Food habits of a small black bear population in an endangered ecosystem Authors: Sean M. Murphy, Wade A. Ulrey, Joseph M. Guthrie, David S. Maehr, Warren G. Abrahamson, et. al. Source: Ursus, 28(1) : 92-104

Published By: International Association for Bear Research and Management

URL: https://doi.org/10.2192/URSU-D-16-00031.1

BioOne Complete (complete.BioOne.org) is a full-text database of 200 subscribed and open-access titles in the biological, ecological, and environmental sciences published by nonprofit societies, associations, museums, institutions, and presses.

Your use of this PDF, the BioOne Complete website, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/terms-of-use.

Usage of BioOne Complete content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research.

Downloaded From: https://www.bioone.org/journals/Ursus on 1/2/2019 Terms of Use: https://www.bioone.org/terms-of-use Access provided by Norwegian University Library of Life Sciences Food habits of a small Florida black bear population in an endangered ecosystem

Sean M. Murphy1,3,4, Wade A. Ulrey1,5, Joseph M. Guthrie1,6, David S. Maehr1,7, Warren G. Abrahamson2, Sutton C. Maehr1,8, and John J. Cox1

1Department of Forestry, University of Kentucky, Lexington, KY 40546, USA 2Department of Biology, Bucknell University, Lewisburg, PA 17837, USA

Abstract: The Highlands–Glades subpopulation (HGS) of Florida, USA, black bears (Ursus ameri- canus floridanus) is small, genetically depauperate, and resides primarily within the endangered Lake Wales Ridge ecosystem, which has lost >85% of native habitat to land development. Habitat loss can reduce availability of critical natural foods and cause bears to increase reliance on anthropogenic foods (i.e., human-sourced); lands supporting the HGS are expected to lose >50% of remaining Florida black bear habitat in coming decades. We used scat analysis to describe seasonal food habits, inves- tigate potential dietary responses to food shortages, and inform habitat conservation and human–bear conflict management. Florida black bears in the HGS mostly relied on native soft and hard mast and invertebrates, which are all available in endangered scrub habitat communities. Corn dispensed at hunter-operated feeding stations was a dominant food item in scats; and other alternative foods, such as citrus fruit and white-tailed deer (Odocoileus virginianus), were found in summer-collected scats when soft mast should have been prevailing. Results indicate bears may respond to soft mast shortages caused by mast failures or habitat loss by consuming anthropogenic foods (e.g., corn, deer chow, citrus fruit, and garbage), which could increase human–bear conflicts. Florida carpenter ants (Camponotus floridanus) appeared to provide a reliable compensatory food during such shortages, but they are ar- boreal and largely dependent on imperiled bear habitat for proliferation. We strongly suggest remnant scrub and other communities rich in soft and hard mast-producing flora be targeted for acquisition and protection to ensure persistence of Florida black bears in this diverse ecosystem. We also suggest non-lethal actions to mitigate bear habituation to anthropogenic foods be implemented to minimize human–bear conflicts and prevent unnecessary losses to the already small HGS. Our study should be repeated to investigate whether dietary shifts occur in response to impending habitat loss and to further inform population conservation, habitat protection, and conflict management.

Key words: anthropogenic, diet, Florida, Florida black bear, food habits, habitat fragmentation, habitat loss, human– bear conflict, Lake Wales Ridge, Ursus americanus floridanus

DOI: 10.2192/URSUS-D-16-00031.1 Ursus 28(1):92–104 (2017)

Widespread habitat destruction caused by urban devel- land uses subdivided a formerly contiguous statewide opment, agriculture expansion, and other anthropogenic Florida black bear (Ursus americanus floridanus) popula- tion into 7 disjunct subpopulations with varying levels of connectivity among them (Dixon et al. 2007). State-level 3email: [email protected] protection and habitat restoration efforts facilitated abun- 4Present address: Carnivore Program, New Mexico dance increases and some range reoccupation by most Department of Game and Fish, Santa Fe, NM 87507, USA Florida black bear subpopulations during the past decade 5 Present address: The Nature Conservancy, Swanton, OH (Humm et al. 2017). However, results from a recent 43558, USA non-invasive genetic spatial capture–recapture study that 6Present address: Nelson Byrd Woltz Landscape Architects, Charlottesville, VA 22902, USA used sampling and analytical methods similar to those 7Deceased used by Murphy et al. (2016) indicated the Highlands– 8Present address: Large Carnivore Program, Louisiana Glades subpopulation (HGS) in south-central Florida, Department of Wildlife and Fisheries, Baton Rouge, LA USA, remains small (N = 98 bears) and genetically de- 70898, USA pauperate (effective population size [NE] = 25 bears;

92

Downloaded From: https://www.bioone.org/journals/Ursus on 1/2/2019 Terms of Use: https://www.bioone.org/terms-of-use Access provided by Norwegian University Library of Life Sciences r BEAR FOOD HABITS IN AN ENDANGERED ECOSYSTEM Murphy et al. 93

