Paolo Ciucci, Luigi Boitani, Department of Animal and Human Biology, Universite di Roma "La Sapienza", Viale dell'Universite 32, 00185 Rome, Italy

As a preliminary phase of a long-term wolf study, we carried out a simulated howling survey to census and localize reproductive packs in the Pollino National Park (PNP), southern Italy. Based on a systematic sampling procedure the use of simulated howling surveys has been advocated to census wolf packs over relatively small areas (e.g. National Parks; cf. Fuller and Sampson 1988). However, according to topographic and habitat features, and because the distribution of centers of human activity is not homogeneously distributed within PNP, we systematically sampled only the southern portion (800 km2) of the Park; an opportunistic sampling design was instead considered to be more feasible in the rest of PNP.

The Park (2,200 km2) is the largest in Italy and is featured by mountainous topography and dense forest cover. From 28 July through 16 September 1999 we sampled about 60% of the Park using holws recorded from Italian wolves (courtesy of F. Rivoira). Following indications by Harrington and Mech (1979) and Fuller and Sampson (1988), 12 fixed howling routes (range in length: 61-214 km) were each travelled by car for 3 consecutive nights, and simulated howls were broadcasted from 5-11 howling stations located along each route (average distance between stations = 3 km). Each night 6 operators, in teams of 2, covered simultaneously 3 adjacent routes to avoid double-counting due to wolf movements or shifts of rendez-vous sites. We elicited howls from adult wolves and/or pups in 14 different occasions, and localized rendez-vous sites of 5 different packs. To complete the survey, we travelled 4,400 km and spended 294 hours in the field, with an average of 115±54 Km travelled and 8±1 hours spent each night. Requiring an average of 8-9 hours in the field to run the survey, each operator worked for 33 consecutive days for a total of 198 days/6 operators. On the basis of differences reported in the effort for the systematic and the opportunistc sampling designs, we conclude with cost-benefit considerations on the optimal sampling procedure to adopt in a human-dominated landscape matrix.

References:

Harrington, F.H., L.D. Mech. 1979. J. Wildl. Manage 46:686-693
Fuller, T.K., B.A. Sampson. 1988. J. Wildl. Manage 52:60-63.