p m r- r-' ME FYF-11 - - T r r.- 1. 4,6*. of the FLORIDA MUSEUM OF NATURAL HISTORY THE COMPARATIVE ECOLOGY OF BOBCAT, BLACK BEAR, AND FLORIDA PANTHER IN SOUTH FLORIDA David Steffen Maehr Volume 40, No. 1, pf 1-176 1997 == 46 1ms 34 i " 4 '· 0?1~ I. Al' Ai: *'%, R' I.' I / Em/-.Ail-%- .1/9" . -_____- UNIVERSITY OF FLORIDA GAINESVILLE Numbers of the BULLETIN OF THE FLORIDA MUSEUM OF NATURAL HISTORY am published at irregular intervals Volumes contain about 300 pages and are not necessarily completed in any one calendar year. JOHN F. EISENBERG, EDITOR RICHARD FRANZ CO-EDIWR RHODA J. BRYANT, A£ANAGING EMOR Communications concerning purchase or exchange of the publications and all manuscripts should be addressed to: Managing Editor. Bulletin; Florida Museum of Natural Histoty, University of Florida P. O. Box 117800, Gainesville FL 32611-7800; US.A This journal is printed on recycled paper. ISSN: 0071-6154 CODEN: BF 5BAS Publication date: October 1, 1997 Price: $ 10.00 Frontispiece: Female Florida panther #32 treed by hounds in a laurel oak at the site of her first capture on the Florida Panther National Wildlife Refuge in central Collier County, 3 February 1989. Photograph by David S. Maehr. THE COMPARATIVE ECOLOGY OF BOBCAT, BLACK BEAR, AND FLORIDA PANTHER IN SOUTH FLORIDA David Steffen Maehri ABSTRACT Comparisons of food habits, habitat use, and movements revealed a low probability for competitive interactions among bobcat (Lynx ndia). Florida panther (Puma concotor cooi 1 and black bear (Urns amencanus) in South Florida. All three species preferred upland forests but ©onsumed different foods and utilized the landscape in ways that resulted in ecological separation. Further, panthers exhibited crepuscular activity whereas black bears were predominantly diurnal. Diet, movernents, and reproduction varied seasonally among speciel Subadults of all three species demonstrated extensive dispersal abilities, but only male black bears were documented to have crossed the Caloosahatchee River, a potential landscape barrier that may restrict effective dispersal northward in bobcats and panthem BecAllte bobcat and black bear in South Florida occur at relatively high densities. anthropogenic changes to thclandscape and sea level rise willaffect them less severely thanpanther. The problems associated with the habitation of a naturally fragmented and patchy forest are exacerbated by the conversion ofproductive habitat types to types that amavoided. Another factor'' - ' '' ' "'yofcoological relations among this carnivore community is the range ell:adon ofthc coyote (Cani, latians) into South Florida. This canid is known to exhibit interference competition with bobcats, black bears, and panthers in other parts of North America. The diet ofthe coyote in Florida may overlap with the dicts of the three native carnivores by at least 38 percent and asmuch as 64 percent The highest concentrations of black bears and panthers in South Florida Loincide with an extensive forest, a landscape feature that accounts for only a small proportion of public land. Increasing forest fragmentation from the Sarasota area southeastward suggests that most public lands are relatively unimportant to the two larger species Because the demographics of even the.mallest ofthese populations (panther) are shown to be typical of healthy populations, creative management, such as flexible reserve boundaries and the enlistment ofprivate property owners in conservation efforts, may be of more immediate value than symptom-oriented management practices such as genetic introgression. RESUMO Comparagaes do habito alimentar, uso de habitat e movimentos revelaram uma bain probabilidade de interaf6es competitivas entre o «bobcat"(Lynx ndiu), o puma americano (Puma concolor cood, e o urso negro (Urms amencanus) no sul da Florida. As tr68 cap6cies preferiram florestas de tcrras altas mas consumiram diferentes tipos de alimento e utilizaram o espa90 de uma modo que regultou em separa40 =016gica. 0 puma Boresentou atividades crepusculares, enquanto que o urso· negro fbi predominantemente de habito diurno. Dieta, movimentos, e reprodu#o variaram sasonalmente entre as espkies. Subadultos das ' The author & Assistant Professor, Foresuy, C - _ - -- ' '„Lecington KY 40546-0073 U.SA MAEHR, D.S. 1997. The comparative ecology ofbobcat. black bear, and Florida panther in South Florida Bull. Florida Mus. Nat. Hist. 40(1):1-176. 2 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) trBs esp6cies demonstraram grande habilidade de dispersao, mas somente machos do uno negro foram documentados atravessando o Rio Caloosahatchee, unia potencial bamira na paisagem que talvez'limite uma dispesao efetiva pan o norte em "bobcals"e purnai Uma vez que no sul da Florida "bobcats" e ursos negros ocorrem em densidades relativame~e altas, mudangas antr6picas na paisagem, somadas a um possivel aumento no nivel das mards. irao afeta-los de maneira menos severa doquc o puma. Os problemas associados com a habits* dedreas naturalmente fragmen,9,1.. e florestas isoladas sao exacerbados pela conversao de habitas produtivos para tipos habitas que sao evitados por cssas ess,6cies. Outro fator queameagaaestabilidade das relag6cs cool6gicas entre esta comunidade de carnivoros 6 a expansao da distribui~ao do coyote (Cani, latrans) que vem ovorrendo no sul da Florida. £ sabido que esse canideo compete com "bobcats" ursos c pumss em outras parts (la An*ica do Norte. A dieta do coyote na F16rida talvez se sobreponha com a dieta desses tres carnivoros natives em pelo menos 38% podendo chegar a16 a 6404 Incais de alta concentragao de urses negrosepumas no gui da F16rida coincidem com florestas extenvas, uma caracteristica de paisagem que ocorrem em pequena proporgao em terms pOblicas. 0 mini%to da fragmentafao das florestas na drea de Sarasota em diregao sul, sugere que a maioria das terras pablicas sao relativamente menos importantes para as duas csp6cics maiores. Devido aos fatores demogr~ficos at,1 para a menor dessas popula~es (punia) se mostrarem ser tipicos de popula0es sadias, o mancio criativo. tais como limits mais fiexiveis das reservas, e o cadastramento de proprietdrios de terras privadas em esforfos conservacionistas, talvcz tenham valor mais imediato do que pdticas de manejo sintomdticas, como por exemplo introgressao genatica. TABLE OF CONTENTS Acknowledgemmk 3 1. Introduction........................................................................................................................................... Overview ofCamivore Community Studies.................................................................................. - Study A- 9 Summary of Previous Large Carnivore Work in Florida 13 General Methodology 18 2. Dietary Overi,r 21 Methods................................................,..............„........,................................................................. 21 Results and Discussioa................................................................................................................ 22 Species Compari- 27 3. Habitat Use Among South Florida's Large Mammalian Carnivores. ..........'........~.........~...........~........ Mel]¤Is........................................................................................................................................... Results and Discussion.................................................................................................................... 4. Habilat Pattern around Florida Panther Natal Dens and Black Bear Winter Dem............................. = * A S S =Method. Results and Discussion..............................„......„.......„........................................,......................... Influence of Roark 65 5. Home Range, Activity, and Land Tenure 73 Methods 74 Results and Discussion. 75 Seasonal Effects 76 Home Range Fidelity and Replacement 81 Overlap Within and Among Rperi- 83 6. Dispersal Characteridie« 117 118 Results and Discussion...................................................,*............................................................... 118 Adults and Dispersers Compared............................................................................................... 120 7. Synthesis and Conclusions................................................................................................................... 144 The Rellim of Wild Canids to Florida......... .. .......................................................... 147 A Longer Range View 150 Literature Cited......................................................................................·········································· 161 Appendix A. .................I.-M-....*..........*.............A.............."..m...."m...................~....'....................'.......... 173 Appendix B. ......~.. --m.-I~~eee~~,m.. -le-m-g---*mil-Mm~Mm-mM--m#-m-----I-'-Immmm. 175 MAEHR: ECOLOGY BOBCAT, BLACK BEAR, PANTHER 3 ACKNOWLEDGEMENTS Many people have contributed to our current understanding of large carnivore ecology in South Florida. Notable in this endeavor are E.D. Land, Todd Logan, J.C. Roof J.W. McCown, O.L Bass, and J.N. Layne. R.T. McBride. and frequently his sons Rocky and Rowdy, linked its with study in;„,i# and maintained long-term contact with many ofthem. I amalsogratefulfortheopportunity tohave woked with M.E. Roelke, C. Glass, M.Lamm, T. Ruth, S. Citino, and especially J. Lanier, all of whom helped to look after anesthetized bobcats and panthers. J. Kappes, R. Bell, and P. Baronoff were important field assistants who lent their unique personalities to the project Diane Maehr. C. Machr. and E Machr were special assistants during several captures ofbobcats. panthers and black beam I thank the Florida Game and Fresh Water Fish Commission for allowing me to be a part of its celebrated research efforts on the Florida panther and for consistent funding of this important field work. The U.S. Fish and Wildlife Service, Florida Department of En< ' 1 Protection, National Park Savice. and the Seminole Indian Tribe facilitated access to their vast holdings of publicly owned land. Of equal, if not greater, importance was the access to 1 . , including B. Lester, B.H. Griffin, E Carlson, E Engiish, L Gable. 11 Collier, M. Scofield, ~ Fitzpatfick, T. Baker, J. McI)aniel, D. Pylant. M. R.amsey. J. Hendric, E. and J. Smoak, J. Hilliard, and J. Billie M. Sunquist and J. Eisenberg challenged me with provocative ideas and helped draw new ideas out of me. I am especially indebted to LD. Harris who was one of my first acquaintances in Florida and *m continues to be a friend and my most ardent supporter. His visions ofthe way things were and the way they should be have benefitted many ecological disciplines and everyone he has worked with. His abilities to compel, cajole, and convince are equally matched by his willingness to li,den, learn, and ,*s.,ge. I thank G. Tanner and S. Humphrey, and Bulletin editors J. E;=,her& R. Bryant. and It Franz for facilitating the production of this publication. Reviewers J. Seidensticker and H. Quigley provided valuable comments that improved the manuscript Others who have helped me along this path in meaningful ways include T. "Birdman" Hector. K. Deagan, J. Co~ F. Percival, P. Hall, M. Moulton, C. Arnold. R Hellgren. C. Abercrombie. J. Mullahey, E. Jacobson, W. Kitchens, F. Mazzotti, and J. -I,oquacious" Schorterneyer. Publication of this bulletin was made possible through the generous cont,ibutions of Scott Harris Naples, Florida, a grant from the Turner Foundation, Inc., by way of the Florida Stewardship Foundation and the University of Florida. This is journal series number R-05638 ofthe Florida Agricultural Experiment Station. It is connected with a project of the Kentucky Agricultural Station (No. 97-09-62) and is published with the approval ofthe director. 4 BULLETIN FLOIUDA MUSEUM NATURAL HISTORY VOL 40(1) 1. INTRODUCTION The terrestrial mammalian carnivore communities of pre-European temperate North America consisted of 35 species belonging to five families (Table 1.1). The prehistoric distribution, abundance, and associations of these species were a function of the interaction of climate, plant succession, competitors, and prey demographics (Harris 1988), factors that are all directly linked to geography and productivity of the landscape (Harris 1984:11-23; Zonneveld 1990). Although primitive humans (Homo sapiens) were a significant competitor with and predator on many of these species post-Columbian humans have been much more effective than their predecessors in reducing native carnivore abundance and diversity throughout the continent (Diamond 1992). Modern local extinctions of large carnivores and a relative overabundance of medium-sized mammals uiroughout the southeastern United States have resulted from anthropogenic influences that include species introductions, over-harvest, and habitat fragmentation. The study of large carnivores in Noith America during the 20th century has evolved from a predator control philosophy to an ecological paradigm that includes carnivores as integral components of community and landscape processes. Leopold's (1949) vision of enlightened wolf and landscape management seems to have become a modern standard for several ecological disciplines. Indeed special volumes of Transactions of the North American Wildlife and Natural Resources Conference (Vol. 56; 1991), Conservation Biology (Vol. 10, No. 4; 1996), and the Wildlife Society Bulletin (Vol. 24, No. 3; 1996) emphasize the scientific and popular roles that predators now play in both domestic and international environmental policy. Restoration of carnivore populations has now replaced efforts to eradicate them as conservation professionals and the public learn to accept the value of large, natural areas and the wide-ranging animals that live in them (Clark et al. 1996; Mech 1996). Eleven mammalian carnivores existed in post-Pleistocene Florida (Table 1.1). The red wolf (Canis rufus) has been extirpated due to overharvest and habitat alterations (Nowak 1991; Robson 1992), and the Caribbean monk seal Bfonac/ms tropicalis) is extinct primarily because of overharvest (Wing 1992). The remaining species represent a nearly intact assemblage that has persisted to the present, despite Florida's quickly growing human population. It is a testimony to the difficulties of settling a wet, hot, and flat landscape that three out of four large (>10 kg) terrestrial carnivores persist in South Florida. It has been only 70 years since highways were built to bisect this previously impenetrable wilderness (Carter 1974), and there is nowhere else in eastern North America where bobcats (Lynx rufus), black bears (Ursus americanus), and Florida panthers (Puma concolor cocy,) continue to co-exist Although the three large carnivores native to South Florida have been studied previously, none of the ecological studies exceeds more than a few years. Further, MAEHR: ECOLOGY BOBCAT, BLACK BEAR PANTHER 5 no analyses have examined them as an interacting community that inhabits the same landscape. Therefore, the objectives of this study were as follows: 1) Describe the spatial dynamics and habitat requirements of resident adult panthers, black bears, and bobcats in a rapidly developing South Florida landscape; 2) Analyze, compare, and contrast the use of space of resident Carnivores with that of dispersing subadults; and 3) Discuss the long-term prospects for the large carnivore community in South Florida with respect to landscape and biotic changes that are under way. Overview of Carnivore Community Studies Because of the difficulties in studying large carnivores, investigations of multi-species predator communities are sparse relative to studies of individual species. In addition, few detailed studies of multiple-species carnivore assemblages have