Variable impacts of alien mink predation on birds, mammals and amphibians of the Finnish Archipelago: a long-term experimental study Peter Banks Mikael Nordström, Markus Ahola, Pälvi Salo, Karen Fey, Chris Dickman, Erkki Korpimäki Universities of New South Wales and Sydney, Australia University of Turku, Finland Metsähallitus
American mink (Mustela vison) in Fennoscandia Escaped from fur farms to Finnish ecosystems >6 years ago Continued deliberate release Widespread generalist predator seabirds small rodents amphibians What is their impact on native fauna?
Project background Concern of the impact of mink predation on populations of archipelago birds raised in 197s and 198s 1992: the Finnish Forest and Park Services started a pilot project to eradicate mink from a large area in the outer archipelago Bird, frog census started at the same time 1998: project extended with one more removal-control pair Mammal census added 23-26 Nessling Foundation grant
Aim of this talk Overview of the results of 13 years of experimental mink removal Responses of native archipelago fauna Birds, rodents and amphibians Effects on spatial patterns in distribution Effects on temporal patterns in abundance Summary of current directions
Study area and mink removal 2 Control, 2 Removal Areas (36km 2 total) Mink removal on 7 islands in R1 from 1992-25 New area R2 added in 1998, 62 islands
Control 2 Removal 2 Removal 1 1 kms Control 1 Finland
Island Characteristics Typical island traits Size: 1-6ha Ponds: 1-5 Within 5m of another 1-2 islands within 2km Island size and isolation included in all analyses
Occurrence of mink signs and island characteristics No signs Island isolation Signs of mink Island size 5 4 4 3 2 3 2 Isolation Size (ha) 1 1 1998 1999 2 21 1998 1999 2 21 Year Year
Pälvi Salo
Female, no kits Female, 1 kit Male Female, 4 kits 1 km
Experimental design: Birds Bird counts three times per breeding season all islands used Nest counts for most species except for waders and passerines 22 species for breeding data
Turnstone, Arenaria interpres Razorbill, Alca torda Common eider, Somateria mollissima Tufted duck, Aythya fuligula White-tailed Sea Eagle, Haliaeetus albicilla William Velmala/ www.tarsiger.com
Rock pipit, Anthus petrosus Common gull, Larus canus Mute Swan, Cygnus olor Arctic Tern, Sterna paradisaea William Velmala/ www.tarsiger.com
Impacts on bird populations = Removal 1 = Removal 2 + = Control 1 = Control 2 = Removal = Control 3 25 Tufted Duck 3 Long-term areas Short-term Arctic tern 3 2 1 2 1 Pairs/km 2 2 15 1 5 1993 1995 1997 1999 21 1998 1999 2 21 5 4 3 2 1 Turnstone 5 4 3 2 1 Pairs/km 2 3 2 1 1992 1994 1996 1998 2 22 4 Velvet Scoter 1993 1995 1997 1999 21 1998 1999 2 21 1992 1994 1996 1998 2 22 Nordström et al. 22 (Ecography) Nordström et al. 23 (Biological Conservation) Pairs/km 2 Pairs/km 2
Not all birds responsed... Mute Swan 4 Oystercatcher 3 2 1 1993 1995 1997 1999 21 Year 1992 1994 1996 1998 2 22 4 3 2 1 1998 1999 2 21 35 3 25 2 15 1 5 Common Eider Pairs/km 2 Pairs/km 2 4 35 3 25 2 15 1 5 5 4 3 2 1 Great black-backed gull 5 4 3 2 1 Pairs/km 2 Pairs/km 2 1992 1994 1996 1998 2 22 Year 1993 1995 1997 1999 21 1998 1999 2 21 Year Nordström et al. 22 (Ecography) Nordström et al. 23 (Biological Conservation)
Impacts on bird distributions 8 6 4 2 species / island 3 Removal Control 4 3 2 1 1 2 3 4 log e (isolation) log e (number of species) loge (number of species) 14 of 22 breeding birds populations increased Small size and late breeding most affected Control area diversity greatest on isolated islands no effect of isolation in removal area 4 2 1 1994 1996 1998 2 22 1 2 3 4 log e (isolation) a) Removal b) Control Korpimäki Increasing & Nordström isolation 24 Suomen Riista
Experimental Design: Rodents Voles trapped over 4 nights in Autumn 1998, 1999, 2 2 groups of 5 islands in each area (4 total) 18m Bank vole, Clethrionomys glareolus 4m Considered metapopulation influences Field vole, Microtus agrestis
Summary of rodent trapping Field Vole Bank Vole 697 individuals marked in 3 y 424 individuals marked in 3y -42 removal islands, -2 control islands absent from R2 (most remote area) 1 colonization's, 9 extinctions -43 on removal -19 on control 4 colonization's, 5 extinctions
larger island more island area nearby No association with mink Influenced by.5 1-7 -3-1 1 removal 3 Isolation index (island area nearby) -5 control Field vole distribution Probability of occupancy
Bank vole distribution Influenced by 1.5 Probability of occupancy more island area nearby No association with mink control removal -7-5 -3-1 1 3 Isolation index (island area nearby)
1 2 3 4 removal 1998 control B A 1999 Field vole abundance numbers of voles B 2 Banks et al 24 (Oikos)
Bank vole abundance 4 3 2 1 B control A removal numbers of voles 1998 1999 2 B Banks et al 24 (Oikos)
wet dry Rainfall variation 1 5 4 3 2 1-1 -2 rainfall residual index rain (mm) 345mm 285mm 242mm Mar Aug Feb Aug Feb Aug Feb Aug Feb 1997 1998 1999 2 21
Extinction and density High density populations more prone to extinction Suggestive of density dependent pressure on resources on small islands 1 8 6 4 2 % extinction events 3 13 7 2 5 1-1 11-2 21-3 31-4 >4 prior vole abundance Banks et al 24 (Oikos)
Inter-island dispersal 7 inter-island movements, all during high density 2m - 4.3km Maximum inter island movement of 14m
Different factors drive abundance and distribution Occupancy related to isolation extinction's also related to isolation accords with metapopulation predictions Abundance related to presence of mink and rainfall generalist mink may suppress boom-bust dynamics driven by rain
Mink Absent average summer rain high summer rains average summer rain Mink Present time
Experimental Design: Amphibians
Amphibian responses Common frog (Rana temporaria) Common toad (Bufo bufo) Ahola et al. 26
12 1 8 6 4 Common Frog Number of females Proportion islands occupied Temporal and spatial patterns 1..8.6.4 2.2 1992 1994 1996 1998 2 22 Year. Ahola et al. 26
Conclusions Feral American mink is an important predator for birds, voles and amphibians in the Baltic Sea archipelagos Impacts are highly variable in space, time and with species Interaction with fragmented environment Stochastic weather effects Some species take time to respond Affecting populations and communities Distributions and abundances
Sea eagle-mink-prey interactions Top predator White-tailed sea eagle (Haliaeetus albicilla) Mesopredator American mink (Mustela vison) Prey Seabirds Voles Producers/ prey Mussels, crustaceans, fish etc. Plants
Key research questions Effects of sea eagle predation risk on mink behaviour Can a native top predator suppress the feral meso-predator? Mink predation and vole metapopulations vole interactions with food plants (mink and the trophic cascade) test predictions of extinction model Mink impact on interspecific competition in voles Sub-lethal effects on habitat use & foraging behaviour or prey naïveté
Acknowledgements Maj & Tor Nessling Foundation Jouko Högmander, Jukka Nummelin & Metsähallitus William Velmala, Jayne Tipping and many volunteers