Bobcat (Lynx rufus floridanus) Population Dynamics Research in Fragmented Landscapes

Introduction

This paper presents preliminary ecological data and analysis on four bobcats (two males and two females) that were captured, radio-collared, and monitored between January 1 and June 30, 1998 in a study being conducted by Coryi Foundation, Inc. The purpose of the study is to examine various ecological and biomedical aspects of bobcat populations in fragmented landscapes in order to understand associated dynamics and to create habitat conservation guidelines that are based on real time local science. This paper is brief and preliminary. The database amassed so far from both living and dead bobcats is much larger than presented in this paper. However, due to the infancy of the study and current small sample size, sufficient data has not yet been gathered to allow neither statistically rigorous analysis nor the development of valid conclusions. Nor are all analyses of the existing data included in this paper. Furthermore, papers will frequently present analyses by season. Since only one complete season (spring: March 22 - June 21) had occurred in the January-June time frame, only findings from the spring season are presented in this paper. Data and analysis of bobcat use of the landscape in terms of corridors, habitat patchiness, etc., as well as the results of population viability analyses, will be provided at a later time after sufficient data that can be rigorously quantified is obtained. The ultimate message to be conveyed is that this information is limited, reflects the infancy of the research, is not considered fully useful in its present state and should not be used by any person as a basis for any management decision. Research is ongoing and will eventually produce a data set that is sufficiently complete for the establishment of conclusions.

Methods

For the purposes of habitat selection analysis and home range size estimation, radio-collared bobcats are located a minimum of 30 times per calendar season. Only locations that are temporally independent are used in these analyses (Anderson 1982, Jacobsen and Wiggins 1982, Swihart and Slade 1984). Radio-collared bobcats are located via triangulation with radio-telemetry equipment at distances of less than 500 meters between the observer and animal. The difference in bearings at each location is 75 to 105 degrees.

Home range sizes were estimated by the 100% minimum convex polygon (MCP) (Mohr 1947) and 95% harmonic means (HM) (Dixon and Chapman 1980) methods. An estimate of density was not calculated due to the small sample size at this time.

Habitats were classified according to the modified Florida Land Use Cover Classification Scheme system from May (1993) and are coarsely classified as: urban and built up; agriculture; range land; upland forests; water; wetlands; barren land; and transportation, communication, and utilities. Within these categories are numerous sub-classifications. Selection indices were calculated for the following categories: males, females, and sexes combined. The habitat selection index for a given cover type was computed as the ratio of the number of observed locations in that type to the number of expected locations in that type. The number of expected locations in a given type was calculated as the product of the percentage of the home range area comprised of that cover type and the total number of locations. An index greater than one indicates that a habitat type is preferred. An index less than one indicates that a habitat type is avoided. An index equal to indicates that the animal does not care one way or the other about the habitat type - it randomly selects to use that type. Digitized maps were used in the selection index computation process.

Range sizes were correlated with habitat type for each sex by computing the slope of the line that describes the home range size as a function of the percentage of the home range that possesses the type.

Home Range Sizes

During the period, male and female MCP home range sizes averaged 17.8 and 8.4 square kilometers, respectively (Table 1). Respective HM range sizes averaged 10.1 and 6.9 square kilometers. The percent difference between the two home range size estimation methods was 43.1 and 17.4 for males and females, respectively. Male to female range size ratios were 2.1 and 1.5 for the MCP and HM methods, respectively.

Habitat Selection

Bobcats in the period used the following habitats (Table 2 and Figure 1): wetland forest mixed, pine flatwoods (upland coniferous forests), upland hardwood forest, shrub and brushland, wetland hardwood forest, low density residential, freshwater marsh, and mixed scrub-shrub wetland. Male, females, and combined sexes preferred wetland forest mixed, pine flatwoods, upland hardwood forest, and shrub and brushland. Indices for use of preferred types by males were (in order of ranking): 2.28 (wetland forest mixed), 2.02 (shrub and brushland), 1.98 (pine flatwoods), and 1.39 (upland hardwood forest). Indices for use of preferred types by females followed an identical ranking with values of: 2.99 (wetland forest mixed), 2.48 (shrub and brushland), 2.25 (pine flatwoods), and 1.48 (upland hardwood forest). Combined, male and female rankings were 2.71 (wetland forest mixed), 2.23 (pine flatwoods), 1.61 (upland hardwood forest), and 1.52 (shrub and brushland). Bobcats in all categories avoided freshwater marsh and mixed scrub-shrub wetland. Indices for males, females, and combined sexes for these two types were 0.91, 0.4, 0.52, 0.35, 0.61, and 0.35, respectively. Wetland hardwood forest and low density residential exhibited varying preferences in each category. Males randomly selected both of these types, while females avoided these types. Combined, both sexes avoided these types. Cover types not used included urban/built-up, agriculture, herbaceous and mixed rangeland, tree plantations, wetland coniferous forest (cypress), vegetated non-forested wetlands (excluding freshwater marsh and scrub-shrub wetland), barren land, and transportation, communication, and utilities. However, some of these types were not always represented in home ranges.

