2006 SOUTHEASTERN NATURALIST 5(2):277–284 Changes in Abundance of Gopher Tortoise Burrows at Cape Sable, Florida J. Hardin Waddle1,*, Frank J. Mazzotti2, and Kenneth G. Rice3 Abstract - The southernmost population of Gopherus polyphemus (Gopher Tortoises) occurs at Cape Sable, FL. The burrows of this population were surveyed in 1979 using strip transects to estimate Gopher Tortoise burrow density. We present data from a 1990 strip-transect survey and a 2001 line-transect survey of burrows for comparison. We found a significantly lower density of active burrows and a significantly higher density of abandoned burrows in 2001 compared to 1979 and 1990, but the overall density of burrows did not differ significantly over time. The distribution of burrow widths in 2001 was not significantly different from the 1982 distribution, and the bimodal pattern suggests that reproduction has occurred at the site. The 76% decline of active burrows at the site suggests that Gopher Tortoises have been subject to mortality. Reduction of habitat quality and tropical storms are possible explanations for the apparent decline in the Gopher Tortoise population at Cape Sable, but more monitoring is required to understand the long-term trends in this coastal population. Introduction The southern extent of the range of Gopherus polyphemus Daudin (Gopher Tortoise) is Cape Sable, FL. This population of Gopher Tortoises occurs in the palm savanna of East and Middle Cape Sable within Everglades National Park. Kushlan and Mazzotti (1984) surveyed the burrows of this population in 1979 and reported an estimated density of 11.3 tortoises/ ha using the burrow-to-tortoise conversion factor of 0.614 from Auffenberg and Franz (1982). McCoy and Mushinsky (1992) compared burrow abundance estimates at Cape Sable from 1978–79 and 1987–88, and found no evidence of a change in tortoise density during that time period. Many estimates of tortoise burrow densities appear in the literature (Auffenberg and Franz 1982, Breininger et al. 1994, Mushinsky and McCoy 1994, Stewart et al. 1993), but few offer published accounts of long-term trends of burrow density at a single site. Kushlan and Mazzotti (1984) suggested that the inherent instability of the coastal habitat and the periodic occurrence of hurricanes and other tropical storms could lead to fluctuations in the population size of Gopher Tortoises at Cape Sable. Our objective in 1Florida Cooperative Fish and Wildlife Research Unit, PO Box 110485, University of Florida, Gainesville, FL 32611. 2Department of Wildlife Ecology and Conservation, University of Florida-FLREC, 3205 College Avenue, Fort Lauderdale, FL 33314. 3US Geological Survey, Florida Integrated Science Centers, Center for Water and Restoration Studies, 3205 College Avenue, Fort Lauderdale, FL 33314. *Corresponding author - hardin_waddle@usgs.gov. 278 Southeastern Naturalist Vol. 5, No. 2 this study is to examine possible trends in Gopher Tortoise burrow density, activity status, and width at Cape Sable during a 22-year period from 1979 to 2001, using data from the survey by Kushlan and Mazzotti (1984) and two surveys conducted in 1990 and 2001. Methods Cape Sable is the southernmost point of mainland Florida, consisting of three main sandy capes (Northwest Cape, Middle Cape, and East Cape; Fig. 1) described in detail by Kushlan and Mazzotti (1984). Gopher Tortoises currently occur on both Middle Cape and East Cape. Although East Cape was surveyed in 2001, we only consider the burrows at Middle Cape in this analysis because it is the only area with three comparable samples and contains the largest population of tortoise burrows (Kushlan and Mazzotti 1984). The burrow surveys were concentrated in the open palm savanna located between the beach and the dense mangrove forest that fringes Lake Ingraham (Fig. 1). Using the vegetation classification of Everglades National Park available in a geographic information system (GIS; Madden et al. 1999), we calculated the area of open savanna grassland with scattered Sabal palmetto Lodd. ex Schultes (palms) at Middle Cape Sable to be 72.3 ha. Figure 1. Map of southern Florida showing location of Cape Sable within Everglades National Park with Northwest Cape, Middle Cape, and East Cape indicated. 