Latest Version, 2018 Assessment

Invasive mammal presence on island seabird colonies

The UK target of a reduction in the risks to island seabird colonies from invasive predatory mammals could not be assessed because there has been no previous assessment of this risk. A total of 30 Special Protection Areas did not have invasive non-native or native mammals on offshore islands. Six had effective biosecurity measures. Twelve Special Protection Areas had rats and or feral cats present on shore islands and did not have effective biosecurity measures.

Background

UK target on bird population condition: Invasive mammals

This indicator aims to assess progress against the following target, which is set in the UK Marine Strategy Part One (HM-Government, 2012): “At the scale of the Marine Strategy Framework Directive sub-regions, the risks to island seabird colonies from non-native mammals are reduced.” However, there has been no previous assessment, so it was not possible to monitor progress towards the target.

Key pressures and impacts

The invasion of seabird colonies on offshore islands by non-native predatory mammals such as brown rats, cats, and American mink and native mammals introduced by humans like the hedgehog and fox has had severe impacts on populations of ground-nesting seabirds in the UK including petrels and puffins. Predation of eggs and young birds can cause reductions in breeding success and could lead to the desertion of whole colonies.

Measures taken to address the impacts

The UK Marine Strategy Part Three (HM Government, 2015) states that future implementation of a UK-wide programme of quarantine (also referred to as ‘biosecurity’) against invasive, non-indigenous mammals from island seabird colonies and the strategically targeted removal of mammals from some islands will increase our confidence in meeting the above target.

Further information

The risks to island seabird colonies from invasive predatory mammals

Predation of ground-nesting marine birds, their eggs and chicks by mammals can have impacts at a population scale. The main strategy adopted by seabirds to cope with mammalian predation is avoidance. They tend to nest on cliffs, offshore island or remote beaches where predators are scarce or absent (Birkhead and Furness, 1985). The introduction of mammal predators to places where they would not naturally occur without human assistance has had catastrophic impacts on bird populations around the world (see for example: Courchamp and others, 2003; Jones and others, 2008; Russell and Clout, 2005; Towns and others, 2006).

In the UK, Stanbury and others (2017) identified 9,688 distinct islands around the coast of the UK. The vast majority of these are very small, and only 506 of them are larger than 10 hectares Many of these islands have ground-nesting birds that can breed successfully in the absence of predatory mammals. Some of these islands are home to the largest seabird colonies in Europe. However, some islands do have invasive predatory mammals (see Table 1 for species list). The most widespread species on islands are the American mink (Mustela vison) and brown rat (Rattus norvegicus ), which occur on 23 % and 21 % of islands (>10 hectars), respectively (Stanbury and others, 2017). Both species are non-native to the UK. Brown rats are thought to have originated on the steppes of central Asia and spread, possibly naturally, to eastern Europe in the early eighteenth century. Brown rats were first reported arriving in England on ships around 1728 to 1729 (Non-native Species Secretariat, 2012). The American mink, escaped from fur farms following the expansion of the industry in the UK during the 1950s (Dunstone, 1993). They have since colonised almost all of mainland Britain and some of the larger Scottish islands including Skye, Lewis and Harris (see data and maps for Neovison vison in the National Biodiversity Network Atlas). American mink are adept swimmers and can easily reach seabird colonies on inshore islands, where they cause complete breeding failures, adult mortality, and eventually site abandonment. Along the west coast of Scotland, mink predation within the last 10 to 30 years has led to the redistribution and decline in numbers of common and Arctic terns, common and black-headed gulls and black guillemots (Craik 1997; 1998; Mitchell and others, 2004).

Brown rats are less adept swimmers than mink but can be accidentally introduced on to island by boats. On islands where rats have been recently introduced, the impacts on the resident seabirds is evidenced through. the extinction of puffins on Ailsa Craig and Puffin Island and of Manx Shearwaters on Canna. However, on most islands where rats are present, their impact is not as visible. This is because the presence of rats for almost 300 years in some places has led to an equilibrium between the impact of rats and the number and distribution of seabirds. This is evidenced by the fact that European and Leach’s storm-petrels are only found breeding on islands that are free of rats (Mitchell and others, 2004; Mitchell and Newton, 2004). Furthermore, seabird species have returned to breed on some islands following the eradication of rats, for example, Manx shearwaters on Lundy (Lock, 2006).

On islands where seabirds and rats co-exist, impacts are evidenced by the distribution and abundance of nesting seabirds. For example, in Orkney and Shetland, black guillemots nesting on islands with rats and stoats avoid predation by nesting in crevices high off the ground and on cliffs that are inaccessible to the rats and stoats, rather than in the boulder beaches that they use as nest sites on islands that have no rats or stoats (Ewins and Tasker, 1985). Mitchell and Ratcliffe (2007) found that throughout the UK there were significantly more puffins breeding on islands without rats, compared to islands of a similar size that do have rats.

Table 1 lists species of predatory mammals that are of a high risk to ground-nesting marine birds. In contrast to rats and mink, other species of invasive predatory mammals appear to have a much more limited distribution on islands and therefore have a more localised impact on seabirds. In addition to non-native brown rats and American mink, native species can be introduced to islands that they would not naturally get to by swimming. In such cases, these native species are considered invasive. The most widespread invasive native species on UK islands is the hedgehog (Table 1).

Table 1: Species of predatory mammals present on islands in the UK. All species are considered a high risk to ground-nesting birds (Stanbury and others, 2017).

