The Pacific fisher (Martes pennanti pacifica ) is a Federal Species of Concern, at risk of extinction from logging, fur trappng and habitat loss.  However, politics have prevented its official listing at the federal level.  The fisher is endangered in Washington , a species of concern in Idaho , sensitive-critical in Oregon .   Only two populations remain in California , one in the western part of the California/Oregon border and the other in the southern Sierra Nevada Mountains , believed to be of lower densities than the northern California population. 

The fisher is elusive, complicating efforts to track its population status.  Previous explorations in California to determine fisher population size and distribution relied on the functionality of track plates and photographs from scented or baited stations placed throughout the mountains of northern and central California . While these methods are commendable for providing nonlethal detection of a species, detectability was low.  Use of scented or baited stations can also lure targeted animals from their natural micro-habitat, and can potentially bias data toward areas of non-essential habitat.

The Center for Conservation Biology (CCB) has employed their scat detection dog methodology as an alternative non-invasive method of monitoring fisher, and applied this to both the northern and southern fisher populations in California .  The northern population study is in collaboration with, and supported by, the US Fish and Wildlife Service.  The southern population study is in collaboration with the US Forest Service, and Pacific Southwest Research Station, with contributions from the U.S. Forest Service, Pacific Southwest Region. 

Our initial goals were to assess the utility of scat detection dog methods for monitoring fisher populations across their range in California , hoping that results will better inform decisions regarding a federal listing.  The southern California study is additionally examining impacts of fuel reduction programs on the fisher.

In northern California , 2 detection dogs, trained to find fisher scat, were used to sample a 250 km² area over a six week period in July and August 2005.  Nine 3 X 3 km cells were surveyed on Green Diamond and twelve 3 X 3 km cells were surveyed on the adjacent Hoopa Indian Reservation.  Each cell was surveyed four times on the Hoopa Land . However, time constraints enabled only three sessions per cell to be conducted on Green Diamond Land .

Seven hundred scat samples were collected in the 6 week period, demonstrating a remarkably high detection rate relative to other methods.   The sampling rate was very comparable between sessions for any given cell.  Collection rates were approximately 1 sample/ km² on both Green Diamond and Hoopa Land (Fig. 1).

Fishers are typically associated with dense forests with high canopy cover, multiple layers, and large trees, snags, and logs. Some of these habitat elements are also those associated with high fire risk—dense trees with thick undergrowth and dead and downed woody material that acts as fuel ladders for ground fires to escalate to crown fires. These forest characteristics are also likely to be altered to some extent by forest management activities designed to reduce the high risk of fire in forests where fire has been suppressed for decades.

Although fishers are expected to ultimately benefit from the reintroduction of fire, the short-term risks associated with fuels management treatments are poorly understood. As long as the effects of fuel reduction activities remain unknown there is a potential conflict between the need to reduce fuels and short-term habitat needs of fishers.

Designing a scientifically credible study design that is not cost-prohibitive poses a challenge for fisher conservation given their low densities, secretive nature and large home ranges.

In fall 2006, we tested the efficacy of scat detection dogs to detect fisher habitat use at six sites in the Sierra and Sequoia National Forests . This study also sought to examine the use of these tools to monitor impacts of fuel reduction programs on the Pacific fisher. Thus, 3 of the 6 sites were scheduled for fuel reduction treatments and three served as controls.

Two detection dogs were trained to identify fisher scat and to exclude scats from other species, including martens and ringtails. Based in part on our high sampling rate in northern California during 2005, this study reduced cell size, allowing us to reduce the number of surveys per cell without loss of precision.  Each 1 x 1 km cell within the six sites was sampled twice by dog teams. Upon detection, scat samples were collected, GPS location recorded, and the scat stored for hormone analysis and individual identification.

The fisher population in the Sierra Nevada Mountains is believed to be markedly lower than in northern California .  Nevertheless, fisher scat detection by dogs was also high in this population. Two dogs collected over 600 samples in just four weeks, once again demonstrating the suitability of these methods for studying fisher.

Cost permitting, a subset of samples from both northern and southern populations is also being analyzed for gender using the zinc finger and sry markers. Data will provide baseline informations on sex and physiological and reproductive status of fishers using the sites. Identification of individuals will eventually provide information on the number of individuals using each site, their behavior, and site tenacity.

Paetkau (personal communication) recently analyzed 36 tissue samples collected from individual fisher, radiocollared on the Hoopa Reservation.  He found very low variability in the samples and concluded that a cost prohibitively large number of loci would be required to assign individual identities to these samples.  Based on that information, we are focusing our efforts on the use of matching dogs to determine individual identities of the collected scat samples.  Sample matching dogs are trained to use scent to match scat samples collected from the same individual.  We estimate that dogs can match samples from the same individual at 25% the cost and time required for microsatellite DNA analyses, given moderate to high genetic heterozygosity.  The savings should be many times greater with samples of low genetic heterozygosity, as is the case for the California fisher.