Karen L. Pope

Garter snake distributions in high-elevation aquatic ecosystems: Is there a link with declining amphibian populations and nonnative trout introductions?

Abstract The dramatic amphibian population declines reported worldwide likely have important effects on their predators. In the Sierra Nevada, where amphibian declines are well documented and some are closely tied to the introduction of nonnative trout, the mountain garter snake, Thamnophis elegans elegans, preys predominately on amphibians. We surveyed 2103 high-elevation lakes in the Sierra Nevada, quantified the distributional relationship between the mountain garter snake and anuran amphibians (Pseudacris regilla, Rana muscosa, and Bufo spp.) and used this information to evaluate the possibility that amphibian declines lead to declines of garter snakes. We observed a strong association between amphibian presence and garter snake presence. The probability of finding snakes in lakes with amphibians was 30 times greater than in lakes without amphibians. Lakes with snakes had higher numbers of amphibians within 1 km (mean = 4018.8) than did lakes without snakes (mean = 642.1). On a landscape scale, in Kings Canyon National Park (where 40% of larger lakes contain nonnative trout) amphibians were found in 52% of lakes, and 62 garter snakes were found in 33 of the 1059 lakes surveyed. In contrast, in the John Muir Wilderness (JMW; where 80% of larger lakes contain nonnative trout), amphibians were found in 19% of lakes, and no snakes were found in any of the 1044 lakes surveyed. Based on these data, we suggest that the introduction of nonnative trout has led not only to the decline of amphibians but also to the decline of garter snakes. This study supports the hypothesis that the presence of amphibians is a prerequisite for garter snake persistence in high-elevation portions of the Sierra Nevada and that the introduction of trout into an ecosystem can have serious effects, not just on their prey but also on other predators in the ecosystem.

A Telemetric Study of the Movement Patterns and Habitat Use of Rana muscosa, the Mountain Yellow-legged Frog, in a High-elevation Basin in Kings Canyon National Park, California