2006 J.H. Waddle, F.J. Mazzotti, and K.G. Rice 279 The first survey of burrows at Middle Cape used in this study was conducted in August 1979. This survey was described in Kushlan and Mazzotti (1984), but only the density of all burrow types combined was reported. They searched 37 seven-m-wide strip transects completely for burrows. Transects were spaced 20 m apart, and ran from the beach across the open savanna to the line of dense mangrove forest. Each transect was oriented at a compass bearing of 48° from the beach. Because of the triangular shape of Middle Cape, transects varied in length from 10–1013 m. In February 1990, F.J. Mazzotti and others again surveyed Middle Cape for burrows. In this survey, 30 seven-m-wide strip transects were searched for burrows using the same technique as the 1979 survey. Although the 1990 transects were not in the exact locations of the 1979 transects, they were arranged in a similar manner along the same orientation. The length of transects in 1990 varied from 8–1208 m. In May 2001, we used a modification of the distance-estimation technique developed by Lohoefener (1990) and similar to that used by Hermann et al. (2002). Observers walked 31 line transects arranged in the manner of the strip transects in the 1979 and 1990 surveys. Transect lengths varied from 11–998 m. The perpendicular distance to all burrows observed from the line was measured to the nearest 0.1 m Each burrow in all three surveys was assigned to an activity status category following descriptions by Auffenberg and Franz (1982). Active burrows were distinguished by the presence of fresh tracks, scat, or other signs of recent use. Inactive burrows maintained the characteristic shape of a tortoise burrow, but lacked signs of recent use, and abandoned burrows were at least partially collapsed or filled with debris (Breininger et al. 1991, 1994; Mushinsky and McCoy 1994; Stewart et al. 1993). The density estimates of Gopher Tortoise burrows at Middle Cape Sable in 1979 and 1990 were calculated by dividing the number of burrows observed by the total area of the strip transects that was searched in either survey. This density estimate was multiplied by the total area of the tortoise habitat at Middle Cape Sable to produce an estimate of the number of burrows. The variance for this estimate and a 95% confidence interval was calculated using the equation for the ratio-estimator-of-abundance method described by Cochran (1977) for use when complete detectability of individuals is assumed on sample units of unequal area. Burrow densities for the 2001 sample were calculated using the program DISTANCE (version 4; Thomas et al. 1998), with encounter rate and detection function estimates pooled across transects. Density estimates of each type of burrow were compared among survey years using the compare surveys feature in the program AERIAL SURVEY (version 2000; Leban and Garton 1999), which performs a chi-square test on the null hypothesis that the density of burrows was the same in each year. The width of 93 burrows located within a 1.45-ha plot at Middle Cape Sable was measured in 1982 (Kushlan and Mazzotti 1984). Burrows were measured at 50 cm inside the entrance to adjust for the often variable width 280 Southeastern Naturalist Vol. 5, No. 2 of the mouths of burrows (Alford 1980). No burrow measurements were made in 1990, but a set of comparable burrow-width measurements was obtained from active and inactive burrows during the 2001 burrow survey. The distributions of burrow widths from 1982 and 2001 were compared using a two-sample Kolmogorov-Smirnov Z test, which tests whether two samples come from the same distribution (SPSS 2001). Results During the 1979 survey at Middle Cape, 11.16 ha was searched for Gopher Tortoise burrows. A total of 270 burrows was located, of which 162 were active, 69 were inactive, and 