Mammal species	Status in the UK	% of islands (>10 hectares) in the UK with confirmed or probable presence
Brown rat (aka Norway rat) Rattus norvegicus	Non-native	21%
Black rat (aka ship/roof rat) R. rattus	Non-native	2%
American mink Neovison vison	Non-native	23%
Ferret Mustela furo	Feral	4%
Stoat Mustela erminea	Native	4%
Weasel Mustela nivalis	Native	1%
European badger Meles meles	Native	1%
Hedgehog Erinaceus europaeus	Native	12%
Red Fox Vulpes vulpes	Native	2%
Cat Felis catus	Feral / free-roaming domestic pet	17%

Assessment method

Assessment of UK target

This first assessment measures the number of protected islands on which impacts from invasive mammals are already occurring and the number of predator-free islands on which the risk to seabirds from invasive mammals has been minimised by effective quarantine and other biosecurity measures. Future assessments will investigate whether the numbers of these islands have changed sufficiently to reduce the overall risk from invasive mammals and thus meet the UK target.

The presence or absence of invasive predatory mammals (see species list in Table 1) was determined at all 42 of the UK offshore Islands within the Greater North Sea and Celtic Seas that are designated as protected Special Protection Areas designated under the Birds Directive (European Commission, 2009). These Special Protection Areas contained a total of over 700 islands. All Special Protection Areas in the assessment were designated because of the national or international importance of their breeding seabird populations, and all are on islands). On those islands that do not have invasive predatory mammals, the assessment looked at whether biosecurity measures were in place and how effective they are at minimising the risk of these mammals arriving and becoming established. Brown rat and American mink are the most widespread predatory mammals on UK islands (Table 1) and are the most likely of the non-native species to arrive at an island. Mink are most likely to arrive on islands by swimming. Rats can also swim to islands close by but can also stow away on boats without being detected and can be accidentally introduced onto islands, potentially a long way from the mainland. If measures are sufficient to prevent incursion and establishment by rats and mink, then they will most probably be effective against other species.

On islands where invasive predatory mammals are currently present, this assessment focused on whether any biosecurity was in place to prevent further incursion by other species. This is important, since the invasion by new species may impact further on the resident bird communities. Sufficient biosecurity measures will also be important in ensuring the success of future eradication schemes. At each site, a risk assessment was carried out to determine if the risk from invasive predatory mammals had been minimised by effective biosecurity. The results of the risk assessment were categorised as follows:

• ‘Yes’: the potential risk (without any biosecurity) had been a reduced to a low actual risk by having in place (or planned) ‘sufficient’ quarantine measures, invasive monitoring scheme and rapid response.
• ‘Partially’: the potential risk had been a reduced to a lower actual risk, but not all the measures were considered ‘sufficient.’
• ‘No’: the potential risk had NOT been reduced to a lower actual risk and measures were absent or ‘below standard.’

Defining an ‘island’

Islands were defined as being entirely distinct from the mainland at low tide, small islands that are part of bigger mainland Special Protection Areas have mostly been excluded. Exclusions include Steep Holm and Havergate Island. We have only included island Special Protection Areas designated wholly or primarily for their breeding seabird interest features. Offshore islands that are permanently surrounded by the sea are less likely to be invaded by mammals than tidal islands that are connected to the mainland by a land-bridge at low tide. Mammals can potentially get to tidal islands during each low-tide, occurring twice daily. The biosecurity measures required to stop mammals getting onto tidal islands are so intensive that they are often not financially or logistically viable. Therefore, tidal islands were not included in this assessment. For one Special Protection Area (Ynys Feurig, Cemlyn Bay and the Skerries), only the Skerries, which are fully offshore islands, were assessed for biosecurity. A total of 723 islands were included in this assessment.

Risk assessment approach

This assessment was conducted by following the steps laid out in the flowchart in Figure 1. First, the potential risk (without biosecurity) from rats and mink was assessed as low, medium or high (L, M, and H in Figure 1) for each site. The risk was calcuated as the product of the likelihood of incursion and the potential hazard from incursion (see risk matrix in Table 2). The effectiveness of current biosecurity measures in reducing both the likelihood and the hazard were assessed to estimate the actual risk from rats and mink at each site, given the measures in place (see Figure 1 and Table 2). Each step in the assessment was scored. The majority of these scores were taken from interviews of site managers conducted by the Royal Society for the Protection of Birds as part of the Shiant Isles Seabird Recovery Project (Thomas and others, 2016). Additional information was used to supplement the results of these interviews.

Figure 1: Risk assessment framework for assessing whether potential risk from invasive non-native mammals has been reduced to a lower ‘actual risk’ by current biosecurity measures at each site.

Table 2. Risk Matrix used to assess the potential or actual risk to an island seabird population from invasion by mammal predators.

		Likelihood of invasion
		L	M	H
Hazard associated with invasion	H	Medium	High	High
	M	Medium	Medium	High
	L	Low	Medium	Medium

Assessing potential risk

Potential risk from invasion by rats or mink was assessed only for sites that currently have no high-impact mammals. First the likelihood of incursion on the island was considered, then the hazard or potential damage they would cause following arrival. The scores for these two processes were put into the risk matrix in Table 1 to estimate the potential risk from invasion.

The potential likelihood of incursion by rats and mink was scored separately for each mammal as ‘high,’ ‘medium’ or ‘low.’ The score was dependant on the pathway of introduction, be it natural, such as swimming or human-assisted – usually accidentally by boats visiting an island. The likelihood was assessed by considering the following:

1. Human-assisted introductions: Site managers were asked to provide information on the current levels of boat traffic to the island and the nature of the traffic – including types of vessels used and the purposes of visits. Assumptions were made that islands with resident populations would have higher risks of human-assisted pathways than those that were uninhabited. Ratcliffe and others (2009) found no rat-free islands in the UK that were inhabited by more than 100 people and considered islands with a human population less than 10 to be of low risk of invasion, islands with 11 to 50 people being medium risk and over 51 to be high risk.