occurred in settings conducive to direct observation. Schaller's (1972) study of the African lion (Panthera leo) included detailed accounts of interactions among five Serengeti predators. Competition appeared to be reduced in that community by differences in habitat use, temporal separation, prey preferences, prey size, and hunting methods. Kruuk and Turner (1967), Bertram (1979), and Hanby and Bygott (1979) observed similar patterns among lion, leopard (P. pardus), cheetah (Acinonyxjubatus), and wild dog (Lycaon pictus) in the Serengeti. Mills (1984) found that the four large predators in the Kalahari coexisted by reducing competition. Although lion and spotted hyena (Crocuta crocuta) were mostly nocturnal and fed on similar species, they targeted different sex and age classes. Cheetah and leopard, on the other hand, exhibited temporal habitat separation and the leopard had a more varied diet than its likely competitors. On the other hand, in the Kalahari Gemsbok National Park, extremely high dietary overlap among the lion, leopard, cheetah, and spotted hyena was caused by relatively low prey species diversity (Eloff 1973). Mills and Mills (1982) found that the brown hyena (Hyaena brunnea) and spotted hyena became direct competitors only when both were forced to rely on scavenging. Inverse relations in abundance have been observed or inferred between pairs of Old World predators. Myers (1977) found that cheetah were more abundant where spotted hyena were absent or scarce. When both species competed for Thompson's gazelle (Gaze#a thomsoni), the cheetah was at a disadvantage and usually declined in number. Seidensticker (1976) found that tiger (Panthera tigris) and leopard in Nepal exhibited nearly total ecological separation from each other by consuming different-sized prey, by using different habitat, and by exhibiting different patterns ofactivity. Only where prey were abundant did these two species coexist in similar habitats. A study of leopard and caracal (Fe/is caracal) in South Africa showed 6 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) that these species avoided competition by utilizing mutually exclusive habitat types (Norton and Lawson 1985). In South America direct observations of interactions between species are nearly impossible due to the dense vegetation that many forest carnivores inhabit At the same time, the very environmental conditions that hamper the development of ethograms, have likely affected the way these species utilize their landscape. Interestingly, while most studies of carnivore communities in the Old World tropics inferred several levels of conlfetition among sympatric vertebrate predators the opposite appears to be the case in the New World tropics. Konecny (1989) ex:,mined a small carnivore community in Belize that lacked obvious competitive interactions. Jaguarundi (Felis jaguaroundi), tayra (Era barbara), ocelot (Felis pardalis), and margay (Felis weidif) coexisted *Whout using similar habitats and with little dietary overlap. Puma, jaguar, and ocelot avoided competition by means of prey size partitioning and habitat specialization (Schaller and Crawshaw 1980; Rabinowitz and Nottingham 1986; Emmons 1987). Sunquist et at. (1989) found that ocelot, hog-nosed skiink tayra, grison (Galictis vittata), and crab-eating fox (Cerdocyon thous) avoided competition in Venezuelan 11anos through diet partitioning. The importance of the grizzly bear (Ursus arctos) in the sclerophyll community of westcrn North America was considered sufficient to include its common name as part of Shelford's (1963) ecological classification of North America Its predominance resulted in the virtual exclusion of the black bear. But through time and coincident with the decline of its larger competitor, the black bear is now widespread in this part of the grinly bear's former range. Herrero (1978) suggested that evolutionary processes resulted in differences in form and behavior between black bear and brown bear. Larger size, more aggressive behavior, adaptations for digging, and the inability to climb trees suit the brown bear to life in more open habitats than the forest-dwelling tree-climbing, less aggressive black bear. The separation of these species was maintained by historic patterns of forest cover and the brown bear's dominance over the black bear. In most cases where oven interactions have been reported, the brown bear was dominant (Mattson et al. 1992; Ross et al. 1988). Giant panda Wiluropoda melanoleuca) and Asiatic black bear (Unus thibetanus) exhibit a high degree of spatial overlap and similarity in size and form yet exhibit divergent food habits and feeding strategies (Schaller et al. 1989). Despite having similar digestive systems, the giant panda is a food specialist while the Asiatic black bear is a food generalist. No competitive interactions between these species have been reported Johnson et al. (1988) reported that red panda Wilurus fulgens) and giant panda overlapped in space but have evolved very different energetic and behavioral strategies that allow them to utilize different plant parts. Sympatric carnivores throughout the world exhibit a multitude of strategies for separating themselves in environments of limited resources. When a common MAEHR: ECOLOGY BOBCAT, BLACK BEAR, PANTHER 7 resource is utilized by more than one species, co«currence is facilitated by differences in habitat use, and/or activity pattern. These patterns change with sizc of prey, sizc of predator, number of potential competitors (Rosenzweig 1966), climate, group size, and human influences. Schoener (1974) geneplimt thnt resource partitioning was most often accomplished by means of separation along habitat dimensions rather than temporal dimensions. However, his review focused primarily on invertebrates, birds, and small mammals and did not consider the diversity of species aggregations characteristic of mnmmnlian carnivores. CMe and Gilpin (1974:3076) suggested that the relative costs of exploitation- versus interference-competition favored the latter in pat because "the contraction from the fundamental niche to the realized niche is likely to be small for an interference competitor and high.for an exploitation competitor." This pattern appears evident in the Old World tropics where dominance hierarchies among predators have been frequently obser~d, but it is less apparent in the New World tropics. More diverse landscape features, most specifically topography and vegetation, have offered very different milieus for community evolution and likely have exerted a powerful force on the nature of resource partitioning among sympatric carnivores. At the risk of oversimplification, communities evolving in landscapes domiruted by unforested expanses (e.g., East Africa) tend to exhibit more interference competition than species complexes coevolving in dense expanses of forest cover (e.g., South America, Southeast Asia). The aggressive and dominating nature of the brown bear in lightly forested terrain is an example of this process in North America. For other Nonh American carnivore communities, human-caused changes to the lanrkmpe have affected the patterns of community organization and resource partitioning through recent losses and additions to local carnivore faunas. South Florida offers a variety of land cover types including expansive areas of open, herbaceous vegetation and extensive systems of dense forest. The three remaining species of'large terrestrial carnivores native to this area all confine most activities to plant communities contained within or immediately adjacent to forest cover (Maehr et al. 1991ai Foster 1992). The recently extirpated red wolf may have made more use of relatively open terrain, a trait that may have facilitated its demise. In recent years the coyote has become a more noticeable component of the current carnivore assemblage, but little is known about its diet, distribution, and habitat needs in South Florida Studies of large carnivores consistently support the notion that the conservation of these species is a landscape-level issue. Although proposed solutions to the problem of shrinking wildlife habitat have stimulated debate (Harris and Gallagher 1989; Simberloff et al. 1992), all potential approaches are land extensive. Spatial requirements of Florida panthers and black bears are enormous. Annual home ranges of individual adult male panthers can e*ceed 500 km2 (Machr et al. 199la), and one-way dispersal movements of black bears can exceed 140 km (Maehr et al. 1988). Telemetry studies of South Florida carnivores span 15 years, yet government agencies are just now attempting to apply these 8 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) findings to management (e.g., Cox et al. 1994). Progress in landscape-level species management is limited but includes construction of wildlife underpasses and the purchase of the Florida Panther National Wildlife Refuge in Collier County. Unfortunately, the treatment of symptoms will fail to correct the root problem facing terrestrial carnivores in Florida: large scale alteration of the landscape. Inexorable human development of private lands in South Florida has the potential to eliminate 45 percent of presently occupied panther range and reduce the existing population by over 50 percent (Maehr 1990). Although the black bears of Collier County appear to be tolerant of many anthropogenic changes to the landscape, most people will not tolerate their presence. The result is a high rate of mortality and injury to bears inhabiting the urban/wilderness interface. Bobcats are commonly reported as predators of domestic livestock throughout Collier County and can still be legally eliminated consequent to these depredations. All three species have been studied extensively throughout their ranges (Anderson 1983; Anderson 1987; Pelton 1982) in North America. In Florida published bobcat investigations mostly in the vicinity of the Lake Wales Ridge, span two decades. However, concern over the impact of the fur trade on bobcats (National Wildlife Federation 1977) led to extensive fieldwork to detail population status and trends (Florida Game and Fresh Water Fish Commission, unpubl. data). These activities stimulated a statewide food habits analysis that was based on collections of stomachs obtained by trappers (Maehr and Brady 1986). In North Florida Conner (1982) and Progulske (1982) examined population estimation techniques and movements, respectively. Foster (1992) described South Florida. bobcat home range characteristics in conjunction with an evaluation of highway underpass effectiveness. Papers detailing a variety of Florida black bear subjects span four decades. Like bobcats, most of these studies were conducted outside of South Florida, but they covered a greater diversity of topics. Although the greatest political issue following the listing in 1974 of black bear as a threatened species was sport hunting (Maehr and Wooding 1992), most studies in the state have focused on basic natural history and bear/human conflicts. Compared to bobcatq. black bears in Florida have been the subject of more work on diseases, parasites, and non- hunting management issues. In view of their rarity, Florida panthers have received an inordinate amount of scientific attention, .with technical literature dating to 1950. Since then, publications on the basic natural history of this federal- and state-listed endangered subspecies have evolved into discussions of controversial issues ranging from property rights to genetic restoration. More than half of the published literature on Florida panthers appeared after 1990, so few generalizations can be made about the panther's historical distribution. There remain few unstudied aspects of the modern panther's biology or ecology; however, recent management has consisted of sporadic efforts to treat symptoms associated with grnall population Ri72 rather MAEHR: ECOLOGY BOBCAT, BLACK BEAR, PANTHER 9 than addressing the basic reasons for its current status or necessary steps for restoration. Relative to most other species of terrestrial vertebrates in Florida panthers, black bears, and bobcats exist at low densities. A high degree of dispersion has contributed to a low frequency of epizootics although individuals of all three species are susceptible to a number of diseases, and Florida bobcat populations are known to have suffered locally severe disease outbreaks (Wassmer et al. 1988; Progulske 1982). Forrester (1992) examined the disease occurrence in these species and suggested an inverse relation between body size and the likelihood of disease. Bobcats have experienced temporary, local extinctions, while black bears appear relatively disease-free. Because a spatially influenced resistance to disease may improve the survival probabilities of large, solitary carnivores, this same characteristic makes them vulnerable to habitat fragmentation and habitat loss. Maehr (1990) argued that it may be pragmatic to satisfy the habitat requirements of many species by meeting the spatial needs of a single species. The biological rationale for this approach has been debated (Wilson 1987; Terborgh 1988), ·but given their track records, it is unreasonable to expect natural resource agencies, which are traditionally underfunded and often unwilling to address multi-species management, to address the needs of the many wildlife species that are suffering the effects of range constriction. Thus, single-species, or trophic-level management remains as the substitute for a landscape-level approach to biodiversity conservation. Harris and Cropper (1992) suggested that a combination of sea level rise, climate change, and anthropogenic influences have led to the post-Pleistocene faunal collapse that has occurred in Florida. Assuming that current rates of sea level rise and human population growth will continue, it is clear that Florida's most widespread populations of panthers, black bears, and bobcats also may be the most at risk. The displacement of tropical plant communities and the elimination of large tracts of forest may negate landscape conservation efforts even if they are successful in the short term. Study Area Field activities were conducted in extreme South Florida primarily between 82° and 81° 50' W longitude, and below 27° N latitude. The eastern portion of the study area is bounded by the Everglades, sprawling coastal urban development, and the Everglades Agricultural Area. While the development of South Florida is generally equated with the southeast coast, it is ironic that the earliest landscape-altering changes caused by humans occurred in Southwest Florida. These changes began after passage of the Swamp Lands Act of 1850-legislation that was intended to stimulate the reclamation of inundated federal lands of the United States (Carter 1974). Inroads into the interior of South Florida began with the dredging of the Caloosahatchee 10 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) River from the Gulf of Mexico to Lake Okeechobee. This water course, which originally began west of the lake near LaBelle, Florida drained the landscape on either side of the Hendry/Glades county line but left an upland linkage between Southwest and Southcentral Florida near Lake Okeechobee (Fig. 1.1). Although railroads had reached Miami by 1898 (Carter 1974), the modern vision of a drained and productive Everglades did not materialize until the campaign of Governor Napoleon Bonaparte Broward in 1904. The first canal to connect the Atlantic Ocean with Lake Okeechobee was dredged in 1906. This led to booming and sometimes fraudulent farmland marketing lhat hinged upon South Florida's rich muck soils. By 1929 over 730 km of canals aimed at draining the Everglades were in place. Clearing for farmland resulted in the elimination of a vast forest of