S. M. Murphy and W. A. Ulrey, unpublished data). The such as dietary shifts and increased reliance on human- primary limiter to population growth and genetic recov- sourced foods (Hopkins et al. 2014, Ditmer et al. 2016), ery of the HGS is habitat availability (Maehr et al. 2004, and inform habitat and population conservation in areas Florida Fish and Wildlife Conservation Commission where the loss of critical habitat is an imminent threat to [FWC] 2012). long-term population persistence. The HGS is supported by a mosaic of native habi- Limited availability or accessibility of natural foods tat communities that are disjunctly arranged in the en- because of habitat quality and quantity issues can in- dangered Lake Wales Ridge ecosystem and adjacent ar- crease the probability that bears exploit anthropogenic eas across a primarily agricultural and urban landscape foods (Benson and Chamberlain 2006, Merkle et al. 2013, (Pearlstine et al. 1997, Maehr et al. 2004, Weekley et al. Ditmer et al. 2016), which are human-sourced foods 2008; Fig. 1). This includes federally endangered scrub made available to bears via human activities (e.g., agricul- communities that support numerous endemic and imper- ture, hunting, and urban development; Oro et al. 2013). iled flora and fauna (U.S. Fish and Wildlife Service [US- Consumption of anthropogenic foods by bears is asso- FWS] 1999). Although a hotspot for biodiversity (Noss ciated with human–bear conflicts; therefore, food habits et al. 2015), the endangered Lake Wales Ridge ecosys- studies can also inform conflict management (Greenleaf tem has experienced a >85% area loss of native com- et al. 2009, Hopkins et al. 2014). Human–black bear con- munities (Weekley et al. 2008), and lands occupied by flicts increased rapidly throughout Florida over the past the HGS constitute one of the most fragmented areas in decade, particularly in the central portion of the state North America with an established black bear popula- where multiple bear attacks on humans occurred in recent tion (Maehr et al. 2001, 2004). Nonetheless, the HGS years, which prompted the state wildlife management has the potential to serve as a stepping stone that facili- agency to increase bear euthanasia in urban and exurban tates demographic and genetic connectivity between the areas and implement a controversial legal harvest (FWC southern-most Big Cypress subpopulation and all other 2015). Although those are oft-employed agency man- Florida black bear subpopulations to the north (Dixon agement responses to chronic conflicts (Agee and Miller et al. 2007; Fig. 1 inset). Central Florida is, however, 2009, Penteriani et al. 2016), similar actions applied to expected to incur substantial anthropogenic development the HGS could further reduce the number of black bears and the largest percentage of habitat loss (>50%) in the in this already small subpopulation. state in coming decades (Carr and Zwick 2016), and in- To address said issues and provide managers with tervening habitat between those subpopulations is threat- science-based information on HGS black bear ecology, ened by rising sea levels from climate change (Whittle we conducted a food habits study in which we collected 2009, Carr and Zwick 2016). and analyzed Florida black bear scats over a 2-year pe- Contrary to all other Florida black bear subpopula- riod. We also investigated whether a lack of natural foods tions, management of the HGS has relied almost exclu- in bear scats was associated with an increased occur- sively on anecdotal information (FWC 2012). A lack rence of anthropogenic foods in scats. Our rationale was of scientific knowledge about HGS bear ecology has that, in the absence of direct measurements of natural thwarted conservation efforts (Maehr et al. 2004) and foods availability (e.g., rigorous mast surveys; Ryan et al. thus possibly hindered resiliency of the HGS to impend- 2007), an apparent replacement of natural foods with an- ing habitat loss. Food habits studies are an important ini- thropogenic foods may suggest limited availability or tial step toward a more comprehensive understanding of accessibility of the former (Greenleaf et al. 2009, Merkle ecological relationships of bear populations (Beckmann et al. 2013, Hopkins et al. 2014). Collectively, findings and Berger 2003, Lesmerises et al. 2015, Costello et al. from our study should be useful for habitat and subpop- 2016). For small or imperiled bear populations, such in- ulation conservation efforts and for human–bear conflict vestigations may be vital for informing landscape-level management in the HGS. conservation planning (Fortin et al. 2013, Ciucci et al. 2014). Although habitat productivity and therefore food availability change dynamically because of natural eco- Study area logical processes (e.g., transitions between seasons), an- The 998-km2 primary range of the HGS (Simek et al. thropogenic land-use activities can permanently degrade 2005; Fig. 1) is located in Highlands and Glades coun- important habitats and decrease availability of critical ties of south-central Florida, where the climate is humid natural foods. Dietary studies can facilitate monitoring sub-tropical with hot, wet summers and mild, dry winters. of bear population responses to landscape-level changes Average elevation is 47 m above sea level, average annual

Ursus 28(1):92–104 (2017)

Downloaded From: https://www.bioone.org/journals/Ursus on 1/2/2019 Terms of Use: https://www.bioone.org/terms-of-use Access provided by Norwegian University Library of Life Sciences r 94 BEAR FOOD HABITS IN AN ENDANGERED ECOSYSTEM Murphy et al.

Fig. 1. Available habitat and primary range (Simek et al. 2005) of the Highlands–Glades subpopulation of Florida black bears (Ursus americanus floridanus) in south-central Florida, USA, where 166 bear scats were collected during 2004–2006 to describe food habits.

◦ temperature is 21.2 C, and average annual precipitation is Approximately 36% (359 km2) of lands within HGS 136 cm (Archbold Biological Station 2010). Native hard primary range are considered potential Florida black bear mast–producing trees include the Florida-endemic scrub habitat (Fig. 1). Many lands within primary range have hickory (Carya floridana) and 7 species of oak (Quercus been converted to agriculture (41%; 409 km2), primarily spp., including the Florida-endemic Q. inopina). Domi- citrus groves (12%; 120 km2) and open pasture (23%; 230 nant soft mast-producing shrubs and plants are saw pal- km2; FWC 2014). Twelve percent (120 km2)ofprimary metto ( repens), Florida-endemic scrub palmetto range has been developed as urban areas (FWC 2014); (Sabal etonia), blueberry (Vaccinium spp.), and black- average human population density is 37 people/km2 and berry (Rubus spp.; Abrahamson et al. 1984). the largest city is Sebring (10,331 people). Numerous

Ursus 28(1):92–104 (2017)

Downloaded From: https://www.bioone.org/journals/Ursus on 1/2/2019 Terms of Use: https://www.bioone.org/terms-of-use Access provided by Norwegian University Library of Life Sciences r BEAR FOOD HABITS IN AN ENDANGERED ECOSYSTEM Murphy et al. 95

freshwater lakes and wetlands represent the remaining gist who was familiar with species in the study area (M. land area within primary range (11%; 110 km2). A. Deyrup, Archbold Biological Station). We classified all vegetation items (e.g., leaves, grasses, and forbs) into a single category (i.e., vegetation) because they typically Methods have similar nutritional value (Dahle et al. 1998, Klare Field sampling, scat classification, and food et al. 2011). We discarded black bear hair, which was items identification presumably ingested by bears when grooming. From May 2004 to April 2006, we collected Florida black bear scats opportunistically along roads and trails, at locations where bears were live-captured as part of a re- Data analyses source use study (Ulrey 2008, Guthrie 2012), and where Seasonal sample sizes within a bear-year (i.e., May otherwise we had permission to access bear-occupied 2004–Apr 2005 and May 2005–Apr 2006) were rela- lands (Fig. 1). We placed each collected scat in a sealed tively small, so we pooled season-specific data from both plastic bag labeled with the date of collection and Univer- years. Most previous Florida black bear studies only used sal Transverse Mercator coordinates, and froze samples frequency of occurrence (FO) to describe food habits ≤4 hours post-collection. Scats decomposed rapidly in (Maehr et al. 2001), and an objective of our study was to the hot, humid, and wet study area (W. A. Ulrey and J. M. compare food habits among subpopulations. Frequency Guthrie, personal observation); therefore, we presumed of occurrence is also particularly appropriate for inves- all collected scats reflected the month of defecation. tigating diet diversity (Bojarska and Selva 2012), which We classified scats into 3 seasonal periods that corre- was of importance for the HGS because of impending sponded to Florida black bear biology and natural foods habitat loss and the associated potential decline of natu- availability (FWC 2012): winter–spring (Jan–Apr), en- ral foods. Therefore, we calculated seasonal FO of each compassing hypophagia when parturient females den food item at the lowest identifiable taxon and seasonal and also when den emergence occurs; summer (May– FO of major food categories using the same method used Aug) during breeding season and soft mast ripening; by Ciucci et al. (2014) and Larson et al. (2015). We com- and autumn (Sep–Dec) when hard mast becomes avail- pared FO of the major food categories (soft mast, hard able and bears exhibit hyperphagia. We combined winter mast, invertebrates, vertebrates, vegetation, and anthro- and spring into a single period because sample sizes for pogenic) across seasons using a χ2 test (Stenset et al. both seasons were independently small (n < 20 scats 2016). Data pooling was necessary, so we averaged FOs each season/yr) and because radiomonitoring data (n = among seasons and calculated standard deviation as a 64 bears; 2004–2009) indicated many male and non- measure of variance to provide a more reliable approxi- parturient female bears were winter-active (Ulrey 2008, mation of food item importance across a given year in the Guthrie 2012). absence of volume-based metrics (Leger and Didrichsons We thawed and rinsed each scat through a single sieve 1994, Ciucci et al. 2014, Stenset et al. 2016). (1.5-mm mesh) and then mixed the contents. We did not To investigate whether a lack of natural foods in scats subsample for our analyses (e.g., Ciucci et al. 2014), but influenced the odds of a scat containing anthropogenic instead evaluated each scat in its entirety, hand-separating foods, we fit logistic regression models using maximum macro-categories (e.g., grasses, seeds, hair, and bones). likelihood via the glm function in R (Cherry et al. 2016, We sorted and identified all food items to the lowest tax- R Core Team 2016). We used binary (i.e., 1 or 0) oc- onomic resolution possible using field guides and iden- currences of food categories instead of FO because >1 tification manuals (e.g., Martin and Barkley 1961, Bor- item within individual categories was present in most ror and White 1970, Peterson 1999, Marshall 2006). We scats and totals likely did not lack independence. We identified seeds by comparing them with a reference col- used anthropogenic foods as the response variable, sea- lection from the study area (Archbold Biological Station, son and natural foods categories as independent vari- Venus, Florida, USA), and examined mammal hairs as ables, and one level of each variable as a reference. We medulla slides that depicted the outside structures of hair used Akaike’s Information Criterion corrected for small under a 40 × microscope and classified them to species sample size (AICc) for model selection (Burnham et al. using keys and reference collections from the study area 2011), assigned significance to β estimates from the most (Archbold Biological Station). We identified parsimonious model if P < 0.05, and converted β esti- to taxonomic order and to genus or species when possible; mates to odds ratios. We assessed goodness-of-fit of our identifications were verified by a professional entomolo- top model using a le Cessie–van Houwelingen–Copas–