2. Natural introductions: Distances to neighbouring islands containing invasive species, and distance to the mainland determines how likely invasive species could get to the island by swimming there. Brown rats can swim for up to 1 km in open, temperate seas (Russell and others, 2008) and occasionally 2 km and possibly 4 km (Thomas and others, 2017). But Ratcliffe and others (2009) found that many UK islands are rat-free despite being less than 1 km from the mainland, possibly due to strong currents in the channels separating the islands from the mainland. Ratcliffe and others (2009) found a maximum swimming distance of 300 m fitted better with the observed distribution of rats on offshore islands in the UK. American Mink are much stronger swimmers that rats and can reach islands within 2 km of shore along the west coast of Scotland (Ratcliffe and others, 2009).

The potential hazard or potential damage that could be caused following the incursion of an invasive mammal onto an island seabird colony was scored as ‘high,’ ‘medium’ or ‘low’ by considering the following:

1. The likely impact of invasion on resident seabird species – this was scored as ‘High’ for all sites included in this assessment because they are all protected for their important colonies of breeding seabirds, which would certainly be impacted if the site was invaded by a high-impact mammal species.
2. The likelihood of establishment of a breeding population of rats and mink following an initial incursion. This is dependent on the ability to detect the invasive species and to then respond rapidly to remove any animals. Without systematic monitoring it is highly likely that rats that could arrive on an island (either by natural or human-assisted pathways) and remain undetected and then multiply quickly.

Assessing the actual risk

This part of the risk assessment considered the effectiveness of current biosecurity measures in reducing the potential risk of invasion and estimated the actual risk from invasive mammals. Current biosecurity measures were assessed against standards set in the UK Rodent Eradication Best Practice Toolkit (Thomas and others, 2017). They were scored as ‘sufficient,’ ‘below standard’ or ‘none’. At some sites, a biosecurity plan was in place, but the actual implementation of the plan had not yet taken place. In such cases, the assessment was based on the contents of the plan and on the assumption that it would be implemented.

Firstly, using data on actual biosecurity measures (for example, the quarantine) in place at each site, the potential likelihood of incursions by rats and by mink were each reassessed to determine the ‘actual likelihood of incursion’ (as low, medium or high). For example, if quarantine measures were ‘sufficient,’ the actual likelihood of incursion would be assessed as lower than the potential likelihood. There would be no reduction in actual likelihood if there were no measures in place. To meet the standard of sufficient quarantine measures, as many barriers and checks as possible need to be in place along pathways of introduction (Thomas and others, 2017). ‘Barriers’ usually involve placing traps or poison stations at strategic locations that would:

• prevent species getting on to vessels, either directly, for example, by climbing up mooring ropes or indirectly, for example, as a stowaway in cargo
• prevent species dispersing from land within swimming distance of the island
• identify the presence of species on vessels in transit
• prevent species getting off vessels
• prevent species getting out of quarantine areas on the island

Secondly, the potential hazard from invasion was reassessed at each site given the measures in place that could reduce the likelihood of the invasive species becoming established on the island. The first of these measures include an effective surveillance scheme designed to detect the incursion of an invasive species as soon as possible after it has occurred (for example, by using non-toxic wax blocks set inside bait stations, tracking tunnels, rodent motels, and trail cameras).

Once an invasive species has been detected on an island, a rapid response is required to remove all individuals from the island before they spread far from the incursion point. Preparedness for incursion response was assessed through interviews with site managers. The biosecurity plans currently in development contain enhanced monitoring and incursion response plans to deal with possible and probable ordefinite invasive species sightings. Incursion response plans are based on setting up a grid of bait stations with rodenticide bait 250 m in all directions around the sighting or sightings augmented with snap traps and non-toxic monitoring.

The combined assessment of surveillance and rapid incursion response measures was used to determine if the actual likelihood of invasion could be scored lower than the potential likelihood. Finally, the lowest of the scores for likely impact and actual likelihood of invasion was entered as the hazard score into the risk matrix (Table 1). The actual likelihood of incursion was also entered into the risk matrix to determine an overall score for the actual risk from invasive mammals at the site.

Results

Findings from the 2012 UK Initial Assessment

This indicator was not considered as part of the 2012 Initial Assessment (HM Government, 2012).

Latest findings

Status assessment

Invasive predatory mammals were present on 12 out of the 42 Special Protection Areas assessed (see Figure 2) and absent from islands in 30 Special Protection Areas (Figure 3). Note thatthe Isles of Scilly Special Protection Area is included in both figures because high-impact invasive mammals are present on some, but not on all of its 85 islands. Of the 12 Special Protection Areas where high-impact invasive mammals were present (Figure 2), eight had rats. The other four Special Protection Areas had feral cats present. On islands in these Special Protection Areas, seabirds will be restricted to nesting in places that are inaccessible to rats and cats, such as cliffs. Hence, seabird species that don’t nest on cliffs may be scarce or absent from islands where there are rats or cats. Evidence from other islands shows that it will be a major undertaking to remove the rats and the removal of feral cats can be contentious. Biosecurity measures were carried out at only half of these sites (six) and were only partially effective at minimising the risk of other predatory mammal species arriving and becoming established (Figure 2).

Figure 2 – Island seabird colonies within UK Special Protection Areas, where invasive predatory mammals were present and the level to which risk of further incursion has been minimised by biosecurity. Isles of Scilly – refers only to Tresco, St Martin’s, Bryher, Samson, St Helen’s, Northwethel, Men-a-vaur, Tean, Merrick and Round Island. Some contain brown rats and/or feral cats.

Figure 3. Island seabird colonies within UK Special Protection Areas, where invasive predatory mammals were absent. Isles of Scilly – refers only to Annet (successful response to rat incursion in 2004) and St Agnes and Gugh, where rats were eradicated in 2013/14. Firth of Forth Islands include Isle of May, Bass Rock, Inchmickery, Craigleith, Lamb and Fidra.