S.–In a high-elevation (3470 m) lake basin (upper Dusy Basin) in Kings Canyon National Park, California, we used radio transmitters on 24 mountain yellow-legged frogs (Rana muscosa ) to gather basic information on their movement patterns. Rana muscosa have declined throughout their range in the Sierra Nevada and restoration plans require information on their movement ecology. Our study indicates that R. mus cosa had different movement patterns and habitat associations during the 1997 summer period (August and September) compared to October when winter dormancy began. In August, visual surveys found frogs in 10 of the 11 lakes in upper Dusy Basin. During August most tagged frogs moved little (mean movement 77 to over five day periods) and all were found in the lake or adjacent stream where they were originally tagged. During September, movement increased compared to August Frogs moved from the original capture lake mean distances of 145 m,and moved cumulative distances of 315-466 m. By October, frogs were again sedentary (mean distance moved 43 mt and frogs were found in three of the 11 lakes in the basin. Moreover, mean home ranges (adaptive kernel 90% contours) also were different throughout the summer and were highest for frogs tracked during September (5336.2 m2) compared to August (385 m2, and October (52.8 m2. Before this study it was assumed that R. muscosa over-wintered in the deepest portion of the lake, However, most lakes were frozen when our study ended, and tagged frogs were found nearshore under ledges and in deep underwater crevices suggesting that at least some R. muscosa over-winter in these nearshore areas. In this study, we found R. mus cosa in different aquatic habitats over the course of their activity period and that they readily moved between these habitats using both aquatic and overland pathways. The movements appear to be associated with seasonal migrations between summer and over-wintering sites. fish stocking still occurs , mountain ye l l ow-l eg ged frogs were found in only 5% of the lakes (Matthews and Knapp, 1999). In contrast, Kings Canyon National Park, where fish stocking was terminated in the late 1970s, had frogs in 35% (379 of 1083) of the sampled lakes. Presumably, pred ation on tadpoles and young frogs has caused the elimination of R. muscosa from the majority of lakes inhabited by introduced fish (Bradford, 1989; Hayes and Je n n i n g s , 1986; Knapp and Matthews, in press). Remaining populations have become increasingly isolated and, therefore, are more susceptible to local extinctions without the o p p o rtunity for re c o l o n i z ation from neighbori n g populations (Bradford et a]., 1993), especially if movement is minimal. Rana muscosa inhabits high-elevation (1370 to 3660 m) lakes and streams in the Sierra Nevada (Zweifel, 1955; Mullally and Cunningham, 1956). Due to the extreme environment encountered at high elevations (e.g., long winters, sustained freezing, and low temperatures), frogs may only be active for a few months during the summer after snowmelt and before the winter freeze.Tadpoles over-winter for at least two to three years (Zweifel, 1955; Cory, 1963, Bradford et al., 1993) and adults, like other anurans, presumably perform migrations to locate suitable areas for reproduction, feeding, and over-wintering (Baker, 1978; Sinsch, 1990). During the summer, tadpoles and adults seek the wa rmest therm a l regimes (Bradford, 1982) and presumably feed to store fat reserves for winter dormancy that can last up to nine months (Bradford, 1983). Despite the information describing population declines, habitat use information on R. muscosa is lacking or speculative. Rana muscosa is highly aquatic and reportedly is never found more than two or three jumps from water (Mullally and Cunningham, 1956; Stebbins, 1985) suggesting that movement is restricted to a q u atic pat h ways. And although never dire c t ly observed, R. muscosa is believed to over-winter in the deepest portions of lakes below the ice,thus re q u i ring lakes >4 m deep for surv ival (Bra d fo rd, 1 9 8 3 ) . Our study was designed to gather quantitative information on R. muscosa movements and habitat use during summer and fall. With this information, we should be able to more effectively restore some of their habitat in the Sierra Nevada. Movement distances and ranges will also be important for predict ing recolonization. patterns and for metapopulation a n a lysis (Hanski and Gilpin, 1997). Duri n g August-October 1997, we used telemetry to determine R. muscosa home ranges, their movement between lakes, and typical habitat associations during the summer and fall as lakes began to freeze. MATERIALS AND METHODS Study Area.–The study was conducted in upper Dusy Basin, Kings Canyon National Pa rk , C a l i fo rnia (Latitude 37°5’40”, L o n gi t u d e 118°33’45’) at an elevation of 3470 m (Fig. 1). The site supports a large population of R. muscosa of varying age classes. The glacially formed granite basin supports alpine fell field vege t ation with low-growing herbaceous plants, dwarf shrubs, and few krummholzed white-bark pines (Holland and Keil, 1995). There are a series of streams, lakes, and ponds in the basin that are fed by snowmelt. The study area covers approximately 0.75 km2. Our study focused on 11 lakes and ponds in Dusy Basin. All lakes and streams within the study area have been numbered and mapped using a Trimble Pro XL GPS system accurate to 1 m. Only lakes 1 and 3 (all water bodies being considered lakes) support self-sustaining populations of trout. Fish were also found in some of the connected creeks. This scenario will likely be typical of future refuges in National Forest Wilderness areas where, after re-introductions of mountain yellow-legged frogs, self-sustaining fish populations will likely persist in large lakes despite fish stocking changes. Lakes ranged in size from 114 m2 to 5.3 ha and were 0.25 to 10 m deep. Field Techniques.–We attached radio transmitters (Holohil Systems Ltd.; BD-2 transmitters; 15 mm X 7 mm X 4 mm thick) to 24 R. muscosa (snout-vent length > 55 mm) and documented movement from August 2-October 28,1997 (Table 1). We tagged frogs larger than 55 mm to minimize possible effects of transmitter weight. To at t a ch radio tra n s m i t t e rs , a wa i s t-belt made of alum i num ball or beaded chain was used, similar to that used on the Califo rnia re d l egged frog, Rana auro ra d ray t o n i i ( R at h bun and Murp h ey, 1996). The total weight of the at t a ched transmitter and belt was ap p roxi m at e ly 1.5 g, wh i ch is below the 10% rule that at t a ched objects not exceed 10% of body mass (Heye r et al., 1994). Frogs we re hand-c ap t u re d, we i g h e d, m e a s u re d, t agge d,and then released at the cap t u re site. S ex was determined by the enlarged nuptial pad at the base of the inner-most fi n ger found in adult males ( S t ebb i n s , 1985). The transmitter bat t e ries lasted ab o u t one month. To monitor movements over Au g u s t , S ep t e m b e r, and October, we tagged frogs in three diffe rent gro u p s : G roup one consisted of 12 frogs monit o red from 2 Au g u s t-25 Au g u s t , group two consisted of nine frogs monitored from 3 Sep t e m b e r-30 Sep t e m b e r, and group three consisted of three frogs monitore d f rom 3 October through 28 October. We attempted to re m ove tra n s m i t t e rs and belts just befo re the bat t e ry ex p i re d. After frogs we re tagged they we re re l o c ated on K. MATTHEWS AND K. POPE 616