Ursus 28(1):92–104 (2017)

Downloaded From: https://www.bioone.org/journals/Ursus on 1/2/2019 Terms of Use: https://www.bioone.org/terms-of-use Access provided by Norwegian University Library of Life Sciences r 96 BEAR FOOD HABITS IN AN ENDANGERED ECOSYSTEM Murphy et al.

Fig. 2. Seasonal variation in frequency of occurrence (FO) of food item categories found in 166 Florida black bear (Ursus americanus floridanus) scats collected during 2004–2006 in the Highlands–Glades subpopulation of south-central Florida, USA. Results of χ2-tests comparing FO of food item categories among seasons are also presented. Categories were hard mast (Hard), soft mast (Soft), vegetation (Veg), invertebrates (Invert), vertebrates (Vert), and anthropogenic (Anthro).

Hosmer unweighted sum-of-squares test (Hosmer et al. Feral hog (Sus scrofa), white-tailed deer (Odocoileus vir- 1997) via the R package rms (Harrell 2017). ginianus), and gopher tortoise (Gopherus polyphemus) remains in summer scats resulted in the highest seasonal FO for vertebrates. Among anthropogenic foods, corn Results was the most dominant across all seasons, followed by During 2004–2006, we collected 166 Florida black deer chow during winter–spring and citrus fruit during bear scats, 27 (16.3%) of which were from 17 individ- summer and autumn; we documented few occurrences ual bears that were live-captured during our study pe- of items indicative of garbage consumption (e.g., paper riod. We recorded 532 occurrences of 50 different na- and plastic). The unweighted sum-of-squares test demon- tive, non-native, and anthropogenic food items (Table strated adequate fit of the top regression model (SSE = = = 1): winter–spring scats contained 96 occurrences of 28 32.19; Z 1.00; P 0.31), which included both items, summer scats contained 151 occurrences of 35 soft mast and vegetation as explanatory food categories items, and autumn scats contained 285 occurrences of 42 (Table 2). Soft mast was the most important predictor of items. Seventeen (34.0%) of the 50 individual food items anthropogenic food occurrence in scats; the odds of a scat were present in scats from all 3 seasons. containing anthropogenic foods significantly increased in We found significant FO differences among seasons for the absence of soft mast (Table 3). all natural foods categories but not anthropogenic foods (Fig. 2), and average annual FOs generally exhibited con- siderable variation (i.e., SD; Table 1). Among the major Discussion food categories, soft mast, invertebrates, and hard mast Across the southeastern United States, oak acorns pro- were dominant across a year. Oak acorns had the highest vide the calories and macronutrients (e.g., crude fat) in- average annual FO among specific food items, followed tegral to black bear body fat accumulation during au- by anthropogenic corn and native Florida carpenter ants tumn hyperphagia (Pelton 2001), and our findings ex- (Camponotus floridanus). Oak acorns, Florida carpen- tend this generalization to the HGS. Carbohydrate-rich ter ants, and saw palmetto fruit had the highest FOs in saw palmetto also fruits during autumn and is considered winter–spring, summer, and autumn scats, respectively. the single most important natural food item throughout

Ursus 28(1):92–104 (2017)

Downloaded From: https://www.bioone.org/journals/Ursus on 1/2/2019 Terms of Use: https://www.bioone.org/terms-of-use Access provided by Norwegian University Library of Life Sciences r BEAR FOOD HABITS IN AN ENDANGERED ECOSYSTEM Murphy et al. 97

Table 1. Frequency of occurrence of food items in Florida black bear (Ursus americanus floridanus) scats collected in the Highlands–Glades subpopulation of south-central Florida, USA, during the winter–spring (W), summer (S), and autumn (A) seasons of 2004–2006. Dashes (—) indicate the food item was not documented during that season, and # and n correspond to the total number of occurrences of each food item among all collected scats and the scat sample size, respectively.

