Dispersal and depensation in low density culled mink populations

Lead Research Organisation: University of Aberdeen
Department Name: Inst of Biological and Environmental Sci


Understanding the ecological dynamics of populations at range margins where populations may be patchy and at low density is fundamental for our understanding of how range expansion can ensue and how invasive species spread. Species of conservation concern are by definition found at lower than expected densities. Therefore understanding processes causing low density populations to perform poorly is fundamental to conservation and management. While there is a rich body of theory predicting its impact on species spread and persistence, empirical studies of depensation are scarce owing to the inherent difficulty in studying populations at very low density. Allee effects however have been detected in a range of invasive species. Strikingly however, even though depensatory processes define the early demographic spread of invasive species, it is only very recently that modelling studies have started considering depensation in the management of established invasive species.

The aim of this application is to improve the link between theory and practice and, in doing so improve both ecological understanding of depensation and test its potential as an effective tool for invasive species management. Our program of research is integral to a cooperative conservation project that has already removed invasive American mink from 10,000km2. Given the conservation project area is surrounded by uncontrolled mink populations, it is exposed to re-invasion from its periphery in a process not dissimilar to the initial invasion and to the spread of species in a new range. This represents a unique opportunity to test predictions on patterns of immigration and demography over a large spatial scale in a natural setting normally inaccessible to investigation, making it possible, for the first time, to gain an empirical understanding of processes operating at low density.

By combining genetic and age information from all mink caught as part of the conservation project, we can reconstruct genealogies, and subsequently measure the extent to which mink spread through the environment. We will do this for all mink caught since January 2010 and combine these data with information already obtained from 365 mink caught when the mink population was at saturation. Thus we will have information on dispersal patterns from area that were either saturated (pre-control), partially depleted (first year of raft deployment) or severely depleted of conspecifics (subsequently).

We will establish when the spread of mink may be affected by the inability to locate mates at very low density by comparing patterns of dispersal and reproduction in low density mink populations immigrating from the periphery of the (expanding) project area with patterns from saturated populations. We will also consider whether the fact that un-mated females mink remain in heat and after some time may suffer from anemia contribute to heightened mortality of females. We will also determine whether the observation that dispersing mink tend to settle in productive areas could lead to emergent depensation described as "range pinning", whereby isolated colonies fail to spread because population loss due to emigration is not be compensated for by immigration. We will also use our data to test the effectiveness of management options that aim to maximise depensation. It has been suggested that a potential control strategy for some invasive species may be to suppress peaks in population abundance close to range borders. This could increase the likelihood of invasion pinning by reducing the flow of individuals from higher density populations behind the front. If effective, such management might be cost effective given the effort required to detect individuals at low density in vast controlled areas.

Our project is unique in combining ecological theory with practical management of an invasive species in partnership with communities.

Planned Impact

This project will use an excellent system to deliver general insights into fundamental ecological processes operating in low density populations. This work will build upon our in depth knowledge of our system to examine how habitat heterogeneity (variability in the size and spatial distribution of habitat with different productivity potential), distance from source populations, and dispersal rates and mechanisms, define demographic parameters and ultimately the success or failure of nascent populations.

Whilst this will have immediate applied value for the SMI, the potential for broader impact upon conservation and ecological sciences is considerable, as the models developed on our findings will be flexible to parameterisation to fit a range of other scenarios. Thus we predict that our work will have direct value for the expansive range of stakeholders involved in this and other mink eradication schemes (e.g. SNH, SWT, CNPA, RSPB, NTS Scotland, 16 rivers and fisheries trust, the Hebridean Mink Project, > 600 volunteers and SMI staff, and control projects in Italy, France, Spain, Poland and Finland), but will also have a wider relevance to invasive eradication schemes per se and in particular those dealing with low density populations e.g. red foxes in Tasmania, grey squirrels in the UK, stoats and ferrets in New Zealand.

Our findings will also contribute to our general ecological understanding of the dynamics of low density populations and will provide valuable insights well beyond invasion biology e.g. to understand extinction/recovery thresholds for populations driven to very low density; identify ranges of demographic traits that determine species susceptibility to depensation effects; and to understand the role of habitat heterogeneity in population persistence.

A key to the success of this project will be to ensure our findings are effectively disseminated to all stakeholders involved in mink eradication, the scientific and general communities, and policy makers. We will achieve this through a number of channels. Firstly, we will engage SMI stakeholders through: daily contact between the PI, PDRA and SMI POs, and subsequently POs direct contact with volunteers and community talks; lay articles in SMI quarterly newsletters and website; participation in the RAFTS annual Freshwater Biosecurity and INNS Conference. Secondly, we will communicate our findings to the scientific community through: publication in high tier journals; what about journal blogs?; and presentations at international meetings (including the 3rd European Congress of Conservation Biology). Thirdly, we will raise awareness with the wider public through press releases, Interviews and articles targeting popular media (previously e.g. BBC, NERC Planet Earth, US National Public Radio) in coordination with the University Press Dept and SWT. We have an excellent track record in delivering across all these channels of impact.


