Extirpation and reintroduction of fishers (Martes pennanti) in Oregon: implications for their conservation in the Pacific states

1. What they did

Before European settlement, fishers lived in most coniferous forests of Washington, Oregon and California. By the 1990s they had disappeared from much of that range. Losses in Washington and California had been documented, but nobody had pulled together the story for Oregon, where fishers had also been released several times. The authors set out to:

• Review the causes of the losses, using published and unpublished reports, museum and agency records, and the field notes of trappers and early naturalists. Washington records filled gaps, because forests in the two states are similar.
• Document the releases, from Oregon Department of Fish and Wildlife (ODFW) files and interviews with current and retired staff of state and federal agencies and timber companies who took part.
• Map where fishers live now. In 1995 they sent questionnaires to 1781 registered trappers and hound hunters, asking for first-hand accounts of fishers caught, seen, or their tracks, and phoned many who replied to check details and the observer’s experience. They added museum and agency records and the results of standardized remote camera and sooted track-plate surveys run in Oregon from 1990 to 2000.
• Rate each record’s reliability on a scale of 1 to 6: (1) specimens, photographs, video or track-plate prints; (2) fishers trapped or treed by dogs and released; (3) sightings by experienced or careful observers; (4) tracks seen by experienced people; (5) sightings by people of unknown skill; (6) records with poor descriptions or locations. Ratings 5 and 6 were dropped.
• Compare the timing and place of records with the releases, and use published genetic and body weight data to judge where today’s fishers came from.

2. What they found

Why fishers disappeared

• Overtrapping appears to have been the main first cause. In the early 1920s a prime skin could fetch up to $150, fishers are easy to trap, and their forests were reachable all year. Before 1920, a trapper could reportedly take 6 to 8 fishers on one Cascade trap line; from 1924 to 1936, only 2 to 13 a year were trapped in all of Oregon.
• In Washington, two trappers killed 37 fishers on the lower Queets River in the winter of 1920, about 74 fishers per 100 km². Two brothers killed 20 on the upper Quinault River in 1921, about 10 per 100 km².
• Poisoning with strychnine baits set for wolves and coyotes may have added to the losses, though most of this had ended by the 1930s.
• Habitat loss: clearcutting of low- and mid-elevation old-growth forest, and settlement along rivers and estuaries, probably stopped fishers recovering after they were protected (Oregon 1937, Washington 1933, California 1946).

The releases

• 1961: 24 fishers from south-central British Columbia, 11 near Klamath Falls and 13 in the Wallowa Mountains, to control porcupines that were damaging forest plantations. Two newborn kits were found dead in a crate after the Wallowa release. No record exists of what happened to the tagged animals.
• 1977 to 1980: 11 fishers from British Columbia released west of Crater Lake by a group of timber companies and Douglas County; and in fall 1977, 6 more from British Columbia released southwest of Crater Lake by a timber company biologist.
• 1981: 13 fishers from northern Minnesota released northwest of Crater Lake.
• None of the later releases was followed up. Two fishers found during that time, a female killed in a chicken coop near Roseburg (1979) and a male killed on the road near Medford (1980), were the first Oregon specimens since 1913 and were probably released animals that had wandered into towns.

Where fishers live now

• The authors gathered 244 records from 1954 to 2001. After screening, the high-reliability records showed fishers only in southwestern Oregon from 1977 onward: 8 records from 1977 to 1992 and 36 from 1993 to 2001. Only one record (1967) came from the Wallowa area.
• Unscreened records made fishers look widespread and continuous across the Pacific states. Camera and track-plate surveys told a different story: no fishers were detected in Washington, and in Oregon only in the southern Cascade Range near the 1977 to 1981 release sites and in the northern Siskiyou Mountains near California.
• The two Oregon groups are about 75 km apart, separated by Interstate 5, the city of Medford, farmland and open grassland and oak savannah. Microsatellite DNA shows they are genetically isolated from each other.
• The southern Cascade fishers carry two mitochondrial DNA types otherwise found only in the British Columbia and Minnesota source populations, and none of those found in northwestern California.
• They are also much heavier: adult males averaged 5.91 kg and females 2.80 kg, against 3.82 kg and 2.08 kg in northwestern California. Food supply alone does not explain the difference, because it is not expected to affect female weights.

3. What they concluded

• The southern Oregon Cascade population is a reintroduced population, descended from the fishers released between 1977 and 1981. The northern Siskiyou fishers may be the northern edge of the native California population, but this needs more genetic work.
• The paper links no surviving fishers to the 1961 releases: fishers appear to be gone from all other parts of their presumed historical range in Oregon.
• Fishers in the Pacific states now survive only in a few small, separate populations. The Oregon Cascade fishers are more than 650 km from the nearest population in southern British Columbia.
• Unscreened sighting records are not a reliable basis for conservation decisions. A 1994 petition to list fishers was rejected partly on the strength of such records, on the grounds that fishers’ range was continuous through Canada. That continuity no longer exists.
• After more than 20 years, the reintroduced fishers have not spread beyond a fairly small area, which suggests surrounding habitat may not be good enough.

4. How tracking was or was not used

Tracks entered the study in three ways:

• Questionnaires asked trappers and hound hunters for first-hand accounts of fishers or their tracks.
• Track records from experienced people were rated as moderately reliable (rating 4). They were kept in the “unscreened” records but not in the high-reliability records used to map the current distribution.
• Sooted track-plate prints from standardized surveys counted as the highest reliability (rating 1), alongside specimens and photographs. These detections, together with remote camera photos, were central to showing where fishers survive.

No trailing was reported.

5. Tracker or observer certification

Reported but not validated. The authors phoned many questionnaire respondents to assess their qualifications, and track observations were accepted only from “experienced individuals”, but no certification or test of their skill is described. Who identified the track-plate prints is not reported.

6. Other technologies, skills or innovations

• Standardized surveys with remote 35 mm or video cameras and sooted track-plate boxes, set for 12 to 28 days in sample units of 10.4 km².
• A six-level reliability rating for occurrence records.
• A spatially referenced database of all records and survey effort, mapped by township.
• Genetic evidence (mitochondrial DNA types and microsatellite alleles) from earlier studies, and radio telemetry of the southern Cascade population from 1995 to 2001 (Aubry and Raley 2002).

7. Implications for conservation (from the paper)

• Decades of protection from trapping and poisoning have not brought fishers back. Fishers can rarely disperse more than 100 km, and highways, large rivers and unsuitable land make it even harder for the separate populations to support one another or recolonize old range.
• Restoring connectivity with fishers in southern British Columbia is likely to be key to conserving fishers in the western United States.
• Continued surveys and monitoring are needed to see whether fishers are spreading on their own. If not, further reintroductions may be appropriate, but only after full feasibility studies covering habitat, genetics of source animals, timing, social and economic limits, and the age, sex and number of animals.
• Past releases taught little, because no one followed up. The authors urge that future reintroductions include monitoring and research.