cuslard apple (Annona glabra), a landscape feature of Lake Okeechobee's south rim that once hid Seminole Indians from Union troops and likely facilitated the east-west movements of many species of South Florida's vertebrate wildlife. As human access to the lake increased, and roads and railways were built as far south as Miami, construction began on the Tamiami Trail. Just as the dredging of east- west canals did for ships, this highway linked the east and west coasls for automobiles in 1928 and opened the Big Cypress Swamp to development With agriculture dominating much of the drainable wetlands and farmable uplands south of the Caloosahatchee River, some land preservation in South Florida was initiated. Everglades National Park and Collier-Seminole State Park were established in 1947, and Corkscrew Swamp Sanctuary was dedicated in 1954. Through the second half of this century the interplay between land preservation and land development became a chess match whereby each advance in creating a new preserve was countered with a new farm, pasture, or housing development For example, before the Fakahatchee Strand came under state ownership, all of its merchantable timber, primarily large cypress (Tarodium distichum), was removed by Lee Tidewater Cypress Company after the construction of an extensive network of elevated railroad beds, and its closed canopy forest was returned to an early successional stage (Burns 1984). Although the majority (9920 ha) of the strand was acquired by the state of Florida in 1974, 18,522 ha bordering the new preserve were marketed by Gulf American Land Corporation as an expansive residential development known as Golden Gate Estates. An intricate network of roads and canals, built to accommodate residents who may never construct homes, has now left an indelible mark on this part of the South Florida landscape. Controversy surrounding the construction of a regional jetport in western Dade County during the late 19605 and early 19705 resulted in a state-sponsored land purchase that was second in size only to Everglades National Park. The 230,770 ha Big Cypress National Preserve was established by an Act of Congress in 1974 to conserve natural resources and recreational opportunities. Approximately two decades transpired before the next significant wave of additions to conservation lands occurred in South Florida. In the meantime, the human population of Collier County more than doubled in each of the decades starting with 1960, 1970, and MAEHR: ECOLOGY BOBCAT, BLACK BEAR PANTHER 11 1980 (Fernald and Purdum 1992), and Alligator Alley (the precursor to Interstate 75) was built to serve as the second high-speed roadway to connect the southeast and southwest coasts. The 10,120 ha Florida Panther National Wildlife Refuge was created in 1989, and over 40,000 ha are scheduled to be added to the Big Cypress National Preserve as the result of an unprecedented land swap between the private sector and the federal government (Maehr 1992). As of 1990, over 1.4 x 106 ha in South Florida were held in public ownership (Fig. 1.1) and dedicated to conservation purposes (Machr 1990). On the surface. this appears to be a significant portion of South Florida under government stewardship, and indeed, nearly 60 percent of Collier County alone is in some form of government protection (which has led many local officials and community leaders to proclaim that no more land-saving actions are necessary). However, the vast majority of government land in South Florida is not conducive to agriculture nor urban development because of harsh soil and/or hydrological conditions (Leighty et al. 1954). In contrast to Collier County, neighboring Hendry County withstood a conversion between 1900 and 1973 of over 50 percent of its native cover to agricultural and urban uses (DeBellevue 1976). In the 23 years since then, South Florida has sustained continued increases in citrus, cattle production, sugar cane, and other agricultural land uses (Fernald and Purdum 1992) that, when combined with the dredging of the Caloosahatchee River, the clearing of the Lake Okeechobee custard apple forest highway construction, and the impounding of much of the Everglades, have effectively isolated the forests of South Florida from the rest of the state. The Natural Landscape Although it lacks topographic relief, South Florida supports many recognized vegetation communities. The high variability inherent in South Florida vegetation communities and their descriptions (Craighead 1971; Soil Conservation Service 1981; McPherson 1984; Olmstead and Loope 1984; Myers and Ewel 1990) results in part from the many zones of transition from one community to another, as well as the interests of the authors. Gradations between plant communities are also suggestive of the constant changes in species composition that have been, and continue to be, influenced by climate. The implication of this slow but inexorable landscape process is that vast expanses are necessary to accommodate not only the peregrinations of wildlife populations, but also the migration of plants and entire vegetative communities. Florida is considered more at risk from sea level rise due to global warming than is any other state (Henry et at 1994), and South Florida has experienced more subsidence and consolidation of soils than any other region of the state. Effects of recent increases in salinity were observed in Everglades National Park by Craighead (1971), and have had negative consequences on a variety of economically valuable environmental resources. Although sea level is not expected to rise beyond 65 cm by the year 2100 (Henry et at 1994), even a 12 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) change of this magnitude will cause widespread constrictions to the distribution of cover types that are important to the terrestrial vertebrate carnivores of South Florida Although South Florida lies within the Great Desert Belt of the earth (Henry et al. 1994), its climate is classified as tropical savannah (Koppen 1931: cited in Robertson 1955; Hela 1952) and monsoon rainforest (Trewartha 1943). A distinct warm wet season is typical from May through October when 60-80 percent of the annual average 1525 mm of rainfall occurs (Craighead 1971). The mean annual temperature is 23°C with extremes of-2°C to 38°C (Duever et al. 1986). Southern Florida is often spared the effects of continental winter cold fronts due to the influence of warm air originating from over the Gulf of Mexico and the Caribbean (Henry et at. 1994). This has permitted the existence of a high diversity of tropical plants including palms, epiphytic orchids, and bromeliads. Most of South Florida is below 7.6 m above mean sea level (Wade et at. 1980). Floods, fires, freezes, and droughts are considered to be the most important natural environmental influences on the distribution and kinds of plants in South Florida (Robertson 1955; Craighead 1971; Wade et al. 1980). Most non-anthropogenic fires are caused by lightning strikes associated with summer thunderstorms. Davis (1943) assembled the most exhaustive account of vegetation communities in South Florida Although he used only nine broad categories, these were fuKher divided into 64 subclasses. This classification was used as the basis for descriptions of bobcat, black bear, and Florida panther habitat use and home range composition. Several of the communities that were described by Davis (1943) were combined in order to match a current observer' s ability to correctly identify plant communities from 152 m elevation in a fixed-wing aircraft. For example, the subtle differences among 'oak and cabbage:lm hammocks.' 'cabbage palm hammocks,' and 'low hammocks' were not consistently discernible from a fixed-wing airplane; nor were they remarkably different even at ground level. Other groupings were not used because they were unique to Southeast Florida or they were not found in the study area. This resulted in the use of only 11 cover types for delineating the habitats of Southwest Florida terrestrial carnivores. These are given on the following pages. Pine fiatwoods are dominated by slash pines (Pinus elliottii) growing in open forests on moderately well=drained soils. Saw palmetto (Serenoa repens) is a common and often dominating understory shrub. Pine and cabbage palm woods are relatively limited in distribution and contain slash pine and cabbage palm (Sabal palmetto) in similar abundance. Saw palmetto is usually absent from this community. Pine scrub is dominated by sand pine (Pinus clausa) with thickets of scrub oaks (Quercus spp.) and other xeric shrubs. Hardwood hammocks are found on well to poorly drained soils and are dominated by broad-leaved deciduous oaks in association with cabbage palm and many temperate and tropical shrubs. MAEHR: ECOLOGY BOBCAT, BLACK BEAR, PANTHER 13 Mixed swamps are inundated forests of hardwoods such as red maple (Acer rubrum) and laurel oak (Quercus laurefolia) with cypress present but not dominant Standing water can persist for a few months to the entire year depending upon drainage conditions and minfall patterns. Cwress swamos range from the remnant stands of large specimens such as Corkscrew Swamp to the dwarf cypress forests of the eastern Big Cypress Swamp. Most cypress forests are characterized by long periods of inundation and low primag productivity. Thicket swamDS are shrub forests dominated by elderberiy (Sambucus canadensis), ¥Allow (Salix caroliniand), pop adi (Fraxinus caroliniand), wax myrtle (Myrica cerifera), or buttonbush (Cephalanthus occidentalis). These areas are usnally transition zones between swamp forests and marshlands and often follow clearing or the abandonment of agricultural lands. Bav tree forests are composed of broad4eaved evergreen trees including red bay (Persea borbonia), sweetbay (Magnolia virginiand), and dahoon (Ilex cassine) on poorly drained soils, and primarily in Highlands and Glades counties. Freshwater marshes are treeless wetlands domingted by sawgrass (Cladium jamaicense), flags (Thalia geniculata, Sagittaria spp., and Pontederia spp.) or wetland grasses and sedges. Manmves are found in coastal estuaries in saline to brackish water and are composed of red mangrove (Rhizophora mangle), black mangrove Wvicennia germinans), and white mangrove (Laguncularia racemosa). These spedes are usually divided into distinct zones with buttonwood (Conocarpus erectus) inhabiting the inner-most high salinity zone. Airricultural/disturbed areas were once occupied by vegetation as described above but have been converted to croplands, improved pasture, rock mines urban areas, and roadsides. Summary of Previous Large Carnivore Work in Florida Until the middle of the 20th century, most published literature on or relevant to Florida's large terrestrial carnivores dealt with general distribution, taxonomy, or economic status (Merriam 1896; Bangs 1898; Hamilton 1941; Young and Goldman 1946; Young 1946a). Recent studies have focused more on biological, management, and conservation topics (Anderson 1983; Anderson 1987; Eagar and Stafford 1974; Pelton 1982; Tumilson et al. 1982). Bobcat Descriptions from Florida are similar to those from other parts of the bobcat's range: it is a secretive, solitary carnivore that specializes on small prey-especially rabbits, rodents, and to a lesser extent birds. Maehr and Brady (1986) Analyzed food habits throughout Florida and determined that there were no sex-related food 14 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) preferences, but that seasonal variation in diet involved an increased use of artiodactyls and birds during fall and winter, respectively. The use of birds is apparently in response to increases in over-wintcring populations of migrants. Land et al. (1993) and Wassmer et al. (1988) reported similar proportions of prey species in bobcat diets from Southwest and Southcentral Floridai however, neither study revealed statistically significant seasonal variation. From a range-wide perspective, Florida bobcats utilize deer less frequently and birds more frequently th=n bobcats from other regions. Wassmer (1982) and Guenther (1980) in Southcentral Florida, and Foster (1992) in Southwest Florida found that annual home range size varied from 11.6 to 31. l kme for adult males, and 5.8 to 21.6 km2 for adult females. Progulske (1982) reported a mean of 44.4 km2 for two adult male bobcats in North Florida Florida bobcat home range sizes fall well within the extremes reported for the species (Foster 1992). Bobcats in Southcentral Florida preferred dense forest cover in an uplands- dominated landscape matrix (Wassmer et al. 1988). Foster (1992) hypothesized that bobcats in Southwest Florida preferred upland habitats although she did not compare frequency of use to habitat availability. Female bobcats appear to prefer thickets of saw palmetto for their natal dens (Wassmer 1982; Foster 1992). Winegarner (1985b) documented a bobcat natal den in a gopher tortoise burrow located in a dense saw palmetto thicket Foster (1992) reported considerable overlap among adult male bobcat home ranges in Southwest Florida, but little overlap among adult females. Wassmer et al. (1988) recognized similar patterns in Southcentral Florida. Both studies reported extensive overlap between males and females, and Wassmer et al. (1988) described a pattern of home range replacement and social ecology resembling the land tenure system described by Seidensticker et al. (1973) for mountain lions. Reproductive characteristics of Florida bobcats have been described only as isolated anecdotes referring to a few individuals (Winegarner and Winegarner 1982; Winegarner 1985; Foster 1992) and for a local population in Highlands County (Wassmer et al. 1988). Based on these observations, litter size in Florida averages between two and three. Most reports of mortality cite natural causes. Foster (1992) found that most of the study animals in Southwest Florida avoided paved roads and, thus, avoided highway collisions. Wassmer et al. (1988) and Progulske (1982) found that disease was the major cause of death in Southcentral and North Florida bobcat populations. Although no statewide population estimates have been made, the Florida Game and Fresh Water Fish Commission considers bobcats sufficiently abundant to allow hunting and trapping throughout the state. Bobcats are likely still found in every county of Florida MAEHR: ECOLOGY BOBCAT, BLACK BEAR, PANTHER 15 Black Bear The black bear is Florida's largest terrestrial mammal, and it exhibits patterns of behavior and ecology that are typical of the species throughout North America Early studies by Harlow (1961, 1962a) summarized body measurements that suggested black bears from Florida were as large or larger than individuals from northern populations. Schemnitz (1974) estimated a South Florida population of 145, while Harlow (1962b) estimated that 800-1000 inhabited the entire state. The earliest telemetry studies occurred in Northcentral Florida in response to concerns over proposed phosphate mining in Osceola National Forest (U.S. Dept Interior 1979). This study resulted in an evaluation of home range estimation techniques (Mykytka and Pelton 1989) and a habitat analysis that suggested the importance of large swamps and pine natwoods ecosystems (Mykytka and Pelton 1990). Home ranges of two females in Osceola National Forest were 93.4 and 39.4 kin2 while six males ranged from 35.9 to 457.2 km2 (Mean=171.l km~) Wooding and Hardisky (1988) estimated male and female black bear home ranges in Ocala National Forest at 170 and 26 km2, respectively. The black bear's ability to tolerate anthropogenic alterations to the landscape is reflected in its widespread contemporary distribution (Maehr 1984; Brady and Maehr 1985), as well as a statewide beehive-depredation problem (Machr 