ournal of Applied Ecology, 91 - PDF document

ournal of Applied Ecology, 91Ð98Ecological SocietyBlackwell Publishing, Ltd.Indirect negative impacts of radio-collaring: sex ratio ariation in water volesTOM P. MOORHOUSE and DAVID W. MACDONALD W. MacdonaldEcological Society, ournal of Applied Ecology 91Ð98ter voles Arvicola terrestris(Lacepede) areendangered in the UK (Woodroffe, Lawton & Davidson1990; Macdonald & Strachan 1999; Macdonald &Harrington 2003). We therefore set out to study thepopulation dynamics, dispersion and movement pat-terns of water voles in two contrasting populations togain information relevant to their conservation andpreservation. We employed radio-tracking amongstmethods.Several studies of rodent populations, including wateroles (Leuze 1976, 1980), have reported no direct impactsof radio-collaring (Smith 1980; Ormiston 1985; Reynolds1992; . 1996; Johannesen, Andreassen &study, however,observed a rapid decline in the number of present in one of our populations. This led us to con-sider the possibility that radio-tracking inßuenced theumber of females at this site, potentially by affectingnatal sex ratios.Sex ratio variation is commonly observed as aconsequence of varying survival and value of offspringof different sexes. Evolutionary theory suggests thatparental investment may favour the sex expected toproduce the most grandchildren per unit investment,here this investment inßuences the Þtness of one sexmore than the other.emale condition is an important determinant of natalsex ratios. The local resource competition hypothesispredicts that mothers with access to poor resourcesshould produce offspring of the sex that is most likelyto disperse, in order to reduce competition for resourcesin the natal range (Clark 1978; Silk 1983). Similarly,ers & Willard (1973) proposed that the optimaleproductive strategy for a mother is to produce off-spring of the sex that can beneÞt most from the level ofesources that she can provide. Both models connectfspring sex ratio with the availability of food resources,xpressed through maternal condition.There is evidence of natal sex ratio variation amongter voles. Bazhan, Makarova & Yakovleva (1996)showed that food deprivation during pregnancy almosthalved the number of weaned young produced by cap-tive female water voles, and litter sex ratios were skewedin favour of males by� 2 : 1. They also demonstratedincreased levels of corticosterone in food-deprivedfemales, suggesting a stress response. Nazarova & Evsikov2000) demonstrated that captive female water volesgaining relatively little weight during pregnancy weanedsmaller, male-biased litters compared with females thatgained more weight.Our aim was to identify the causes of a decline inumbers of female water voles, which was observed ina population that we were radio-tracking, by analysingsurvival rates, recruitment rates and offspring sex ratios.Our analysis revealed that the most likely cause for thefemale decline was a shift in the sex ratio of young raisedadio-collared females. This result has implicationsor conservation research, especially for the monitoringof water vole populations.


The study was conducted at two sites, the Bure MarshesNational Nature Reserve (NNR), Norfolk, UK, and theLittle Bedwyn stretch of the Kennet and Avon Canal,iltshire, UK.The Bure Marshes NNR (UK grid reference TG335 165) is a fenland habitat comprising a network ofditches in which water voles hold territories. Within thestudy area, we deployed live traps (Sherman XLK; 30 cm; H.B. Sherman Traps, Inc., Tallahassee,FL, USA) along 4á07 km of ditch edge. The populationas open to immigration and emigration in that occupiedditches were accessible from three sides of the site. Thissite was trapped from April to September in 2000 and2002 and during April 2003. Public health restrictions,wing to an outbreak of fdisease, pre-ented us from trapping between April 2001 and June 2001.The Little Bedwyn site (UK grid reference SU295 665) comprises a length of canal running parallelto the River Dun, separated from it by a mown Þeldarying between 20 and 60 m in width. Live traps weredeployed along 1á04 km of the canal and 0á96 km of theriver, making a total trapped length of 2á0 km. The siteas linked to further patches of suitable habitat via theriver, and therefore open to dispersal. The Little Bedwynsite was trapped from June 2002 until July 2003. -    ps were positioned 20 m apart and their positionsecorded with a 10-Þgure grid reference using a globalpositioning system (GPS Garmin ÔEtrexÕ, Garmin Ltd,omsey, Hants, UK) to a maximum error of 10 m. Trapsbaited with apple and contained approximately 150 gof carrot as an overnight food source (Efford 1985).During 2000, half of the Bure Marshes site was trappedin alternate months. During 2001 and 2002, the entiresite was trapped each month. Traps were open for fourconsecutive trap nights per trapping session in 2000and 2001, and for Þve consecutive trap nights in 2002.The entire Little Bedwyn site was trapped with one ses-sion per month of Þve trap nights during 2002.    The radio-collars (TW-4 transmitter, Ag386 cell; BiotrackLtd, Wareham, Dorset, UK) weighed 4á5 g and werettached only to water voles weighing more than 180 g,epresenting less than 2á5% of the individual bodyeight. The collars were of a type commonly used forsmall transmitter and battery imbedded in dental acrylicand fastened in place with a cable tie, trimmed to pre-ent post-release tightening. The collars had no sharpedges, all joins between components were smoothed andno signs of neck abrasion were found on any animals W. MacdonaldEcological Society, ournal of Applied Ecology 91Ð98(see the Results). The third assumption is reasonable ifjuvenile female survival rates are similar to those ofadults after entering the trapped population. No dataxist to support or refute this assumption.hen calculating the actual proportion of the breedingfemale population that was collared, females weighingmore than 140 g were assumed to be of breeding weight;ollowing Stoddart (1971), nearly all captured femalese this weight had perforate vaginas.How the exact proportion of collared voles in eachcapture session would affect the recruitment for the nextonth was unknown. The collared proportion was there-eraged across all capture sessions, simplifyingthe model and removing this source of uncertainty.
    
