An evaluation of territory mapping to estimate fisher density

We evaluated winter-territory mapping as a method for estimating fisher (Martes pennanti) density in a 210-km2 survey area in north-central Massachusetts in 1994 and 1995 by comparing estimates with simultaneous camera mark–resight estimates. Assuming intrasexual territoriality and accounting for all occupied habitat, territories of resident radio-marked fishers were mapped (mean = 54% of all territories in the study area), and those of unmarked resident fishers were identified from tracks and photographs. The total number indicated a population of 40 (19/100 km2) and 49 (23/100 km2) residents for 1994 and 1995, respectively. Results from replicated automatic-camera capture–mark– resight surveys suggested slightly higher total numbers and densities of fishers in 1994 (44.5; 21/100 km2) and 1995 (52.9; 25/100 km2), but these estimates likely also included nonresident juveniles. Territory mapping and automatic-camera mark–resight methods resulted in very similar population estimates, but both require large numbers of radio-marked fishers to effectively detect small population changes (e.g., such as the 20% observed in this study). Individually marking animals would enhance mark–recapture estimates. Résumé : Nous avons évalué la cartographie des territoires d’hiver comme méthode d’estimation de la densité des Martres (Martes pennanti) dans une zone d’inventaire de 210 km2 dans le centre nord du Massachussetts en 1994 et 1995 en comparant les valeurs estimées avec des estimations simultanées de marquage–réobservation sur vidéo. En supposant l’existence d’une territorialité intra-sexuelle et en tenant compte de tous les habitats occupés, les territoires des Martres résidantes porteuses d’un émetteur ont été cartographiés (moyenne = 54 % de tous les territoires dans la région étudiée) et ceux des Martres non porteuses ont été identifiées au moyen de pistes et de photographies. Le dénombrement total indique que la population comptait 40 résidants 19/100 km2) en 1994 et 49 (23/100 km2) en 1995. Les résultats obtenus au moyen de caméras automatiques en réplicat et selon la méthode marquage–réobservation indiquent des nombres totaux et des densités légèrement plus élevés tant en 1994 (44,5; 21/100 km2) qu’en 1995 (52,9; 25/100 km2), mais ces estimations comprennent sans doute aussi des juvéniles non résidants. La cartographie du territoire et la méthode marquage–réobservation donnent des estimations semblables, mais les deux méthodes nécessitent de grands nombres de Martres porteuses d’émetteurs pour détecter efficacement les petites fluctuations dans la population (e.g., telle que le 20 % enregistré au cours de cette étude). Le marquage individuel d’animaux améliorerait les estimations obtenues par marquage-recapture. [Traduit par la Rédaction] 1696 Fuller et al. Introduction Ongoing changes in habitat conditions and trapping pressure necessitate the use of reliable monitoring techniques for conservation of carnivores in North America. Fisher (Martes pennanti) populations are potentially vulnerable because of their relatively low density and high fur value (Douglas and Strickland 1987). Methods for monitoring fisher abundance have included analyses of harvest records (Cardoza 1986; Douglas and Strickland 1987), livetrapping (Paragi 1990), the use of sooted track panels (Barrett 1983) and track transects in snow (de Vos 1952; Thompson et al. 1989), and camera surveys (Bull et al. 1992; Raphael 1994). However, accurate estimates of density using these methods are difficult, if not impossible, to obtain (de Vos 1952; Douglas and Strickland 1987), and none have been verified independently. Radiotelemetry has allowed a few researchers (Arthur et al. 1989; Kohn et al. 1993; Garant and Crete 1997) to quantitatively estimate density by assuming that fishers are intrasexually territorial (Powell 1994:111), but this assumption, and thus the utility of even these estimates, has rarely been tested in order to understand any of its limitations (Raphael 1994; but see Garant and Crête 1997). More generally, the development of methods to adequately estimate fisher densities was recently identified as an important research need (Powell and Zielinski 1994). Territory mapping is probably the most accurate method of estimating population density of breeding birds (Verner 1985), but is a less commonly used technique for mammals Can. J. Zool. 79: 1691–1696 (2001) © 2001 NRC Canada 1691 DOI: 10.1139/cjz-79-9-1691 Received February 21, 2001. Accepted July 31, 2001. Published on the NRC Research Press Web site at http://cjz.nrc.ca on September 12, 2001. T.K. Fuller,1 E.C. York, and S.M. Powell. Department of Natural Resources Conservation, University of Massachusetts, Amherst, MA 01003-4120, U.S.A. T.A. Decker.2 Massachusetts Division of Fisheries and Wildlife, Field Headquarters, Westboro, MA 01581, U.S.A. R.M. DeGraaf. Northeastern Research Station, USDA Forest Service, University of Massachusetts, Amherst, MA 01003-4120, U.S.A. 1Corresponding author (e-mail: [email protected]). 2Present address: Vermont Fish and Wildlife Department, 184 Portland Street, St. Johnsbury, VT 05819, U.S.A. J:\cjz\cjz79\cjz-09\Z01-129.vp Thursday, September 06, 2001 10:16:36 AM Color profile: Generic CMYK printer profile Composite Default screen