1982; Brady and Maehr 1982; Maehr and Brady 1982a). Although such interactions may lead to bear poaching, highway collisions are the most common form of human- related mortality (Wooding and Brady 1987). Natural mortality has been documented infrequently but may include occasional predation by Florida panthers (Maehr et al. 1990a) and cannibalism (Wooding and Hardisky 1988). Black bears are susceptible to a variety of diseases and parasites (Conti et al. 1983; Pirtle et al. 1986; McLaughlin et al. 1993), but none has been demonstrated to be a significant mortality factor (Forrester 1992). Food habits have been examined from statewide and regional perspectives. Maehr and Brady (1984a) measured seasonal changes that were consistent with findings from other parts of the species' range. Food availability and diversity vary geographically (Machr and Brady 1982b; Maehr and Brady 1984b), however, foods are consistently dominated by fruits, insects, and occasionally vertebrates (Maehr and DeFazio 1985). Like their North American conspecifics, black bears in the Ocala National Forest exhibit seasonal movement patterns (Wooding and Hardisky 1988). While some males remain active throughout the year, restricted movements during winter are particularly pronounced among pregnant females. Although black bears were listed by the State of Florida as threatened in 1974 (Maehr and Wooding 1992), fall and winter hunting were permitted in Apalachicola National Forest and Baker and Columbia counties until 1994 when an experimental moratorium was imposed. Despite the black bear's recently documented occurrence in at least 50 of Florida's 67 counties (Brady and Maehr 16 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) 1985), its distribution is almost exclusively confined to five disjunct populations. Occasional long-distance dispersal movements (Maehr et al. 1988) have the potential to occasionally cross the gulfs between some of these populations. Florida Panther The existence of the Florida panther was debated for several decades until its "official" rediscovety in 1973 by Nowak and McBride (1973), who estimated the South Florida population at 20-30 individuals. Although evidence of panthers has continued to emanate from Southcentral Florida (Layne and Wassmer 1988; Maehr et al. 1992; Maehr 1994) and the St Johns River drainage (Maehr 1992), concerted research efforts have been restricted to South Florida Prior to the capture of an old female in Glades County in 1973 (Nowak and McBride 1973), the South Florida population was estimated to be 92 (Schemnitz 1974). Williams (1978) placed the population at 30-50 individuals and stated that credible but unsupported sign originated only from Collier County. Roof and Maehr (1988) developed a standardized survey method for field verification of panther sign. Based on such sign surveys and extrapolations from radio-collared study animals, the contemporary population in Southwest Florida (exclusive of the Everglades and eastern Big Cypress Swamp) is estimated at 70-80 individuals (Maehr et at 199la). Young and Goldman (1946) described the Florida subspecies and its distribution in the southeastern United States. Allen (1950) discussed vocalizations of captive panthers, and a specimen examined by Belden and Forrester (1980) solidified the stereotypic description of the modern subspecies- especially the characteristic crook at the end of the tail and a whorl of dorsal hair near the scapulae. Similarities between some panther and bobcat scats (feces) led to an unsuccessful effort to differentiate the two species with chromatographic bite assays (Johnson et al. 1984). Specialized capture techniques (McCown et at 1990) have been a pat of intensive and invasive examinations of individual panthers during anesthesia and necropsies since the mid-1980s. As a result, a detailed catalog of parasites (Forrester et al. 1985; Greiner et al. 1989; Machr et al. 1995) and diseases (Forrester 1992; Roelke et al. 1993b; Glass et al. 1994) has been assembled Although no parasites have been demonstrated to be pathogenic, Notoedric m=nge has the potential to cause mortality at least in juveniles (Machr et al. 1995). Diseases that have caused mortality include bacterial infections rabies (Roelke et al. 1993b), and pseudorabies (Glass et al. 1994). Mortality caused by highway collision is well documented but is less important to the wild population than natural mortality-especially within-species aggression (Maehr et al. 199 lb). Illegal killing has not been documented for over a decade and capture-related deaths are rare. MAEHR: ECOLOGY BOBCAT, BLACK BEAR, PANTHER 17 Panthers prefer native uplands over other habitat types in South Florida (Belden et al. 1988; Maehr et al. 1991ai Maehr et al. 1992). Maehr and Cox (1995) found that large patches of forest cover were important in explaining panther occurrence and that natural and unnatural habitat fragmentation reduced the value of these forests to panthers. Thickets of saw palmetto are important as daytime rest sites and natal dens (Maehr et al. 199Ob). Age at first reproduction in females is 18 months (Maehr et al. 1989b) and males have not been observed to breed before 3 years of age (Maehr et al. 199la). Average litter size is 2.25 (Maehr and Caddick 1995), natal sex ratios are approximately 50:50, and bitths have occurred in almost every month. Survival of kittens between birth and 12 months of age is greater than 0.80 (Machr and Caddick 1995), and Annual momlity of all sex and age classes combined is less than 0.20 (Machr et al. 199lb). Florida panthers exhibit a system of land tenure typical of solitary carnivores and their home ranges average 519 km2 and 193 km2 for adult males and adult females, respectively (Maehr et al. 199la). Activity of both solitary and denning panthers follows a bimodal pattern with crepuscular peak~ (Maehr et al. 199Ob). Female panthers exhibita regular pattern of den attendance during the two months that kittens are unable to travel (Maehr et al. 1989a). Litter size appears to be a function of prey abundance and habitat productivity. Attempts to artificially augment prey availability were unsuccessful (Maehr et al. 1989c). Concerns over nutritional status led to a preliminaly conclusion that female panthers were consistently undernourished and anemic (Roelke et al. 1985). However, food habits analyses indicated that prey abundance and distribution varied geographically (Maehr et al. 1990a). Panther prey in South Florida follows a northwest to southeast gradient of declining abundance (McCown et al. 1991) that appears to be a product of soil quality and primary productivity. Wild hogs (Sus scrofa) are the most frequently taken prey; however, this species is sparse south of Interstate 75. White-tailed deer (Odocoi/eus virginianus) and wild hogs combined account for 70 percent of the frequency of occurrence in panther scats. Where both species are abundant (i.e., the north part of the range), panthers are larger, more abundant, and produce more kittens. This led Maehr (1990) to emphasize the importance of private lands in Collier and Hendry counties to the future of the subspecies. At least some rationales argue that low numbers increase the extinction probabilities for small populations. In addition to demographic stochasticity and unpredictable climatic events, genetic problems have been suggested as a primary threat to panthers (Roelke et al. 1993a). O'Brien et al. (1990) identified two distinct lineages of wild panthers in South Florida--one originating in situ, the other originating from captivity. Despite this apparent introgression, they concluded that the South Florida population was suffering from reduced allozyme variation when compared to other cougar populations. Spermatozoa abnormalities (Barrone et al. 1994), heart defects infectious diseases, and matings between close relatives (Roelke et al. 1993a) have been suggested as symptoms of a collapsing 18 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) population. However, Maehr and Caddick (1995) reminded that philopatry is expected in a land tenure social system, and thus matings between close relatives especially males and their female offspring, are typical. O'Brien et al. (1990) led Harris (1990) to point out that the reason for the genetic variation documented in South Florida may date to the Pleistocene connection between Florida and the Yucatan Peninsula. Regardless of the panther's lineage in South Florida normal demographics, as indicated by behavior typical of the species-low mortality, high natality, and kitten survival, may be indicative of a high level of local adaptation. These factors may be ameliorating'the effects of small population size and reducing the potential for genetic problems to be immediate conservation concerns (Maehr and Caddick 1995). General Methodology The data upon which this study is based were collected primarily while I was the supervisor of research for the Florida Game and Fresh Water Fish Commission in South Florida from 1985 to 1994. Field activities focused primarily on Florida panther captures and the monitoring of their radio-collar signals. This work began in 1981 and continues to this day. Methods of panther capture and restraint have been described in detail by McCown et al. (1990) and Barrone et al. (1994). Bobcats were captured opportunistically primarily during 1986 and 1987 while panther hunting with hounds trained to trail and tree cats. Bobcats were handled in a fashion similar to panthers, but due to their smaller size, they were fitted with smaller radio collars with functional lives of no more than one year. Black bears were captured with Aldrich spring-activated snares and with culvert traps (Erickson 1957) primarily during the three-year span 1991-1993. Bears were anesthetized with a pole syringe using Ketamine hydrochloride at a dosage of 20 mg/kg. Radio collars worn by adult black bears and adult panthers hmi battery lives of at least two years. Florida panthers were captured as necessafy to replace aging transmitter batteries and maintain contact with study animals. All study animals were regularly monitored from fixed-wing aircraft at an altitude of about 150 m following the methodology of Mech (1983). Florida panthers were located three times per week, black bears were located three times every two weeks, and bobcats were located approximately once each week. Location data included Universal Transverse Mercator coordinates, habitat type, time of day, animal identification number, and an index to activity if the study animal carried a transmitter that contained a motion-sensitive switch (Telonics, Inc., Mesa AZ). Additional telemetry data used in these analyses for panthers monitored by the National Park Service or following the termination of my involvement with field activities were obtained through requests to the Florida Game and Fresh Water Fish Commission's Office of Environmental Services. More specific field methods and the statistical analyses used are described in detail within subsequent chapters. MAEHR: ECOLOGY BOBCAT, BLACK BEAR, PANTHER 19 Table 1.1. The terrestrial mammalian camivora ofNorth America north of Mexico presumed to be present at the time of European colonization. Mass estimates represent species means (female and male) and were derived from Walker (19751 Burt (19751 and Chapman and Feldh.mmer (1982). Species Mass Status in Florida URSIDAE Black bear (Ursus americanus) 150 kg threatened Brown bear (Unus arcros) 250 kg not native Polar bear (Thalarctos man'timusj 300 kg not native CANIDAE Gray wolf (Conis lupus) 42 kg not native Red wolf (Canis ndits) 25 kg eldirpated Coyole (Canis latrans) 15 kg naturalized/introduced Red f6x (Vulpes vulpes) 5 kg naturalized Swift fox (Fulpes velar) 3 kg not native Kit fox (Fulpes macrons) 2 kg nol native Gray fox (Urocyon cinereoargenteus) 4 kg resident Arctic fox (Aloper lagopus) 5 kg not native PROCYONIDAE Cacomisne(Bassariscus astutus) Ikg not native Raccoon (Procyon lotor) 10 kg resident Coatimundi (Nasua nasua) 9 kg not native MUSTELIDAE Ermine qtustela erminea) 100 g not native Least weasel (Mustela nivahs) 50 g not native Long-tailed weasel (Mustelaftenam) 200 g resident American mink (Mustela vison) 1.1 kg resident Black-footed ferret (Afusrela mgripes) 1 kg not resident American pine marten (Marres amencana) 750 g notresident 'Asher (Martes pennant) 3 . 5 kg not resident Wolverine (Guto guto) 20 kg not resident American badger (Taxidea tarus) 8 kg not resident Spottedskunk(Spilogale putorius) 700 g resident Western spotted skunk (Spilogalegracihs) 700 g resident Striped skunk (Vephitis mephitis) 2.5 kg resident Hooded skunk (Mephins macroura) 2 kg not resident Hog-nosed skunk (Conepatus maoleucus) 3 kg not resident Hog-nosed skunk (Conepams teuconoms) 3 kg not resident River otter (Lutra canadensis) 8 kg resident FELIDAE Lynx(Lynx canadensis) 10 kg not native Bobcat (Lynx ndia) 10 kg resident Ocelot(Felis pardalis) 13 kg not resident Margay (Felis weidif) 2.5 kg not resident laguarunm(Felis yagouaroundf) 7. 5 kg not resident Puma(Felis concolor') 60 kg resident Jaguar (Panthera onca) 80 kg not resident 20 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) cf Avon park III Lake Okeechobee C ''FCato att: 4'ver *- Laselle Ft. Myers Naples [fl 1 r«»94,2 4-- - , f -1 -11~~ ;f' 4-'1 ~S, It~ ~ !*,-B' '~1'1 -lili 11:1,~UrI ; A Florida Panther National Wildlife Refuge T. 1+ .' 1-11--- Miami B Southern Golden Gate Estates C Fakahatchee Strand State Preserve D Big Cypress National Preserve E Big Cypress Addition Lands ..F Big Cypress Seminole Indian Reservation . ~1 G Rotenberger/Holy Land Wildlife Areas f I H Water Conservation Areas 1 Everglades National Park 05081 Figure 1.1. Study ana and large tracts of publicly owned land in South Florida MAEHR: ECOLOGY BOBCAT, BLACK BEAR, PANTHER 21 1 DIETARY OVERLAP Although the actual nutritional requirements of most wildlife species are unknown, food remains the most commonly examined resource for assessing the degree of interspecific competition (Hutchinmn 1957; Caughley and Sinclair 1994). The food habits of native large terrestrial carnivores in Florida have been well described (Maehr and Brady 1982, 1984a; 1984b; Maehr and DeFazio 1985; Maehr and Brady 1986; Maehr et al. 1990; Wassmer et al. 1988). However, they have not been examined for potential dietary overlap. Further, published food habits data are primarily from North Florida for black bears, Southcentral Florida for bobcats, and from South Florida for panthers. Black bears in North Florida exhibit only modest differences in food habits (Maehr and Brady 1984b), and bobcat diets in South Florida are similar to those reported for the entire state (Land et al. 1993). I examined previously published data for bobcats and Florida panther food habits, and compared these with unpublished food data for black bears in South Florida Methods Collection, identification, and analytical procedures for foods of bobcats and panthers can be found in Maehr and Brady (1986) and Maehr et al. (1990). Because volumetric measurements were made for contents of bobcat stomachs, these data were converted to percent frequency in order to allow for direct comparisons with other species. Black bear scats were collected throughout South Florida at trap sites and during routine field activities associated with radio telemetry studies. Scats were rinsed with water through a 1 mm sieve and individual food items werc separated in a white enamel wash pan. Reference collections and guides (Martin and Barkley 1961; Schopmeyer 1974; Tomlinson 1980; Arnett 1985) were used to identify foods as close to species as possible. Black bear foods were then examined by month in order to detect seasonal patterns . related to food availability or preference. These patterns formed the basis