Bellard, C; Rysman, J-F; Leroy, B; Claud, C; Mace, GM; (2017) A global picture of biological invasion threat on islands. Nature Ecology & Evolution , 1 (12) pp. 1862-1869. 10.1038/s41559-017-0365-6.

Preview

Text Mace_Bellard network submitted.pdf - Accepted Version Download (4MB) | Preview

Abstract

Biological invasions are among the main drivers of biodiversity losses. As threats from biological invasions increase, one of the most urgent tasks is to identify areas of high vulnerability. However, the lack of comprehensive information on the impacts of invasive alien species (IAS) is a problem especially on islands, where most of the recorded extinctions associated with IAS have occurred. Here we provide a global, network-oriented analysis of IAS on islands. Using network analysis, we structured 27,081 islands and 437 threatened vertebrates into 21 clusters, based on their profiles in term of invasiveness and shared vulnerabilities. These islands are mainly located in the Southern Hemisphere and many are in biodiversity hotspots. Some of the islands share similar characteristics regarding their connectivity that could be useful for understanding their response to invasive species. The major invaders found in these clusters of islands are feral cats, feral dogs, pigs and rats. Our analyses reveal those IAS that systematically act alone or in combination, and the pattern of shared IAS among threatened species, providing new information to implement effective eradication strategies. Combined with further local, contextual information this can contribute to global strategies to deal with IAS. Islands harbour a significant proportion of the Earth’s species and have an unusually high rate of endemism1. However, many species on islands are now recorded as threatened, and most of the recorded extinctions of vertebrate species have occurred on islands2. IAS are considered the second most important driver of species extinctions on islands, and are associated with nearly 25% of birds and amphibians that are currently threatened worldwide3. Island ecosystems are particul arly vulnerable to biological invasions2. To date, there have been over 700 attempted eradications of invasive alien mammals4 that have benefited 600 local populations, leading to larger populations or increased distributional areas4. Specifically, 236 species have benefited from those eradication programs, including 62 species that are at risk of extinction, and four species had their extinction risk reduced as a direct result of these eradications4. Despite these encouraging results, the threat posed by IAS remains substantial and widespread for native species3. Therefore, prioritization of research efforts and eradication strategies needs to be more effective5 and there are many more opportunities to decrease extinction risk for island species by eradicating IAS. Because the amount of funding that is allocated to conservation is limited, it is important that these interventions target those islands where the conservation benefit will be highest. Efforts to prioritize research, management and policies for IAS have traditionally relied on expert judgments, and have been limited to either single IAS6 or subsets of islands7. In the absence of a systematic analysis of IAS impacts, it is difficult to see how an efficient and comprehensive strategy can be developed. Although several recent papers have examined the threats posed by IAS globally8 or for entire taxonomic groups (for example, mammals9), all of these studies implicitly assume that the distribution of IAS reflects their impacts (for example, in terms of the decline in native species populations that they cause). However, there is a lot of evidence that the impacts of IAS are context dependent3,10,11 and spatially variable. Other large-scale studies have analysed IAS impact on specific archipelagoes12 or on a particularly problematic subset of IAS13. However, all of these approaches are piecemeal and cannot deliver the evidence that is needed to support an efficient approach to focus IAS research and action at a global scale. Here we use the comprehensive data compiled by the International Union for the Conservation of Nature (IUCN) database, BirdLife International database and the Global Invasive Species Database (GISD) to extract information about vertebrates that are threatened by IAS, and the identity of the threatening IAS to address the following three questions: (i) How are the impacts of IAS distributed among islands and threatened species? (ii) Are there combinations of IAS for which targeted actions may have high conservation impacts? (iii) Within the targeted areas, what are the characteristics of the network between IAS and IAS-threatened species that can improve strategies to deal with IAS? Some species occur on multiple islands whereas others are endemic to a single island. Moreover, some IAS are shared as threats by multiple species on the same or different groups of islands, whereas other IAS are very localized and specific. To gain insights into the structure of this complex interconnected system we apply network approaches. Many systems can be represented as networks of interconnected nodes. Networks are mathematical objects in which a node is linked (connected) to zero, one or several other nodes. The links highlight a relationship between two nodes. For instance, each inhabitant of the world can be represented as a node in a network graph and each email sent by those inhabitants to others can be represented as a link. Of course with increasing node and link number, the complexity of network graphs grows exponentially and limits our ability to identify structure. Network theory aims at extracting insightful patterns from networks. In our study, we built two kinds of networks. In the first, the nodes are either species or islands and the links represent the presence of a species on a given island. We clustered this network to identify islands that are highly interconnected by the co-occurrence of many IAS-threatened species and selected the clusters with highest number and densities of IAS-threatened species. The aim of this clustering approach is to identify groups of islands that share a similar pattern of threatened species based on biogeographical knowledge. In the second set of networks, the IAS and their threatened species were linked and related to the island clusters identified in the first analysis (see Methods). This allowed us to determine the IAS that are responsible for most of the threats in those clusters of islands and we used this to provide insights to deal with IAS threat.

Download activity - last month

Export as CSVXMLJSON

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item