USDA Forest Service

Pacific Southwest Research Station

Pacific Southwest
Research Station

800 Buchanan Street
Albany, CA 94710-0011
(510) 883-8830
United States Department of Agriculture Forest Service. USDA logo which links to the department's national site. Forest Service logo which links to the agency's national site.

Research Topics Wildlife & Fish

About this Research:
Research Topics:
Contributing Scientists and Staff
Participating Programs:

Mill Creek Restoration:
Distributions of Zero-order and Headwater Amphibian Assemblages Relative to the Landscape Mosaic at Mill Creek, Del Norte, California

Picture of redwood forest in Del Norte Coast Redwoods State Park The restoration of the Mill Creek property, in far Northwestern California, offers an excellent opportunity to increase our understanding of the components that contribute to a healthy ecosystem. In recent years, forest amphibians have moved to the forefront of this process of selecting meaningful bio-indicators. Because of their high site fidelity, sensitivity to environmental perturbations and ease of detection, amphibians make excellent indicators of ecosystem health (Welsh and Ollivier 1998, Welsh and Droege 2001).

On the Mill Creek property the main objective is to return heavily managed forest stands to late- seral conditions by reducing the extensive road system and applying silvicultural treatments to maximize tree growth and augment stand structure while protecting watercourses and minimizing fire hazards. Ultimately, success or failure will be partly determined by the response of the native animal communities. To that end, managers need to know the current status of the animal communities as a baseline to compare with future distributions. For general information on the Mill Creek property visit the California State Parks Mill Creek site.

Picture of degraded stream in Mill Creek
Degraded stream
Picture of recovering stream
Recovering stream
Picture of late-seral stream
Late-seral stream
Our objective is to quantify the distributions of the zero-order (terrestrial/headwater stream interface) amphibians at headwaters sites in representative sub-basins in both the East and West fork basins of Mill Creek. This sampling design is intended to provide baseline data on current conditions and provide sites that can be used for future monitoring. In the long-term, we hope to document the responses of the amphibians as the forest structure recovers and develops toward late-seral conditions.

Picture of cover/crack boards
Cover/crack boards

Our study design was strongly influenced by the recent findings of Chris Sheridan and Deanna Olson in their study of Amphibian Assemblages in Zero-order Basins in the Oregon Coast Range. They defined the term "zero-order basin" and confirmed our suspicions that these areas are hot-spots for amphibian diversity.

We randomly selected 37 zero-order basins among four forest age classes. 25 sites were on the Mill Creek property in second-growth stands, and 12 were in adjacent parklands of Del Norte Redwoods State Park and Jedediah Smith Redwoods State Park (8 late-seral, 4 historically thinned late-seral).

To sample the headwater amphibian assemblage, we employed three techniques. At each site, we performed one hour of upland time-constrained searching (TCS), deployed cover/crack boards, and performed stream channel area-constrained searches (ACS).

The preliminary results from the upland data of our pilot study, conducted in the spring and summer of 2006 showed significant differences in individual species, and in the amphibian community between the late-seral parkland and the second-growth Mill Creek property (MCP).

Box plot of terrestrial amphibain captures by property Box plot of aquatic amphibain captures by property

Richness (number of terrestrail species observed) was significantly higher in the late-seral forest (seral stage 4) then in the youngest stands (seral stage 1) (F = 4.15, P = 0.0168), with an overall trend of more amphibian species in older stands.

Box plot of terrestrial richness by seral stage Legend:  1 & 2 are on the MCP and 3 & 4 are on parklands

For the California slender salamander (Batrachoseps attenuatus), we found significantly higher densities in the adjacent late-seral forest (seral stage 4) then in either of the Mill Creek stands (seral stage 1 and 2) (F = 4.12, p = 0.0172).

Ensatina salamander (Ensatina eschscholtzii) densities were significantly higher in the older thinned sites of Del Norte Coast Redwood State Park, then in the youngest stands on the Mill Creek property (F = 3.56, p = 0.0291). Although not significant at this sample size, it appears that both the age classes from the Mill Creek property (Seral stage 1 and 2) have lower Ensatina densities then the adjacent parkland (seral stage 3 and 4).

Interestingly, we did not see the same pattern in the Del Norte salamander (Plethodon elongatus) densities. There were not enough data on the wandering salamander (Aneides vagrans) or red-legged frog (Rana aurora) to compare between forest age categories.

Box plot of California Slender salamander captures by seral stage Box plot of Ensatina salamander captures by seral stage Box plot of Del Norte salamander captures by seral stage

We found no significant differences in animal densities between any of the stand age categories for aquatic amphibians, but this with a minimum sample size of 28. Although not significant, coastal (Pacific) giant salamander (Dicamptodon tenebrosus) numbers did show a trend of higher densities in the older stage classes (F = 2.04, p = 0.1354). Overall, coastal giant salamander densities were high, with an average of nearly three animals per square meter of headwaters stream. Late seral forest densities averaged five animals per square meter, and the highest density of any stream was over 13 animals per square meter.

Southern torrent salamander (Rhyacotriton variegatus) had similar densities across all the seral stage categories. This species is significantly impacted by forest management practices. That we did not detect differences, attests to the surprising resiliency of the Mill Creek geology; which may be why it still supports five salmonid species. However, confounding our results were the differences in steam slopes and aspects between the seral stage categories. The younger, Mill Creek streams were higher gradient then the parkland streams. High gradient streams tend to scour fine sediments, which can be detrimental to southern torrent salamander after harvest. Another factor is the cool, moist coastal environment that allows these salamanders to persist despite the forest canopy being removed by timber harvest.

There were not enough data for the nocturnal tailed frog (Ascaphus truei) to compare numbers between treatments. Our protocol was not designed to sample the adults, and there was insufficient water velocity at ourheadwater sites for tailed frog breeding.

Box plot of Southern torrent salamander captures by seral stage Box plot of Coastal Giant salamander captures by seral stage

Picture of Monty with large Douglas Fir log in Jedediah Smith Redwoods State Park

Overall, the study was successful in demonstrating the applicability of amphibian assemblages as indicators of forest ecosystem status. This approach is gaining greater acceptance as others test and verify the use of amphibians as biometrics (e.g., Ashton et al. 2006, Perkins and Hunter 2006). We recommend continued monitoring of these amphibian assemblages in five or ten-year intervals at these same sites in order to track the recovery of the Mill Creek redwood ecosystem as it returns to late-seral conditions.

Literature cited: