Terrestrial herpetofaunal community structure
along a faunal break in north central Florida

Jennifer S. Staiger & William J. Barichivich
USGS-Florida Integrated Science Center
7920 NW 71^st St., Gainesville, FL 32653

Presented at the Joint Meeting of Ichthyologists and Herpetologists
the week of July 10, 2006 in New Orleans, Louisiana.

Introduction and Background

In response to global declines and threats to amphibian populations, the US government implemented the national Amphibian Research and Monitoring Initiative (ARMI). The ARMI goal is to monitor amphibian populations on federal lands and evaluate potential causes of declines. The Southeast ARMI selected Lower Suwannee National Wildlife Refuge (NWR) as a research site in 2002. The refuge is of interest biogeographically because it spans a major river and is in a region where several amphibian and reptile species and subspecies reach the limits (at least along the Gulf coast) of their ranges.

Due to a paucity of historical data on the herpetofauna of the refuge, an inventory was initiated which included terrestrial sampling of both amphibians and reptiles. Preliminary results of this sampling suggested a difference in herpetofaunal community structure between sites on opposite sides of the river. The purpose of the analysis presented here is to determine if the river influences the structure of the terrestrial herpetofaunal community of Lower Suwannee NWR.

Fig. 1. Map of Florida with detail showing Lower Suwannee
NWR boundaries and Southeast ARMI drift fence sites.

Study Site

Lower Suwannee NWR stretches 42 km north to south in Levy and Dixie Counties along Florida's Big Bend region on the Gulf of Mexico (Fig. 1). The refuge's 21,425 hectares encompass a diversity of wetland and upland habitats at the lower reach of the Suwannee River, from salt marsh and maritime hammock to pine flatwoods and sandhill scrub. Historic agroforestry activity affected large parts of what is now the refuge. Up to 36 amphibian and 65 reptile species (excluding nonindigenous species) may be found in the area.

Methods

Amphibians and reptiles were sampled using drift fences with associated funnel traps and PVC pipe refugia at 10 sites, five north and five south of the Suwannee River (Fig. 1). The north sites represented three habitat types: pine plantation/hydric hammock (DC1), pine plantation/forest regeneration (DC2, 3), and forest regeneration sites (DC4, 5). The south sites represented four habitats: bottomland hardwood swamp (LC1), hydric hammock (LC2), pine plantation (LC3, 5), and maritime hammock (LC4). At eight sites, fence sections were arranged in a "Y" pattern, with six funnel traps and six PVC pipes. At two sites (DC3, LC3), arrays consisted of four fence sections, with eight funnel traps and eight PVC pipes, arranged parallel to the edges of isolated wetlands. Traps were checked daily during sampling periods between May 2003 and Feb 2005.

Bray-Curtis Similarity Index values were computed for amphibians and reptiles by site using non-transformed count data in program Primer-E (Beals, 1984; Clarke & Gorley, 2001; see Sorenson quantitative measure in Magurran, 1988). Similarity relationships were examined using two ordination methods, cluster and MDS, in this program. Additionally, the SIMPER routine in Primer-E was used to compute the overall percentage contribution each species made to the dissimilarity between groups determined in the cluster analysis.
 

Fig. 2. Hierarchical cluster dendrogram based on Bray-Curtis Index
values showing relationship of site and amphibian community.

Fig. 3. Hierarchical cluster dendrogram based on Bray-Curtis Index
values showing relationship of site and reptile community.

Table 1. SIMPER similarity percentages and species contributions to similarity.

Anolis carolinensis

Results

646 captures (excluding recaptures) of 34 species (20 reptile, 14 amphibian) at the 10 sites over 4876 trap nights.

North (Dixie) sites (Table 1):

• 167 captures of 22 species (11 reptile, 11 amphibian).  
• Hyla squirella captured most often (average abundance = 10.0).
• 7 species recorded from single captures.
• 7 species captured at these sites were not recorded from the south sites, though 2 of these were detected there by other methods.
• Average similarity of 51.28% for amphibians, 34.10% for reptiles.

South (Levy) sites (Table 1):

• 479 captures of 27 species (16 reptile, 11 amphibian).
• Rana sphenocephala with most captures (average abundance = 23.60) followed closely by H. squirella (average abundance = 23.40).
• 10 species recorded from single captures. 
• 12 species captured at these sites were not recorded from the north sites, though 4 were detected there by other methods.
• Average similarity of 68.84% for amphibians, 34.49% for reptiles.

The overall similarity between north and south sites is less than 50% for both amphibians and reptiles (Table 1).

Amphibians (Table 1, Fig. 2):

• 4 species contribute over 90% of the similarity between north sites.
• 5 species contribute over 90% of the similarity between south sites.
• The 3 shared dominant species contribute 53.15% of the dissimilarity across sides.
• Cluster analysis shows 2 north sites (DC4 & DC5) group with the south sites and 1 north site (DC1) is least similar to any other site.

Reptiles (Table 1, Fig. 3):

• 3 species contribute over 90% of the similarity between north sites.
• 6 species contribute this percentage for south sites. 
• The 2 shared dominant species contribute 20.32% of the dissimilarity across sides.
• Cluster analysis shows 1 north site (DC1) is least similar to any other site.

Conclusions

Average similarities, both overall and between sites on the same side of the river, are higher for the amphibian community than for reptiles.

Differences in capture rates across sites are greater for amphibian species than for reptile species. 

The dissimilarity in amphibian and reptile community structure at the sites may be influenced more by differences in capture rates than by the presence of the river.

There does seem to be a trend in the amphibian community of increased dissimilarity across sides moving downriver. This is possibly related to the increasing salinity gradient toward the mouth of the river.

The lower Suwannee River does not appear to act as a faunal break based on our drift fence sampling of the terrestrial herpetofauna of the refuge.

The limited scale of this study and suite of species encountered precludes a complete evaluation of the river as a regional faunal break.

Literature Cited

Beals, E.W. 1984. Bray-Curtis ordination: an effective strategy for analysis of multivariate ecological data. Advances in Ecological Research 14:1-56.

Clarke, K.R., & R.N. Gorley. 2001. PRIMER v5: User manual/tutorial. PRIMER-E, Plymouth, UK.

Magurran, A.E. 1988. Ecological diversity and its measurement. Princeton University Press, Princeton, NJ.

For more information: jstaiger@usgs.gov