The effect of road crossings on stream-associated salamanders within Holly Springs National Forest

Road crossings have significant effects on wildlife, but there is limited information on how road crossings affect stream-associated salamanders. Stream-associated salamanders are vital to their ecological communities and are likely to experience the effect of roads more readily than other species due to their physiological characteristics. To test the effects of road crossings on stream-associated salamanders, I surveyed 12 pairs of confluent streams – one stream crossed by a road and the other not in each pair – within Holly Springs National Forest, Mississippi. Surveys in the summer of 2015 were used to measure abundance and species richness of stream-associated salamanders. Transects were established across streams and abundance and species richness were measured at different distances from the road crossing (or stream midpoint where no road was present). Although salamanders were not abundant and only about 40% of samples contained any salamanders, abundance and richness varied between road-free and road-crossed streams, having 2.5 times as many individuals and 2.4 as many species per transect in road-free streams. The effect on abundance associated with roads was greater than previously found in Appalachia where salamanders are more abundant and species rich. There was no significant difference in abundance or species richness with distance (0–36 m) from the road crossing, but greater statistical power was needed to detect significance for the small effect size observed. Road-free and road-crossed sampling locations varied from 300–2500 m apart and no relationship existed for distance between and difference in abundance or species richness for paired locations. This lead to the conclusion that the extent of the road-effect zone lies between 36 and 300 m. When the data were re-analyzed without including the least stream-associated species (Plethodon mississippi), the results were essentially the same. There was a significant period × position interaction effect, suggesting that the first effort might have affected salamander abundance in the late summer sampling. Abundance and species richness were not associated with microclimate factors that differed between road-free and road-crossed streams. Considering the difference in magnitude for road-effects between this and Ward et al. (2008), characteristically similar disturbances in different regions may vary in their magnitude and extent – this could help explain the effect of sampling disturbance and no association between microclimate factors and salamander abundance and species richness. Additional sampling in other seasons and surveys that include additional, further from the road or midpoint might help to resolve uncertainty about spatial and temporal variation. Future research might also include components of salamander movement using mark-recapture techniques, since there are some suggestions that there are substantial temporal changes in location of the salamanders in this community.

Name thesis:degree_name

M.S. in Biological Science

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Biology

Year dc:date.available

2017

Author dc:creator

• Aldridge, Caleb Ashton

Contributors dc:contributor

• Stephen Threlkeld
• Rebecca Symula
• Marjorie M. Holland

• edge effects
• road effects
• road-effect zone
• stream-associated salamanders
• Biology

Repository record dc:identifier

https://egrove.olemiss.edu/etd/1274

OAI identifier oai:identifier

oai:egrove.olemiss.edu:etd-2273