Thursday, 12 April 2007

Water vole survival and the role of mink control
Lessons from the River Chess

In ECOS 28 (1), Catherine Shellswell and Allan Beechey reported on work along the River Chess in the Chilterns to provide habitat management advice and mink control for riparian landowners, in order to counteract a decline in the water vole population.

Here, in unabridged form, Trevor Lawson, author of an earlier article on the same subject*, challenges the authors (in black text below), and Shellswell and Beechey reply (in blue text).

1. Trevor Lawson:
The authors comment on their provision of fencing to provide "buffer zones" for water voles and other wildlife, yet the linear extent of these buffer zones is not quantified. They are very limited. Criticisms about the wholly inadequate depth of the buffer zones are not addressed. Further, the authors make no attempt to distinguish between the contribution that the alleged buffer zones made to water vole recovery, compared to the benefits delivered by mink culling. It is, of course, impossible to distinguish between the benefits of habitat and culling using the methodology adopted in this study (and adopted in other Environment Agency-supported projects). The Chilterns Conservation Board was warned of this weakness in the methodology prior to initiating culling and strongly advised to address it, but ignored the advice.

Catharine Shellswell & Allan Beechey reply:
Whilst in an ideal world the desirable buffer zone width would be 6m, in practice a pragmatic approach must be taken to negotiate a buffer zone. In the majority of cases the landowner will be unwilling to lose a 6m deep section of his/her land. In this instance a fence was erected along a 500m stretch of river 1m away from the edge of the watercourse. The adjacent field downstream had already been fenced by the landowner. The first two metres are the most important for water voles and the fence has stopped horses and cattle entering the river enabling lush marginal vegetation to develop. The decreased disturbance has also benefited the water voles. Placement of this fencing has clearly enabled the water vole population to increase and it is hoped that the area of bank fenced from the field could be extended in the future. Currently there is a project being undertaken by WildCRU investigating water vole population growth and individual vole growth compared to amount of bankside habitat. This type of project has not been undertaken before and hopefully it will provide weight to the provision of more bankside habitat for water voles.

2. Trevor Lawson:
One landowner allegedly trapped nine mink in 2003-4, prior to the culling scheme. None were made available to WildCRU and no records are available viz the age, sex or health of this sample. Yet a trout fishery less than two miles downstream trapped no mink. Polecats have returned to the Chilterns since 1997. What attempt was made to validate this sample?

Catharine Shellswell & Allan Beechey reply:
The eight mink trapped by the landowner were caught in a trap that he had been running for a number of years. We know that they were a female and young. Unfortunately we do not have any other information on the mink as they were caught prior to the start of the trapping scheme organised by BBOWT and the Chiltern Chalk Streams Project, and therefore did not fall under the remit of providing them to WildCRU. At the time when the first mink was caught (September 2003) the young mink would have just started to disperse away from their mother, but would still stay fairly close to the natal territory (which is approximately 3km in length). This would have restricted the range of the juvenile mink and they may not have ventured downstream to the fishery. There is also the possibility that the mink dispersed across land rather than along the watercourse (males tend to live away from the watercourse for the majority of the year and come back to breed with females in autumn, winter and early spring). The ninth mink was killed on a nearby road in March 2004 and it is believed that this was the last mink to be caught from this group. It was very difficult to determine the gender of this individual once it had been flattened and it was also considered unsuitable for WildCRU to autopsy. The two mink sent to WildCRU were incorporated into their mink and polecat project which has been running since 2003.

3. Trevor Lawson:
A 2003 water vole survey found that the voles had "dramatically" declined since 2001. Yet their disappearance was reported from sites where I continued to observe them.

Catharine Shellswell & Allan Beechey reply:
Unfortunately, it is very hard to prove a negative survey as there is always an element of observer error. For example, surveys carried out from the bankside may miss signs close to the water edge (this is only undertaken where the watercourse is too deep, and once the vegetation has grown field signs may have been missed. This is stated in the survey results. BBOWT collects sighting reports from people who have seen water voles, and if these occur along stretches of watercourse which haven’t been surveyed every effort is made to survey these areas to confirm the presence of water voles. Water voles can also live at low population densities, and this reduces the number of field signs that they leave. Might I suggest that Mr Lawson sends in his water vole sighting to the Water Vole Recovery Project (please see the online record form at for details) as we would be interested in any information that he could provide.