W S A Annual x¯ Food item # n = 32 n = 39 n = 95 n = 166 SD Hard mast 79 50.0 5.1 51.6 35.6 26.4 Oak acorn (Quercus spp.) 63 50.0 2.6 47.4 33.3 26.6 Hickory nut (Carya spp.) 16 — 2.6 11.6 4.7 6.1 Soft mast 126 18.7 71.8 67.4 52.6 29.5 Saw palmetto (Serenoa repens) 61 3.1 10.2 56.8 23.4 29.2 Unknown fruit 25 12.5 20.8 10.5 14.6 5.5 Grape (Vitis spp.) 17 — 41.0 1.0 14.0 23.4 Gallberry (Ilex glabra) 7 3.1 2.6 6.3 4.0 2.0 Blueberry (Vaccinium spp.) 5 3.1 7.7 1.0 3.9 3.4 Blackberry (Rubus spp.) 4 — 7.7 1.0 2.9 4.2 Swamp tupelo (Nyssa biflora) 3 — — 3.2 1.1 1.8 Cabbage palm () 2 — — 3.2 1.1 1.8 Dahoon holly (Ilex cassine) 1 3.1 — — 1.0 1.8 Brazilian pepper (Schinus terebinthifolius)a 1 — — 2.1 0.7 1.2 Vegetation 51 46.9 41.0 15.8 34.6 16.5 Leafy green vegetation 11 31.2 2.6 — 11.3 17.3 Vegetation (general) 12 9.4 15.4 5.3 10.0 5.1 Plant fiber 10 9.4 7.7 5.3 7.5 2.1 Grass 11 — 12.8 6.3 6.4 6.4 Palm heart (Sabal palmetto or Serenoa repens) 3 3.1 2.6 1.0 2.2 1.1 Ragweed (Ambrosia spp.) 2 — — 3.2 1.1 1.8 Smartweed (Polygonum spp.) 1 3.1 — — 1.0 1.8 Lichen 1 — — 1.0 0.3 0.6 Invertebrate (Arthropoda) 163 40.6 66.7 48.4 51.9 13.4 Hymenoptera 79 12.5 61.5 28.4 34.1 25.0 Florida carpenter ant (Camponotus floridanus) 49 12.5 48.7 26.3 29.2 18.3 Acrobat ant (Crematogaster spp.) 7 3.1 15.4 — 6.2 8.1 Bumble bee (Bombus spp.) 5 — 12.8 — 4.3 7.4 Unknown bee 5 3.1 7.7 1.0 3.9 3.4 Unknown ant 11 — 2.6 2.1 1.6 1.4 Yellow jacket (Vespula spp.) 1 — 2.6 — 0.9 1.5 Guinea wasp (Polistes exclamans) 1 — 2.6 — 0.9 1.5 Coleoptera 43 25.0 12.8 25.3 21.0 7.1 Unknown 17 28.1 10.2 17.9 18.7 9.0 (Curculio spp.) 15 15.6 — 10.5 8.7 7.9 Scarab beetle (Scarabaeidae) 2 6.2 7.7 6.3 6.7 0.8 Bess bug (Odontotaenius disjunctus) 9 6.2 5.1 — 3.8 3.3 Unknown 20 9.4 20.5 12.6 14.2 5.7 Isoptera 13 3.1 12.8 7.4 7.8 4.9 Subterranean termite (Rhinotermitidae) 8 3.1 5.1 5.3 4.5 1.2 Unknown termite 5 — 7.7 2.1 3.3 4.0 Metastigmata 5 — 2.6 3.2 1.9 1.7 Tick (Ixodidae) 5 — 2.6 3.2 1.9 1.7 Odonata 1 3.1 — — 1.0 1.8 Dragonfly (Anisoptera) 1 3.1 — — 1.0 1.8 Diptera 1 — — 1.0 0.3 0.6 Unknown fly 1 — — 1.0 0.3 0.6 Vertebrate 30 6.2 23.1 15.8 15.0 8.5 Mammalia 21 6.2 20.5 11.6 12.8 7.2 Feral hog (Sus scrofa)a 4 — 7.7 1.0 2.9 4.2 White-tailed deer (Odocoileus virginianus) 3 — 7.7 — 2.6 4.4 Unknown hair or tissue 4 3.1 — 3.2 2.1 1.8 Nine-banded armadillo (Dasypus novemcinctus)a 4 — 2.6 3.2 1.9 1.7 Raccoon (Procyon lotor) 6 3.1 — 2.1 1.7 1.6

Ursus 28(1):92–104 (2017)

Downloaded From: https://www.bioone.org/journals/Ursus on 1/2/2019 Terms of Use: https://www.bioone.org/terms-of-use Access provided by Norwegian University Library of Life Sciences r 98 BEAR FOOD HABITS IN AN ENDANGERED ECOSYSTEM Murphy et al.

Table 1. Continued.

W S A Annual x¯ Food item # n = 32 n = 39 n = 95 n = 166 SD Reptilia 4 — 2.6 3.2 1.9 1.7 Gopher tortoise (Gopherus polyphemus) 4 — 2.6 3.2 1.9 1.7 Aves 5 — — 1.0 0.3 0.6 Unknown feathers 5 — — 1.0 0.3 0.6 Anthropogenic 75 40.6 35.9 23.2 33.2 9.0 Corn 41 40.6 28.2 18.9 29.2 10.9 Deer chow 13 31.2 7.7 1.0 13.3 15.9 Unknown grain 7 15.6 — 3.2 6.3 8.2 Citrus 12 — 10.2 7.4 5.9 5.3 Paper or plastic 2 3.1 — 1.0 1.4 1.6 Unknown food item 8 9.4 2.6 3.2 5.1 3.8

aNon-native natural food.

Florida black bear range (Maehr et al. 2001; Table 4). 23,000 joules/gm dry mass; Abrahamson and Abraham- Collectively, energy-rich oak acorns and saw palmetto son 1989). Although autumn mast failures can have neg- fruit dominated autumn scats in the HGS (∼18,000– ative demographic consequences on bear populations (Rogers 1976), Florida black bears can overcome such failures via a compensatory reliance on oak acorns and Table 2. Logistic-regression model selection inves- tigating whether a lack of seasonal occurrences of saw palmetto fruit (Stratman and Pelton 1999). As a re- natural food categories in Florida black bear (Ursus sult, productivities of those masting species likely act as americanus floridanus) scats influenced the odds primary predictors of HGS reproductive rates and popu- of a scat containing anthropogenic foods for the lation growth (Rogers 1976, Maehr et al. 2001, Obbard Highlands–Glades subpopulation in south-central and Howe 2008). Considering the severe habitat loss that Florida, USA (2004–2006). We used anthropogenic foods as the response variable, and season and oc- is expected to occur in the Lake Wales Ridge ecosystem currence of soft mast (Soft), hard mast (Hard), inver- (Carr and Zwick 2016), protection of remaining scrub tebrates (Invert), vertebrates (Vert), and vegetation communities that harbor both saw palmetto and most oak (Veg) as explanatory variables. We used occurrence species (Weekley et al. 2008) may be critical to long-term of each natural foods variable as the reference level persistence of the HGS. to evaluate non-occurrence. For brevity, only models Winter-active male and non-parturient female black ≤2 AICc are presented. bears are common in the milder climate of the South- a  b c d Model AICc AICc wi logLik eastern Coastal Plain where natural foods are available ∼ Soft + Veg 194.90 0.00 0.10 −94.36 ∼ Soft + Season 195.43 0.48 0.07 −93.55 ∼ Soft + Veg + 195.61 0.68 0.06 −93.65 Table 3. Model-averaged coefficient estimates (β) Hard from the top logistic-regression model predicting ∼ Soft + Veg + 195.68 0.69 0.06 −92.59 occurrence of anthropogenic food items in Florida Season black bear (Ursus americanus floridanus) scats col- ∼ + + − Soft Veg 196.00 1.12 0.05 91.72 lected in the Highlands–Glades subpopulation of Invert + Season ∼ + + − south-central Florida, USA (2004–2006). Occurrence Soft Invert 196.12 1.26 0.05 93.94 of each natural foods variable was used as a refer- Veg ence; thus, non-occurrence of those variables was ∼ Soft + Hard 196.60 1.68 0.04 −95.20 estimated. Standard errors (SE), odds ratios (eβ) ∼ Soft + Invert + 196.65 1.71 0.04 −93.10 and their 95% confidence intervals, z-values, and Season β ∼ Soft + Veg + 196.69 1.74 0.04 −94.18 the probabilities that differed from 0 are also Vert presented. ∼ Soft 196.83 1.93 0.03 −96.37 Variable β SE eβ 95% CI Z Pr (>|Z|) aAkaike’s Information Criterion corrected for small sample size. Intercept −0.69 0.34 0.50 0.26–0.97 −2.05 0.04a b Difference between AICc of model and model with the lowest Soft mast 0.88 0.35 2.42 1.20–4.86 2.49 0.01a AICc. Vegetation −0.76 0.38 0.47 0.22–0.98 −2.01 0.08 cModel wt. dLog likelihood. aα < 0.05.