At the Special Protection Areas that had no invasive predatory mammals, the risk of invasion was minimised by effective biosecurity at six sites and was partially reduced at a further ten sites (Figures 3 and 4). At the remaining 14 Special Protection Areas, biosecurity was either not carried out or was below the standard required to reduce the risk of invasive mammals arriving.

Figure 4. The number of UK Special Protection Areas, where invasive predatory mammals were absent (as listed in Figure 3) and where biosecurity measures have minimised the risk (yes, partially, or no) from predatory mammals arriving and becoming established.

Trend assessment

The trends are unknown. The indicator was not considered as part of the UK Initial Assessment (HM Government, 2012).

Further information

Table 3 shows the results of the step-by-step process to assess risk from invasive brown rats and American mink on the 30 island Special Protect Areas where both species are currently absent. The column headings in Table 3 correspond to the steps outlined in the risk assessment process in Figure 1. Potential risk (without current biosecurity) and actual risk (with current biosecurity) were estimated using the risk matrix in Table 2. The majority of the scores in Table 2 were taken from interviews of site managers conducted by the Royal Society for the Protection of Birds as part of the Shiant Isles Seabird Recovery Project (Thomas and others, 2016). The results of the risk assessment in Table 3 are summarised in Figures 2 and 3 and below in Tables 4 and 5.

Table 3a.– Risk assessment of non-native rats and American mink and the effectiveness of biosecurity measures at island seabird colonies within UK Special Protection Areas where they are not currently present.

MSFD Sub-region	Country	Special Protection Area name	Approximate size of islands (hectars)	Key islands within SPA	Total number of islands within SPA
Celtic Seas	England	Isles of Scilly	170	Annet, Gugh, St Agnes
Celtic Seas	Northern Ireland	Carlingford Lough	8	Green Island	2
Celtic Seas	Northern Ireland	Copeland Islands	156	Copeland, Lighthouse	7
Celtic Seas	Northern Ireland	Larne Lough	0.75	Swan, Blue Circle	2
Celtic Seas	Scotland	Ailsa Craig	90	Ailsa Craig	1
Celtic Seas	Scotland	Canna and Sanday	1327	Canna, Sanday	30
Celtic Seas	Scotland	Flannan Isles	54	Mor, Tighe	27
Celtic Seas	Scotland	Glas Eileanan	1.43	Glas Eileanan	2
Celtic Seas	Scotland	Mingulay and Berneray	1135	Mingulay, Berneray, Pabbay	21
Celtic Seas	Scotland	North Rona and Sula Sgeir	128	Rona, Sule Sgeir	20
Celtic Seas	Scotland	Priest Island	120	Priest	1
Celtic Seas	Scotland	St Kilda	860	Hirta, Dun, Soay, Boreray	76
Celtic Seas	Scotland	Sule Skerry & Sule Stack	8	Sule Skerry, Sule Stack	2
Celtic Seas	Scotland	The Monach Isles	364	Monach, Siolaigh	16
Celtic Seas	Scotland	The Shiant Isles	177	Rough, House (jointed), Mary	7
Celtic Seas	Scotland	Treshnish Isles	107	Lunga, Fladda	16
Celtic Seas	Wales	Glannau Aberdaron & Ynys Enlli / Aberdaron Coast & Bardsey Island	183	Bardsey	3
Celtic Seas	Wales	Grassholm	9	Grassholm	1
Celtic Seas	Wales	Skokholm and Skomer	399	Skokholm, Skomer and Middleholm	7
Celtic Seas	Wales	Ynys Feurig, Cemlyn Bay and The Skerries	12	The Skerries only (other islands are tidal)	4 (including YF, excluding Anglesey)
Celtic Seas	Wales	Ynys Seiriol / Puffin Island	23	Puffin	1
Greater North Sea	England	Coquet Island	8	Coquet	1
Greater North Sea	England	Farne Islands	35	Inner Farne & Wideopens Knoxes, Brownsman & Stable	13
Greater North Sea	Scotland	Auskerry	84	Auskerry	1
Greater North Sea	Scotland	Calf of Eday	217	Calf of Eday	1
Greater North Sea	Scotland	Copinsay	78	Copinsay, Horse of Copinsay	3
Greater North Sea	Scotland	Firth of Forth	75	Isle of May, Craigleith, Bass Rock, Fidra, Inchmickery, Lamb	10
Greater North Sea	Scotland	Mousa	171	Mousa	7
Greater North Sea	Scotland	Noss	344	Noss	14
Greater North Sea	Scotland	Ramna Stacks & Gruney	12	Gruney	15

Table 3b.– Risk assessment of non-native rats and American mink and the effectiveness of biosecurity measures at island seabird colonies within UK Special Protection Areas where they are not currently present, assessment continues in Table 3c.

