February 2004

At least 13 amphibian species are found on the Colorado Plateau, although some have very limited distributions in this bioregion. Amphibian declines have been documented in many parts of the world, but knowledge of amphibians on the Colorado Plateau is inadequate to determine population status or trends for most species. We have yet to even delineate distribution across the Colorado Plateau for some species, much less be able to state whether a species is declining. At present, we do not know of acute threats to most amphibians on the Colorado Plateau, however, habitat loss is a chronic problem as it is in most places.

The central portion of the Colorado Plateau was selected to be part of the USGS Amphibian Research and Monitoring Initiative, and we are now developing monitoring methods that will work in areas where amphibian habitat is ephemeral in time and space. Identifying breeding habitat for amphibians on the Colorado Plateau is difficult because most breeding sites, even in cañons, arroyos and washes, are not permanent bodies of water, and the animals are only active for brief periods of the year. Thus, we have a very small window of time in which to survey a large and rugged area for the presence of amphibians. We are trying to streamline this effort by identifying potential habitat in advance (i.e., when neither water nor amphibians are present) in order to know where to conduct surveys when conditions are right.

Our monitoring program is designed to document the Proportion of habitat Areas Occupied (PAO) by amphibians. This is a metric being tested by ARMI, and entails surveying randomly selected habitat locations and recording where amphibian species are found. If amphibians are not found, it may be because they are actually absent from the site, but can also be because they were simply not detected by surveyors. Re-visiting sites provides additional data to help estimate the detectability or probability of finding an amphibian if it is, indeed, present at a site.

By selecting our sites at random within a monitoring area, we can extrapolate PAO estimates across a larger landscape and assess the status of each amphibian species for the entire monitoring area. As data are collected over years, trends in habitat occupation by amphibians can be estimated as well. This information is important for resource managers in the National Park Service, Bureau of Land Management, and U.S. Forest Service to effectively manage and protect amphibian species on the Colorado Plateau.

We have been working on protocols for monitoring amphibians within the Canyonlands National Park survey area since 2001. We began with an area-based method of selecting survey sites. The entire area (approx. 3050 km2) was divided into 32 subareas (based on hydrologic reporting units), and 12 were selected at random for survey on a rotating basis (3 each year for four years before the first three would be re-surveyed). Each unit must be surveyed at least twice within the activity period of the amphibians to determine the detection rate for each species.

Surveying hydrologic units involves covering large areas of non-habitat. Since most amphibian habitat on the Colorado Plateau occurs in drainages, we are now concentrating survey efforts along drainage courses (i.e., solid or dashed blue lines), as depicted on USGS 7.5' quadrangle maps. We use a Geographic Information System to partition all drainage courses into 500-meter segments, then randomly select about 10% of the segments for surveying. The advantages of this approach are that field efforts are oriented toward the areas of the landscape where most amphibian habitat occurs, drainage segments take less time to locate in the field than points away from drainages, and traveling up and down drainages (e.g. canyons, arroyos, etc.) is usually easier than trying to move across the rugged landscape (i.e., we gain increased efficiency for surveys and re-visits). There are disadvantages to this technique. Because the "blue-line" approach limits surveys to drainages, the statistical universe to which our data apply is also limited to the drainages of the Canyonlands survey unit.

Breeding habitats outside of drainages (e.g., potholes, rock pools, stock ponds, swales in undulating topography, and wet meadow pools) are not included as potential habitat to be surveyed. Not all drainages contain habitat (e.g., 22% of segments surveyed in 2003 contained no amphibian habitat), so we still lose time surveying non-habitat, but less than before. We are still working on the temporal dynamics of amphibians on the Colorado Plateau relative to requirements of using PAO. We may have to exclude the summer monsoon activity period in order to meet constraints on repeat visits for PAO. We are working to resolve these issues as we continue in the protocol development phase of the project. Full implementation of a Colorado Plateau monitoring program is anticipated to begin in 2005 or 2006.

In 2003, we surveyed 94 drainage segments in the Canyonlands survey area. We also surveyed a few segments in Arches National Park (12) and Natural Bridges National Monument (8) to evaluate whether the sampling approach worked equally well in smaller parks. We found amphibians in only 23 segments (15 in Canyonlands, 1 in Arches, and 7 in Natural Bridges). We re-visited 43 of the segments to determine species detection rates. There were no cases of finding amphibians on the second visit that were not already detected on the first visit.

The National Park Service Northern Colorado Plateau Network (NCPN) has provided additional funding for us to refine these survey protocols, after which we will expand our monitoring area to include core parks of the NCPN: Canyonlands, Arches and Capitol Reef National Parks, and Natural Bridges National Monument. In addition, we are exploring the possibility of expanding our efforts to include Glen Canyon National Recreation Area, as well as the Grand Staircase-Escalante National Monument and other lands administered by the Bureau of Land Management.