for seasonal comparisons among the three native large carnivores in South Florida Although the possibility for gender-related differences in diet exists, Maehr and Brady (1986) found that male and femgle bobcats hmci fimilar food hnhits, Anderson (1983) did not report sex-related differences in cougar diets, and Schwartz and Franzmann (1991) found that black bears exhibited no sex-related differences in rate of moose calf (Akes alces) predation. Therefore, gender was ignored in comparisons of diets. Sorensen's similarity coefficient (Sorensen 1948; Greig-Smith 1964) was used to quantify the amount of annual dietaky similarity between bears, bobcats, and panthers using species occurrence data from Machr et al. (1990), Land et al. (1993), and this study. Dietary overlap between species was 22 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1 ) estimated by using Pianka's (1986) variation of an algorithm first presented by MacArthur and Levins (1967): 0,=Ipa pa VIP'#IP:a j j where P, is the proportion of food resource in the ;th species class, and n is the total number of species categories. O, then, has a possible range of values from 0 to 1. Higher values equate to higher levels of overlap. Shannon-Weiner indices of diversity (Shannon and Weaver 1963) were calculated using percent frequency of foods. Because not all cited studies identified small mammals, birds, and lagomorphs to species these items were grouped into three categories. Nutritional analyses were conducted on selected black bear foods that were frequently consumed in South Florida. Food items were collected in occupied black bear range in South Florida, oven-dried at 600°C, and ground in a Wiley mill. Analyses conducted at the Forage Evaluation Support Laboratory, Animal Nutrition Lab, University of Florida, Gainesville, measured dry matter, organic matter, total neutral-detergent fiber, ash-free neutral-detergent fiber, total nitrogen, crude protein, in vitro organic matter digestibility, and total phosphorous. Crude fat was measured by A&L Plains Agricultural Laboratories, Inc., Lubbock, Texas, using ether extract methodology. Ants were washed, separated, and dried at 100°C for 48 hours, and milled through a 1 mm screen at the Caesar Kleberg Wildlife Research Institute, Texas A&M University, Kingsville, Texas. Milled samples were dried again and 2-g aliquots of ground ants were extracted for eight hours with petroleum ether. Samples were dried again, weighed, ashed at 600°C for five hours, dried and weighed Results and Discussion The three native carnivores in South Florida exhibited distinct trends in their utilization of important species or groups of species. Results of food habits studies from South Florida were comparable to the results of food habits studies of the same species in more northern Florida locates (Table 2.1). Bobcat Maehr and Brady (1986) found that Florida bobcats specialized on small prey and killed white-tailed deer infrequently. Land et al. (1993) confirmed this general pattern for Southwest Florida although in their study of female deer mortality within the range of the Florida panther, bobcats killed more deer thgn did panthers. MAEHR: ECOLOGY BOBCAT, BLACK BEAR. PANTHER 23 Maehr and Brady (1986) attributed bobcat consumption of deer to crippling and mortality caused by sport hunting, whereas T 3,nd et al. (1993) believed thst predation on deer was caused by a small proportion of the local population that occasionally ambushed prey in habitats that panthers generally avoided (i.e., freshwater marsh). Seasonal variation in Florida bobcat diets may be due to an influx of overwintering migrant birds, hunter harvest of large prey. and annual patterns in small prey reproduction (Machr and Brady 1986). Much of South Florida is closed to doe hunting, and mild winters maintain food supplies for small mammals. Thus, hunter crippling/mortality, and annual nuctuations in small mammal abundance are likely of lesser influence on bobcat nutrition than the sometimes five-fold increases in overwintering bird densities that occur each year (Robertson and Kushlan 1974). Florida Panther Maehr et al. (1990) found that Florida panthers specialized on large prey, especially wild hog and white-tailed-deer. Proportions of food items in the diet did not change seasonally; however, smaller prey such as raccoons were consumed in areas with low densities of large prey. These areas also support lower panther densities with only sporadic reproduction (see Maehr et al. 1989). In Everglaa National Park, where panthers became effectively extinct in 1991 (Bass and Machr 1991), panthers occasionally consumed river otters (Lutra canadensis), bobcats, and alligators Wiligator mississippiensis) 6)alrymple and Bass 1996), apparently because deer occurred at low densities (Smith and Bass 1994). Like populations of cougar throughout North America, the Florida panther seems inextricably tied to a resident deer population. Black Bear Despite copious food habits data on Florida black bears, none of the studies was conducted in South Florida where tropical climate and seed sources have created a much different milieu than the habitats available to bears in Central and North Florida Analysis of 739 scats collected from July 1991 through 1993 indicated that black bears in South Florida consumed at least 40 species or distinct parts of plants, insects, and mammals. Several species including saw palmetto, cabbage palm, giant palm weevils (Rynchophorus cruentatus), and social insects provided at least two plant parts or life stages as food. Apical meristems and seeds were available on both of the palm species, whereas eggs, larvae, adults, and honey (when available) of colonial insects were eaten. Several species did not appear in scats but were observed to be eaten, such as Florida damp-wood termites (Prorhinotennes simplex), or have been reported previously, for example, alligator eggs (Maehr and Brady 1984a). 24 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) Monthly nndyses of food habits (Table 2.2) were used to separate the year into categories that corresponded to plant phenology and bear behavior related to denning (Fig. 2.1). This resulted in the subjective division of the year into three seasons: winter (January-April), summer (May-August), and fall (September- December). Winter foods were characterized by a preponderance of soft mast, primarily the fruits of Brazilian pepper (Schinus terebinthe/blius), and plant fibers such as alligator flag (77:alia geniculata), and pickerel weed (Pontederia cordata) (Table 2.3). Summer foods were dominated by plant fibers and insects, and included the fruit of swamp dogwood (Comus foemina) and lantana (Lantana spp.). Swamp dogwood fruits are locally abundant and very high in crude fat relative to other plant foods (Table 2.4). Such high-energy foods are unusual for this time of year in other parts of the black bear's range. Although summer and winter diets appear similar when only major categories are compared gable 2.3), these two seasons are distinguished primarily because of the widespread availability of Brazilian pepper seeds in winter when few other seeds are available, and because ant consumption increases during May. In addition, lantang which is a tropical genus that contains at least one native species as well as several naturali7cd forms (Nellis 1994:136), is available primarily during early summer. Fall foods were mostly seeds of saw palmetto, cabbage palm. and to a lesser degree, live oak (Quercus virginiana). This seasonal pattern in food habits is similar to other descriptions of black bear food habits in the southeastern U.S. (Hardy 1974; Landers et al. 1979; Beeman and Pelton 1980; Eagle and Pelton 1983; Machr and Brady 19848; Smith 1985b; Garner 1986; Hellgren and Vaughan 1988), where an abundance of hard mast, such as acorns, follows a period of soft fruit availability and precedes winter denning and hibernation. This period has been described as critical to determining reproductive output the following winter, and mast failures have caused measurable impacts to cub production the following year (Rogers 1976). South Florida, however, has a strong tropical influence and has been invaded by a very abundant winter-fruiting exotic shrub (Brazilian pepper). Further, native plants, such as saw palmetto and cabbage palm provide food year- round, a sharp contrast to nutritional opportunities for black bears in those parts of North America devoid of native palms and without introduced, food-producing tropical shrubs. This study recorded several food items that were previously unreported for black bears. These included stems of sawgrass, leaves and flowers of bromeliads (Tillandsia spp.), flowers of thistle (Cirsium horridulum), seeds of marlbeny (Ardisia escallonioides), seeds of royal palm (Roystonea elata), ~ant palm weevils, and colonial semi-aIboreal ants (Crematogaster pilosa). The first two of these new species may be important foods during times of limited soft and hard mast availability. The latter two are of interest because they are highly'nutritious and involve adaptive feeding strategies. Giant palm weevils feed and reproduce in damaged palms (Woodruff 1967), and I have frequently observed weevils colonizing recently damaged cabbage palms MAEHR: ECOLOGY BOBCAT. BLACK BEAR PANTHER 25 and palmettos throughout Collier County. The damage was caused by the deliberate extraction of the apical meristems for human consumption-an activity that mimics the damage caused by black bear feeding. Given that other palm- damaging forces are rare in South Florida (i.e., lightning and wind do not appear to damage either species), the creation of feeding and egg4aying conditions for giant palm weevils may depend largely on the activities of humans and black bears. If these weevils benefit from the creation of feeding and egg-laying conditions, and black bears occasionally return to previously damaged plants and consume the weevils, this may represent a symbiotic relation thai hinges on South Florida's native palms. In other words black bears are known to utilize «hearts-of-palm" as foodstuff. The damaged palms apparently release a volatile pheromone-like compound that attracts palm weevils. Black bears that return to a palm or palmetto that they had previously fed upon are further rewarded by the presence of nutritious palm weevils. The small, semi-arboreal ant Crematogaster pilosa, builds wasp-like nests in shrubs and stout herbaceous vegetation in wet prairies and thicket swamps. While tracking radio-collared bears from aircraft trails were visible in marshes surrounding the Fakahatchee Strand, and individual bears were occasionally seen standing in open settings along these trails. Subsequent field investigations revealed the trails were made by black bears venturing out of the dense mixed swamp forest to feed on these tiny (3-5 mm long), but abundant insects. Ether extract analyses dndicated that the crude fat content of adult Crematogaster exceeds the fat content of white-tailed deer flesh (Table 2.4) and was nearly four times greater than the fat content of the commonly consumed Florida carpenter ant (Campanotus jloridanus). The availability of alternative high-energy foods may help explain the low frequency of mammalian prey in the diets of South Florida black bears compared to bears from other parts of the species' range. For example, Schwartz and Franzmann (1991) found that early summer diets of black bears in Alaska would provide insufficient nutrition without the frequent consumption of moose. Although Rogers (1987b) concluded that black bears specialize on plant foods because they are poor predators, their carnivore dentition still offers this alternative when animal prey are available. Although domestic livestock, wild hogs, and armadillos are common, noisy, and often unwmy potential prey in South Florida they were rarely consumed by bears. Thus, the year-round availability of a variety of plant and insect foods (Table 2.3) may not only reduce potential conflict with livestock owners and apiarists but may also reduce the likelihood for competition with bobcats and panthers. Nutritional analyses of frequently consumed foods indicated that a high energy and moderate-protein diet can be maintained by black bears throughout the year without the consumption of vertebrates which are high in both crude fat and crude protein (Robbins 1983). In Florida, 70 percent of panther and 51 percent of bobcat diets are composed of two species of mammals (Maehr and Brady 1986; 26 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) Maehr et al. 1990). The ability of these cats to subsist on the much lower diversity diet is certainly a reflection of the constant availability of prey, which, in turn reduces seasonal variation in home range use. Most black bears in this study and elsewhere (Pelton 1982) exhibit extensive fall movements. Such movements appear to be a universal, if not annual phenomenon that precedes denning, hibernation, and successful overwintering in temperate North America. In South Florida however, most bears remain active through the winter and feed on a variety of fibrous plant material as well as Brazilian pepper, which is not only widely distributed throughout South Florida but is also one of the highest energy plant foods anywhere in the black bear's range. South Florida black bears ate the apical meristems of palms and the vegetative parts of several emergent plants in every season. Although these species are low in fat and protein, they are highly digestible (in vitro organic matter digestibility=93.8 percent for cabbage palm heart, and 83.6 percent for 77,alia), and there was always at least one abundant fruit high in available energy. This year-round availability of plant and insect foods, many of which are high in lipid content, may also help explain the rapid mass growth and early age of first reproduction that has been documented for South Florida black bears (Maehr et al. inpress). Compared to black bears from North Florida, the fall diet in South Florida bears contains a considerably greater variety (Table 2.5). Maehr and Brady (1984b) suggested that saw palmetto fruit may not be preferred because bears in Northwest Florida consumed black gum (A*so bulora) and odorless baybeny (A*rica inodora) with greater frequency when all three species were available. This may be true for a short period of time in North Florida inasmuch as Treichler et al.(1946), Bonner (1971), Short et al.