Graduation Year


Document Type




Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Biology (Integrative Biology)

Major Professor

Thomas L. Crisman, Ph.D.

Committee Member

Stephen Deban, Ph.D.

Committee Member

Joni Downs Firat , Ph.D.

Committee Member

Mark Rains, Ph.D.

Committee Member

Jason Rohr, Ph.D.


silviculture, plantation, enclosure, land use, management


Once dominant, longleaf pine forests of the southeastern United States have been modified by 97 percent, resulting in several animal species being listed as endangered and threatened. Pine plantation silviculture (tree plantings) now occupies half of the original longleaf range where several animal species of conservation concern have experienced recent local population declines. In North America, the accepted practice of pine plantations is to plant pines densely in rows for wood production. Given that land use is considered a primary local driver for the 30% of amphibian species currently at risk of extinction, and planted pine is predicted to expand coverage by 2020, sustainable land management will require integrating ecological and economic goals, including conservation objectives. To understand how amphibian species characteristic of longleaf pine forest are affected by planted pine forestry, it is necessary to understand how associated shifts in habitat structure associated with aging pine stands influence species composition across a wide geographic area, especially populations of rare species. The purpose of this dissertation is to examine how forest structure (natural regeneration vs plantation) affects amphibian species composition and occupancy of small isolated wetlands embedded within the forest. Particular emphasis is placed on assessing several potential causal mechanisms of regional declines in amphibian species. This study was performed on Florida public forests where active forest management is a potential conservation tool and historic populations of rare amphibians occur. Sites where occupancy was assessed included: Goethe State Forest, Ocala National Forest, Jennings State Forest, Saint Marks National Wildlife Refuge, and Apalachicola National Forest. Withlacoochee State Forest was used for examining potential causal mechanisms of amphibian declines because of close proximity of reference condition forest to planted pine.

Chapter One of this dissertation uses Objective Based Vegetation Monitoring (OBVM) metrics to quantify differences in forest habitat structure surrounding study sites (among planted pines and naturally regenerated second growth pine) and relationships between vegetation metrics. Increased basal area of planted pine resulted in decreased canopy openness and a significant decrease of ground cover, especially wiregrass. Serenoa repens petiole counts and percent cover of woody shrubs also significantly decreased wiregrass cover, and variance partitioning indicated that the effects of woody shrub invasion and pine basal area on wiregrass were independent on public forests. Absence of bare ground because of pine needle litter was a significant predictor of wiregrass absence in a zero-inflated negative binomial model (ZINB). ZINB predictions for wiregrass cover along a gradient of pine basal area and woody shrub abundance demonstrated the importance of habitat management for native groundcover on public forests. OBVM metrics were often positively correlated within the canopy, subcanopy, and groundcover categories.

Chapter Two presents the results of site occupancy modeling comparing the presence of 10 species of amphibians in planted pine and naturally regenerated forest, focusing on the influence of canopy and groundcover habitat structure on occupancy of individual amphibian species. Amphibian species of greatest conservation need (SGCN) in Florida adapted to xeric soils, i.e. the gopher frog (Lithobates capito) and striped newt (Notophthalmus perstriatus), were found where basal tree areas were below 10.3 m2/ha, and a mesic amphibian SGCN, the ornate chorus frog (Pseudacris ornata), was found at pine basal areas up to 13.1 sq. m2/ha. All SGCN were found at sites with average woody shrub cover below four percent, and litter cover below 80 percent. Wiregrass cover was higher than four and six percent per m2 at sites with L. capito and N. perstriatus, respectively. Of 33 site detections for SGCN, only three sites had < 2.5 m2 wiregrass cover. Only one species, the pinewoods treefrog (Hyla femoralis), was more common on planted pine stands, and both Lithobates sphenocephala and L. grylio had constant probability of occupancy regardless of forest type. Percent wiregrass cover was a significant predictor of occupancy for five species, particularly for State listed species L. capito and N. perstriatus, suggesting it may be a useful indicator of habitat quality for longleaf-dependent amphibians.

Chapter Three focuses on experimental release of juvenile southern toads into two types of terrestrial enclosures to independently determine if amphibian movement and desiccation is determined by forest management. Movement enclosures consisted of four 50 m x 2.5 m unidirectional runways joined at the center to determine movement rate, distance, and behavior among forests of varying habitat structure. Movement rates were relatively consistent among forest types and positively related to rainfall, which itself did not vary among forests. Canopy closure and ground slope were predictors of behavior as toads move preferentially toward canopy openings and negative slopes, particularly when in planted pine habitat. Ten desiccation enclosures, each 15 cm diameter by 45 cm tall, were used to determine water loss and survival of toads for up to 72 hours. The proportional water loss from toad bodies was significantly related to ambient soil moisture at enclosures, with moisture consistently less at planted pines sites in xeric soil, suggesting a potential source of mortality for species specialized to sandhills. Juvenile survival was particularly low at sandhills planted with pine where dry duff replaced native groundcover and likely prevented successful water conservation behavior.

The results of these studies suggest that land management decisions related to planted pine forests will determine the species composition at embedded isolated wetlands for both rare and common amphibians. As natural disturbance regimes that limit woody shrub invasion are replaced by plantation silviculture tree plantings that further decrease light transmittance, native groundcover is reduced to greater extent than stands allowed to naturally regenerate following past timber harvest. Current and predicted expansion of pine plantation will particularly limit occupancy for amphibian SGCN endemic to sandhills, where planting dense pines lowers ambient soil moisture and juvenile survival. The vegetation metrics presented will allow land managers to guide forests toward conservation goals, to predict suitability of forests for amphibian species, and enhance success when repatriation efforts are needed. The significant relationship of amphibian occupancy to sensitive herbaceous vegetation (wiregrass) highlights that greater emphasis on forest groundcover is needed where amphibian SGCN occur and that cumulative impacts of forest management on native groundcover should be considered.