Habitat selection by males correlated well with females (CF=0.922). There was no significant difference in the way males and females ranked habitats (T=66,70, p=0.05, two-tailed, Wilcoxon Rank Sum). Male ranges possessed 41.9% of preferred types, while female ranges possessed 63.3%. Combined, bobcat ranges possessed 52.6% of preferred types. The percentages of developed types (urban/built-up excluding low density residential, agriculture, unnatural barren land, transportation, communication, and utilities) within home ranges were 30.6%, 45.0%, 35.0% for males, females and the combined sexes, respectively. The percentages of avoided or unused undeveloped types were 27.5%, 7.5%, and 21.4% for males, females, and combined sexes, respectively. There was no significant difference in the composition of male and female ranges (T=9,12, p=0.05, two-tailed, Wilcoxon Rank Sum).

Home Range Size as a Function of Habitat Composition

The percentage of pine flatwoods in a home range was the type that was most strongly correlated with the inverse of the home range size for males and females (Figure 2). That is, the greater the percentage of pine flatwoods that existed in a range, the smaller the range. However, only female home ranges exhibited a strong inverse correlation with the percentage of wetland forest mixed type found within their ranges. Male and female range sizes were strongly and positively correlated with the percentage of upland hardwood forest contained within a range. That is, the more upland hardwood forest contained within a range, the larger the range. However, only females exhibited a strong and positive correlation of range size with all other habitat types (those that were not used at all). All other correlation of range size and habitat type was weak.

Discussion

The findings presented in this paper represent a small sample size and merely one season. Thus, these findings can not be interpreted as conclusive nor be used as a basis for any management scheme. Conclusiveness and the development of a habitat conservation plan will occur as the result of an increased sample size and continued research. Be that as it may, the findings do allow some useful interpretation in terms of the fragmented landscape that foreshadows how the final analysis may look. However, the reader is again reminded of the preliminary nature of this discussion.

Home range sizes in the study reflect a typical density for this species (McCord and Cardoza 1982). However, there is an appreciable difference in the estimates between the MCP and HM methods. And it is with regards to this aspect that the bulk of discussion of bobcat ecology in the fragmented landscape is most insightful. This would primarily be due to the difference in the ways the two methods estimate range sizes.

The MCP method is a non-statistical, quick and simple estimation of range size that is determined by calculating the area formed by the minimum convex polygon that possesses apexes at the outermost locations. That is, one simply establishes a range boundary by drawing straight lines between the outermost locations and then determines the area enclosed by the boundary. Regardless of how locations are distributed within the boundary, the enclosed area is assumed (albeit erroneously) to be used uniformly by the animal. The MCP’s obvious shortcoming is that it does not represent a true use area because it includes appreciable unused regions within the boundary. These unused areas consist of avoided cover types. This bias can be most pronounced in landscapes where severe heterogeneity (chiefly due to habitat fragmentation via developed areas) leads to an appreciably uneven distribution of locations across cover types on which animals strongly differentiate preferences. For instance, where the landscape forces the animal to occupy a home range that is U-shaped or crescent-shaped, the arithmetic mean center will lie entirely outside of the home range, whereas the MCP center will appear inside the MCP boundary. The MCP tends to be most accurate in landscapes where the animal uses areas more uniformly. Such landscapes are largely undisturbed and/or may contain more or less evenly distributed preferred cover types.