Assessing changes in amphibian population dynamics following experimental manipulations of introduced fish.

Sport-fish introductions are now recognized as an important cause of amphibian decline, but few researchers have quantified the demographic responses of amphibians to current options in fisheries management designed to minimize effects on sensitive amphibians. Demographic analyses with mark-recapture data allow researchers to assess the relative importance of survival, local recruitment, and migration to changes in population densities. I conducted a 4-year, replicated whole-lake experiment in the Klamath Mountains of northern California (U.S.A.) to quantify changes in population density, survival, population growth rate, and recruitment of the Cascades frog (Rana cascadae) in response to manipulations of non-native fish populations. I compared responses of the frogs in lakes where fish were removed, in lakes in their naturally fish-free state, and in lakes where fish remained that were either stocked annually or no longer being stocked. Within 3 years of fish removals from 3 lakes, frog densities increased by a factor of 13.6. The survival of young adult frogs increased from 59% to 94%, and realized population growth and recruitment rates at the fish-removal lakes were more than twice as high as the rates for fish-free reference lakes and lakes that contained fish. Population growth in the fish-removal lakes was likely due to better on-site recruitment of frogs to later life stages rather than increased immigration. The effects on R. cascadae of suspending stocking were ambiguous and suggested no direct benefit to amphibians. With amphibians declining worldwide, these results show that active restoration can slow or reverse the decline of species affected by fish stocking within a short time frame.

Effects of Nonnative Trout on Pacific Treefrogs (Hyla regilla) in the Sierra Nevada

Abstract We used analyses based on surveys of > 1700 water bodies in a 100,000-ha area in the John Muir Wilderness (JMW) and Kings Canyon National Park (KCNP) to determine the influence of nonnative trout on the distribution and abundance of Hyla regilla in the High Sierra Nevada. At the landscape scale (JMW compared to KCNP), a negative relationship between trout and frogs in lakes was evident. In the JMW study area where trout are more abundant, only 7.2% of all water bodies contained H. regilla versus 26.6% in the KCNP study area. Also, the percentage of the total water body surface area containing H. regilla was 19.4 times higher in the KCNP study area than in the JMW study area. Hyla regilla were most abundant in portions of KCNP where the probability of finding lakes with trout is lowest and least abundant in the northern part of the JMW where the probability of finding lakes with trout is highest. At the water body scale, after accounting for the effects of all significant habitat and isolation variables, the odds of finding H. regilla in water bodies with no trout was 2.4 times greater than in water bodies with trout, and the expected number of H. regilla in water bodies with H. regilla and without trout was 3.7 times greater than in water bodies with both H. regilla and trout. Hyla regilla were significantly more likely to be found at the lower elevations (3000–3400 m) compared to higher elevations (3400–3800 m) and in shallow water bodies with high percentages of silt in near-shore habitats. Our study demonstrates a negative relationship between fish presence and H. regilla distribution and abundance in lakes and suggests that H. regilla has declined in portions of the High Sierra with high numbers of trout-containing lakes. It adds an additional native species to the mounting evidence of landscape-scale declines of native species resulting from the introduction of predatory fish.

