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5016778&req=5

f3: Scatter plot obtained from Principal Component Analysis of the sites and variables considered in this study.The diagonals (in green) represent the variables analyzed (red squares) and are proportional in length to the relative contribution of each variable to the total variance. Site acronyms (M for Moorea, green ellipses; V for Vancouver, orange ellipses): PP, Pao-Pao; Pa, Papetoai; Th, Tiahura; Fa, Faarehau; Vn, Vai’ane; Vr, Vai’are; Ti, Tiki; At, Atiha; Hu, Hauru; Mh, Maharepa, Te, Temae; Ma, Maatea; EB, Entre-2-Baies; Op, Opunohu; Af, Afareitou; CI, Cortes Island; Cr, Crofton; FB, Fanny Bay; NB, Nanoose Bay; PI, Portland Island; Ld, Ladysmith; Sd, Sydney; Vi, Victoria; Sk, Sooke; CB, China Beach; Bf, Bamfield; PA, Port Alberni; SB, Salmon Beach; LB, Long Beach.

Mentions: The principal component analysis (PCA) allowed us to identify two outlier sites (Fig. 3): China Beach and Long Beach from Vancouver Island. These two sites are anomalous because their distance to the nearest port is much longer than that of any other sampling point, thus this factor (distance to port) exhibited a discontinuous distribution if they were included. They were removed from our models in further analyses. Consistently with the correlations explained above, NIS proportion and local PS (Status-local in Fig. 3) were placed in quadrants 4 and 2 respectively and exhibited the longest diagonals, indicating opposite trends and maximum contribution to the dataset variance. Aquaculture was almost in the same position as NIS. In Fig. 3 it can be observed that despite very different conditions in the two islands (Table 1), the sites were not grouped by island but interspersed, with many sites from Moorea and Vancouver distributed in similar zones of the plot. Exceptions were four sites from Moorea (Hauru, Maatea, Temae, Tiki) and three sites from Vancouver Island (Cortes Island, Crofton, Fanny Beach), grouped by islands and located apart from the rest of the sites. Those four sites from Moorea had the maximum level of local PS, no NIS, and at the same time low number of native species, while the three sites from Vancouver Island had minimum level of protection and high proportion of NIS. The four principal components in the dataset accounted for 65% of the variance (Table 2), with NIS, local PS, 10-km PS and the number of native species providing the maximum load in PC1, PC2, PC3 and PC4 respectively.


Rate of biological invasions is lower in coastal marine protected areas
Scatter plot obtained from Principal Component Analysis of the sites and variables considered in this study.The diagonals (in green) represent the variables analyzed (red squares) and are proportional in length to the relative contribution of each variable to the total variance. Site acronyms (M for Moorea, green ellipses; V for Vancouver, orange ellipses): PP, Pao-Pao; Pa, Papetoai; Th, Tiahura; Fa, Faarehau; Vn, Vai’ane; Vr, Vai’are; Ti, Tiki; At, Atiha; Hu, Hauru; Mh, Maharepa, Te, Temae; Ma, Maatea; EB, Entre-2-Baies; Op, Opunohu; Af, Afareitou; CI, Cortes Island; Cr, Crofton; FB, Fanny Bay; NB, Nanoose Bay; PI, Portland Island; Ld, Ladysmith; Sd, Sydney; Vi, Victoria; Sk, Sooke; CB, China Beach; Bf, Bamfield; PA, Port Alberni; SB, Salmon Beach; LB, Long Beach.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5016778&req=5

f3: Scatter plot obtained from Principal Component Analysis of the sites and variables considered in this study.The diagonals (in green) represent the variables analyzed (red squares) and are proportional in length to the relative contribution of each variable to the total variance. Site acronyms (M for Moorea, green ellipses; V for Vancouver, orange ellipses): PP, Pao-Pao; Pa, Papetoai; Th, Tiahura; Fa, Faarehau; Vn, Vai’ane; Vr, Vai’are; Ti, Tiki; At, Atiha; Hu, Hauru; Mh, Maharepa, Te, Temae; Ma, Maatea; EB, Entre-2-Baies; Op, Opunohu; Af, Afareitou; CI, Cortes Island; Cr, Crofton; FB, Fanny Bay; NB, Nanoose Bay; PI, Portland Island; Ld, Ladysmith; Sd, Sydney; Vi, Victoria; Sk, Sooke; CB, China Beach; Bf, Bamfield; PA, Port Alberni; SB, Salmon Beach; LB, Long Beach.
Mentions: The principal component analysis (PCA) allowed us to identify two outlier sites (Fig. 3): China Beach and Long Beach from Vancouver Island. These two sites are anomalous because their distance to the nearest port is much longer than that of any other sampling point, thus this factor (distance to port) exhibited a discontinuous distribution if they were included. They were removed from our models in further analyses. Consistently with the correlations explained above, NIS proportion and local PS (Status-local in Fig. 3) were placed in quadrants 4 and 2 respectively and exhibited the longest diagonals, indicating opposite trends and maximum contribution to the dataset variance. Aquaculture was almost in the same position as NIS. In Fig. 3 it can be observed that despite very different conditions in the two islands (Table 1), the sites were not grouped by island but interspersed, with many sites from Moorea and Vancouver distributed in similar zones of the plot. Exceptions were four sites from Moorea (Hauru, Maatea, Temae, Tiki) and three sites from Vancouver Island (Cortes Island, Crofton, Fanny Beach), grouped by islands and located apart from the rest of the sites. Those four sites from Moorea had the maximum level of local PS, no NIS, and at the same time low number of native species, while the three sites from Vancouver Island had minimum level of protection and high proportion of NIS. The four principal components in the dataset accounted for 65% of the variance (Table 2), with NIS, local PS, 10-km PS and the number of native species providing the maximum load in PC1, PC2, PC3 and PC4 respectively.

View Article: PubMed Central - PubMed

ABSTRACT

Marine biological invasions threaten biodiversity worldwide. Here we explore how Marine Protected areas, by reducing human use of the coast, confer resilience against the introduction of non-indigenous species (NIS), using two very different Pacific islands as case studies for developing and testing mathematical models. We quantified NIS vectors and promoters on Vancouver (Canada) and Moorea (French Polynesia) islands, sampled and barcoded NIS, and tested models at different spatial scales with different types of interaction among vectors and between marine protection and NIS frequency. In our results NIS were negatively correlated with the dimension of the protected areas and the intensity of the protection. Small to medium geographical scale protection seemed to be efficient against NIS introductions. The likely benefit of MPAs was by exclusion of aquaculture, principally in Canada. These results emphasize the importance of marine protected areas for biodiversity conservation, and suggest that small or medium protected zones would confer efficient protection against NIS introduction.

No MeSH data available.