(1975), and Hellgren and Vaughan (1988) found that the fruit of the closely related Ayssa sy/vatica is higher in crude fat than is saw palmetto. However, North Florida food habits studies were conducted where black bears may have avoided upland habitats when hunter activity along roads was highest during the legal fall bear season. Further, the extensive home range shifts of radio-collared bears to areas of abundant saw palmetto, and the dominance of its fruit in bear diets in this study suggest that saw palmetto fruit is in fact preferred in South Florida It has been suggested that black bears in South Florida avoid such potential foods as gallberry (Ilex glabra), Amedcan beauty berry (Callicarpa americand), and Florida trema (Trema micrantha) because these species are extremely common, regularly produce fruit, and are widely distributed yet contdbute little or nothing to the bear's diet. Although nutritional analyses were not conductcd for these three species, Landers et al. (1979) found that gallberly was low in both protein and fat contenL A similar phenomenon was found among North Florida bears where gallberty is very abundant produces fruit regularly, but is consumed relatively infrequently compared to species of higher nutritional quality (Machr and Brady 1984a). MAEHR: ECOLOGY BOBCAT, BLACK BEAR, PANTHER 27 Species Comparisons Given their similar lifestyles bobcats and panthers had more prey in common with each other (0.57) than did either species with bears (Table 2.6). In addition the diets of both cat species were comparably diverse (Table 2.5). Black bears overlapped less than 0.20 with either bobcats or panthers, and this overlap was almost totally due to the occasional consumption of armadillo, white-tailed deer, and wild hog flesh. While the Sorensen similarity coefficient is useful as an index to prey species overlap among predators, it does not take into account the differential use of these foods. The Pianka algorithm also has its shortcomings, but it utilizes frequency of occurrence data and thus, from a potential competition perspective, is likely better for portraying the actual degree of resource-use overlap between species. These calculations suggest that even between cat species there is little food overlap rrable 2.6, Fig. 2.2). The predatory lifestyle of black bears in South Florida appears to be purely opportunistic and is clearly demonstrated by Pianka overlap coefficients of less than 0.02 with both cat species. Similarly, the apparent overlap between panthers and bobcats was diminished when frequency of occurrence was considered. While considerable mortality is inflicted on some populations of large ungulates in other regions by black bears (Behrend and Sage 1974; Franzmann et al. 1980; Ozoga and Verme 1982; Verspoor 1983; Wilton 1983; Franzmann and Schwartz 1986; Matthews and Porter 1988) and bobcats (Hamilton and Hunter 1939; Pollack 1951; Westfall 1956; Beale and Smith 1973; Turkowski 1980; Toweill and Anthony 1988), conditions in South Florida do not compel these carnivores to consistently prey upon large vertebrates. Prevailing climatic conditions prolong growing seasons for temperate plant species allow the establishment of both naturally and artificially established tropical plants. and encourage stability in prey populations. These factors may help explain the infrequent predation by South Florida black bears on large vertebrate prey, infrequent killing of livestock by panthers, and the predominance in South Florida bobcat diets of small-sized prey throughout the year. Moreover, influxes of wintering migrant birds boost prey opportunities without requiring bobcats to shift their home ranges in order to maintain nutrition during temporary periods of small mammal scarcity. The black bear, Florida panther, and bobcat trace their ancestries to well before the Pleistocene (Webb 1974) at a time when very different environmental conditions prevailed. Stirling and Derocher (1989) suggested thnt the black bear has remained virtually unchanged for 1x 106 years, and all three species have adapted to changes in food availability that can be measured over geological time. Florida's three native carnivores are similar because they are all adaptable species that can fill distributional extremes. It is of interest that interference competition has been documented in the relatively open Everglades (where panthers killed and consumed bobcats, Daliymple and Bass 1996), and bobcats were the major source 28 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) of mortality on a deer population that inhabited an open, freshwater marsh in Big Cypress National Preserve (Land et al.1993). In the latter case, panthers were also permanent residents but avoided the open marshes where bobcats preyed on deer. Hence, utilization of a similar food source was facilitated by habitat separation in the Big Cypress Swamp, but in the relatively treeless Everglades, panthers exened interference competition over the bobcaL Despite the panther's competitive dominance, bobcats are now the only native cat species permanently inhabiting southeastern Florida. It is possible that the existing carnivore community in southeastern Florida where the largest two species are now rare, is an artifact of human-induced landscape changes in the region. The elimination of the Atlantic Coastal Ridge forest that once bordered the herbaceous expanse known as the Everglades, may be the most significant factor in the decline of black bears and Florida panthers in southeastern Florida, and may explain the local abundance of bobcats. In general, however, available food appears to be effectively partitioned among the region's three largest carnivores where forests dominate the landscape. Table 2.1. Comparison of food habits among South Florida's naive large mammalian carnivores. Data summarized from Maehr and Brady 1986", Land et al. 19932, Machr et at. 19902, Machr and DeFazio 1985; and this study'. SPECIES Percent frequency in diet Food type Bobcati Bobcal' Panther' Black Black Beari bea2 Rodents & insectivores 36 36 2 0 0 I.agomorphs 25 37 4 0 <1 Birds 16 14 <1 1 0 Opossum 1 0 0 0 0 Armadillo 0 0 8 1 <1 Raccoon <1 7 12 0 0 W.L deer 2 3 28 <1 <1 Black bear 0 0 <1 0 <1 Feral hog 1 0 42 <1 <1 Livestock 0 0 2 0 <1 Alligator 0 0 <1 <1 0 Other herps <1 1 <1 <1 0 Insects 0 0 0 30 16 Plant fiber 11 - - 6 27 Soft mast 0 0 0 22 23 Hard mast 0 0 0 25 30 Table 2.2. Frequency of occurrence by month offoods eaten by black bean in south Florida 1991-1993. Numbers below months represent seats per month. JFMAMJ J A S OND Species 14 8 12 7* 27 « 35 9 131 110 1. Plant fibers Serenoa revens 1 1 9 4 19 6 4 3 Sabal palmetto 5 9 6 1 15 1 4 4 15 5 Pontederia cordata 11 2 2 3 14 1 112261116 M A E H R : EC O LO G Y B O B C AT, B LA C K BEAR , PAN TH ER 29 Cladium iamaicense 21 36 7 2 1 12 Thalia feniculata 16 2 1 1 1 75 3 14 16 33 Potamozeton spp. 1 Cirsium horridulum 23 Tilandsia app. 3 4 22 1 3 2 Graminae 1 2 Unknown 1 4 2 5 2 7 SoR most Psychotria nervosa 2 5 6 7 Ps¥chotria sulmeri 1 Lantana involucrata 1 9 16 1 Vitis spp. 10 20 2 Persea borbonia 38 1 Celtis laevitzata 41 Callicama americana 23 1 Ardisia escalloniodes 1 Schinu5 terebinthefolius 67 8 2 1 1 2 31 65 Ilex cassine 4 52 Con,us foemina 17 12 19 3 1 Smilax spp. 1 1 Rubus spp. 71 Unknown 2 Hard mast Serenoa ret,ens 1 1 8 3 9 45 89 36 13 Sabal palmetto 14 1 1 3 36 77 74 Quergua spp. 1 1 4 16 31 30 B U LLE TIN FLO R ID A M U SEU M N ATU R A L H ISTO R Y V O L 40(1) Table 2.2. (continued) SPECIES JFMAMJJHA SOND Insects Cam pallida 1 Roystonea elata 1 Odontotaenius disiunctus 1 2 2 1 2 6 Rynchophorus cruentatus 2 4 4 3 10 1 Polistes spp. 11 3 2 1 1 Apis mellifera 33 Xylocooa spp. 2 1 1 3 2 12 33 3 5 Campanotus floridanus 5 21 5 93 9 20 5 Vespula sauamosa 1 5 1 2 8 1 Unknown Coleoptera 241 3 44 Cicadidae 1 Unknown wasp 1 1 1 Crematoeaster pilosa 1 1 4 3 8 2 Unknown insect 1 1 2 Vertebrates Das¥pus novemcinctus 1 2 3 1 Odocoileus virenianus 3 1 2 3 Sus scrofa 1 3 Ursus americanus 1 1 Sylvilagus spp. 1 1 1 £228 5» 1 Bone 1 2 Eggshell Human origin 162 1 2 MAEHR: ECOLOGY BOBCAT, BLACK BEAR PANTHER 31 Table 2.3. Seasonal changes in percent frequency offood types in South Florida black bear diets. Percent frequency Season Plant fiber Soft mast Hard mast Insects Vertebrate Artificial Winter 42.6 25.9 7.4 18.5 5.5 0 Summer 47.6 22.5 3.4 22.5 1.4 2.5 Fall 19.4 20.1 42.6 15.5 2.1 0.3 32 B U LLE TIN FLO R ID A M U SEU M N ATU R A L H IS TO R Y V O L 40(1) Table 2.4. Nutrient composition of foods that account for 70.2 percent ofthe diet ofblack bean from south Florida % Ash- % % Cnade Dly % Total See % Total Total % Food Item Pd Diet protein matter NDF' NDF N, P' CM:de fat Cabbage palm Fruit 13.3 6.27 90.2 66.3 65.8 1.00 - 2.99 Saw palmetto Fruit 12.7 4.92 89.0 48.3 46.4 0.78 0.119 9.42 Brazilian pepper Fruit 11.5 730 910 36.2 35.6 1.16 0.219 10.28 Alligator flag Fiber 6.4 18.02 88.9 40.3 39.3 2.88 - 0.85 Carpenter ant Adult 5.0 - . . - - - 11.68 Saw palmetto Fiber 4.5 17.83 87.0 45.7 44.5 2.85 - 0.46 Cabbage palm Fiber 4.5 19.69 %1.6 28.9 27.6 3. 15 - 1 .99 Swamp dogwood Fruit 3.4 6.13 93.2 54.6 52.3 0.97 0.168 17.16 Oak* Fruit 3.4 5.90 - 18.7 23.8 - - 4.30 Wild grape Fruit 2.2 7.75 91.6 36.2 36.6 1.24 0.202 4.50 Thistle Bloom 1.5 9.20 94.4 47.1 46.9 1.46 0.333 4.03 Deer' Flesh 0.6 47.4 - 0 0 - -· 41.30 Arboreal ant6 Adult 1.2 - - - - . - 45.45 Gallberry' Fruit 0 4.90 - 12.8 - - - 3.40 'neutrat-de¢¤gent Sber '1.=nde,3 et al 1979 fum'tn 'phos*,horus 'MoCutlogh and Unrey 190 *Shortand Epps 1976 Table 2.5. Shannon and Weaver (1963) indices of diversity fir large native terrestrial carnivore diets in Florida. Higher values indicate greater dietary variation. Fall Species Location Reference Annual H' H' Florida panther Southwest Florida Machr et at 1990 1.51 Bobcat South central Florida Wass,ncretal. 1988 1.88 - Bobcat Southwest Florida Land et at. 1993 1.39 - M AEH R : EC O LO G Y B O B C AT. B LA C K BEAR PA N TH E R 33 Bobcat North Florida Machr and Brady 1986 1.51 Black bear South Florida This audy 3.00 2.60 Black bear Northeast Florida Machr and Brady 1982 - 1.49 Black bear Northwest Florida Maehr and Brady 1984b - 1.69 Table 2.6. Dietary overlap among native large terrestrial carnivores in south Florida. Values approaching 1.0 indicate higbly similar diets between species Sorensen's similarity coefficient comparisons (based on species occurrence in diet) Species Bobcat Florida panther Black bear Bobcat 1 Florida panther 0.57 1 Black bear 0.17 0.19 1 Pianka's resource overlap algorithm comparisons (based on percent occurrence in diet) Species Bobcat Florida panther Black bear Bobcat 1 Florida panther 0.13 1 Black bear 0.002 0.015 1 34 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) 100 0 Bo Z Ul C 60 U. Z 40 UJ 0 Ul 20 0 Jan Feb lia, Ap, Ilay Am J~ Aug Sep Oct Nov Dec MONTH [3 Human origin G Vertebrates [5 Insecm G Son ma,t ~ Hard mast ~ Plant fibers Figure 2.1. Temporal variation in black bear food consumption in Southwest Florida from 1991-1993. 50 Black Bear Rorlda Panther Bobcd --6-- 0. -4- 040 · * ue nc y of D ie t 30 a- 6 / 1 :/-4 1 1 1 ' / ir / 1 4 *10 ; . q ' 0 4 0 0 0 0 4'r<-A--*4 2 1 2-s·" SU+*. S ..9 Food Categories Figure 2.2. Food habits overlap among black bears, bobcats, and Florida panthers in Southwest Florida MAEHR: ECOLOGY BOBCAT. BLACK BEAR, PANTHER 35 3. HABITAT USE AMONG SOUTH FLORIDA'S LARGE MAMMALIAN CARNIVORES A diverse array of large mammalian herbivores in East Africa uses the same space by partitioning food resources (Lamprey 1963; Jarman and Sinclair 1979). In the same landscape, five large mammalian predators utilize a common herbivore biomass by a variety of social strategies preying on different species, or even partitioning the same species 63ertram 1979). These are but two examples of resource partitioning among similar species and illustrate coexistence among organisms with similar life styles (see Caughley and Sinclair 1994:145). They provide evidence that the prospect of competition can result in intricate resource putitioning mech,nisms. Most of the aforementioned studies occurred in settings where food habits and habitat preferences could be determined by direct observation. In South Florida terrestrial carnivore behavior is hidden beneath a canopy of dense forest and/or darkness; thus interpretations of habitat use are complicated by secretive species. Aerial radio-telemetry eliminates the problem of error implicit in ground triangulation, but flights restricted to daylight hours may uncover only a portion of these species' spatial activities (Mech 1983), and most behavior must be inferred from interpretation of spoor and telemetry data. Thus inferences made about habitat use must be viewed with caution, especially if the study species are nocturnal or crepuscular. Maehr et al. (199la) found that nine Florida panthers monitored from 1985 to 1989 preferred upland forests to open and/or denatured habitats. However, this analysis ignored the potential impacts of annual variation and social status of the study animals. Habitat use analyses in Maehr et al. (199la) included two non- reproductive adult females (#08, #18), two non-resident adult males (#13, #20), and an adult female lhat was captured as a kitten (#19). Foster (1992) examined home range use patterns in South Florida bobcats but did not consider seasonal efrects, nor habitat use relative to its availability. Wassmer et al. (1988) examined bobcat ecology in Southcentral Florida and found that among natural habitats closed-canopy forests were preferred over more open forests. Habitat preferences in South Florida black bears have not previously been examined. Maehr and Cox (1995) utilized over 10 years of telemetry data to illustrate the importance of forest cover in explaining panther distribution and abundance - throughout South Florida. At such a large scale, inherent biases in landscape data such as inaccurate cover type identification, are likely insignificant. But even at the scale of individual study animals observations of field sign, and the trailing behavior of hounds have not revealed patterns of habitat use that differ from previous analyses of telemetry data (Maehr et al. 1990, 199la). Although I have seen tracks of panthers in unforested, settings, their discovery in such open areas is unusual, and the appearance of these tracks usually indicates directional travel to 36 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) the nearest forest patch. Black bears were found to be active primarily during daylight in South Florida (see home range and land tenure chapter). The discovery of bobcat sign is most often associated with trails through wooded terrain (pers. observ.), suggesting that bobcats do not typically venture far from forest cover. Therefore, even though this study did not depend upon data collected at night there appears to be reasonable evidence to suggest that daytime radio locations are a reasonable representation of habitat use patterns among panther, black bear, and bobcat in South Florida. This chapter focuses on patterns of habitat use exhibited by South Florida's large mammalian carnivores as revealed by radio-telemetty. Methods Habitat use was determined by identifying the vegetative cover types associated with each radio-location. While radio-locations were fixed points defined as Universal Transverse Mercator (UTM) coordinates error associated with observer giance and map accuracy precluded the determination of habitat type solely from this cartographic system. Using the aerial tracking methodology described by Mech (1983), vegetative cover types were recorded as independent variables that were associated with a pair of UTM coordinates. Thus, even though some coordinate pairs may have imprecisely pinpointed a two-dimensional location, the habitat type was not affected by as many potential sources of error. Annual variation in habitat use was examined by comparing years for those individuals that were monitored for longer than one year. Several Florida panthers were monitored for more than three years whereas most black bears were monitored for fewer than three years. Bobcat transmitters rarely functioned for more than one year, so this species was not included in this analysis. Chi-square analysis (p=0·05) was used to compare habitat percent frequencies between consecutive years for selected individuals that were monitored for more than one year. Multiple analysis of variance (PROC MANOVA, SAS Institute, Inc. 1988) was used to determine whether habitat variables differentiated species and gender groups within each of three seasons (Winter=January-April, Summer=May- August Fall=September-December). Wilks' Lambda scores (p<0.05) were used to indicate if the total model detected significant differences within seasons because meaningful patterns could exist even when no individual habitat types exhibited significant differences in rates of use among species or gender groups. When differences were indicated by the overall analysis significant differences among species within individual habitat types were determined. These frequency data were used in subsequent factor analyses to help clarify