4. Trevor Lawson:
The subsequent mink culling scheme initiated in February 2004 along virtually the entire catchment, only removed two mink, even though the project had concluded that mink had dispersed along the river and were breeding.

Catharine Shellswell & Allan Beechey reply:
Mink were breeding along the catchment at the time of survey (August-September 2003) near to Bois Mill where eight mink were caught and one killed on the road (as mentioned earlier). The placement of mink rafts allowed the watercourse to be monitored for mink (using a footprint tracking pad) and once footprints were identified a live capture trap was placed on the raft and the mink removed accordingly (these were the two caught in march 2004). In actual fact without any form of mink control eleven mink would have been able to disperse along the river. The mink rafts are primarily a monitoring device that can be used as a trap once mink have been identified in the vicinity.

5. Trevor Lawson:
Mink signs have only been found since at the site where nine mink were allegedly trapped in 2003-4. The authors make no attempt to assess why this site, which is not at the confluence with any other water course or dry valley, and where the Chess is particularly canalised adjacent to a busy road, should be the focal point for mink arrival and dispersal.

Catharine Shellswell and Allan Beechey reply:
The site where the mink were caught is in farmland and is not canalised. It is also fairly wooded at the location where large and small mink footprints were found. It is also near a stocked lake and the confluence of two channels. It is an ideal den site as it is undisturbed.

6. Trevor Lawson:
Water vole signs increased five-fold by 2005 and this improvement is attributed to mink culling. Yet no other variables – in particular the impact of winter flooding - have been factored into this analysis. Much of the river Chess has no suitable banks for water voles to burrow in and where those banks do exist there is a risk of flooding due to the rapid run-off from intensively managed land. Flooding is a known major cause of water vole mortality.

Catharine Shellswell & Allan Beechey reply:
Generally the land alongside the River Chess is not intensively managed. Very little is cultivated and most of the pasture is remnant water meadows. The water meadows are grazed, but on the whole stocking levels remain low as the land does not support a high proportion of animals. Where stocking densities are slightly higher fences have been erected to protect the banks. Therefore, flooding due to runoff from intensively managed farmland along the River Chess is not an issue. Between 2001 and 2003 there were no large flood events that affected the entire valley which would have caused the demise of the water vole population by 98%. There were also no local flood events which would have had a greater impact than normal. Water vole populations can reduce by up to 70% during the winter period and flooding is a cause of death. The lack of flood events in this time period and the extent of the decline points to other factors affecting water voles such as the dispersal of American mink along the river rather than a large flood event.

7. Trevor Lawson:
The authors should have considered other possibilities before drawing their conclusions. For example, that before the 2001 baseline survey, water voles had expanded their range to less optimal parts of the river. Subsequent flooding or another factor (mink are a possibility, of course) then reduced the population to those optimum parts of the river identified in the 2003 study. It is surely no coincidence that these sites also have the very best vegetative habitat and structural diversity.

Catharine Shellswell & Allan Beechey reply:
Water voles follow a metapopulation structure and do experience population rises, where they disperse along the river corridor, and falls, where their range contracts. The water vole population along the River Chess decreased by 98% according to the surveys. The surveys were both carried out at the same time of year; this avoids seasonal population density variation when comparing the two surveys and also points to the decline being extreme between 2001 and 2003. All three surveys were carried out using the standard methodology for surveying water voles (described in the water vole conservation handbook). It would be beneficial if we had information about the water vole population prior to this date to assess whether the water vole population in 2001 was unusually high, unfortunately this is not the case. As stated earlier, flooding had already been discounted as a cause of the decline.

The mink trapped in March 2004 were caught between the two remaining colonies of water voles, and if the female had stayed in this area (which is a high possibility as mink were seen entering a good den site under a nearby barn, and the river at this location is fairly undisturbed), both colonies were within an average territory size of a female. Furthermore, not all the water vole field signs were located along the optimal habitat. Some were located in back gardens which had close mowing regimes. BBOWT and the Chiltern Chalk Streams Project have been providing advice to landowners to help them encourage water voles through sympathetic management of the bankside. This includes leaving areas of uncut vegetation along the river which provides cover from predators.