39 were abandoned. In 1990, a total area of 11.25 ha was searched for burrows and 321 were found; of these, 208 were active, 97 were inactive, and 16 were abandoned. The methods employed in the 2001 survey do not provide an exact area searched, but 275 burrows were detected of which 71 were active, 114 were inactive, and 90 were abandoned. The overall effective strip half-width (the effective detection area for line transects) was 6.71 m (95% C.I. = 6.00–7.50), and the burrow detection probability was 0.25 (95% C.I. = 0.22–0.28; Buckland et al. 2001). The estimated densities of active Gopher Tortoise burrows at Middle Cape in 1979 and 1990 were 14.51 and 18.49 burrows/ha respectively (Table 1). The estimated density of active burrows in 2001 was 4.38 burrows/ ha. When these estimated burrow densities are multiplied by the area of suitable habitat at Middle Cape, the estimated total number of active burrows is 1049, 1337, and 317 in 1979, 1990, and 2001, respectively (Table 2). The survey comparison test (Leban and Garton 1999) concluded that the density of active burrows in 2001 was significantly lower than densities in 1979 and 1990 ( χ2 = 14.0955, df = , p < 0.001). The density of Table 1. Estimated density of burrows in each activity status category at Middle Cape for the 1979, 1990, and 2001 surveys. Survey Est. density of Est. density of Est. density of Est. density of year active burrows inactive burrows abandoned burrows all burrows 1979 14.51/ha 6.18/ha 3.49/ha 24.19/ha 1990 18.49/ha 8.62/ha 1.42/ha 28.54/ha 2001 4.38/ha 7.84/ha 6.93/ha 19.08/ha Table 2. The estimated number of burrows of each activity status category occurring in the 72.3- ha palm savanna habitat at Middle Cape in 1979, 1990, and 2001. The 95% confidence interval of the estimate is shown in parentheses. Survey Est. number of Est. number of Est. number of Est. number of year active burrows inactive burrows abandoned burrows all burrows 1979 1049 (566–1533) 447 (203–690) 253 (97–408) 1749 (923–2575) 1990 1337 (615–2059) 624 (356–891) 103 (4–201) 2063 (1121–3005) 2001 317 (186–538) 566 (392–819) 501 (370–678) 1379 (1089–1747) 2006 J.H. Waddle, F.J. Mazzotti, and K.G. Rice 281 abandoned burrows was significantly greater in 2001 than in 1979 and 1990 (χ2 = 19.1289, df = 2, p < 0.001). No significant differences in burrow density existed among years for the inactive-burrow status category (χ2 = 0.9862, df = 2, p = 0.6107) or for the estimate of all burrows (χ2 = 2.2638, df = 2, p = 0.3224). We measured the width of 180 active and inactive Gopher Tortoise burrows in 2001. The frequency distribution of burrow widths in 2001 was not significantly different in shape (Kolmogorov-Smirnov Z = 1.019, p = 0.25) from the 1982 distribution measured by Kushlan and Mazzotti (1984). The 2001 distribution is bimodal, with peaks at 10 cm and 30 cm (Fig. 2). Discussion The density of all burrows was not significantly lower in 2001 than in 1979 or 1990, but we found significantly fewer active Gopher Tortoise burrows at Middle Cape in 2001. We have no direct estimates of Gopher Figure 2. Frequency distribution of 180 Gopher Tortoise burrow widths measured at Middle Cape Sable in 2001. 282 Southeastern Naturalist Vol. 5, No. 2 Tortoise abundance from any of the three samples, but the 76% decline in the number of active burrows during the 11-year period from 1990 to 2001 suggests that there was a decline in the number of Gopher Tortoises at Middle Cape or a decrease in tortoise activity. Air temperatures were warm during the 2001 sample (maximum 32 ºC) and May is well into the tortoise activity season (Eubanks et al. 2003), making it unlikely the occupied burrows were scored as inactive or abandoned. The insular nature of Middle Cape Sable makes it unlikely that emigration of Gopher Tortoises from the area occurred, and the distribution of burrow widths suggests that recruitment into the population from reproduction is occurring at the site. Thus, emigration and lack of reproduction are unlikely explanations for the population decline. We conclude