Global losses of amphibian populations are a major conservation concern and their …

• Abstract
• Introduction
• Materials and Methods
• Results
• Discussion
• References
• Figures

Biology Articles » Conservation Biology » High dispersal in a frog species suggests that it is vulnerable to habitat fragmentation » Results

Results

- High dispersal in a frog species suggests that it is vulnerable to habitat fragmentation

3. Results

Marked Columbia spotted frogs showed high dispersal rates over long distances in both basins. Juveniles moved significantly more than adults (pfigure 1b). Twenty-five per cent of recaptured juveniles moved 200m (n=108) or further, 14% moved 1000m (n=60) or further, 9% moved 2000m (n=39) or further, and 2% moved 5000m (n=7) or further. In contrast, only 4% of adults moved 200m (n=13) or further, 2% moved 1000m (n=6) or further, and 1% moved 2000m (n=4) or further. The maximum distance moved was 5750m, the maximum elevation gain was 770m and the greatest incline traversed was 36° (700m elevation gain over 1930m horizontal distance), all by juveniles (figure 2).

Annual juvenile movement probabilities between the lower and upper groups of ponds were exceptionally high in some years. In Keeler Creek, juvenile movement probabilities were 0.290.12 (s.e), 0.000.00 and 0.490.19 in 2000, 2001 and 2002, respectively. In Marten Creek, juvenile movement probabilities were 0.120.11, 0.090.04 and 0.020.01 from the lower to the upper group of ponds and 0.620.31, 0.030.04 and 0.260.16 from the upper to the lower group in 2000, 2001 and 2002, respectively. Annual adult movement probabilities between the lower and upper groups of ponds approximated to zero for all years in both basins. (Movement, survival and capture probability estimates are found in tables 3 and 4 in Electronic Appendix A.)

Ninety-five per cent of frogs (21 out of 22) that were marked, recorded in a new location in a subsequent year and then caught again in another year remained in the site to which they immigrated. This indicates that almost all movement represents permanent dispersal rather than temporary migration. Moreover, annual juvenile survival rates were fairly high in both basins (mean=0.33), suggesting that juveniles often survive long enough to reproduce in the sites to which they immigrate. We found no difference in movement distributions between basins (p=0.59 for juveniles and p=0.29 for adults) or sexes (p=1.00), nor any bias towards upstream or downstream movement (0.10p

F_st was low in Keeler Creek (0.0640.011) and in Marten Creek (0.0160.002), as expected if historical dispersal rates and gene flow are high. This degree of subdivision is expected if there are on average 2.5 and 10.5 dispersers (genetic 'migrants') entering each population during each generation in Keeler and Marten Creeks, respectively, assuming an island model of migration corrected for a finite number of populations (Wright 1969; Slatkin 1995). Moreover, the island model estimate of the number of dispersers is probably biased low for Keeler Creek because of decreasing gene flow with increasing geographical distance in this basin (p=0.01). Distance does not predict gene flow in Marten Creek (p=0.21).