 Nine females slipped collars or died within a week ofbeing collared. Of these, Þve were deÞnitely taken bypredators, three were probably preyed upon (they werenever recaptured) and one slipped its collar, remainingalive. Seven male voles slipped collars or died within aeek of collaring. Of these, four collars were slipped,one individual was deÞnitely taken by predators and twoothers were probably preyed upon. For voles that retaineda collar for more than 1 week, the mean time spentcollared was 69 days for females ( = 9, SD 33, range20Ð119) and 76 days for males ( = 13, SD 45, range 

    hereas the robust design population estimates for bothsexes for the Bure Marshes and Little Bedwyn sitesstable during live trapping-only periods (the beforeand control periods), the numbers of females in theBure Marshes site declined throughout 2002 (pertur-bation), reaching a minimum of Þve in April 2003 (Fig. 1).emale numbers at the Bure Marshes site declinedsteadily during April to September 2002, from 20 to nine,cept for an estimated increase from 13 to 15 from Juneto July. In contrast, between April and September 2002the estimated number of males increased from 28 to 35.
   
Survival rates were higher for females than for males atthe Bure Marshes site from the outset of the pertur-bation period (April 2002 onwards) and higher thanthe survival rates of either sex at Little Bedwyn up tougustÐSeptember (Fig. 2). The decline in numbers offemales at Bure Marshes therefore occurred while theirsurvival rates were the highest of either sex or population.The best Þtting model that estimated survival ratesdid not incorporate emigration. We therefore excludedthe possibility that the female decline at the BureMarshes derived from unacknowledged emigrationbecause any emigration of females would lower thesurvival probabilities. Rather, the observed decline musthave resulted from low recruitment rates of femalesinto the population.   estimates for the number of recruits for anysite or sex in a given month were within three individualsof the number of new animals captured in all cases, suggest-ing that the estimates were a good representationof the data.After AprilÐMay 2002 (the perturbation period), theumber of females recruited to the Bure Marshes siteas consistently lower than during the before periodor during any trapping session at Little Bedwyn (thecontrol site; Fig. 3).At the Bure Marshes site, 51 females and 45 malesecruited during AprilÐSeptember 2000, and 19females and 15 males from June to September 2001 (thebefore period). At Little Bedwyn (the control site), 27females and 46 males were recruited from June toeptember 2002. During the same period at the BureMarshes site (JuneÐSeptember 2002; the perturbation Estimated population sizes for water voles from 2001 to 2003 at Bure Marshes and Little Bedwyn, UK. Solid linesepresent Bure Marshes populations, dotted lines represent Little Bedwyn populations. Triangular points represent males,rhomboid points represent females. Error bars represent SE of population estimate. ater vole and sex ratiosEcological Society, ournal of Applied Ecology 91Ð98tes unaltered but reducing a theoretically stablefemale recruitment rate by the proportion of the breed-ing females that are collared was sufÞcient to accountor the perturbation female decline.interpret these data to suggest that the 2002 declinein female numbers at the Bure Marshes site resultedfrom male-skewed offspring sex ratios mediated by adeterioration in maternal condition in response tothe attachment of radio-collars to breeding females.Evidence for a detrimental inßuence of collaring uponsmall mammals has been shown in a number of studies,including effects upon body mass (Berteaux, Duhamel& Bergeron 1994), activity levels (Berteaux, Duhamel& Bergeron 1994), social interactions (Pouliquen,Leishman & Redhead 1990) and mortality (Wolton &In our study there was mortality in the Þrst weekollowing collaring (Þve conÞrmed and three suspectedcases of predation for females, vs. two and one, respec-tively, for males). For voles surviving the Þrst week, themean time spent collared was 69 days for females and76 days for males. This suggests immediate short-termadverse effects from collar attachment.Leuze (1980) found no effects of radio-collars uponbody weight or fecundity of water voles. LeuzeÕs (1980)collars were of a similar construction to those used inthis study, comprising a cable tie attachment with thecomponents embedded in silicone rubber, but theyeighed slightly more, at a maximum of 6á1 g, as opposedto 4á5 g. Leuze (1980) recorded only numbers of wean-lings seen with the mother, not offspring sex ratios orumbers weaned. Nazarova & Evsikov (2000) showedthat low-weight females not only had male-biased littersut a reduced ability to wean offspring successfully.Therefore, Leuze (1980) could have missed alterationsin sex ratios or fecundity resulting from collaring.Skewed sex ratios have been described in stressed andmalnourished females of different species (McGinley1984; Labov . 1986). Female water voles exhibitdecreased litter sizes, male-biased offspring sex ratiosand stress responses in response to decreased foodailability and body weight (Bazhan, Makarova &vleva 1996; Nazarova & Evsikov 2000).emale water voles may inherit their motherÕs territory(Leuze 1976; Strachan 1998), whereas males usuallydisperse (Leuze 1976). Females inhabiting suboptimalterritories may therefore beneÞt by differentiallyproducing more males (Clark 1978; Silk 1983; Julliard2000). Radio-collars have the potential to cause stressto water voles: it is possible that such stress mightstimulate sex-ratio adjustment as part of an evolutionarymechanism mitigating impacts of suboptimal habitats,similar to the sex-ratio bias and stress response in food-deprived water voles (Bazhan, Makarova & Yakovlevahether the sex ratios in the present study were maleskewed from birth or skewed during and post-weaningbecause of poor female pup survival is unknown.However, in either case the effect would be poor femaleecruitment and male bias in the juveniles entering thetrapped population.
 