8. Trevor Lawson:
The authors state: "Mink control should always be implemented where habitat advice is also provided to landowners." Many landowners receive habitat advice but it is clearly impractical for mink control to be implemented in every instance. Perhaps the authors actually mean: "Mink control should only be implemented if habitat advice is also provided to landowners." This, too, is unwise. Habitat advice is quite different from actual habitat creation or restoration. Advice alone will not contribute to water vole recovery.

Catharine Shellswell & Allan Beechey reply:
We stand corrected in that mink control should always be implemented (where practical) where water voles are found to be present and habitat advice should always be provided to landowners. Mink control is not always practical for example in areas with high human activity as they tend to disturb mink traps and rafts. However, the rafts are placed at 1km intervals (optimum density according to the Game Conservancy Trust) and usually these areas can be avoided. Mink were found to be present during the 2003 water vole survey and are considered the larger threat to water voles in Berks, Bucks and Oxon. The water vole colonies have shrunk to such a degree that there is a great amount of habitat available for inhabitation by water voles, but their population is controlled by mink. This means that in most cases a monitoring scheme to investigate the presence of mink is implemented at the same time that habitat advice is provided and mink are removed (resources and landowners allowing) once their presence has been found. Where areas of unsuitable habitat provide a barrier to dispersal (such as heavily scrubbed sections of watercourse over 1-2km in length or artificial banks) habitat advice is provided and where funding allows habitat enhancement or restoration is undertaken. Farmers are encouraged into agri-environment schemes, homeowners are encouraged to leave uncut strips and mitigation is provided as part of planning applications. Habitat management is carried out to create a mosaic of habitats along a watercourse which would benefit a range of species rather then just one, and it is not suitable to make every stretch of watercourse ideally suited to water voles, otherwise you would be destroying the optimum habitat for otters. There is no legislation to enforce provision of habitat for water voles, only to protect water vole burrows, and therefore we provide advice to landowners where they can be persuaded to carry forward action. This may be provided through agri-environment schemes or other local grant bodies interested in water vole conservation. As there is a large amount of habitat available along the River Chess, it was felt that habitat was not the influencing factor in the water vole decline and since mink were found during the water vole survey it was determined that resources were best placed in a monitoring and trapping programme along the watercourse.

9. Trevor Lawson:
Loss and fragmentation of habitat is the primary cause of water vole decline and impacts on a wider range of species, too. Had this project been planned strategically, with rigorous scientific methodology, with the needs of other wildlife in mind and for the long term, the authors would state: "Mink control should only be considered when habitat restoration has been successfully implemented but has failed to reverse the decline in water voles."

Catharine Shellswell & Allan Beechey reply:
The statement ‘loss and fragmentation of habitat is the primary cause of water vole decline’ can not be verified scientifically. It is currently considered that habitat loss and the dispersal of American Mink throughout the countryside are jointly considered to have caused the decline (see the Water Vole UK BAP). Habitat loss has caused the decline over the greater part of the 20th Century, but the decline was particularly severe after the Second World War when the government encouraged increased agricultural production and during the 1970s-80s due to increased urbanisation and installation of flood defences. During the 1980s it was realized that American mink were spreading throughout Britain (as well as Europe where they threaten the European mink) and the water vole population appeared to be declining in their wake. There are numerous examples of mink predating water voles to extinction through radio-telemetry studies and mink scat analysis. It is unlikely that we would still be losing water voles without mink as habitat is not sufficiently poor in many areas to explain their more recent demise. The biggest single barrier to their recovery would seem to be a non-native predator which has an almost unique ability to exploit this species. Concentrating on habitat to the exclusion of mink will unfortunately not stop the decline of the water vole. Implementation of mink control is not taken lightly and all factors are considered prior to any mink rafts and traps being deployed. Certainly in regard to the River Chess the mink control has stopped the demise of the water vole. If the two mink were not caught in March between the remaining water vole colonies, it would probably be another sad story where water voles have become extinct. The mink rafts now provide a valuable monitoring tool for both ad hoc water vole presence and also to provide an early warning system of any mink that may disperse back to the river. The increase in the water vole population between 2003 and 2005 shows that mink control and habitat enhancement has been successful at the moment and we hope that the water vole population will continue to increase. The survey will be repeated next year (with the landowners permission) to assess the range of the water vole population.

* Lawson. T (2005) 'Seen a mink? Unleash the dogma!', ECOS 26 (3/4) pp81-85.