High Lakes -
Ecology of the Cascades Frog and Interactions with Garter Snakes and Nonnative Trout in the Trinity Alps Wilderness, California
The global decline of amphibians has become a concern over the last two decades with the ranid frogs of western North America among the most seriously impacted of all species. In light of investigations documenting rapid declines of Cascades frogs (Rana cascadae) in California (Fellers and Drost 1993, Jennings and Hayes 1994, Davidson et al. 2002, Welsh et al. 2006, Fellers et al. 2008) renewed attention has been placed on the conservation of this species. One central problem regarding the management of Cascades frog populations in California is the lack of information on general life history requirements of the species, as well as its role in the ecological community. This study provides a much needed detailed ecological study on Cascades frogs in California . The primary objectives of this project were to determine key aspects of Cascades frog life history at a population scale, as well as to determine its role in a community with both native and non-native predators. We specifically focused on age-based seasonal habitat use, movement patterns, and reproductive ecology of an entire population of Cascades frogs. In addition, we studied the role Cascades frogs have in local food web dynamics including novel associations with non-native brook trout.
Based on California Department of Fish and Game (DFG) region one stocking records, the earliest written record of non-native trout introduced into Echo Lake occurred in 1930 (B. Aguilar, pers comm.). This record indicates that 10,000 eastern brook trout (Salvelinus fontinalis) and 5,000 rainbow trout ( Oncorhynchus mykiss ) were planted during this event, representing densities of 1.3 fish/sq. meter of lake surface area. During the summer of 1942, three stocking events occurred at Echo Lake, which totaled an impressive 18,400 brook trout, or 1.6 fish/sq. meter. In the following decades, stocking numbers were reduced drastically, ranging from 780 to 4,000 fish per year, though the lake was not stocked on an annual basis. The vast majority of fish planted were brook trout; rainbow trout were stocked on only a few occasions. In 1999, Echo Lake was pulled off the stocking rotation by DFG after being assessed multiple times as a poor fishery (B. Aguilar, pers. comm.) though a self sustaining population of brook trout has since persisted. Based on our recent snorkel surveys, and gill net surveys conducted by Karen Pope, rainbow trout appear be extirpated from Echo Lake basin. Unlike brook trout, it appears rainbow trout are unable maintain a self-sustaining population in Deep Creek basin without periodic stocking efforts.
We found Cascades frogs used a variety of habitats for breeding, summer foraging and over-wintering, but it was common for these habitats to be spatially or temporally separated and frogs were observed to move seasonally among them. Migrations and dispersal events were common among isolated habitats indicating that, in many cases, single sites are not likely to be self-sustaining, but contribute to a matrix of required resources across a patchy landscape. Furthermore, based on extensive migrations and dispersal we found the population dynamics of Cascades frogs to be operating across a whole-basin, so conservation of this species will require making decisions that reflect this scale.
Our food web results indicate one species of garter snake (T. atratus) was found to occur in high densities in areas with trout, and has adopted eating introduced trout as a subsidized food source. This species also feeds on amphibians. Conversely, the other garter snake species (T. sirtalis) foraged exclusively on amphibians and had lower overall densities across the landscape. These observations suggest introduced trout could be impacting native amphibians indirectly through altered food-web dynamics.
Like many other anuran species, Cascades frogs exhibit sexual size dimorphism as adults (Monnet and Cherry 2002) , with females attaining larger sizes than males. Although male Cascades frogs developed nuptial pads at 45mm SUL in this study area, we considered this as a secondary sexual characteristic and used reproductive behavior to determine age class instead. During each breeding season (2003-2006) only active adult frogs exhibiting courtship behavior congregated around breeding sites. Based on minimum sizes of these animals, we determined the smallest size of adult frogs to be 50 mm SUL for males and 58 mm for females. Furthermore, all measured amplexing individuals ( n = 66) as well as fall females showing gravidity, exceeded our minimum size cutoffs.
Adult frogs had the highest diversity of macrohabitat use out of all age classes. In contrast to immature frogs, adults were found regularly at Echo Lake and in large deep ponds as well as small ponds and streams. Among adults, females used streams twice as much as males. Based on annual captures, adult male Cascades frogs were found in lentic habitats more often (70-87%) than any other post-metamorphic group. Adult females were captured at stream habitats on average 40% (23-57%) of the time compared to 21% (13-30%) of males. Although Cascades frogs appeared to use habitats in different proportions based on age and sex, overall use varied from lentic to lotic habitats throughout the annual active period for all post-metamorphic age classes. For example, frogs were captured at significantly higher proportions in stream habitats during the summer than in the spring. During the spring, the majority of captures were associated with lentic sites used for breeding and over-wintering. As summer progressed, many frogs moved to stream habitats. Fall captures showed no significant preference for either lentic or lotic habitats. Since Cascades frogs are considered strict lentic breeders in this region, this shi ft in habitat use from lentic to lotic sites during the summer further suggests that post-metamorphic frogs use a variety of habitats during the summer when habitat availability is at is peak.