On the other hand, the harmonic means method is a statistical estimation that generates home range boundary contours that are not restricted to the convex shape of outermost locations. This method evaluates location clusters as spatial representations of activity intensity (Dixon and Chapman 1980). Range boundaries are established as isopleth contours based on the location clusters and their harmonic means in an areal distribution. These contours represent activity cores that are correlated with areas of "equal" activity. The final products are a home range size and boundary that are more representative of actual use and are a greater approximation of true activity because unused space is more effectively excluded. Indeed, in some cases, the HM home range may actually be represented by two or more activity cores that are separated by "hostile" (or largely unused) areas through which the animal quickly travels. The cores tend to represent the preferred, and therefore, most used areas. Because unused (avoided) space is more effectively excluded from the home range boundary, HM ranges are generally smaller than MCP ranges. This method is very useful in fragmented landscapes where land cover diversity is great and habitat preferences strongly differentiated between preferred cover types and avoided types (i.e., developed portions) of a home range.

Comparing estimates of both methods can demonstrate the severity of internal home range patchiness and fragmentation. Generally (though not rigidly true), one may deduce that the greater the difference in range size estimates between the two methods for an "urban" bobcat, the greater the level of fragmentation and patchiness due to development (or other avoided cover types) within its home range.

Indeed, the home ranges in the study are severely fragmented internally as a result of development, as is the landscape at large. Bobcats are restricted in their use of the landscape to the undeveloped portions of their ranges, chiefly those that are forested or brushy as indicated in Table 2. Locations never fell within developed cover types, except in low density residential (seldomly) or when preferred types were bisected by a road (an avoided type). That the percentage of home range areas that possessed developed cover types ranged from 30.6-45% is indicative of appreciable internal patchiness. This intrarange heterogeneity results in the appreciable percent difference in home range size estimates between the MCP and HM methods.

The selection data suggests that bobcats prefer habitats that are forested (wet or dry), or possess dry and sufficient shrub and brush. Specifically, the wetland forest mixed, pine flatwoods, upland hardwood forest, and shrub and brushland types were preferred. Preferred cover types are probably selected as preferred due to higher availability of prey and more superior cover relative to avoided types. The data suggests that females may place a greater emphasis of selection on preferred cover types than do males. This may be attributed to the greater need of a female to easily acquire prey and readily access adequate cover while engaged in the demanding tasks of rearing kittens.

Among the preferred types, wetland forest mixed is a community in which hardwoods and conifers are present in the canopy. This wetland type may partly serve to provide bobcats with relatively cool rest sites during times or seasons of high temperatures. This type is not always inundated with flooding, as hydroperiods are a function of high groundwater, surface runoff or lake flooding. Pine flatwoods in the study area are typically inundated with saw palmetto (Serenoa repens), many patches that are thick, and other understory species. Such cover in the flatwoods is likely a principal benefit for bobcats. That home range sizes for both sexes are reduced the most by an increase in the percentage of enclosed pine flatwoods may suggest that this type is the most important in terms of prey. It is commonly known that range sizes are inversely proportional to prey availability. This response on the part of the bobcat implies that where there is abundant food, there is less need to travel widely in search of it, just as there is less need to defend a larger home range to conserve those prey resources. Shrub and brushland, an association where upland community types (presumably many of which were former pine flatwoods or other upland forests) have been disturbed as the result of clear-cutting, land clearing, or fire, are recovering through natural succession. Void of dense cover, bobcats probably favor this type for the abundant prey that can be found therein. This early successional community possesses a myriad of shrubs, herbs and grasses - conditions favorable for rabbits (Sylvilagus spp.) and small rodents such as cotton rats (Sigmodon hispidus) (Myers and Ewel 1990). Upland hardwood forests in the area include a variety of associations where the crown canopy is dominated by hardwood tree species. Included are types such as xeric oak, temperate hardwood, and hardwood conifer mixed. Temperatures, soils, hydrology, botanical diversity and density, and prey abundance and diversity vary among these types. As such, this wide suite of factors variably influences how bobcats use the upland hardwood forests.

The avoidance of freshwater marsh and shrubby wetlands is likely due to a combination of a lack of preferred prey, adequate cover, and suitably dry denning sites. With the above preference for forested or dry shrub and brush types, it is no surprise that freshwater marsh would be consistently avoided. This type is neither forested nor shrubby and is frequently too wet. Furthermore, all locations in this type were in the immediate vicinity of the marsh edge where the marsh meets a preferred habitat. That is, the interiors of freshwater marshes are not what bobcats use. It is speculated that the use of freshwater marsh edges is associated with high prey availability at an ecotone (upland-wetland interfacing edge). This speculation is plausible because species diversity is high at ecotones where species from the different adjoining habitats can be found to commonly occupy the narrow transition zone. Additionally, some prey species do have a preference for ecotones. All locations in the mixed shrub-scrub wetland were also near the edge of that wetland type and a preferred type.

