Effect of Sub-Canopy on Habitat Selection in the Blue-spotted Salamander (Ambystoma laterale-jeffersonianum unisexual complex)
Elucidating the mechanisms that influence spatial distribution patterns is vital to understanding how populations persist. We examined distribution in one of the most common salamanders in southeastern Michigan, the Blue-spotted Salamander (Ambystoma laterale-jeffersonianum unisexual complex). Three major tree species dominate the sub-canopy in the study site and form monospecific patches, which may have an effect on the distribution of terrestrial phases of local amphibian species. To examine this, we tested whether adult salamanders are significantly associated negatively or positively with patches of particular tree species. We then examined two potential causal factors: habitat quality, using leaf litter macroinvertebrate biomass as a proxy, and behavioral site choice in the Blue-spotted Salamander. We found that this species is positively associated with patches of red maple and negatively associated with patches of black cherry. We also found that, when presented with a choice, salamanders choose red maple leaf litter over black cherry leaf litter. In the absence of differences in physical leaf litter characteristics between patches of red maple and black cherry, we suggest that the Blue-spotted Salamander may be utilizing chemical cues to select habitat. Our findings imply that salamander populations may be affected not only by habitat loss, but also by changes in forest composition. These results provide evidence for a more complex model than the traditional amphibian metapopulation concept, where even fully forested habitat may form a matrix of optimal and sub-optimal, or even intolerable, patches.
Distribution of Blue-spotted Salamanders and tree species within the sampled area. Blue-spotted Salamanders are marked with a “+” and understory trees with a dot: green are witch hazel (Hamamelis virginiana), red are red maple (Acer rubrum), purple are black cherry (Prunus serotina), and gray are all other understory tree species. The two blue polygons represent the ponds within the plot. Sampling quadrats for leaf litter measurements are indicated with boxes, and quadrats for macroinvertebrate collection are indicated with triangles. The treeless stripes are walking trails that have become overgrown in recent years.
Histograms of Blue-spotted Salamander distance to nearest pond. (A) Frequency of distances of observed salamanders to nearest pond. (B) Frequency of distances of an equal number of ‘salamanders’ randomly placed within the survey plot 10,000 times.
In each of the three subfigures, solid black lines show mean number of actual Blue-spotted Salamanders found within 1-meter ‘donuts’ of various radii of different focal tree species within the study plot. Using 10,000 random placements of the same number of ‘salamanders’ we constructed an expectation due to random distribution, with 95% confidence interval, which are shown by solid and dashed gray lines, respectively. (A) The focal tree is black cherry. There were significantly fewer salamanders found 2 to 20 meters from black cherry trees than expected. (B) The focal tree is red maple. There were significantly more salamanders observed within 0 and 15 meters of red maple trees than expected. (C) The focal tree is witch-hazel. The distribution of salamanders is no different from random with respect to distance to witch-hazel trees.
Contributor Notes
Associate Editor: J. W. Snodgrass.