that mortality at the site is the best explanation for the apparent decline in Gopher Tortoises from 1990 to 2001. Although the sampling methods employed in our third survey of Gopher Tortoise burrows at Cape Sable are slightly different than the first two surveys, the results are directly comparable. Both methods produce estimates of the density of burrows. Additionally, we calculated confidence intervals for the estimated numbers of burrows at Cape Sable from the first two surveys for statistical comparison to the 2001 data. The variance estimates from the 1979 and 1990 strip transects resulted in wider confidence intervals compared to the confidence intervals estimated with distance sampling. The compare surveys feature in the program AERIAL SURVEY is considered to be a conservative test, especially when confidence intervals are wide (Leban and Garton 1999). This underscores the importance of the significant decrease in the numbers of active burrows and increase of abandoned burrows between 1990 and 2001. McCoy and Mushinsky (1992) examined changes in Gopher Tortoise abundance at Cape Sable by comparing data on burrow abundance and activity status collected in 1978–79 to data collected in 1987–88. They found an increase in the total number of burrows and a slight increase in the percentage of active burrows during the 10-year period, a pattern similar to the one observed in this study between the 1979 and 1990 samples. Both this study and that of McCoy and Mushinsky (1992) indicate that the Gopher Tortoise population at Cape Sable was stable or slowly increasing in the decade prior to 1990. Aresco and Guyer (1999) found that Gopher Tortoises in southern Alabama abandoned their burrows at a rate of 22% per year during a fiveyear study, and they found a significant positive correlation between age of an active burrow and tree density and basal area. With the exception of a few small hammocks, the Gopher Tortoise habitat at Middle Cape Sable is essentially treeless, so canopy closure is an unlikely cause of the increased number of abandoned burrows at Cape Sable in 2001 relative to earlier surveys. Guyer and Hermann (1997) found no difference in the longevity of Gopher Tortoise burrows based on soil type or root structure at pine forest sites in Georgia and Alabama. There is evidence that 2006 J.H. Waddle, F.J. Mazzotti, and K.G. Rice 283 tortoises in disturbed sites will readily abandon burrows to migrate to areas of better forage as conditions change (Aresco and Guyer 1999, Guyer and Hermann 1997). It is possible that tortoises at Cape Sable abandoned burrows at a higher rate in 2001 than previously described because of deteriorating habitat conditions. Kushlan and Mazzotti (1984) noted that large tropical storms might be an important cause of mortality in this coastal population of Gopher Tortoises. From 1970 to 1990, no major tropical storms passed within 25 km of Cape Sable. However, in 1992 the eye of Hurricane Andrew passed within 45 km of Cape Sable as a category 3 hurricane with sustained winds of 202 km/h (Pimm et al. 1994), and in 1999 the eye of Hurricane Irene passed within 10 km of Cape Sable as a category 1 hurricane with sustained winds of 120 km/h (Avila 1999). It is possible that a storm surge associated with either or both of these hurricanes caused tortoise mortality at Cape Sable, but because no survey of the Gopher Tortoises at Cape Sable was conducted directly before or after these hurricane events, the impact of hurricanes on the Gopher Tortoises at Cape Sable remains speculative. It appears that the suggestion by Kushlan and Mazzotti (1984) that instability in coastal areas may cause fluctuations in tortoise populations was correct. There has been a significant decrease in active Gopher Tortoise burrows despite evidence of reproduction at Cape Sable between 1990 and 2001. Reduction of habitat quality or the occurrences of tropical storms are possible causes of the apparent decline in Gopher Tortoises at Cape Sable. Continued monitoring of this population would provide more