Although some adult frogs exhibited strong seasonal and annual site fidelity to specific patches, others had distinct migration patterns among separate breeding, summer foraging and wintering habitats. Over the duration of this study a substantial proportion of the adult Cascades frog population in Echo Lake basin completed extensive movements among and between different habitat patches.
Perhaps the most unexpected movement information gathered from this study was the dispersal of individual Cascades frogs between neighboring basins. We documented inter-basin dispersal of 17 individual frogs that were recaptured in one of four neighboring basins (Stony Creek, Echo, Siligo and Deer basins). Minimum air distances traveled by these individuals between successive captures ranged from 769 to 1558 m. The majority of distances traveled were over steep and rocky terrain lacking permanent aquatic features, with at least 12 individuals moving greater than 500 m over land. Although dispersal movements of these frogs occurred between 03 July and 14 September, the shortest travel time captured was a movement within 19 days (03 July and 22 July). This movement is evidence that inter-basin movements may have been facilitated in early to mid-summer when the ground was still saturated from snowmelt. Routes most likely followed low points in saddles associated with mountain passes and avoided steep jagged ridges and peaks.
This study reveals that the two observed garter snake species appear to have different life history strategies. These differences may explain why these closely related species can co-exist in a relatively small and simplified food web. Specifically, we found the diet, distribution, movement and density differed greatly between the two species. The most striking observation was the utilization distributions of common garter snakes (Thamnophis sirtalis) and aquatic garter snakes (Thamnophis atratus) in Echo Lake basin. These distributions mirrored the distribution of their respective primary prey. Common garter snakes were strict amphibian predators whereas aquatic garter snakes consumed both amphibians and trout in near equal proportions. The majority of amphibians found in the basin were lentic breeders (Cascades and Treefrogs), and may explain why common garter snakes were found predominantly in lentic habitats, especially around amphibian breeding sites containing larvae. In contrast, aquatic garter snakes were found almost exclusively in and around streams containing brook trout. Since many Cascades frog breeding locations were near sites with fish, and predation by aquatic garter snakes was highest on frogs here in general, we suspect frogs are experiencing predation levels above what would be expected naturally. Possible landscape level consequences from our data have been incorporated into a much larger analysis that addresses this phenomenon across the Trinity Alps Wilderness (see Pope et al. 2008). At minimum, our results indicate that predators are influencing Cascades frog populations and introduced fishes have the potential to alter garter snake roles in food web dynamics (more).
The results from this study fill important knowledge gaps with regard to the life history of the Cascades frog, and its role in a community with both native and introduced predators. We provide recommendations to assist stakeholders in designing management strategies more tailored to the ecological requirements of Cascades frogs which will benefit future generations of this declining amphibian.
Davidson, C., B. H. Shaffer, and M. R. Jennings. 2002. Spatial tests of the pesticide drift, habitat destruction, UV-B, and climate-change hypotheses for California amphibian declines. Conservation Biology 16: 1588-1601.
Fellers G.M., K.L. Pope, J.E. Stead, M.S. Koo, and H.H. Welsh, Jr. 2008. Turning Population Trend Monitoring into Active Conservation: Can we save the Cascades Frog in the Lassen region of California? Herpetological Conservation and Biology (in press).
Fellers, G. M., and C. A. Drost. 1993. Disappearance of the Cascades frog, Rana cascadae at the southern end of its range, California, USA. Biological Conservation 65: 177-181.
Jennings, M. R., and M. P. Hayes. 1994. Amphibian and reptile species of special concern in California. California Department of Fish and Game, Sacramento, California. 255pp.
Monnet, J. M., and M. I. Cherry. 2002. Sexual size dimorphism in anurans. R. Soc. Lond. B. 269: 2301-2307.
Pope, K. L., J. G. Garwood, H. H. Welsh, Jr., and S. P. Lawler. 2008. Evidence of indirect impacts of introduced trout on native amphibians via facilitation of a shared predator. Biological Conservation, Vol. 141, No. 5, pp 1321-1331
Welsh, H. H. Jr., K. L. Pope, and D. Boiano. 2006. Sub-alpine amphibian distributions related to species palatability to non-native salmonids in the Klamath Mountains of northern California. Diversity and Distributions 12: 298-309.
NOTE: Results on this page have not been peer reviewed (DO NOT CITE). To cite information on hyperpredation, refer to the following publication.
Pope, K. L., J. G. Garwood, H. H. Welsh, Jr., and S. P. Lawler. 2008. Evidence of indirect impacts of introduced trout on native amphibians via facilitation of a shared predator. Biological Conservation, Vol. 141, No. 5, pp 1321-1331.
Garwood J. M., C. A. Wheeler, R. M. Bourque, M. D. Larson, and H. H. Welsh, Jr. 2007. Egg mass drift increases vulnerability during early development of Cascades frog ( Rana cascadae). Northwestern Naturalist, 88:95-97.
Garwood, J. M. 2006. Rana cascadae (Cascades frog). Tadpole predation. Herpetological Review 37: 76.
Garwood, J. M., and H. H. Welsh, Jr. 2005. Rana cascadae (Cascades frog). Predation. Herpetological Review 36:165.
This research was conducted by Justin Garwood during his employ with PSW.