Ursus 28(1):92–104 (2017)

Downloaded From: https://www.bioone.org/journals/Ursus on 1/2/2019 Terms of Use: https://www.bioone.org/terms-of-use Access provided by Norwegian University Library of Life Sciences r BEAR FOOD HABITS IN AN ENDANGERED ECOSYSTEM Murphy et al. 99

Table 4. Dominant food items in the annual diets of all 7 Florida black bear (Ursus americanus floridanus) subpopulations (Florida, USA), with foods listed in decreasing order of frequency of occurrence (FO). Sample sizes for each study area (n) and total number of unique food items found (#) are presented, as are FO values in parentheses. Modified from Maehr et al. (2001).

Subpopulation n # Food 1 Food 2 Food 3 Eglina 259 30 (46%) Oak acorns (38%) Saw palmetto fruit (17%) Apalachicolab,c 54 9 Swamp tupelo fruit (47.6%) Bayberry fruit (21.2%) Oak acorns (7.6%) Osceolad 703 27 Corn (29%) Saw palmetto fruit (17%) Blackgum fruit (11%) Ocala–St. Johnse 676 36 Oak acorns (59.6%) Saw palmetto fiber (27.4%) Sabal palmetto fruit (20.3%) Chassahowitzkaf 212 20 Vegetation (76%) Sabal palmetto fruit (70%) Saw palmetto fruit (37%) Highlands–Gladesg 166 50 Oak acorns (33.3%) Corn (29.2%) Florida carpenter ant (29.2%) Big Cypressh 739 44 Sabal palmetto fruit (29.4%) Saw palmetto fruit (22.7%) Brazilian pepper (22.1%)

aStratman and Pelton (1999). bMaehr and Brady (1984). cBased on analysis of stomach contents, not scats. dDobey et al. (2005). eRoof (1997). fOrlando (2003). gThis study. hMaehr (1997).

year-round (Hightower et al. 2002, Garrison et al. 2007), intended to supplement white-tailed deer diets, which ex- and radiomonitoring data confirmed some HGS bears plains the presence of both of those anthropogenic foods were winter-active during our study (Ulrey 2008, Guthrie in scats across all seasons. Hightower et al. (2002) posited 2012). Although oak acorns dominated winter–spring that hunter-dispensed corn allowed some Louisiana black HGS scats, which exemplifies acorn overwintering, mast bears (U. a. luteolus) to remain winter-active while main- surveys conducted in the study area indicated acorn taining a positive energy balance, and corn positively yields during our years of sampling were 54.8% higher influenced vital rates of the Osceola black bear subpop- than the 27-year average (1988–2014; Bowman 2015). ulation in northern Florida (Dobey et al. 2005). Thus, Scrub communities contain 5 of the 7 oak species of the hunter-dispensed corn at feeding stations has the poten- Lake Wales Ridge ecosystem and adjacent areas within tial to sustain at least some winter-active bears, which HGS primary range (Abrahamson and Layne 2003), but could increase skeletal–lean body mass and improve in- substantial scrub has already been lost (Weekley et al. dividual fitness (Robbins et al. 2004, Steyaert et al. 2014). 2008). The magnitude of acorn overwintering that oc- Bears can quickly habituate to corn and other anthro- curred is likely uncommon, and we therefore believe our pogenic foods, leading to increased human–bear conflicts annual acorn FO is an overestimate. Our short 2-year that could endanger humans and cause financial losses sampling period precluded an identification of the foods (Baruch-Mordo et al. 2014, Ditmer et al. 2016, Pente- that winter-active Florida black bears may rely on during riani et al. 2016). Florida black bears visiting feeding years of average or poor oak productivity, but winter– stations and depredating or damaging agricultural crops spring FOs suggest vegetation, anthropogenic foods, and collectively account for a prominent human–bear conflict invertebrates may be important. complaint in the area occupied by the HGS (FWC 2012). Most green vegetation (e.g., leaves, grasses, forbs, Citrus fruit is the most widespread agricultural crop in etc.) generally is considered of tertiary diet importance the Lake Wales Ridge ecosystem, groves of which were to black bears in the southeastern United States because important predictors of female bear road-crossings in the those items provide little nutritional value compared with HGS (Guthrie 2012). Despite the close proximity of nu- other natural foods (Maehr et al. 2001, Inman and Pel- merous groves to established bear subpopulations, our ton 2002). In contrast, anthropogenic foods are calorie- study is the first to document citrus fruit in the diet of rich, typically require little energy expenditure to access, Florida black bears (Maehr et al. 2001). Although not and are often exploited by bears when natural foods are seasonally or annually important, the limited macronutri- scarce (Baruch-Mordo et al. 2014, Ditmer et al. 2016). ents provided by citrus fruit (<1gfat;<1 g protein; ≤12 Many hunters in the HGS area dispensed deer chow (a g carbohydrates) are insufficient for supporting mainte- nutrient-rich feed) and corn year-round at feeding stations nance or growth of lean body mass in bears. Therefore,

Ursus 28(1):92–104 (2017)

Downloaded From: https://www.bioone.org/journals/Ursus on 1/2/2019 Terms of Use: https://www.bioone.org/terms-of-use Access provided by Norwegian University Library of Life Sciences r 100 BEAR FOOD HABITS IN AN ENDANGERED ECOSYSTEM Murphy et al.