Movement Ecology and Seasonal Distribution of Mountain Yellow-Legged Frogs, Rana muscosa, in a High-Elevation Sierra Nevada Basin

Abstract Movement ecology and seasonal distribution of mountain yellow-legged frogs (Rana muscosa) in Dusy Basin (3470 m), Kings Canyon National Park, California, were characterized using passive integrated transponder (PIT) surveys and visual encounter surveys. We individually PIT-tagged 500 frogs during the summers of 1997 and 1998 and monitored these individuals during seven recapture surveys in 1997 and 15 recapture surveys in 1998 from the time they emerged from overwintering in July until the lakes froze for the winter in October. Probability of movement between lakes was associated with abundance of Hyla regilla larvae in the different lakes of origin, activity of the frogs (overwintering, breeding, feeding), and time of year. Overland movements exceeding 66 m were observed in 17% of the tagged frogs. Movement between lakes 1 km apart was detected. Site fidelity from 1997 to 1998 was high, and 97% of the tagged frogs recaptured in October of both years were found in the same overwintering lakes. Frogs were more narrowly distributed in spring and fall than in summer. Summer frog densities (number per meter of shoreline) were positively related to water temperature, air temperature, maximum lake depth, and presence of H. regilla larvae and negatively related to presence of trout. Mountain yellow-legged frogs use a range of aquatic sites throughout their activity period, and basins with a variety of deep lakes and shallow ponds may be the most appropriate reserves for this declining species.

Correlates of virulence in a frog-killing fungal pathogen: evidence from a California amphibian decline

The fungal pathogen Batrachochytrium dendrobatidis (Bd) has caused declines and extinctions in amphibians worldwide, and there is increasing evidence that some strains of this pathogen are more virulent than others. While a number of putative virulence factors have been identified, few studies link these factors to specific epizootic events. We documented a dramatic decline in juvenile frogs in a Bd-infected population of Cascades frogs (Rana cascadae) in the mountains of northern California and used a laboratory experiment to show that Bd isolated in the midst of this decline induced higher mortality than Bd isolated from a more stable population of the same species of frog. This highly virulent Bd isolate was more toxic to immune cells and attained higher density in liquid culture than comparable isolates. Genomic analyses revealed that this isolate is nested within the global panzootic lineage and exhibited unusual genomic patterns, including increased copy numbers of many chromosomal segments. This study integrates data from multiple sources to suggest specific phenotypic and genomic characteristics of the pathogen that may be linked to disease-related declines.

INFLUENCE OF ANURAN PREY ON THE CONDITION AND DISTRIBUTION OF RANA MUSCOSA IN THE SIERRA NEVADA

Mountain yellow-legged frogs (Rana muscosa) at high elevations of the Sierra Nevada must obtain enough food during summer to survive 7–9 winter months when their aquatic habitats are frozen and food is presumably unavailable. Adults of R. muscosa prey on a variety of organisms, including aquatic and terrestrial invertebrates and anuran larvae. To determine if anuran larvae influence the condition, distribution, and survival of adults of R. muscosa, we assessed body condition (variation from the expected mass for given length) and distribution of R. muscosa in relation to abundance and distribution of larval Hyla regilla, Bufo canorus, and Bufo boreas. Using three years of mark-recapture data, we found that abundance of larval H. regilla in a water body was positively correlated with the condition of R. muscosa. Moreover, R. muscosa had higher relative mass (Wr) values in areas with other anuran larvae than in areas with no other amphibian species. High Wr values significantly increased the probability of recapturing frogs in following years. Using surveys of over 1700 lakes and ponds in the John Muir Wilderness and Kings Canyon National Park, we also found that the presence of larval H. regilla or Bufo in a water body was a significant indicator of the presence of adults of R. muscosa after accounting for significant habitat and isolation variables. These findings taken together suggest that some adults of R. muscosa actively seek out water bodies with other anuran species during the critical summer months and that the anuran larvae provide a nutritious food source that increases the body condition and, therefore, survival of the frogs.