	Likelihood of invasion		Potential hazard from invasion		Biosecurity assessment		Actual likelihood of incursion given biosecurity
Special Protection Area name	by rats	by ferrets and mink	likely impact	likelihood of invasion/ difficulty of removal	Potential risk from invasive mammals	Current quarantine measures	by rats	by ferrets and mink
Isles of Scilly	High	Low	High	Medium	High	Sufficient	Medium	Low
Carlingford Lough	High	High	High	High	High	None	High	High
Copeland Islands	Low	Medium	High	Medium	High	None	Low	Medium
Larne Lough	Medium	High	High	Medium	High	None	Medium	High
Ailsa Craig	Low	Low	High	Medium	Medium	None	Low	Low
Canna and Sanday	High	Medium	High	High	High	Below standard	Medium	Low
Flannan Isles	Low	Low	High	Medium	Medium	None	Low	Low
Glas Eileanan	Medium	High	High	Medium	High	Below standard	Low	Medium
Mingulay and Berneray	Low	Medium	High	High	High	None	Low	Medium
North Rona and Sula Sgeir	Low	Low	High	High	Medium	Below standard	Low	Low
Priest Island	Medium	High	High	Medium	Medium	Sufficient	Low	Low
St Kilda	Medium	Medium	High	High	High	Sufficient	Low	Low
Sule Skerry & Sule Stack	Low	Low	High	High	Medium	None	Low	Low
The Monach Isles	Medium	Low	Medium	High	Medium	None	Medium	Low
The Shiant Isles	Medium	Medium	High	High	Medium	Sufficient	Low	Low
Treshnish Isles	Low	Medium	High	High	High	None	Low	Medium
Glannau Aberdaron & Ynys Enlli / Aberdaron Coast & Bardsey Island	Medium	Low	High	Medium	Medium	Below standard	Low	Low
Grassholm	Low	Low	Medium	Low	Low	Sufficient	Low	Low
Skokholm and Skomer	High	High	High	High	High	Below standard	Medium	Medium
Ynys Feurig, Cemlyn Bay and The Skerries	High	High	High	Medium	High	Below standard	Medium	Medium
Ynys Seiriol / Puffin Island	Low	High	Medium	High	High	Below standard	Low	Medium
Coquet Island	High	High	High	Medium	High	Sufficient	Low	Low
Farne Islands	High	High	High	High	High	Sufficient	Low	Low
Auskerry	Low	Low	High	Medium	Medium	None	Low	Low
Calf of Eday	High	High	Medium	High	High	None	High	High
Copinsay	High	High	High	Medium	High	Sufficient	Low	Low
Firth of Forth	High	High	High	Medium	High	Below standard	Medium	Medium
Mousa	Medium	Medium	High	Medium	Medium	Sufficient	Low	Medium
Noss	High	High	High	Medium	High	None	High	High
Ramna Stacks & Gruney	Low	Low	High	Medium	Medium	None	Low	Low

Table 3c.– Risk assessment of non-native rats and American mink and the effectiveness of biosecurity measures at island seabird colonies within UK Special Protection Areas where they are not currently present, continued from Table 3b.

	Biosecurity assessment		Actual hazard from mammal invasion		Actual Risk from invasive  mammals	Has risk been minimised by effective biosecurity
Special Protection Area name	current surveillance measures	current preparedness for rapid response	Actual likelihood of invasion given the surveillance and response measures	likely impact of invasion on resident seabird species
Isles of Scilly	Sufficient	Sufficient	Low	High	Low	Yes
Carlingford Lough	None	None	High	High	High	No
Copeland Islands	None	None	High	High	Medium	No
Larne Lough	Below standard	None	Medium	High	High	No
Ailsa Craig	Below standard	None	Medium	High	Medium	partially
Canna and Sanday	Sufficient	Below standard	Medium	High	Medium	partially
Flannan Isles	Below standard	None	Medium	High	Medium	No
Glas Eileanan	None	Below standard	Low	High	Medium	partially
Mingulay and Berneray	None	None	High	High	High	No
North Rona and Sula Sgeir	Below standard	None	High	High	Medium	No
Priest Island	Sufficient	Sufficient	Low	High	Low	Yes
St Kilda	Below standard	Sufficient	Medium	High	Medium	partially
Sule Skerry & Sule Stack	None	None	High	High	Medium	No
The Monach Isles	None	None	Medium	Medium	Medium	No
The Shiant Isles	Sufficient	Sufficient	Low	High	Low	Yes
Treshnish Isles	Below standard	None	High	High	High	No
Glannau Aberdaron & Ynys Enlli / Aberdaron Coast & Bardsey Island	Below standard	Below standard	Medium	High	Low	partially
Grassholm	Sufficient	Below standard	Low	Medium	Low	partially
Skokholm and Skomer	Below standard	Below standard	Medium	High	Medium	partially
Ynys Feurig, Cemlyn Bay and The Skerries	Below standard	Below standard	Medium	High	Medium	partially
Ynys Seiriol / Puffin Island	None	None	Medium	Medium	Medium	No
Coquet Island	Sufficient	Sufficient	Low	High	Low	Yes
Farne Islands	Sufficient	Sufficient	Low	High	Low	Yes
Auskerry	None	None	Medium	High	Medium	No
Calf of Eday	None	None	High	Medium	High	No
Copinsay	Sufficient	Sufficient	Low	High	Low	Yes
Firth of Forth	Below standard	None	Medium	High	Medium	partially
Mousa	Below standard	None	Medium	High	Medium	partially
Noss	Below standard	None	Medium	High	High	No
Ramna Stacks & Gruney	None	None	Medium	High	Medium	No

Table 4. Celtic Seas: summary assessment of the effectiveness of biosecurity measures at reducing risk from invasive predatory mammals at island seabird colonies within UK Special Protection Areas where they are not currently present. Notes: ¹Isles of Scilly – refers only to Annet (successful response to rat incursion in 2004) and St Agnes and Gugh, where rats were eradicated in 2013 to 14; ²Ailsa Craig – rat eradication in 1991; ³Canna & Sanday - rat eradication in 2005; ⁴The Shiant Isles – rat eradication in 2015 to 16; ⁵Puffin Island - rat eradication in 1992.