the presence of citrus in summer scats anecdotally sug- rich corn and deer chow are easily accessible and hunter- gests higher quality natural foods were either limited in dispensed corn provided approximately 8,000% of the availability or inaccessible during summer (Costello et al. amount of hard and soft mast available to bears in Vir- 2016). ginia, USA, during years of mast failures (Gray et al. Blueberry and blackberry fruits that other Florida black 2004). Anthropogenic foods were, however, not among bear subpopulations relied on during summer (Maehr the most dominant summer-food categories in HGS scats. and DeFazio 1985, Maehr et al. 2001) had nominal FO, Invertebrates were the only category with consistently whereas lower nutrition grapes (Vitis spp.) had the high- moderate to high FO across all seasons, a finding that was est FO among soft mast in summer scats. Interspecific largely influenced by myrmecophagy of Florida carpen- competition for quality summer soft mast is extraordinary ter ants during summer. Ants are a protein- and lipid-rich among vertebrates and could rapidly erode fruit availabil- food that can support lean mass accumulation in bears ity despite high productivity (Inman and Pelton 2002). if consumed in large, concentrated amounts, such as at Heightened occurrence of vertebrate remains in summer formicaries (Lopez-Alfaro´ et al. 2013). Studies of black scats further supports limited soft mast availability during and brown bear (U. arctos) populations elsewhere noted our sampling. Those remains included white-tailed deer, the importance of ants and other invertebrates in bear indicating predation of neonates rather than kleptopara- diets, which can offset the metabolic consequences of sitism of hunter kills or scavenging (Zager and Beecham mast shortages and reduce consumption of anthropogenic 2006); and predation of neonate ungulates by black bears foods (Auger et al. 2006, Coogan et al. 2014, Stenset et al. has been posited as a response to shortages of natural 2016). Although no other Florida black bear subpopu- vegetal foods (Belant et al. 2006, Svoboda et al. 2011). lation exhibited appreciable myrmecophagy, our results Other studies noted the impacts that soft mast short- indicate that bears in the HGS primarily compensated ages can have on black bear populations, such as de- for summer soft mast shortages by consuming Florida pressed vital rates, and the relationship of shortages with carpenter ants and other invertebrates, which are likely human–bear conflicts (Inman and Pelton 2002, Obbard reliable food sources with minimal abundance fluctua- and Howe 2008, Romain et al. 2013). We found height- tions. Florida carpenter ants and , including the ened dietary reliance on anthropogenic food sources as giant palm weevil ( cruentatus), are ar- soft mast occurrence declined, a response that would boreal species that are acutely dependent on remaining increase human–bear conflicts assuming anthropogenic bear habitat for proliferation (Tedder et al. 2012). Thus, foods are a reliable proxy (Howe et al. 2010, Hopkins impending habitat loss could also reduce populations of et al. 2014). Our FO estimates indicate corn is the most those important invertebrates and increase consumption likely substitute for soft mast among anthropogenic food of anthropogenic foods by HGS bears. items, but human–bear conflicts at feeding stations lo- Much of the uncertainty in our results is a function of cated in rural areas may be less severe relative to those sampling duration, which was too short to account for associated with other human-sourced foods (Merkle et al. super-annual mast fluctuations (Craighead et al. 1995), 2013, Ditmer et al. 2016). Despite not collecting scats in and small seasonal sample sizes of scats. Pooling within- the urbanized north-central part of HGS primary range season scats across years to overcome the latter issue where human population density is highest, we docu- resulted in sample sizes that were skewed toward autumn mented a few occurrences of garbage in scats. Bears when bear food consumption and defecation rates are frequenting urban areas to access garbage can result in typically highest (Dahle et al. 1998, Ciucci et al. 2014), conditioning to human presence, loss of fear of humans, thereby attributing more weight to food items in autumn and injurious aggression toward humans (Elfstrom¨ et al. scats (i.e., pooling fallacy; Machlis et al. 1985, Leger 2014, Penteriani et al. 2016). Citrus fruit is Florida’s top and Didrichsons 1994). Therefore, FO probably overes- cash crop (∼US$1.10 billion income annually; Gitzen timated the importance of smaller sized oak acorns while 2015); therefore, contentious and costly conflicts with underestimating the importance of larger sized foods, humans could also develop if HGS bears increase citrus such as vertebrates (Klare et al. 2011). Locations of our fruit consumption or cause damage to citrus trees. sample collections were also clustered in 3 areas of high Whether or not the anthropogenic foods available to Florida black bear use (Ulrey 2008, Guthrie 2012), which HGS bears can support a sufficient portion of the popula- likely exacerbated said issue by violating the assump- tion necessary to ensure long-term persistence in the face tion of sample independence (Klare et al. 2011). Further- of imminent habitat loss is unclear. Although nutrient- more, because some natural foods are endemic to remnant poor citrus fruit is incapable of doing so, carbohydrate- habitat communities (e.g., scrub palmetto fruit) that are

Ursus 28(1):92–104 (2017)

Downloaded From: https://www.bioone.org/journals/Ursus on 1/2/2019 Terms of Use: https://www.bioone.org/terms-of-use Access provided by Norwegian University Library of Life Sciences r BEAR FOOD HABITS IN AN ENDANGERED ECOSYSTEM Murphy et al. 101

difficult to access without substantial risk to bears (e.g., to present; therefore, we strongly suggest repeating our crossing roads) as a result of habitat loss and fragmen- study over a longer duration to investigate whether di- tation (Guthrie 2012), our clustered sampling may not etary shifts occur in response to impending habitat loss have detected all of the foods consumed by bears. Al- and to further inform habitat and subpopulation conser- though increasing our winter–spring and summer scat vation and human–bear conflict management. sample sizes would have been difficult because of the small number of bears in the HGS, future dietary studies should conduct sampling over a longer temporal period Acknowledgments (e.g., >3 consecutive yr) to account for mast fluctuations Without support from multiple private landowners, (Romain et al. 2013). Estimating volume-based metrics particularly the Smoak, Hendrie, and Lightsey families, in our study would have been unlikely to facilitate more this study would not have been possible. We are grateful insightful comparisons among Florida bear subpopula- for the support from J. N. Layne, C. R. Abrahamson, H. tions, but should be used in conjunction with FO for M. Swain, and the late L. D. Harris and late L. V. Layne. future studies to better approximate dietary importance We thank the numerous staff at Archbold Biological Sta- of specific food items (Klare et al. 2011, Ciucci et al. tion who provided assistance—including E. S. Menges 2014, Stenset et al. 2016). and M. A. Deyrup—and the Florida Fish and Wildlife Conservation Commission for supporting our efforts. We especially thank R. Bowman for providing access to his Management implications long-term data set of oak productivity in the Lake Wales Federally endangered scrub communities contain nu- Ridge ecosystem. We appreciate the useful comments merous high-quality hard and soft mast–producing flora and constructive criticism provided by reviews from P. (e.g., saw palmetto, blueberry, and 5 species of oak; Ciucci, J. L. Belant, B. K. Scheick, and 2 anonymous Abrahamson et al. 1984, Layne and Abrahamson 2010); reviewers, all of which improved this manuscript. We therefore, we suggest managers consider intensifying ef- also thank T. E. Boal for copyediting this manuscript. forts to protect remnant patches of scrub via state and This study was funded by the Disney Worldwide Con- federal agency acquisition or conservation easements to servation Fund; supplementary funds were provided by thwart further loss. Soft mast generally appears to be Archbold Biological Station. We dedicate this paper to D. the most important food category to Florida black bears S. Maehr and M. G. Smoak, both of whom tragically per- in the HGS, and the lack thereof is associated with in- ished while conducting aerial radiotelemetry monitoring creased consumption of anthropogenic foods; therefore, of black bears in the HGS. Any opinions, findings, con- habitat conservation should also include other commu- clusions, or recommendations expressed herein are those nities with quality soft-masting shrubs and plants, such of the authors and do not necessarily reflect the views as flatwoods and sandhills (Abrahamson and Abraham- of the University of Kentucky. Use of trade, product, or son 1989, Layne and Abrahamson 2010). An indirect firm names is for descriptive purposes only and does not byproduct of those habitat conservation efforts is main- imply endorsement by the University of Kentucky. tenance of arboreal Florida carpenter ant populations for bears to consume during years of poor mast yields or Literature cited in the event that masting flora abundance and distribu- ABRAHAMSON, W.G., AND C.R. ABRAHAMSON. 1989. Nutri- tion is reduced by anthropogenic development. Although tional quality of dispersed fruits in Florida sandridge hunter-dispensed corn at wildlife feeders provides a nu- habitats. Bulletin of the Torrey Botanical Club 116:215–228. tritious, easily accessible supplemental food for bears, ———, A.F. JOHNSON,J.N.LAYNE, AND P.A. P ERONI. 1984. such feeding could result in habituation and lead to bears Vegetation of the Archbold Biological Station, Florida: An increasing exploitation of other anthropogenic foods that example of the southern Lake Wales Ridge. Florida Scientist are in closer proximity to humans (e.g., garbage) or are 47:209–250. agricultural commodities (e.g., citrus fruit). Small size of ———, AND J.N. LAYNE. 2003. Long-term patterns of acorn production for five oak species in xeric Florida uplands. the HGS should limit human–bear conflict management Ecology 84:2476–2492. to primarily non-lethal methods, so we suggest managers AGEE, J.D., AND C.A. MILLER. 2009. Factors contributing to- increase public education and outreach (Pienaar et al. ward acceptance of lethal control of black bears in Central 2015) and expand agency-assistive anthropogenic-food– Georgia, USA. Human Dimensions of Wildlife 14:198–205. management programs to the HGS area (Barrettetal. ARCHBOLD BIOLOGICAL STATION. 2010. Station fact 2014). Finally, our samples were collected a decade prior sheet: Archbold Biological Station. http://www.archbold-

