Predicting island biosecurity risk from introduced fauna using Bayesian Belief Networks.

Around the globe, islands are the last refuge for many threatened and endemic species. Islands are frequently also important sites for recreation, cultural activities, and industrial development, all of which facilitate the establishment of invasive species. Surveillance is employed on islands to detect the establishment of invasive species after their arrival, leading to decisions about follow-up actions. Unless surveillance is prioritised according to risk of establishment of invasives, it may be infeasible to implement efficiently over large tracts of publicly accessible land, especially in data-deficient areas. The key biosecurity problem for many regions is one of prioritizing sites for surveillance activities and identifying invasive species most likely to disperse to, and establish, and proliferate on those sites. We created a series of Bayesian Belief Networks (BBNs), linked by Java computing code and the freely available GeNIe application to automate the creation and computation of species- and site-specific biosecurity BBNs. The BBNs require data on island attributes, recreational or industrial visitor load, infrastructure, habitat availability, and animal behaviour and dispersal via swimming, flying, human movement, land bridges, or flood plumes. We used this biosecurity BBN to estimate the risk of 11 invasive faunal species arriving and establishing on 600 islands along the Pilbara coastline, Western Australia. Sensitivity analyses were conducted to identify nodes within the BBNs that required refined data inputs. Propagule pressure was the node with the greatest influence over the number of arrivals. Other nodes such as the number of visitors to islands and swimming capabilities of invasive animals greatly influenced the model results. Across the 11 species studied, our models predicted one arrival per 300 visitors. The biosecurity BBN can be used to identify the islands at highest risk from establishment of invasive species within any archipelago/s, and the invasive species most likely to establish on each island.

[1]  T. Blackburn,et al.  The more you introduce the more you get: the role of colonization pressure and propagule pressure in invasion ecology , 2009 .

[2]  Diana S. Jones.  The Burrup Peninsula and Dampier Archipelago, Western Australia: an introduction to the history of its discovery and study, marine habitats and their flora and fauna , 2004 .

[3]  Barbara L. Marks,et al.  Zero-tolerance biosecurity protects high-conservation-value island nature reserve , 2017, Scientific Reports.

[4]  L. C. van der Gaag,et al.  Practicable sensitivity analysis of Bayesian belief networks , 1998 .

[5]  D. Yemshanov,et al.  Representing uncertainty in a spatial invasion model that incorporates human-mediated dispersal , 2013 .

[6]  Hugh P. Possingham,et al.  Protecting islands from pest invasion: optimal allocation of biosecurity resources between quarantine and surveillance , 2010 .

[7]  G. Keighery,et al.  Prioritising weed management activities in a data deficient environment: the Pilbara islands, Western Australia , 2015, Heliyon.

[8]  Robert L Pressey,et al.  A novel approach to model exposure of coastal-marine ecosystems to riverine flood plumes based on remote sensing techniques. , 2013, Journal of environmental management.

[9]  D. Kriticos,et al.  Pest Risk Maps for Invasive Alien Species: A Roadmap for Improvement , 2010 .

[10]  K. Morris.  DAMPIER ARCHIPELAGO NATURE RESERVES , 2000 .

[11]  M. S. Hoddle,et al.  Population biology of invasive species. , 2001 .

[12]  Marnie L. Campbell,et al.  A review of international, regional and national biosecurity risk assessment frameworks , 2011 .

[13]  D. Simberloff,et al.  ECOLOGICAL RESISTANCE TO BIOLOGICAL INVASION OVERWHELMED BY PROPAGULE PRESSURE , 2005 .

[14]  R. Shine,et al.  The extra‐limital spread of an invasive species via ‘stowaway’ dispersal: toad to nowhere? , 2009 .

[15]  V. Cramer,et al.  Research priorities for the northern quoll (Dasyurus hallucatus) in the Pilbara region of Western Australia , 2016 .

[16]  Aminuddin Md Arof,et al.  The Application of a Combined Delphi-AHP Method in Maritime Transport Research-A Review , 2015 .

[17]  M. A. Tabak,et al.  Modeling the distribution of Norway rats (Rattus norvegicus) on offshore islands in the Falkland Islands , 2015 .

[18]  S. A. Morrison,et al.  Strategies for Biosecurity on a Nearshore Island in California , 2014 .

[19]  Phillip Cassey,et al.  Managing the risk of wildlife disease introduction: pathway-level biosecurity for preventing the introduction of alien ranaviruses , 2017 .

[20]  Achim Zeileis,et al.  Diagnostic Checking in Regression Relationships , 2015 .

[21]  S. Carpenter,et al.  Stability and Diversity of Ecosystems , 2007, Science.

[22]  J. Hodder,et al.  Maritime mammals: terrestrial mammals as consumers in marine intertidal communities , 2003 .

[23]  Peter Whittle,et al.  A method for designing complex biosecurity surveillance systems: detecting non‐indigenous species of invertebrates on Barrow Island , 2013 .

[24]  R. Somaweera,et al.  The (non) impact of invasive cane toads on freshwater crocodiles at Lake Argyle in tropical Australia , 2012 .

[25]  James C. Russell,et al.  Review of rat invasion biology: implications for island biosecurity. , 2008 .

[26]  Gregory P. Brown,et al.  Identifying optimal barriers to halt the invasion of cane toads Rhinella marina in arid Australia , 2013 .

[27]  Bryan F. J. Manly,et al.  Mammal extinctions on Australian islands: causes and conservation implications , 2002 .

[28]  P. Pheloung,et al.  A weed risk assessment model for use as a biosecurity tool evaluating plant introductions , 1999 .

[29]  J. Ross,et al.  Integrative Analysis of the Physical Transport Network into Australia , 2016, PloS one.

[30]  A. Zeileis.  Econometric Computing with HC and HAC Covariance Matrix Estimators , 2004 .

[31]  S. Harris,et al.  Turning the tide: The eradication of invasive species , 2002 .

[32]  M. Rouget,et al.  Prioritising surveillance for alien organisms transported as stowaways on ships travelling to South Africa , 2017, PloS one.

[33]  W. M. Lonsdale,et al.  Eradicating invasive plants: hard-won lessons for islands. , 2003 .

[34]  Kevin B. Korb,et al.  Bayesian Artificial Intelligence , 2004, Computer science and data analysis series.

[35]  E. García‐Berthou,et al.  Invasive species are a leading cause of animal extinctions. , 2005, Trends in ecology & evolution.

[36]  Martin A. Nuñez,et al.  Propagule pressure hypothesis not supported by an 80-year experiment on woody species invasion , 2011 .

[37]  J. Settele,et al.  Butterfly dispersal in inhospitable matrix: rare, risky, but long-distance , 2014, Landscape Ecology.

[38]  Daphne A. Onderdonk,et al.  Assessing the invasive potential of biofuel species proposed for Florida and the United States using the Australian Weed Risk Assessment. , 2011 .

[39]  Jeff Short,et al.  Surplus killing by introduced predators in Australia—evidence for ineffective anti-predator adaptations in native prey species? , 2002 .