observed patterns of habitat overlap based on percent frequency occurrence. Factor analysis (SAS Institute, Inc. 1988) was performed for each sex by season combination (n=6) using habitat variables that were used regularly by all three species. Mean factor scores were MAEHR: ECOLOGY BOBCAT. BLACK BEAR, PANTHER 37 then calculated and plotted for each sex by season combination for factors that accounted for at least the average amount of model variance. Habitat use patterns were determined by comparing the percent frequency of habitat use of individual resident adult study animals with the composition of concave polygons that encompassed all resident home ranges of each species from 1986 through 1994 (Fig. 3.1). Concave polygons were determined using McPaal (Stuwe 1985) and were used in order to exclude areas that resident adults did not use. Thus, areas such as subuIban Naples, Florida, were excluded from the habitat available to resident adults even though dispersers of all three species were occasionally found in such peripheral range. Chi-square analysis (p=0.05) was used to compare the composition of concavepolygons of a given species (expected habitat use exhibited by all individuals) with lhe composition of habitat types determined from radio telemetry for each individual within that species (observed habitat use). Although patterns of habitat selection and avoidance have previously been based on a comparison of used versus available habitat within individual home ranges (Maehr et al. 199la), 1his method presumes that the selection of a home range is independent of the spatial arrangement of conspecifics and distribution of landscape features that make up habitat for that particular species. As was observed among dispersing subadults of all three species, some individuals traveled widely within the permanently occupied range for that species and occasionally ventured into areas that were unsuitable for permanent occupation. Although individuals such as female panther #52 expanded the known breeding distribution for the species, most dispersal movements resulted in only temporary occupation outside of the core breeding area I examined seasonal home ranges of all adult bobcats (4 males, 4 females), 5 adult male black bears, 5 adult female black bears, 5 adult female panthers, and 5 adult male panthers, and calculated habitat use patterns for each season. Individuals were selected in order to represent as much of the stu* area as possible. Subadults and dispersing individuals were disqimlified from these analyses. Habitat availability was determined by uging 1985-1989 geo-referenced Landsat Thematic Mapper imagety (Kautz et al. 1993), and SPANS geographic information system software (TYDAC 1991). Kautz et at. (1993) recognizcd 22 land cover types. but not all of them were represented in South Florida, and several could not be consistently differentiated from others with similar reflective qualities. For example, pine/cabbage palm forests were not distinguishable from pine natwoods, cabbage palm woodlands were similar to hardwood hammocks, and grasslands and dry prairies were not distinguishable from each other. Asa result. I used nine cover types readily distinguishable from each other for chi-square analyses requiring comparisons of used versus available habitat This clg=ification is similar to the one used by Maehr et al. (199la) for the Florida panther, the principal difference being the inclusion of mangroves as a separate category and the incorporation of salt marshes into the hetbaceous marsh category. The nine 38 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) cover types recognized include pine natwoods (PP), hardwood hammock (HH), mixed swamp (MS), cypress swamp (CS), thicket swamp frS), herbaceous marsh (MA), dry prairie (DP), agricultural/disturbed (AD), and mangroves (MG). Habitat overlap between species was determined by comparing overlap coemcients based on percent frequencies of seasonal habitat use (Pianka 1986). Values approaching 1.0 represent maximum similarity in habitat use. Results and Discussion The sum total areas of radio-collared, permanent resident adults was 468 km2 for bobcats, 2982 km2 for black bears, and 1735 kin2 for Florida panthers (Fig. 3.1). The bobcat polygon was completely contained within the bear and panther polygons; however, black bear use of coastal areas created a zone of non-overlap with both cat species. As a result, salt marshes and mangroves were available to black bears but not to bobcats and panthers. Bobcats are known to range throughout South Florida and trapping them was opportunistic, so the polygon for this species certainly underrepresents occupied range. In other words, this area reflected only that portion of occupied range that encompassed resident study animals and included gaps between individuals that were likely used by uncollared bobcats. Because of their smaller home ranges and the variability inherent in the landscape, each bobcat Mmpled in thiR study was unlikely to expdrience the plant community diversity that was typical of the larger species. Thus, this sample is likely insufficient to reflect overall habitat use patterns for the species in South Florida. Black bears, on the other hand, were captured over a large area and this coupled with their larger movements and larger sample sizes improved the reliability of generalizations about bear habitat-use patterns in South Florida. Similarly, the panther polygon represents extensive capture efforts and contains the core of permanently occupied range (Maehr et al. 199124 Maehr and Cox 1995). Thus, the estimates of panther and black bear habitat use are probably more representative of 1heir species than are the estimates for bobcats. Florida panthers exhibited little variation in habitat use patterns between years (Table 3.1). Differences between consecutive years that were significant or 1hat approached significance for individual panthers were usually related to social dynamics that influenced home range shape and size. For example, during 1987 and 1988 female panther #09 gradually shifted her home range to the more heavily forested Fakahatchee Strand as the result of the removal of female panther #08 and the home range vacancy that was created by this artificial abandonment in April 1987. Male panther #12 exhibited several habitat-use shifts that were related to the colonizations and deaths of resident males in adjoining home ranges during 1989 and 1991 (Table 3.1) (see home range and land tenure chapter). This interpretation is supported by the observation that panther #12's use of mixed swamp declined in 1hose years following a withdrawal from the Fakahatchee MAEHR: ECOLOGY BOBCAT, BLACK BEAR PANTHER 39 Strand, a system dominated by this forested wetland Number 12's final habitat- use shift that preceded his death in 1994 was apparently related to failing health and a shrinking home range. While it is possible that differences in panther habitat use also may be influenced by annually Ructuating food supplies previous studies suggest that deer populations have remained stable in space and time in South Florida (Land et al. 1993) and are thus unlikely to influence panther habitat use. Female black bears exhibited the greatest variability between years. These differences were probably related to reproduction and denning when females spend several months in a restricted area and small cubs restrict their movements. Although mast supplies were not measured in this study, q,mlitative comparisons suggest that concentrations of important foods such as saw palmetto fruit and acorns changed in spatial distribution annually. Such vafying nutritional opportunities and subsequent movements to access alternative food supplies may explain the variation in habitat use demonstrated by male black bear M13 (Table 3.1). Significant differences (p=0.001) in seasonal habitat use patterns were observed among all species/gender groups for several habitat variables (Tables 3.2 and 3.3). Females tended to differ on more habitat dimensions than males gable 3.2). Habitats that were used differently by females in all seasons included hardwood hammock, mixed swamp, and cypress swamp. Habitats that were used differently by males in all seasons were hardwood hammock and agricultural/distulbed (Table 3.3). It is likely that female bear and panther habitat use was influenced by the interrelated effects of smaller home ranges (relative to males), pregnancy, denning, and dependent young. Localized movements due to dependent cubs and kittens undoubtedly increased the use of areas near den sites and reduced the use of more distant preferred habitat that would otherwise have been used had these females been solitary. Percent frequency of habitat use based on individual analyses indicate that forest cover is used by panthers (Table 3.4). black bears (Table 3.5), and bobcats (Table 3.6) nearly to the exclusion of maritime and unforested cover types (Fig. 3.2). Further, in every season, South Florida's large carnivores used land cover types in different proportions than they were available (p<0.05) rrable 3.7). This was due primarily to an avoidance by most individuals of herbaceous or otherwise unforested habitats. For example, hefbaceous marshes covered from 12.8 percent to 24.0 percent of each habitat availability polygon. yet 3.5 percent was the most this cover type was used (Table 3.8) by any species/gender group. Pine flatwoods (including pine/cabbage palm) were universally preferred by all three species in every season. Bears used hardwood hammock slightly more than it was available, whereas male bobcats exhibited a marginal aversion to this cover type. The latter may be a consequence of the small sample size, because male bobcat M02 used hardwood hammock in proportions more than twice its availability (Table 3.6). Cypress swamps appeared to be used less than available by all species, while mixed swamp, a more thickly vegetated plant community, was used in varying 40 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) proportions by study animals. Bobcats appeared to avoid mixed swamp communities which were not abundant in the bobcat habitat availability polygon. Most mixed swamp in Florida is found in the Fakahatchee Strand, where panthers have resided at least since field searches were first conducted (Nowak and McBride 1973). This cover type was used slightly more than it was available to panthers. even though nutrition may be a problem for some individuals that reside there (Maehr et al. 1989a). Black bears especially females that were known to establish dens in this cover type, also used mixed swamp in a higher proportion than it occurred in the environmenL Further, important black bear foods, such as Thalia geniculata, and early mast-producing shrubs such as swamp dogwood were abundant in the Fakahatchee Strand. Two components common to the six habitats that were used regularly by all three species explained more than an average amount of variation in habitat use during each season (Table 3.9). Among these habitats, pine flatwoods accounted for lhe least variation suggesting that it was used consistently by all of South Florida's large carnivores (Table 3.10). The greatest variability during summer, however, was related to differential use of pine flatwoods and hardwood hammocks. Hammocks tended to be used more by female bobcats during summer than during other seasons and by male panthers during every season. The disproportionately high use of this habitat by male panthers and ferrinte bobcats may explain the consistent divergence between male and female panthers throughout the year and between male and female bobcats during summer. The conclusion of apparent differences between bobcat gender should be tempered, however, because of small sample size. On the other hand, Machr and Cox (1995) noted that similar trends in male panther preferences for hardwood hammock may represent resource partitioning that reduces the potential for competition between male and female panthers. The panther was the only species thai exhibited consistent divergence in seasonal habitat-use across seasons (Fig. 3.3). Although food habits studies have not demonstrated differences related to gender (Maehr et al. 1990), Harlow (1959) found that hardwood hammocks were the most productive habitats for white-tailed deer, and sign of wild hogs is more frequently encountered in this habitat type than elsewhere (pers. observ.). If male panthers use hardwood hammocks disproportionately because larger prey are more available in this upland plant community, this may represent the causal mechanism that facilitateR niche separation between male and female panthers. Such a pattern was not consistently found in the other two species. Overall, considerable habitat overlap among species was apparent, but male and female panthers exhibited more niche separation than bears or bobcats, and panthers differed more by gender within species during summer than with the other species. When compared to panthers, bobcats and black bears exhibited more within-species similarity, suggesting that gender had less influence on habitat use- patterns for these species. If seasonal habitat shifts were made, both sexes experienced them. In general, panthers used pine flatwoods and hardwood MAEHR. ECOLOGY BOBCAT, BLACK BEAR, PANTHER 41 hammocks more than was expected based on 1heir availability in the landscape, whereas cypress swamp, thicket swamp, freshwater marsh, grasslands, and agricultural areas were used less than expected (Table 3.8). These findings also support the conclusions of Maehr et al. (199la) that all panthers prefer upland forests. A mduction in bear use of mixed swamp and an increase in the use of pine flatwoods during fall was probably related to the wide-ranging excursions that many bears undertook to access abundant supplies of saw palmetto fruit Black bears also used agricultural/disturbed areas more than expected from mere abundance in the landscape. This likely was due to the abundance of Brazilian pepper thickets that often are associated with distu1bed sites and agricultural edges. Brazilian pepper thickets frequently replace agricultural fields when they are abandoned, and this species is a common colonizer of roadsides, ditchbanks, and spoil (Loope and Dunevitz 1981; Ewel et al. 1982; Abrahamson and Hartnett 1990). Black bears were the only species observed to utilize mangrove forests. Although use of maritime habitats by bears occurred year-round, the only activity that could be confirmed in these coastal swamps was denning by adult females. Male bears used mangroves proporuonally less than their availability while females used this cover type proportional to availability. Perhaps mangroves are more important for cover than for the procurement of food although black bears likely obtained some foods in mangroves. Foods that draw non-denning black bears to mangroves remain unidentified. The densely branched growth form of mangroves, often dense swarms of mosquitoes, and deep organic soils make these areas nearly impenetrable to humans and other South Florida cursorial mammals. Extensive impenetrable areas made up of Rhododendron spp. and Smilax spp. are similarly impenetrable and are typical in much of the black bear's inh:abited range (Pelton 1982:507). Despite the avoidance of unforested habitats by black bears, heIhncenus wetlands adjacent to large mixed swamps and mangrove forests occasionally were used to obtain nutritious foods, such as the nest-building colonial ant Crematogaster (see dietary overlap chapter). An avoidance of mangroves by panthers was also observed by Smith and Bass (1994) in Everglades National Park; an area where forested cover is sparse relative to forb- and sedge-dominated communities. Even though Foster (1992) observed