information on the long-term dynamics of survival and reproduction at this site. More frequent burrow surveys and surveys after major storm events would provide insight about the vulnerability of Gopher Tortoises at Cape Sable to tropical storms. Acknowledgments M. Deacon, A. Watts, C. Zweig, and two anonymous reviewers provided valuable editorial comments on this manuscript. This project would not have been possible without help in the field provided by M. Caudill, M. Cherkiss, G. Cook, M. Crockett, M. Delong, A. Dove, P. George, J. Graham, L. Hord, S. Howarter, B. Jeffery, A. Maskell, L. McKercher, C. Vischer, and J. Williams. Surveys were conducted under NPS permit #2000074. Literature Cited Alford, R.A. 1980. Population structure of Gopherus polyphemus in northern Florida. Journal of Herpetology 14:177–182. Aresco, M.J., and C. Guyer. 1999. Burrow abandonment by Gopher Tortoises in slash pine plantations of the Conecuh National Forest. Journal of Wildlife Management 63:26–35. Auffenberg, W., and R. Franz. 1982. The status and distribution of the Gopher Tortoise (Gopherus polyphemus). Wildlife Research Reports 12:95–126. 284 Southeastern Naturalist Vol. 5, No. 2 Avila, L.A. 1999. Preliminary report: Hurricane Irene 13–19 October 1999. National Hurricane Center, Miami, FL. Breininger, D.R., P.A. Schmalzer, and C.R. Hinkle. 1991. Estimating occupancy of Gopher Tortoise (Gopherus polyphemus) burrows in coastal scrub and slash pine flatwoods. Journal of Herpetology 25:317–321. Breininger, D.R., P.A. Schmalzer, and C.R. Hinkle. 1994. Gopher Tortoise (Gopherus polyphemus) densities in coastal scrub and slash pine flatwoods in Florida. Journal of Herpetology 28:60–65. Buckland, S.T., D.R. Anderson, K.P. Burnham, J.L. Laake, D.L. Borchers, and L. Thomas. 2001. Introduction to Distance Sampling: Estimating Abundance of Biological Populations. Oxford University Press, New York, NY. Cochran, W.G. 1977. Sampling Techniques. Wiley, New York, NY. Eubanks, J.O., W.K. Michener, and C. Guyer. 2003. Patterns of movement and burrow use in a population of Gopher Tortoises (Gopherus polyphemus). Herpetologica 59:311–321. Guyer, C., and S.M. Hermann. 1997. Patterns of size and longevity of Gopher Tortoise (Gopherus polyphemus) burrows: Implications for the longleaf pine Ecosystem. Chelonian Conservation and Biology 2:507–513. Hermann, S.M., C. Guyer, J.H. Waddle, and M.G. Nelms. 2002. Sampling on private property to evaluate population status and effects of land use practices on the Gopher Tortoise, Gopherus polyphemus. Biological Conservation 108:289–298. Kushlan, J.A., and F.J. Mazzotti. 1984. Environmental effects of a coastal population of Gopher Tortoises. Journal of Herpetology 18:231–239. Leban, F., and O. Garton. 1999. Aerial Survey 2000. University of Idaho, Wildlife Resources, Moscow, ID. Available at: http://www.cnr.uidaho.edu/fishwild/programs/ Aerial%20Survey%206.zip. Lohoefener, R. 1990. Line transect estimation of Gopher Tortoise burrow density using Fourier series. Pp 44–69, In C.K. Dodd, Jr., R.E. Ashton, R. Franz, and E. Wester (Eds.). Proceedings: Eighth Annual Meeting of the Gopher Tortoise Council. Florida Museum of Natural History, Gainesville, FL. Madden, M., D. Jones, and L. Vilchek. 1999. Photointerpretation key for the Everglades vegetation classification system. Photogrammetric Engineering and Remote Sensing 65:171–177. McCoy, E.D., and H.R. Mushinsky. 1992. Studying a species in decline: Changes in populations of the Gopher Tortoise on federal lands in Florida. Florida Scientist 55:116–125. Mushinsky, H.R., and E.D. McCoy. 1994. Comparison of Gopher Tortoise populations on islands and on the mainland in Florida. Fish and Wildlife Research 13:39–48. Pimm, S.L., G.E. Davis, L. Loope, C.T. Roman, T.J. Smith III, and J.T. Tilmant. 1994. Hurricane Andrew. BioScience 44:224–229. SPSS Inc. 2001. SPSS for Windows, Release 11.0.0. Chicago, IL. Stewart, M.C., D.F. Austin, and G.R. Bourne. 1993. Habitat structure and the dispersion of Gopher Tortoises on a nature preserve. Florida Scientist 56:70–81.