The Importance of Obtaining Verifiable Occurrence Data on Forest Carnivores and an Interactive Website for Archiving Results from Standardized Surveys

1. What they did

In the western United States, fishers and American martens live in old, complex forests. Trapping, clearcutting and settlement wiped fishers out of much of their former range in Washington, Oregon and California, and martens from most of the coast. To protect these animals, managers first need to know where they still live.

The authors, both with the United States Department of Agriculture (USDA) Forest Service, make the case that only verifiable records should be used for this. Verifiable records are evidence that others can check for themselves: specimens, photos and track-plate impressions. Anecdotal records, such as sighting reports and descriptions of tracks, cannot be checked.

They:

• Reviewed the standardized survey protocols published in a 1995 “detection manual” (Zielinski and Kucera 1995), which use remote cameras and sooted track-plates in fixed sample units.
• Compared anecdotal and verifiable records for fishers in the Pacific states, using maps from an earlier study (Aubry and Lewis 2003).
• Built a website and database, the Forest Carnivore Surveys website, to permanently store the results of standardized surveys and let anyone map and query them.
• Made recommendations for future surveys, so that their results can be combined across a whole region.

2. What they found

The survey protocol

• The sample unit is a square of 4 sections of the Public Land Survey System (PLSS), 10.4 km² (4 square miles), about the size of an American marten home range. Each township holds 9 sample units.
• Sensor-triggered 35 mm cameras: 2 per sample unit, at least 1.6 km apart, for at least 28 days, checked weekly.
• Line-triggered cameras or track-plate boxes: 6 per sample unit, at least 0.8 km apart, for at least 12 days, checked every other day.
• Snow-tracking was the third method in the manual, but it only works where snow falls often enough, few biologists in the Pacific states used it, and nothing had been published on how well it works. The paper does not discuss it further.

Anecdotal versus verifiable records

• Anecdotal fisher records were numerous and spread widely, suggesting a range little changed from historical accounts.
• Standardized surveys covered most forested areas, many placed on purpose where people had reported fishers. Yet fishers were detected only in parts of southwestern Oregon and several separate areas in California.
• So many anecdotal reports did not mean a resident population was there.
• The authors put this down to strong public and professional interest in rare carnivores: even fleeting glimpses by people unfamiliar with the species get reported and filed. Many reporters identify the animal by elimination from a field guide, and even experienced biologists can misidentify an animal seen briefly at night.

What survey data achieved

• Survey results helped list the fisher as endangered in Washington, and supported the federal finding of “warranted but precluded” for listing fishers in the Pacific states.
• They showed American martens had been lost from large parts of their coastal range, and gave new information on the Humboldt marten.
• They guided genetic research on fisher populations.

The website

• Survey data were sitting in local agency files, and biologists often moved on after a few years, so records could be lost.
• The database holds survey effort, results and contact details for every survey, records all carnivores detected, and stores images of photos or track-plate impressions for the four species of greatest concern: coastal American marten, fisher, wolverine and Canada lynx. Other verifiable records (museum specimens, chance photos, DNA identifications) for these four species are also included.
• Data are shown at the scale of the sample unit, not exact points, which also protects sensitive species.
• Users can map surveys, search by species, years, source and land unit, and save results.

3. What they concluded

• Anecdotal records of fishers and other forest carnivores should not be used for conservation decisions or management recommendations. They can still point to places worth surveying with devices that give verifiable evidence.
• Standardized surveys let any agency collect scientifically valid presence and absence data, and these data are most valuable when combined across large areas.
• Two parts of the 1995 protocol do not suit a regional archive. The authors recommend that surveys use the fixed grid of sample units set by township boundaries, and that surveys run for the full period whatever species is detected, because a survey stopped early after a fisher detection says nothing about whether martens are absent.
• The website gives a permanent home for these data, for future generations and for planning new surveys.

4. How tracking was or was not used

Tracks were central to the paper’s argument in two opposite ways. Descriptions of tracks reported by people are listed with sightings as anecdotal records that cannot be verified. Track-plate impressions, footprints left on sooted plates inside open-ended boxes, are listed with photos and specimens as verifiable records, because the impression itself can be kept and checked by others. The website stores images of track-plate impressions for the four species of greatest concern. Snow-tracking was part of the protocol, but the paper does not discuss it further. No trailing was reported.

5. Tracker or observer certification

Not reported. The paper does not say who identified species from the track-plate impressions or how their skill was checked. Its case rests on the impressions being kept so that others can check them.

6. Other technologies, skills or innovations

• Remote cameras: sensor-triggered 35 mm cameras and line-triggered 110 cameras.
• Sooted track-plates, unenclosed or in open-ended boxes.
• An interactive mapping website built with ESRI ArcIMS, with the USDA Forest Service Geospatial Service and Technology Center in Salt Lake City, Utah.
• Hair-snagging devices with DNA identification were being developed and could be added to the website if standardized protocols are developed.

7. Implications for conservation (from the paper)

• Decisions about rare carnivores can carry social and economic costs, so they need evidence that holds up to scientific and legal challenge.
• Surveys are costly. Archiving their results avoids repeated work between agencies, helps plan and prioritize new surveys, and allows changes in distribution to be followed over time.
• The website could be expanded to species with larger ranges, such as Canada lynx and wolverine.