10 25 50

Description The overall of this project has been to link theory and practice and, in doing so, improve both ecological understanding of low-density 'depensatory' processes and how this may present potential as an effective tool for invasive species management. One of the main challenges faced has been finding robust genetic techniques to identify immigrants. We have found that care must be taken in using pedigree-based techniques in wild populations where only partial sampling of genes and the population are achieved. This has been confirmed by other researchers using simulation techniques to investigate this problem. As such we have steered away from using pedigree-based analysis to identify the precise locations of dispersal sources, when such assignments could strongly affect outcomes and inferences. Instead we have developed a method that examines the number of genetic 'relatives' within a given radius of a focal mink. Mink that have no relatives (i.e. individuals with a relatedness value above a threshold determined by simulations on the observed allele frequency data) within their neighbourhood are assigned as immigrants. An advantage of this is the simple assumption that highly genetically related individuals are likely to be pedigree relatives, whilst potentially erroneous stringent assignments about the exact particular pair-wise relationship are avoided. We have applied this technique to show a 50% probability of being an immigrant for mink captured in areas with 2 years of mink control, versus only 30% for areas in the first year of control. This is evidence of compensatory dispersal in response to mink culling. We are now adapting this method to investigate depensatory processes.

- Our first objective was to compare patterns of dispersal and reproduction in low density mink populations immigrating from the periphery of the (expanding) project area with patterns from saturated populations to quantify the distance travelled by mates. Additionally we aimed to estimate the percentage of unmated females from mating scars, placental scars and age in relation to the presence of males and distance to the source of dispersing males caught, or detected on rafts but not caught. So far, 288 mink (additional to ca. 350 previously analysed) have been genotyped at 15 microsatellite loci, and ca. 200 adult female mink have been dissected and assessed for mating scars and placental scars. All of these mink have additionally had their age determined by cementum analysis of removed canine teeth.

Outcomes: Our preliminary data indicate an increase in the variance of female reproduction in areas that have been subject to longer-term mink control. This is due to some females being unmated, and some females that are mated having higher numbers of offspring (determined by counting placental scars, or embryos). This is indicative of both depensatory processes (e.g. failure to find a mate), as well as compensatory processes (e.g. increased investment in reproduction in low density areas). Whilst this analysis requires the incorporation of multiple covariates into models, we are optimistic that the initial observations point toward findings that will be highly relevant to both pure and applied ecologists.

- Our second objective was to determine whether individual dispersal decisions, as revealed by observed settlement conditions could lead to emergent depensation. Using GLMMs to account for spatial autocorrelation we aim to estimate the likelihood that areas with different productivity, and hence attractiveness, are colonised by males and females for given distances to putative source areas. We are also testing the hypothesis that highly mobile males immigrate to productive patches irrespective of the presence of females.

Outcomes: We are currently in the process of mapping the productivity landscape. This has been delayed slightly by the discovery of new, highly productive, areas during the duration of this grant, which are being incorporated into analysis in order to improve our analyses.

- Our third objective was to compare fluxes of immigrants in situations where culling short of eradication took place and where "attractive sinks" were created in productive areas. We are doing this by using variation in the time when catchment were effectively controlled by modelling immigration rate in relation to productivity and control effort history with immigrants and litters being inferred from genetic relatedness and pedigree reconstruction.

Outcomes: As with objective 2, the mapping of landscape productivity and genetic relatedness matrices are now complete. We are confident that we now have the data to adequately address the focal questions pertaining to low density processes for all objectives, as we have identified > 120 mink (approximately 1:1 sex ratio) that are from areas where density has been considerably reduced due to mink eradication efforts.
Exploitation Route The work is of direct relevance to conservation efforts, often conducted by NGOs or governement agencies and is being used as such The work provide key information that can be used for optimising effort and spend when attempting to control a highly mobile invasive species such as mink. this is a Europe-wide problem as well as a majopr concerns in australia New Zealand where substantial resources are being expanded to minimise the impact of these species

Multiple conference and seminar presentations

Web site accessed by many > 600 volunteers
Sectors Environment

URL http://www.scottishmink.org.uk/
Description Our findings are now fully integrated in management practice by the Scottish mink initiative and sister projects UK and Europe-wide
First Year Of Impact 2013
Sector Environment
Impact Types Economic

Description 2 year extension of Scottish Mink initiative and studentship
Amount £52,000 (GBP)
Organisation Mammals Trust UK 
Sector Charity/Non Profit
Country United Kingdom
Start 03/2014 
End 04/2017
Description 2 year extension of Scottish Mink initiative and studentship
Amount £52,000 (GBP)
Organisation Mammals Trust UK 
Sector Charity/Non Profit
Country United Kingdom
Start 09/2014 
End 10/2017
Description Marie curie fellowship to Yolanda Melero
Amount € 200,000 (EUR)
Funding ID 300288 
Organisation Marie Sklodowska-Curie Actions 
Sector Charity/Non Profit
Country Global
Start 04/2011 
End 05/2013
Description grant
Amount £60,000 (GBP)
Organisation Mammals Trust UK 
Sector Charity/Non Profit
Country United Kingdom
Start 09/2015 
End 10/2018
Description Scottish Mink Initiative 
Organisation Scottish Mink Initiative
Country United Kingdom 
Sector Charity/Non Profit 
PI Contribution All samples we analyse are collected by the partnership SMI which gathers Rivers and Fisheries Trust of Scotland (RAFTS), SNH, Cairngroms National Park authority, SWT and the University of Aberdeen in an inclusive consortium that coordinates networks of volunteers controlling mink
Start Year 2009