Bacterial flora on Cascades frogs in the Klamath mountains of California

Amphibians are experiencing global declines due in part to the infectious disease chytridiomycosis. Some symbiotic bacteria residents on frog skin have been shown to inhibit the growth of Batrachochytrium dendrobatitis (Bd) but few studies have attempted to fully describe the resident bacterial flora of frog skin. We cultured and sequenced 130 bacterial isolates from frogs collected from the California Klamath Range, recovering predominantly Gram-negative bacteria from 20 higher order taxa and 31 genera. There were also a large number of unclassifiable isolates. Forty-three isolates were assessed for their ability to inhibit the growth of Bd in vitro; of these, two had strong and three had slight anti-Bd activity. We suggest that many bacterial species may play a secondary role in Bd resistance, acting synergistically with inhibitory species. Future research is required in order to characterize these interactions. Understanding the relationships between bacterial strains may be important in predicting and managing the effects of future anti-Bd treatments such as antimicrobial compounds or probiotic bacteria.

Factors Mediating Co‐Occurrence of an Economically Valuable Introduced Fish and Its Native Frog Prey

Habitat characteristics mediate predator-prey coexistence in many ecological systems but are seldom considered in species introductions. When economically important introduced predators are stocked despite known negative impacts on native species, understanding the role of refuges, landscape configurations, and community interactions can inform habitat management plans. We measured these factors in basins with introduced trout (Salmonidae) and the Cascades frog (Rana cascadae) to determine, which are responsible for observed patterns of co-occurrence of this economically important predator and its native prey. Large, vegetated shallows were strongly correlated to co-occurrence, and R. cascadae larvae occur in shallower water when fish are present, presumably to escape predation. The number of nearby breeding sites of R. cascadae was also correlated to co-occurrence, but only when the western toad (Anaxyrus boreas) was present. Because A. boreas larvae are unpalatable to fish and resemble R. cascadae, they may provide protection from trout via Batesian mimicry. Although rescue-effect dispersal from nearby populations may maintain co-occurrence, within-lake factors proved more important for predicting co-occurrence. Learning which factors allow co-occurrence between economically important introduced species and their native prey enables managers to make better-informed stocking decisions.

Itraconazole treatment reduces Batrachochytrium dendrobatidis prevalence and increases overwinter field survival in juvenile Cascades frogs.

The global spread of the fungal pathogen Batrachochytrium dendrobatidis (Bd) has led to widespread extirpation of amphibian populations. During an intervention aimed at stabilizing at-risk populations, we treated wild-caught Cascades frogs Rana cascadae with the antifungal drug itraconazole. In fall 2012, we collected 60 recently metamorphosed R. cascadae from 1 of the 11 remnant populations in the Cascades Mountains (CA, USA). Of these, 30 randomly selected frogs were treated with itraconazole and the other 30 frogs served as experimental controls; all were released at the capture site. Bd prevalence was low at the time of treatment and did not differ between treated frogs and controls immediately following treatment. Following release, Bd prevalence gradually increased in controls but not in treated frogs, with noticeable (but still non-significant) differences 3 wk after treatment (27% [4/15] vs. 0% [0/13]) and strong differences 5 wk after treatment (67% [8/12] vs. 13% [1/8]). We did not detect any differences in Bd prevalence and load between experimental controls and untreated wild frogs during this time period. In spring 2013, we recaptured 7 treated frogs but none of the experimental control frogs, suggesting that over-winter survival was higher for treated frogs. The itraconazole treatment did appear to reduce growth rates: treated frogs weighed 22% less than control frogs 3 wk after treatment (0.7 vs. 0.9 g) and were 9% shorter than control frogs 5 wk after treatment (18.4 vs. 20.2 mm). However, for critically small populations, increased survival of the most at-risk life stage could prevent or delay extinction. Our results show that itraconazole treatment can be effective against Bd infection in wild amphibians, and therefore the beneficial effects on survivorship may outweigh the detrimental effects on growth.