Country	Special Protection Area (SPA) name	Risk from invasive mammals		Has the risk been minimised by biosecurity
		Potential (without current biosecurity)	Actual (with current biosecurity)
England	Isles of Scilly1	High	Low	Yes
Northern Ireland	Carlingford Lough	High	High	No
Northern Ireland	Copeland Islands	High	Medium	No
Northern Ireland	Larne Lough	High	High	No
Scotland	Ailsa Craig2	Medium	Medium	Partially
Scotland	Canna and Sanday3	High	Medium	Partially
Scotland	Flannan Isles	Medium	Medium	No
Scotland	Glas Eileanan	High	Medium	Partially
Scotland	Mingulay and Berneray	High	High	No
Scotland	North Rona and Sula Sgeir	Medium	Medium	No
Scotland	Priest Island	Medium	Low	Yes
Scotland	St Kilda	High	Medium	Partially
Scotland	Sule Skerry & Sule Stack	Medium	Medium	No
Scotland	The Monach Isles	Medium	Medium	No
Scotland	The Shiant Isles4	Medium	Low	Yes
Scotland	Treshnish Isles	High	High	No
Wales	Bardsey Island/Ynys Enlli (part of (Glannau Aberdaron & Ynys Enlli / Aberdaron Coast & Bardsey Island SPA)	Medium	Low	Partially
Wales	Grassholm	Low	Low	Partially
Wales	Skokholm and Skomer	High	Medium	Partially
Wales	The Skerries (part of Ynys Feurig, Cemlyn Bay and The Skerries SPA)	High	Medium	Partially
Wales	Ynys Seiriol / Puffin Island5	High	Medium	No

Table 5. Greater North Sea: summary assessment of the effectiveness of biosecurity measures at reducing risk from invasive predatory mammals at island seabird colonies within UK Special Protection Areas where they are not currently present. Notes: ⁶Firth of Forth Islands – includes Isle of May, Bass Rock, Inchmickery, Craigleith, Lamb and Fidra. All are rat-free.

Country	Special Protection Area (SPA) name	Risk from invasive mammals		Has risk been minimised by biosecurity
		Potential (without current biosecurity)	Actual (with current biosecurity)
England	Coquet Island	High	Low	Yes
England	Farne Islands	High	Low	Yes
Scotland	Auskerry	Medium	Medium	No
Scotland	Calf of Eday	High	High	No
Scotland	Copinsay	High	Low	Yes
Scotland	Firth of Forth6	High	Medium	Partially
Scotland	Mousa	Medium	Medium	Partially
Scotland	Noss	High	High	No
Scotland	Ramna Stacks & Gruney	Medium	Medium	No

Rats have been eradicated from some of the islands included in Tables 3, 4 and 5 – a review of these eradication programmes is provided in Stanbury and others (2017) and Ratcliffe and others, (2009). These included a long-running eradication programme on the Isles of Scilly. In 2013, the islands of St Agnes and Gugh were declared rat-free (but work is ongoing to prevent re-incursions of rats at other islands in the archipelago. Most recently, in 2015 work began to eradicate black rats from the Shiants Islands in the Western Isles. So far evidence suggests the eradication on the Shiants may have been successful. There have also been schemes to eradicate and control the spread of American mink – on Lewis and Harris in the Western Isles (and in and around the sea lochs of Argyll and Lochaber) (Craik, 1997; 1998).

Table 6 shows the results of the step-by-step process to assess the biosecurity in place on the 12 Special Protection Areas where high-impact invasive mammals are currently present (listed below in Table 7 and shown in Figure 2). The results of the assessment are summarised in Table 7 below. A full risk assessment (as shown in Table 3) was not carried out since the risk has already been realised on these islands which are already impacted by invasive mammals. The assessment of biosecurity on these sites is still important, since the invasion by new species, may impact further on the resident bird communities.

Table 6a. Risk assessment of non-native rats and American mink and the effectiveness of biosecurity measures at island seabird colonies within UK Special Protection Areas: site information.

MSFD Sub-region	Country	Special Protection Area (SPA) name	Approx size (ha) of islands	Key islands within SPA	Total number of islands within SPA	High impact mammals present
Celtic Seas	England	Isles of Scilly	817	Tresco, St Martin’s, Bryher, Samson, St Helen’s, Northwethel, Men-a-vaur, Tean, Merrick, Round.		brown rat, feral cats
Celtic Seas	Northern Ireland	Rathlin Island	1439	Rathlin	5	Brown rat, ferret, possibly pet cats
Celtic Seas	Scotland	Handa	330	Handa	7	Brown rat
Celtic Seas	Scotland	Rum	10727	Rum	16	Brown rat, feral goat, neutered pet cats (restricted to inhabited zone)
Greater North Sea	Scotland	Hoy	9500	Hoy	3	Brown rat, hedgehog, feral cat, Stoat
Greater North Sea	Scotland	Rousay	4697	Rousay	1	Brown rats, feral cat
Greater North Sea	Scotland	Pentland Firth Islands	129	Swona, Muckle Skerry	3	Possibly Brown rat
Greater North Sea	Scotland	Fair Isle	798	Fair Isle	176	Cats
Greater North Sea	Scotland	Fetlar	4652	Fetlar	61	Cat, hedgehog
Greater North Sea	Scotland	Foula	1305	Foula	62	hedgehog, feral cat
Greater North Sea	Scotland	Papa Stour	569	Papa Stour	2	feral cat assumed present.
Greater North Sea	Scotland	Papa Westray (North Hill and Holm)	884	Papa Westray, Holm of Papa	2	rodents, cats (domestic, possibly feral),

Table 6b. Risk assessment of non-native rats and American mink and the effectiveness of biosecurity measures at island seabird colonies within UK Special Protection Areas.

	Risk of further incursion		Biosecurity Assessment
Special Protection Area (SPA) name	rats	mink/stoats/ferrets	Current biosecurity measures	Current surveillance measures	Current preparedness for rapid response	Has risk of further incursion been minimised by effective biosecurity?
Isles of Scilly	rats	mink/stoats/ferrets	Current biosecurity measures	Current surveillance measures	Current preparedness for rapid response	Has risk of further incursion been minimised by effective  biosecurity
Rathlin Island	High	Low	Below standard	Below standard	Below standard	partially
Handa		High	None	None	None	No
Rum		High	Below standard	Below standard	Below standard	partially
Hoy		Low	None	None	None	No
Rousay		Low	None	Below standard	Below standard	partially
Pentland Firth Islands		High	None	None	Below standard	No
Fair Isle	High	Medium	None	None	None	No
Fetlar	High	Low	None	None	None	No
Foula	High	High	None	None	Below standard	No
Papa Stour	Medium	High	None	None	None	No
Papa Westray (North Hill and Holm)	High	High	None	None	None	No

Table 7. Island seabird colonies within UK Special Protection Areas that have invasive predatory mammals. Isles of Scilly – refers only to Tresco, St Martin’s, Bryher, Samson, St Helen’s, Northwethel, Men-a-vaur, Tean, Merrick and Round Island. Some contain brown rats or feral cats.