  
The implications of this study for further radio-collaringstudies of small mammals are likely to be speciesspeciÞc (Withey, Bloxton & Marzluff 2001), and furtheris required to establish a causal link between col-laring water voles and skew in offspring sex ratios. Thisstudy, however, casts doubt on the assumption that theuse of radio-collars does not fundamentally affect thebiology of collared water voles. This study emphasizesthat the effects of commonplace wildlife marking andtracking techniques may be difÞcult to detect and yetimportant (Amlaner & Macdonald 1980; May 2004).ecommend caution in the selection of the meansused for studying wild populations, especially of raremammals, and, where possible, additional monitoringor negative impacts upon those populations.Acknowledgementsould like to thank the Environment Agency,British Waterways, the PeopleÕs Trust for EndangeredSpecies and English Nature, who all sponsored thisstudy. Many thanks to our collaborators at Bristol Zooand Wildwood, Kent, for measuring weights of known-captive juvenile water voles. We would also like tothank the many volunteers who helped with theÞeldwork during this study. Special thanks to Ian Ellis,becca Dean, Gillian McCoy and Ruth Dalton foreferencesAmlaner, C.J. & Macdonald, D.W. (1980) Biometry and Radio Trackinggamon Press, Oxford, UK.Bazhan, N.M., Makarova, E.N. & Yakovleva, T.V. (1996)Deprivation of food during pregnancy and reproduction inthe water vole (Arvicola terrestrisournal of Mammalogy, 1078Ð1084.Berteaux, D., Duhamel, R. & Bergeron, J.M. (1994) Can radiocollars affect dominance relationships in Microtusdian Journal of Zoology, 785Ð789.Berteaux, D., Masseboeuf, F., Bonzom, J.M., Bergeron, J.M.,Thomas, D.W. & Lapierre, H. (1996) Effect of carrying aadio-collar on expenditure of energy by meadow voles.ournal of Mammalogy, 359Ð363.Burrows, R., Hofer, H. & East, M.L. (1994) Demography,xtinction and intervention in a small population: the caseof the Serengeti wild dogs. Proceedings of the Royal Societyof London BBurrows, R., Hofer, H. & East, M.L. (1995) Populationynamics, intervention and survival in African wild dogsProceedings of the Royal Society of London, 235Ð245.Clark, A.B. (1978) Sex ratio and local resource competition ina prosimial primate. Clobert, J., Wolff, J.O., Nichols, J.D., Danchin, E. & Dhondt,A.A. (2000) Introduction. DispersalE. Danchin, A.A. Dhondt & J.D. Nichols), pp. 1Ð2. OxfordUniversity Press, Oxford.