that the bobcats in her study sample were frequently found in mixed swamp, the sample in either study may not have been representative of the population. However, both studies were similar in demonstrating individual preferences for pine flatwoods that exceeded 60 percent (Foster 1992:72) (Table 3.6). Thicket swamps and freshwater marshes were used less than expected by bobcats in this study, while grasslands and agricultural/disturbed habitats were used to valying degrees in inconsistent gengonal patterns that may reflect individual variation among bobcats or small sample size. 42 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) Saw palmetto thickets appear to be universally impoftant to bobcats in South Florida Foster (1992) suggested that this common pine flatwoods understoty plant was preferred, and Wassmer et al. (1988) found that saw pnlmetto thickets were important as natal dens. Female bobcat F03 denned in a palmetto thicket in the Bear Island Unit of the Big Cypress National Preserve during 1987, as did a female in Foster's (1992) study. Although Foster (1992) found thai bobcats avoided agricultural areas, I found that male bobcats exhibited vengonsilly variable use of this cover type that exceeded its availability (Table 3.6). Wassmer et al. (1988) found that male bobcats used several agricultural cover types more frequently than expected even though citrUS groves were avoided. Capture locations of resident adult bobcats no doubt can bias subsequent interpretations of habitat use. Most animals Foster (1992) studied were captured within the centers of large tracts of public lands whereas several of the animals in this study originated from private lands that bordered or included agricultural fields. Panthers and black bears were captured at widely scattered sites, and they were found to move long distances compared to the relatively sedentary bobcats with smaller home ranges. Thus, by vistue of their smaller spatial requirements bobcats captured on public lands were less likely to utilize agricultural habitats on adjacent private land. Unlike black bears and panthers, ~rnts in this study did not demonstrate a consistent preference for hardwood hammocks; in fact, this cover type was used in lower proportions than its availability during all seasons by males and during fall by females. It is possible that the generally more open understory of hammocks may not provide the cover conditions that are typically found in pine natwoods that contain saw palmetto. The more solid canopy associated with live oak and cabbage palm hammocks creates shaded conditions at ground level that tend to reduce saw palmetto density. Snyder et al. (1990) did not list saw palmetto as a common understory species in South Florida hammocks although it is found in evety variation of this plant community. In addition Loope and Urban (1980) found that saw palmetto occurred in only 2 of 100 tropical hardwood hammocks in Everglades National Pak The consistency with which saw palmetto was used by bobcats in this as well as other studies (Wassmer et al. 1988; Foster 1992) suggests this is a very important component of bobcat habitaL Measurements of dietary overlap indicated little likelihood of competition between any of South Florida's native carnivores. This was not the pattern, however, for measurements of habitat partitioning (Pianka 1986). While males and females from the same species usually exhibited close habitat affinities, black bears showed the least (0.75-0.87) similarities between gender (Table 3.11). Male and female panthers exhibited overlap exceeding 0.90 in all seasons and portrayed a higher level of similarity than was suggested by factor analyses. Overlap ranged from a minimum of 0.49 between female bobcats and female black bears during winter, to a maximum of 0.96 between male panthers and female black bears during winter. Although few patterns were consistent in this analysis, and they did MAEHR: ECOLOGY BOBCAT, BLACK BEAR, PANTHER 43 not universally agree with factor analyses, measurements hneerl on the Pinnkn (1986) algorithm clearly demonstrated a high level of habitat overlap within this group of species. A predominantly vegetarian diet different activity patterns, and winter denning allowed black bears to be involved in more high habitat overlap combinations than pairings between bobcats and panthers, which exhibited higher dietary'overlap (see dietary overlap chapter). When gender was ignored, however. the importance of upland forests, and inland swamps to South Florida's large mnmnulian carnivores is obvious (Fig. 3.4). The habitat overlap analysis (Table 3.11) supports the results of the factor analysis where some pairings of sympatric species exhibited greater overlap than between genders within species. For example, female black bear habitat use was more similar to Rmale panther habitat use during fall (0.965) than it was to male black bears (0.875). Some habitat use differences within species may be the product of evolutionary divergence that has reduced competition between the sexes. Sexual dimorphism occurs in all three species which, when combined with the behavior-altering influence of reproduction, may help to explain the observed patterns in habitat separation within species. While gender-dependent prey selection in bobcats and panthers may not be as dramatic as that reported in weasels (Mustela spp., Hall 1951; Lockie 1966; Erlinge 1974), and several bird species (Darwin 1871:208; Storer 1955; Selander 1966), even slight differences in prey selection may dampen the effects of periodic prey scarcity or allow the predator's population to increase during times of prey abundance. Prey selection differences between male and female bears are likely insignificant despite their high degree of sexual dimorphism. Differences in food habits among South Florida's native carnivores may be sufficient to trivialize the extensive habitat overlap that they demonstrate. But for the two cat species in South Florida, prey selection may be an important factor explaining the differences in habitat use between the sexes. t Table 3.1. Annual 1 L. years using Chi-equare analysis (p=0.05). Probabilities in bold fhce represent significant differences between individual habitat use in one year versus the previous year. Habitat type' ID# Year PP HH MS CS TS MA DP AD MG 2 Female Danthers 09 1986 19.6 20.9 34.8 12.9 1.1 0 0.6 0 - - - 09 1987 12.0 38.7 40.8 7.7 0 0 0.7 0 - 10.0 0.07 09 1988 7.6 26.7 46.5 19.1 0000 - 9.1 0.06 09 1989 3.7 16.1 52.8 26.7 0.6 000 - 6.2 0.18 09 1990 7.0 20.5 59.0 12.2 0.6 0.6 0 0 - 7.9 0.16 09 1991 11.5 26.6 52.5 8.6 0 0 0 0.7 - 4.7 0.57 09 1992 16.6 24.7 53.3 6.7 0 0.7 0 0 - 2.6 0.75 09 1993 12.2 23.0 56.1 8.8 0 0 0 0 - 1.8 0.77 09 1994 11.0 18.1 67,7 3.1 0 0 00 - 4.5 0.21 36 1990 30.7 24.3 7.1 37.9 0 0 0 0 - - 36 1991 43.7 24.3 9.0 22.9 0 0 0 0 - 6.2 0.10 36 1992 54.7 22.0 10.7 12.7 0 0 0 0 - 4.4 0.22 36 1993 50.7 14.9 10.1 24.3 0000 - 5.2 0.16 36 1994 47.6 25.8 14.5 12.1 0 0 0 0 - 7.9 0.05 Male Danthers 12 1986 26.5 39.1 22.3 11.4 0.6 000--- BU LLETIN FLO R ID A M U SEU M N ATU R A L H ISTO R Y V O L 40(1) 12 1957 32.1 22.6 29.8 14.9 0 0 0 0.6 - 7.7 0.17 12 1988 17.4 23.9 34.2 23.2 0.6 0.6 0 0 - 8.3 0.22 12 1989 19.9 27.7 16.3 30.1 5.4 0 0.6 0 - 12.7 0.05 12 1990 21.9 25.6 21.2 28.7 1.9 0 0 0.6 - 3.7 0.71 12 1991 28.7 40.7 14.7 13.3 0 0 0 0 - 12.1 0.03 12 1992 24.5 37.7 14.6 22.5 0.7 0 0 0 - 2.5 0.64 12 1993 32.4 42.7 13.8 10.3 0.7 0 0 0 - 5.9 0.20 12 1994 37.4 48.7 7.3 5.7 0.8 0 00 - 4.1 0.05 17 1987 32.9 36.2 14.1 14.1 0 0 2.7 0 - - - 17 1988 29.7 44.9 5.7 19.6 0 0 0 0 - 8.2 0.08 17 1989 28.4 45.8 7.1 16.8 1.3 0 0.6 0 - 2.3 0.80 Table 3.1. (continued) Hab~tattype' ID# Year PP HH MS CS TS MA DP AD MG Chf p M AEH IR : E C O LO G Y B O B C AT, B LA C K BEAR . P 45 Female black bears F02 1992 50.6 23.9 7.4 12.3 1.2 1.20 1.2 0 - F02 1993 22.4 30.2 21.0 18.4 1.3 0 0 6.6 0 219 a0005 F02 1994 15.5 21.8 6.2 23.4 10.9 0 0 21.9 0 27.0 40001 F03 1993 5.3 20.0 56.0 16.0 0 0 0 2.7 0 - - F03 1994 12.5 26.5 50.0 7.8 1.6 1.6 0 0 0 12.9 &04 F09 1992 12.2 19.5 2.4 8.5 3.7 3.7 0 3.7 46.3 - - F09 1993 16.9 19.7 4.2 7.0 4.2 2.8 0 1.4 43.7 _ 2.7 0.91 F19 1993 26.7 32.0 24.0 8.0 6.7 2.7 0 0 0 - - F19 1994 21.2 30.3 43.9 3.0 0 1.5 0 0 0 8.8 0.26 Male black bears M02 1991 7.3 33.3 4.4 2.9 8.8 8.8 0 0 32.3 - - M02 1992 6.3 30.4 8.9 1.3 6.3 15.2 0 3.8 27.8 8.8 0.26 M MOS 1992 24.7 9.1 14.3 46.7 2.6 1.3 0 1.3 0 - - M08 1993 25.9 6.3 9.1 46.7 0 1.3 0 10.4 0 11.3 0.08 M13 1992 33.7 15.0 3.7 18.7 0 0 0 27.5 1.2 - - M13 1993 39.7 9.6 8.2 12.3 9.6 1.4 0 6.8 12.3 37.3 40001 M13 1994 35.2 25.9 13.0 9.3 1.8 1.8 0 13.0 0 28.9 <0.0001 M18 1992 42.6 8.8 22.1 11.8 2.9 1.5 1.3 5.9 2.9 - - M18 1993 47.7 8.9 14.9 9.0 7.5 1.5 1.5 9.0 0 7.6 0.47 'PP~pine/pa~etto,HH=*,aniwoodhamn,oc~ MS=moted swamp, CS~pressgwamp, TZMhickl,Eswamp, MA-- ' ' andsanmarmh.DP=diypiairic. AD-egficulin,V~stwbed. MG-,ina;,grovas. 46 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 40(1) Table 3.2. Habitat use diffb,ences among female bobcats, black bears and panthers in South Florida as determined bymultipleanalysisof variance (p-0.05) ofhabitatuse frequencies. Only thosehabitats that indicated significant differences are reported. Winter Wilks' I.ambda20.0594 F=6.2 p=0.0001 Habitat type P Hardwood hamineck 12.44 0.0001 Cabbage palm 3.17 0.0113 Mixed swamp 4.05 0.027 Cypre= swa~ 6.40 0.0046 Herbaccous marsh 3.79 0.0333 Mangmve _ 3.35_ 0.048 Summer Wilks' I.ambda=0.0556 F=5.67 p=0.0001 Habitat type F P Pine flatwoods 3.47 0.0433 Pine/cabbage palm 4.17 0.0247 Hardwood hammock 16.97 0.0001 Mi*ed swamp 3.65 0.0374 Cyprm swamp 5.40 0.0095 Dry prairie/grassland 6.57 0.0041 Fail Wilks' I.ambda==0.0585 F=6.00 p=00001 Habitat type F Pine/cabbage palm 13.35 0.0001 Hardwood hammock 7.35 0.0022 Cabbage palm 5.78 0.0069 Mixed swamp 6.82 0.0032 Cypre= swamp _ 6.26 0.0049 MAEH*: ECOLOGY BOBCAT, BLACK BEAR PANTHER 47 Table 3.3. Habitat use differences among male bobcats, black bears and panthers in South Florida as determined by multiple analysis of varianoe (p=0.05) ofhabitat use frequencia Only those habitats that indicated significant differences are reported. Wintef Wilks' I.ambda=0.1064 F=3.443 p=0.0003 Habitat tylle _ F P Pine flatwoods 5.14 0.0118 Haniwood hanimock 17.00 0.0001 Mixed swamp 6.34 0.0049 Agricultural/disturbed 5.29 0.0106 Smnmer Wilks' Lambda=0.012 F=9.665 P=0.0001 Habitat type F Pine flatw=is 4.15 0.0279 Pine/cabbage palm 3.77 0.0370 Hardwood hammock 11.74 0.0003 Agricultural/disturbed 3.61 0.0421 Fall Wilks' Lambda=0.143 F=2.684 P=0.0041 Habitat type F Pine/cabbage palm 4.36 0.0225 Hardwood hammock 11.67 0.0002 Mixed swamp 3.50 0.0440 Agricultural/disturbed 5.29 0.0113 48 B U LLE TIN FLO R ID A M U SEU M N ATU R A L H ISTO R Y V O L 40(1) Table 3.4. Seasonal habitat use by Florida panthers (percent frequencyk 1986-1994. Abbreviations are: PP=pine flatwoods, HH=hardwood hammock. MS=mixed swamp, CS=cypress swamp, TS=thicket swamp, MA=freshwater and saltwater marsh, DP=dry prairie and grassland, AD=agricultural/disturbed, and MG=mangroves. ID# Sex Season Year PP HH MS CS TS MA DP AD Mel 09 F W 1986 14 29 34 21 2 0 0 0 - 11 F W 1986 30 5 8 8 0 0 0 0 - 19 F W 1992 39 24 16 18 0 0 0 2 - 31 F W 1993 51 31 4 10 2 0 0 2 - 32 F W 1992 20 20 33 26 0 0 0 0 - 09 F S 1986 17 25 51 5 2 0 0 0 - 11 F S 1986 31 54 9 3 2 0 0 0 - 19 F S 1992 22 26 16 36 0 0 0 0 - 31 F S 1993 27 19 10 37 4 2 0 0 - 32 F S 1992 44 19 17 19 0 0 0 0 - 09 F F 1986 28 39 18 12 0 0 2 0 - 11 F F 1986 30 32 17 20 0 0 0 0 - 19 F F 1992 40 34 14 12 0 0 0 0 - 31 F F 1993 42 18 16 14 6 0 0 4 32 F F 1992 37 37 10 14 0 0 0 2 12 M W 1986 22 38 27 13 0 0 0 0 - 13 M W 1987 27 46 15 3 5 0 2 0 - 26 M W 1993 53 28 6 12 0 0 0 0 - 46 M W 1993 23 58 4 12 2 0 0 0 - 51 M W 1994 311336140000- 12 M S 1986 19 51 21 9 0 0 0 0 - 13 M S 1987 26 41 13 9 0 4 6 0 - 26 M S 1993 47 36 6 13 0 0 0 0 - 46 M S 1993 23 47 6 21 2 0 0 0 - 51 M S 1994 22 18 53 6 0 0 0 0 - 12 M F 1986 39 28 18 13 2 0 0 0 - 13 M F 1987 42 37 4 16 0 0 0 0 - 26 M F 1993 51 33 8 8 0 0 0 0 - 46 M F 1993 27 57 2 12 0 2 0 0 - 51 M_F 1994 43 43 14 0 0 0 0 0 - Table 3.5. Seasonal habitat use by south Florida black bears (percent frequency). 1991-1993. Abbreviations are: PP=pine flatwooda, HH=hardwood hammock, MS=mixed swamp, CS=cypress swamp. TS=thicket swamp, MA=Reshwater and saltwater marsh DP=dry prairie and grassland. AD=agricultural/disturbed, and MG=mangroves. ID# Sex Season Year pp HH MS CS TS MA DP AD MG F02 F W 1992 73230030000 F03FW199300831700000 F04 F W 1992 3 26 42 22 6 0 0 0 0 FOSFW 199201240400475 F06 FW 1992 55 34 0300070 M AEH R : EC O LO G Y B O B C AT, B LA C K BEAR PAN TH ER 49 F02 F S 1992 23 15 31 8 11 11 0 0 0 F03 F S 1993 4 8 67 12 0 0 0 8 0 F04 F S 1992 8 32 40 20 0 0 0 0 0 FOS F S 1992 0 8 0 0 0 4 0 4 75 F06 F S 1992 41 30 7 22 0 0 0 0 0 F02 F F 1992 3616122804040 F03 F F 1993 11 48 22 18 0 0 0 0 0 F04FF1992 4212331200000 FOSFF 1992 15 15440001546 F06 F F 1992 69 11 8 11 0 0 0 0 0 M06 M W 1991 36 27 18 9 0 0 0 9 0 MOS M W 1992 7 3 17 62 7 0 0 1 0 M13M W 1993 2300429401225 M18 M W 1992 161621105002110 M20MW199374134900000 MOG M S 1992 7 36 36 0 0 0 0 21 0 MOS M S 1992 2512174600000 M13 M S 1993 26 17 22 26 0 0 009 M18 M S 1992 234382344400 M20 M S 1993 32 23 9 32 4 0 0 0 0 M06 MF 1992 26 13 43 13 04000 M08 M F 1992 50 14 9 32 0 4 0 0 0 M13 MF 1993 65 11 4800084 M18 M F 1992 87 9 4 0 0 0 0 0 0 M20MF199340404448000 50 B U LLE TIN FLO R ID A M U SEU M N ATU R A L H ISTO R Y V O L 40(1) Table 3.6. Seasonal habitat use by south Florida bobcats (percent frequency). 1986-1987. Abbreviations arc PP=pine flatwoods, HH=hardwood hammock; MS=mixed swamp, CS=cypress swamp, TS=thicket swamp, MA=freshwater and saltwater marsh, DP-My prairie and grassland, AD=agricultural/disturbed, and MO=mangroves, ID# Sex Season Year PP HH MS CS TS MA DP AD MG, F03 FW 1986 64 00280700- FO6FW 19876436000000- F07F W 1987 64140210000- F03 F S 1986 72 11 3 14 0 0 0 0 - FOG F S 1987 23 61 0 8 0 0 8 0 - F07 F S 1987 65 12 0 18 0 0 6 0 - F09 F S 1987 40 60 0 0 0 0 0 0 - F03 F F 1986 83 14 0 3 0 0 0 0 - FOG F F 1987 40 20 0 0 0 0 40 0 - F07 F F 1987 22 0 0 44 0 0 22 11 - MOI M W 1986 88 6060000- M02 M W 1986 45 45 4 4 0 0 0 0 - M05 M W 1987 77 0 0 9 4 0 0 9 MOSM W 1987 360181800027- M01 M S 1986 76 10 0 13 0 0 0 0 - M02 MS 1986 67 16 284200- MOSMS 198707281400050- MOIMF 1986 8730000100- M02MF19864654000000- MO5 M F 1986 68 11 0 14 0006- MOB M F 1987 28 0 0 28 0 0 43 0 - '11 · Table 3.7. Chi-square values and probabilities for seasonal, . - in south Florida 1986-1994. Winter Summer Fall Species Sex ID# Chiz p Chi' p Chiz p F 09 57.6 <0.001 88.0 <0.001 62.0 <0.001 F 11 59.9 <0.001 92.0 <0.001 56.6 <0.001 F 19 54.5 <0.001 45.3 <0.001 70.1 <0.001 F 31 75.1 <0.001 38.6 <0.001 57.4 <0.001 F 32 53.3 <0.001 60.6 <0.001 62.5 <0.001 M AEH R : EC O LO G Y B O B C AT, B LA C K BEAR . PA N TH E R 51 M 12 67.7 <0.001 77.3 <0.001 65.4 <0.001 Panther M 13 81.5 <0.001 49.9 <0.001 71.0 <0.001 ffi ffi f Firi firif irifir irrir IST / S S/5 51 M 26 76.4 <0.001 75.7 <0.001 82.6 <0.001 M 46 77.6 <0.001 60.4 <0.001 77.5 <0.001 M 51 66.8 <0.001 87.8 <0.001 12.6 0.013 F F02 140.5 <0.001 77.3 <0.001 59.8 <0.001 F F03 128.5 <0.001 49.9 <0.001 77.9 <0.001 F F04 77.8 <0.001 75.7 <0.001 97.5 <0.001 F F05 106.5 <0.001 60.4 <0.001 82.5 <0.001 F FOG 126.8 <0.001 87.8 <0.001 106.4 <0.001 M M06 95.1 <0.001 71.4 <0.001 81.2 <0.001 M MOB 65.8 <0.001 106.7 <0.001 69.6 <0.001 M M13 106.2 <0.001 78.7 <0.001 103.5 <0.001 M M18 72.9 <0.001 102.5 <0.001 145.8 <0.001 M M20 113.9 <0.001 79.0 <0.001 89.2 <0.001 F F03 90.4 <0.001 95.5 <0.001 128.2 <0.001 F FOG 121.0 <0.001 81.6 <0.001 123.8 <0.001 F WI lr;7 <0.001 90.0 <0.001 89. 1 <0.001 F F09 - - 113.4 <0.001 - M M01 133.5 <0.001 107.6 <0.001 153.4 <0.001 M M02 89.7 <0.001 92.2 <0.001 114.7 <0.001 M M05 137.2 <0.001 - - 102.1 <0.001 M MOS 97.0 <0.001 53.1 <0.001 104.2