Ursus 28(1):92–104 (2017)

Downloaded From: https://www.bioone.org/journals/Ursus on 1/2/2019 Terms of Use: https://www.bioone.org/terms-of-use Access provided by Norwegian University Library of Life Sciences r 102 BEAR FOOD HABITS IN AN ENDANGERED ECOSYSTEM Murphy et al.

station.org/station/html/aboutus/factsht.html. Accessed 10 COOGAN, S.C., D. RAUBENHEIMER,G.B.STENHOUSE, AND S.E. Feb 2011. NIELSEN. 2014. Macronutrient optimization and seasonal AUGER,J.,G.L.OGBORN,C.L.PRITCHETT, AND H.L. BLACK. diet mixing in a large omnivore, the grizzly bear: A geomet- 2006. Selection of ants by the American black bear (Ursus ric analysis. PLoS ONE 9:e97968. americanus). Western North American Naturalist 64:166– COSTELLO, C.M., S.L. CAIN,S.PILS,L.FRATTAROLI,M.A. 174. HAROLDSON, AND F.T. VA N MANEN. 2016. Diet and BARRETT, M.A., D.J. TELESCO,S.E.BARRETT,K.M.WIDNESS, macronutrient optimization in wild ursids: A comparison AND E.H. LEONE. 2014. Testing bear-resistant trash cans in of grizzly bears with sympatric and allopatric black bears. residential areas of Florida. Southeastern Naturalist 13:26– PLoS ONE 11:e0153702. 39. CRAIGHEAD, J.J., J.S. SUMNER, AND J.A. MITCHELL. 1995. The BARUCH-MORDO,S.,K.R.WILSON,D.L.LEWIS,J.BRODERICK, grizzly bears of Yellowstone. Island Press, Washington, DC, J.S. MAO, AND S.W. BRECK. 2014. Stochasticity in natural USA. forage production affects use of urban areas by black bears: DAHLE,B.,O.J.SRENSEN,E.H.WEDUL,J.E.SWENSON, AND Implications to management of human–bear conflicts. PLoS F. SANDEGREN. 1998. The diet of brown bears Ursus arc- ONE 9:e85122. tos in central Scandinavia: Effect of access to free-ranging BECKMANN, J.P., AND J. BERGER. 2003. Rapid ecological and domestic sheep Ovis aries. Wildlife Biology 4:147–158. behavioral changes in carnivores: The responses of black DITMER, M.A., D.L. GARSHELIS,K.V.NOYCE,A.W.HAVELES, bears (Ursus americanus) to altered food. Journal of Zool- AND J.R. FIEBERG. 2016. Are American black bears in an ogy 261:207–212. agricultural landscape being sustained by crops? Journal of BELANT, J.L., K. KIELLAND,E.H.FOLLMANN, AND L.G. Mammalogy 97:54–67. ADAMS. 2006. Interspecific resource partitioning in sym- DIXON, J.D., M.K. OLI,M.C.WOOTEN,T.H.EASON,J.W. patric ursids. Ecological Applications 16:2333–2343. MCCOWN, AND M.W. CUNNINGHAM. 2007. Genetic con- BENSON, J.F., AND M.J. CHAMBERLAIN. 2006. Food habits of sequences of habitat fragmentation and loss: The case of the Louisiana black bears (Ursus americanus luteolus)intwo Florida black bear (Ursus americanus floridanus). Conser- subpopulations of the Tensas River Basin. American Mid- vation Genetics 8:455–464. land Naturalist 156:118–127. DOBEY, S., D.V. MASTERS,B.K.SCHEICK,J.D.CLARK,M.R. BOJARSKA,K.,AND N. SELVA. 2012. Spatial patterns in brown PELTON, AND M.E. SUNQUIST. 2005. Ecology of Florida bear Ursus arctos diet: The role of geographical and envi- black bears in the Okefenokee–Osceola ecosystem. Wildlife ronmental factors. Mammal Review 42:120–143. Monographs 158. BORROR, D.J., AND R.E. WHITE. 1970. A field guide to : ELFSTROM¨ ,M.,A.ZEDROSSER,O.-G.STØEN, AND J.E. SWEN- America north of Mexico. Houghton Mifflin, New York, SON. 2014. Ultimate and proximate mechanisms underlying New York, USA. the occurrence of bears close to human settlements: Re- BOWMAN, R. 2015. Annual variation in scrub oak species view and management implications. Mammal Review 44: acorn production. Dataset. Archbold Biological Station, 5–18. Venus, Florida, USA. http://www.archbold-station.org/data/ FLORIDA FISH AND WILDLIFE CONSERVATION COMMISSION. keywordsselect.cfm?dept=Avian + Ecology&sendlab= 2012. Florida black bear management plan. Florida Fish and view + projects + by + Research + Program. Accessed 3 Wildlife Conservation Commission, Tallahassee, Florida, Jan 2017. USA. BURNHAM, K.P., D.R. ANDERSON, AND K.P. HUYVAERT. 2011. ———. 2014. Florida cooperative land cover map. Dataset. AIC model selection and multimodel inference in behavioral Florida Fish and Wildlife Conservation Commission and ecology: Some background, observations, and comparisons. Florida Natural Areas Inventory, Tallahassee, Florida, USA. Behavioral Ecology and Sociobiology 65:23–35. ———. 2015. 2015 Florida black bear hunt summary report. CARR, M.H., AND P.D. Z WICK. 2016. Florida 2070: Map- Technical report. Florida Fish and Wildlife Conservation ping Florida’s future–alternative patterns of development Commission, Tallahassee, Florida, USA. in 2070. Technical Report to Florida Department of Agri- FORTIN, J.K., C.C. SCHWARTZ,K.A.GUNTHER,J.E.TEISBERG, culture and Consumer Services & 1000 Friends of Florida. M.A. HAROLDSON,M.A.EVANS, AND C.T. ROBBINS. 2013. Geoplan Center, University of Florida, Gainesville, Florida, Dietary adjustability of grizzly bears and American black USA. bears in Yellowstone National Park. Journal of Wildlife CHERRY, M.J., K.L. TURNER,M.B.HOWZE,B.S.COHEN,L.M. Management 77:270–281. CONNER, AND R.J. WARREN. 2016. Coyote diets in a longleaf GARRISON, E.P., J.W. MCCOWN, AND M.K. OLI. 2007. Re- pine ecosystem. Wildlife Biology 22:64–70. productive ecology and cub survival of Florida black bears. CIUCCI,P.,E.TOSONI,G.DI DOMENICO,F.QUATTROCIOCCHI, Journal of Wildlife Management 71:720–727. AND B. LUIGI. 2014. Seasonal and annual variation in the GITZEN, R. 2015. Fresh from Florida: 2015 international report. food habits of Apennine brown bears, Central Italy. Journal Florida Department of Agriculture and Consumer Services, of Mammalogy 95:572–586. Tallahassee, Florida, USA.