MSFD sub-region	Country	Special Protection Area (SPA) name	Predatory mammals present	Has risk of further incursion been minimised by effective biosecurity
Celtic Seas	England	Isles of Scilly*	Brown rat, feral cat	Partially
Celtic Seas	Northern Ireland	Rathlin Island	Brown rat, ferret, possibly pet cats	No
Celtic Seas	Scotland	Handa	Brown rat	Partially
Celtic Seas	Scotland	Rum	Brown rat, feral goat, neutered pet cats (restricted to inhabited zone)	No
Greater North Sea	Scotland	Hoy	Brown rat, hedgehog, feral cat, stoat	Partially
Greater North Sea	Scotland	Rousay	Brown rat, feral cat	Partially
Greater North Sea	Scotland	Pentland Firth Islands	Possibly brown rat	No
Greater North Sea	Scotland	Fair Isle	Feral cat	No
Greater North Sea	Scotland	Fetlar	Feral cat, hedgehog	Partially
Greater North Sea	Scotland	Foula	hedgehog, feral cat	No
Greater North Sea	Scotland	Papa Stour	feral cat assumed present.	No
Greater North Sea	Scotland	Papa Westray (North Hill and Holm)	Possibly brown rats, cats (domestic, possibly feral),	No

Conclusions

The UK target of a reduction in the risks to island seabird colonies from non-native mammals could not be assessed because there has been no previous assessment of this risk. This assessment also included invasive native species of predatory mammals as well as non-natives. It focussed on all the 42 Special Protection Areas that were designated to protect the UK’s most important seabird colonies on offshore islands. These sites receive the highest level of statutory protection. Despite this, invasive predatory mammals (brown rats, domestic and feral cats, and hedgehog) were present at 12 Special Protection Areas  The full extent of the impact of these mammals at these sites is currently unknown. There is a medium to high risk of other invasive mammal species arriving at these sites and becoming established because biosecurity measures are either absent or ineffective.

High-impact invasive mammals were absent from 30 Special Protection Areas and absent from some Special Protection Areas on islands in the Isles of Scilly. The risk from high-risk mammals has been minimised by effective biosecurity at only six of these sites (including Isles of Scilly Special Protection Areas) Risk has been partially minimised by current biosecurity at a further ten sites. However, these existing measures need enhancing to meet the standards set in the UK Rodent Eradication Best Practice Toolkit (Thomas and others, 2017).

This assessment was conducted with moderate confidence: confidence in data availability was high, but the methods used in this risk assessment are new.

Future assessments of this indicator must improve on the current assessment to meet the UK target of reducing risks to island seabird colonies from invasive predatory mammals. This assessment supports the conclusions of the UK Marine Strategy Part Three (HM Government, 2015). A UK-wide programme of biosecurity at island seabird colonies will help to achieve this target in the future.

Further information

Most of the Special Protection Areas that require improvements in biosecurity are in Scotland. This is because the majority of seabird island Special Protection Areas are in Scotland – 30 out of 42. Of these Scottish Special Protection Areas ten already have high-risk mammals on them, and biosecurity is currently insufficient to minimise the risk of further invasion by predatory mammals. Management options at these sites could include both mammal eradication and improvements in biosecurity. The risk of invasion at the other 19 Special Protection Areas in Scotland, is currently minimised by effective biosecurity at only three and partially minimised at six sites. Improvements in biosecurity are therefore required at up to 16 Special Protection Areas in Scotland to increase the probability of maintaining them free of invasive predatory mammals.

In Northern Ireland, current biosecurity was assessed as ineffective on all five Special Protection Areas and Rathlin island Special Protection Area already has rats and ferrets. Improvements in biosecurity at Northern Ireland Special Protection Area would contribute to meeting the UK target in the future.

In England, improvements in biosecurity would only be required on some islands in the Isles of Scilly Special Protection Area that have rats or feral cats on them. Elsewhere, current biosecurity was assessed as effective on rat-free islands in Special Protection Areas of the Isles of Scilly, Coquet Island and the Farne Islands.

In Wales, no predatory mammals were present on any of the five seabird island Special Protection Areas that were assessed. Improvements in biosecurity would be required on all sites to minimise risk, but current measures were considered partially effective at four sites.

Knowledge gaps

This assessment clearly identifies where biosecurity needs to be improved to meet the UK target in the future. However, it is not clear how the risk will be reduced on the 12 Special Protection Areas where invasive predatory mammals are already present (Figure 2). A better understanding of the extent of the impacts on seabirds and other ground-nesting birds in these Special Protection Areas will identify the measures needed to reduce these impacts.

Further information

It will not be easy to fully understand the impacts on seabirds at Special Protection Area where invasive predatory mammals, particularly rats, are already present (Figure 1). If rats have been present for many years, a stable state of coexistence may have been established, so that impacts are not that visible. Following the invasion by rats, the seabirds nesting on the islands will have moved to areas that are inaccessible to rats, while some species, such as European storm-petrels, may have deserted the islands completely. There is unlikely to be any visible evidence of rat predation such as predated eggs or chicks on these islands, yet the impacts of rats persist.

Rats are absent from four Special Protection Area where feral cats are present: Foula, Fetlar, Fair Isle, and Papa Stour. Predation by cats is usually more obvious than the impacts of rats including through the presence of predated bird carcasses. However, the extent of impacts by cats on these four Special Protection Area is currently unknown.

