Date of Award

Spring 5-11-2021

Document Type


Degree Name

Master of Science (MS)



First Advisor

Dr. Bruce Snyder

Second Advisor

Dr. Katie Stumpf

Third Advisor

Dr. David Weese


Fire has been a prevalent disturbance on Earth for millions of years. Around the globe there are several regions that have become fire adapted, including the Southeastern United States. There have been few studies examining the effects of wildland fires on soil macroinvertebrates in the Blue Ridge Mountains, in spite of the importance of these animals to soil processes and their contributions to the biodiversity of these ecosystems. During the fall of 2016, the Southeastern USA experienced numerous, large wildfires. These fires offered an opportunity to study the effects of wildland fire on soil macroinvertebrates. We sampled sites from three different wildfires in North Georgia and Tennessee, each site with five burned plots and five unburned plots. These sites were sampled seasonally from fall 2017 through fall 2019. At each plot, on each date, we collected macroinvertebrates by hand sorting both litter (4 m diameter plots) and mineral soil monoliths (30 x 30 x 30 cm) for 30 person-minutes each. All macroinvertebrates were identified to a coarse taxonomic level. One focal taxon, millipedes, were identified to species. We ran three-factor ANOVAs using burn status (burned vs. unburned), site, sampling date, and all the interaction terms as factors and soil fauna richness, soil fauna abundance, litter fauna richness, and litter fauna abundance as dependent variables. We analyzed millipede and macroinvertebrate datasets separately. Because sampling date was a significant main effect, we wanted to determine if it was truly the sampling date or just seasonal differences the fauna experiences. The factors for this set of ANOVAs were site, burn status, and season. Of the sixteen ANOVAs conducted, there was only one where there was a significant difference between the burned and unburned plots as a main effect. Specifically, the mean abundance of leaf litter-dwelling macroinvertebrates was higher in unburned plots than burned plots. However, in almost every ANOVA, time and site had a significant effect on abundance and richness. Pre-fire drought conditions may have masked the effects of the fire by driving the fauna deeper into the soil, thus protecting them from the effects of the fire. Given that the taxonomic resolution for these data was coarse, there could have been responses to the fire by individual taxa. Because soil fauna was not affected by the fire, forest managers may not need to account for adverse effects of fire on soil fauna when planning for prescribed fire.