Biological Invasions

, Volume 12, Issue 8, pp 2559–2571

Predicting establishment success for introduced freshwater fishes: a role for climate matching

Original Paper

DOI: 10.1007/s10530-009-9665-3

Cite this article as:
Bomford, M., Barry, S.C. & Lawrence, E. Biol Invasions (2010) 12: 2559. doi:10.1007/s10530-009-9665-3


We modelled data comprising 1,189 successful and 489 failed introduction records for 280 species of freshwater fishes around the world. We found significant variations in establishment success between genera and families. The number of countries where introductions occurred was a significant predictor of the probability a species would establish in at least one country and all species that had been introduced to nine or more countries (46 species) had established at least one exotic population. We also conducted more detailed quantitative modelling for 135 species introduced to 10 countries to identify factors affecting establishment success. Relative to failed species, established species had better climate matches between the country where they were introduced and their geographic range elsewhere in the world. Established species were also more likely to have high establishment success rates elsewhere in the world. Neither the reason why fish were introduced nor the country where they were introduced was correlated with establishment success. Cross-validations indicated our model correctly categorised establishment success with 78% accuracy. Our findings may guide risk assessments for the import of live exotic fish to reduce the rate new species establish in the wild.


Climate matching Establishment success Exotic species Freshwater fish Prediction Risk assessment 



Food and Agriculture Organisation

ROC curve

Receiver operating characteristic curve

Copyright information

© Springer Science+Business Media B.V. 2009

Authors and Affiliations

  1. 1.Bureau of Rural SciencesCanberraAustralia
  2. 2.CSIRO Mathematical and Information SciencesCanberraAustralia
  3. 3.FraserAustralia

Personalised recommendations