Road crossing data was not presented in this paper but does exist. That data reveals that bobcats cross roadways where preferred habitats exist on both sides of the road. Furthermore, data suggests that certain individuals have adapted to survival strategies that increase the probability of a safe crossing. That data will be presented in a future paper.

The bobcats in the study appear to tolerate moderate levels of internal fragmentation and patchiness. These preliminary findings suggest that the bobcat possesses tenacity in adapting to and surviving in a fragmented landscape and that bobcat conservation in fragmented landscapes is feasible so long as certain conditions persist. These conditions not only include adequate preferred habitat types and associated prey, but also and presumably, that preferred habitat types need to be connected in an area that is large enough to sustain life history functions both at the individual and population level. Though not presented in this paper (to be presented in future papers), the radio-collared bobcats in the study use forested corridors. These corridors provide bobcats with movement pathways that connect larger forested areas within home ranges.

It is commonly assumed that bobcats are highly adapted to most habitats. While this is generally true, the data does suggest a refinement of this view because of the apparent need for forests - both in the early and late stages of the sere. With the exception of low density residential (which was selected randomly or avoided and accounted for a mere 4% of the combined home range areas), the bobcats in the study did not use developed types. It must be pointed out that the low density residential areas that were used by the bobcats were those on which the properties possessed ample intact forests. In contrast, recognizing what is known about bobcats in terms of the preference for forested cover, open short-grass or otherwise barren properties that are largely void of trees and shrubs are of little value for bobcats. Development that removes natural vegetation and replaces it with treeless, well-manicured lawns, sports fields, golf courses, etc., surely diminishes available habitat and degenerates population viability. Refinement of the pervading view is one that utilizes the answer to the question of how tolerant the bobcat is of habitat patchiness within the developed matrix that contains a myriad of these development types. Ideally, within undeveloped areas, the landscape would best meet the needs of bobcats if natural habitats were managed in a mosaic of late and early stages of the sere so that there is ample food supply near the cover bobcats like. The geometric patterns associated with this aspect of conservation and management will be addressed in the research and in future papers, especially with regards to managed corridors.

Ongoing Research

Landscape and population level issues are at the crux of this study in terms of understanding the full spectra of mechanisms that play key roles in population ecology and viability in the fragmented environment. These issues include, but are not limited to, the quantification and evaluation of corridor criteria, demographic stability, movement patterns, dispersal and recruitment success, landscape patterns, energetic responses, reproduction, and biomedical health. We have barely scratched the surface in terms of understanding these aspects. Only with prolonged research and innovative analyses can the mechanisms be fully understood. And it is in such a direction that this research is proceeding. Access to and trapping on new lands has been recently initiated since June 30. As a result, the sample size will increase in the near future. Capture rates in the time frame (4 animals/7 months) are considered normal for this species and is an encouraging start. In time, an increased sample size and data set will produce information that will answer our corridor use and population viability questions.

Acknowledgements: Thanks are extended to the Friends of the Enchanted Forest, Inc. and B. Hoelscher for their financial contributions to the initial phases of this research. Thanks are also extended to the Brevard County EELS program staff for their input at various times during the course of this research.

Literature Cited

Anderson, D.J. 1982. The home range: a new nonparametric estimation technique.
    Ecology 63(1): 103-112.
Dixon, K.R. and J.A. Chapman. 1980. Harmonic mean measure of animal activity areas.
    Ecology 61(5):1040-1044.
Jacobsen, N.K. and A.D. Wiggins. 1982. Temporal and procedural influences on activity
    estimated by time-sampling. J. Wildl. Manage, 46(2):313-324.
May, B. 1993. St. Johns River Water Management District Geographical Information
    System Data Documentation. SJRWMD. Palatka. 146 pp.
McCord, C. and J. Cardoza. 1982. In Chapman, J. and G. Feldhammer (eds.). Wild
    mammals of North America. University of Chicago Press, Chicago, Illinois.
Mohr, C.O. 1947. Table of equivalent populations of North American small mammals.
    Am. Midl. Nat. 37(1):223-249.
Myers, R.L. and J.J. Ewel. 1990. Ecosystems of Florida. University Presses of Florida.
    Gainesville.
Swihart, R.K. and N.A. Slade. 1984. Testing for independence of observations in animal
    movements. Ecology 66(4):1176-1184.

Home Page
Research Page
Population Dynamic Research Page
Conservation Page
Education Page