References

Birkhead TR, Furness, RW (1985) ‘Regulation of seabird populations’ in: Behavioural ecology: ecological consequences of adaptive behaviour (editors: Sibley, RM, Smith, RH), pages 145-167. Blackwell, London.

Courchamp F, Chapuis J-L, Pascal M (2003) ‘Mammal invaders on islands: impact control and control impact’ Biological Reviews, 78:347-383 (viewed on 24 October 2018)

Craik, JCA (1998) ‘Recent mink-related declines of gulls and terns in west Scotland and the beneficial effects of mink control’ Argyll Bird Report, 14: 98-110 (viewed on 24 October 2018)

Dunstone, N (1993) 'The Mink' T & AD Poyser, London.

European Commission (2009) ‘Directive 2009/147/EC of the European Parliament and of the Council of 30 November 2009 on the conservation of wild birds’ Official Journal of the European Union L 20, 26.1.2010, pages 7–25 (viewed 24 October 2018)

Ewins PJ, Tasker ML (1985) ‘The breeding distribution of Black Guillemots Cepphus grylle in Orkney and Shetland’ Bird Study 32:186-193 (viewed on 24 October 2018)

HM Government (2012) ‘Marine Strategy Part One: UK Initial Assessment and Good Environmental Status’ (viewed on 5 July 2018)

 

HM Government (2015) ‘Marine Strategy Part Three: UK Programme of Measures’ December 2015. (viewed on 5 July 2018)

Jones HP, Tershy BR, Zavaleta ES, Croll DA, Keitt BS, Finkelstein ME, Howald GR (2008) ‘Severity of the effects of invasive rats on seabirds: a global review’ Conservation Biology 22: 16-26 (viewed on 24 October 2018)

Lock J (2006) ‘Eradication of brown rats Rattus norvegicus and black rats Rattus rattus to restore seabird populations on Lundy Island, Devon, England’ Conservation Evidence 3: -111-113 (viewed on 24 October 2018)

Mitchell PI (2004) ‘Leach’s Petrel Oceanodroma leucorhoa’ Pages 101-114 in: Mitchell PI, Newton SF, Ratcliffe N, Dunn TE (2004) ‘Seabird Populations of Britain and Ireland’ T & AD Poyser, London (viewed on 24 October 2018)

Mitchell PI, Newton SF (2004) ‘European storm-petrel Hydrobates pelagicus’ Pages 81-100 in: Mitchell PI, Newton SF, Ratcliffe N, Dunn TE (2004) ‘Seabird Populations of Britain and Ireland’ T & AD Poyser, London (viewed on 24 October 2018)

Mitchell PI, Ratcliffe, N (2007) ‘Abundance & distribution of seabirds on UK islands – the impact of invasive mammals’ in Proceedings of the conference on Tackling the problem of invasive alien mammals on seabird colonies – Strategic approaches and practical experience. Edinburgh 2007. The National Trust for Scotland, Royal Zoological Society of Scotland and Central Science Laboratory.

Non-native Species Secretariat (2012) GB Non-native Species Secretariat Factsheet on Rattus norvegicus, Brown Rat.

Ratcliffe N, Mitchell I, Varnham K, Verboven N, Higson P, (2009) ‘How to prioritize rat management for the benefit of petrels: a case study of the UK, Channel Islands and Isle of Man’ International Journal of Avian Science, 151: 699-708 (viewed on 24 October 2018)

Russell JC, Clout MN (2005) ‘Rodent incursions on New Zealand islands’ in Parkes, J., Statham, M. & Edwards, G. (editors) Proceedings of the 13th Australasian Vertebrate Pest Conference Lincoln, New Zealand 2-6 May, 2005. Landcare Research pages 324–330 (viewed on 24 October 2018)

Stanbury A, Thomas S, Aegerter J, Brown A, Bullock D, Eaton M, Lock L, Luxmoore R, Roy S, Whitaker S, Oppel S (2017) ‘Prioritising islands in the United Kingdom and crown dependencies for the eradication of invasive alien vertebrates and rodent biosecurity’ European Journal of Wildlife Research, 63:31 (viewed on 24 October 2018)

Thomas S, Bambini L, Varnham K (2016) ‘Review of biosecurity on Seabird SPAs in the UK’ Technical Report, LIFE13 NAT/UK/000209 Shiants project, Royal Society for the Protection of Birds.

Thomas S, Varnham K, Havery S (2017) ‘UK Rodent Eradication Best Practice Toolkit (Version 4.0)’ Royal Society for the Protection of Birds, Sandy, Bedfordshire (viewed on 24 October 2018)

Towns DR, Atkinson IAE, Daugherty CH (2006) ‘Have the harmful effects of introduced rats on islands been exaggerated?’ Biological Invasions 8: 863–891 (viewed on 24 October2018)

Acknowledgements

Assessment Metadata

Please contact marinestrategy@defra.gov.uk for metadata information

Recommended reference for this indicator assessment

Ian Mitchell¹, Sophie Thomas⁶, Laura Bambini⁶, Karen Varnham, K.⁶, Phillips, R.¹, Gemma Singleton¹, Andrew Douse,² Simon Foster,² Melanie Kershaw,³ Neil McCulloch⁴, Matty Murphy⁵, & Jane Hawkridge¹ 2018. Invasive mammal presence on island seabird colonies. UK Marine Online Assessment Tool, available at: https://moat.cefas.co.uk/biodiversity-food-webs-and-marine-protected-areas/birds/invasive-mammals/

¹Joint Nature Conservation Committee

²Scottish Natural Heritage

³Natural England

⁴Department of Environment, Agriculture & Rural Affairs, Northern Ireland

⁵Natural Resources Wales

⁶Royal Society for the Protection of Birds