Ursus 28(1):92–104 (2017)

Downloaded From: https://www.bioone.org/journals/Ursus on 1/2/2019 Terms of Use: https://www.bioone.org/terms-of-use Access provided by Norwegian University Library of Life Sciences r BEAR FOOD HABITS IN AN ENDANGERED ECOSYSTEM Murphy et al. 103

GRAY, R.M., M.R. VAUGHAN, AND S.L. MCMULLIN. 2004. insight into differences among populations. PLoS ONE Feeding wild American black bears in Virginia: A survey of 10:e0128088. Virginia bear hunters, 1998–99. Ursus 15:188–196. MACHLIS,L.,P.W.D.DODD, AND J.C. FENTRESS. 1985. The GREENLEAF, S.S., S.M. MATTHEWS,R.G.WRIGHT,J.J. pooling fallacy: Problems arising when individuals con- BEECHAM, AND H.M. LEITHEAD. 2009. Food habits of tribute more than one observation to the data set. Ethology American black bears as a metric for direct management 68:201–214. of human–bear conflict in Yosemite Valley, Yosemite Na- MAEHR, D.S. 1997. The comparative ecology of bobcat, black tional Park, California. Ursus 20:94–101. bear, and Florida panther in south Florida. Florida Museum GUTHRIE, J.M. 2012. Modeling movement behavior and road of Natural History 40:1–176. crossing in the black bear of south central Florida. Thesis, ———, AND J.R. BRADY. 1984. Comparison of food habits in University of Kentucky, Lexington, Kentucky, USA. two north Florida black bear populations. Florida Scientist HARRELL, F.E. 2017. Rms: Regression modeling strategies. R 47:171–175. package version 5.1-0. https://cran.r-project.org/package= ———, AND J.T. DEFAZIO. 1985. Foods of black bears in rms. Accessed 4 Jan 2017. Florida. Florida Field Naturalist 13:8–12. HIGHTOWER, D.A., R.O. WAGNER, AND R.M. PACE III. 2002. ———, T.S. HOCTOR,L.J.QUINN, AND J.S. SMITH. 2001. Black Denning ecology of female American black bears in south bear habitat management guidelines for Florida. Florida central Louisiana. Ursus 13:11–17. Fish and Wildlife Conservation Commission, Tallahassee, HOPKINS, J.B., P.L. KOCH,J.M.FERGUSON, AND S.T. KALI- Florida, USA. NOWSKI. 2014. The changing anthropogenic diets of Amer- ———, J.N. LAYNE,T.S.HOCTOR, AND M.A. ORLANDO. 2004. ican black bears over the past century in Yosemite National Status of the black bear in south-central Florida. Florida Park. Frontiers in Ecology and the Environment 12:107– Field Naturalist 32:85–101. 114. MARSHALL, S.A. 2006. Insects—Their natural history and di- HOSMER, D.W., T. HOSMER,S.LE CESSIE, AND S. LEMESHOW. versity with a photographic guide to insects of eastern North 1997. A comparison of goodness-of-fit tests for the logistic America. Firefly Books, Buffalo, New York, USA. regression model. Statistics in Medicine 16:965–980. MARTIN, A.C., AND W.D. BARKLEY. 1961. Seed identification HOWE, E.J., M.E. OBBARD,R.BLACK, AND L.L. WALL. 2010. manual. University of California Press, Berkley, California, Do public complaints reflect trends in human–bear conflicts? USA. Ursus 21:131–142. MERKLE, J.A., H.S. ROBINSON,P.R.KRAUSMAN, AND P. AL- HUMM,J.,J.W.MCCOWN,B.K.SCHEICK, AND J.D. CLARK. ABACK. 2013. Food availability and foraging near human de- 2017. Spatially explicit population estimates for black bears velopments by black bears. Journal of Mammalogy 94:378– based on cluster sampling. Journal of Wildlife Management. 385. doi:10.1002/jwmg.21294. MURPHY, S.M., J.J. COX,B.C.AUGUSTINE,J.T.HAST,J.M. INMAN, R.M., AND M.R. PELTON. 2002. Energetic production GUTHRIE,J.WRIGHT,J.MCDERMOTT,S.C.MAEHR, AND by soft and hard mast foods of American black bears in the J.H. PLAXICO. 2016. Characterizing recolonization by a Smoky Mountains. Ursus 13:57–68. reintroduced bear population using genetic spatial capture– KLARE,U.,J.KAMLER, AND D. MACDONALD. 2011. A com- recapture. Journal of Wildlife Management 80:1390–1407. parison and critique of different scat-analysis methods for NOSS, R.F., W.J. PLATT,B.A.SORRIE,A.S.WEAKLEY,D.B. determining carnivore diet. Mammal Review 41:294–312. MEANS,J.COSTANZA, AND R.K. PEET. 2015. How global LARSON, R.N., D.J. MORIN,I.A.WIERZBOWSKA, AND K.R. biodiversity hotspots may go unrecognized: Lessons from CROOKS. 2015. Food habits of coyotes, gray foxes, and the North American Coastal Plain. Diversity and Distribu- bobcats in a coastal southern California urban landscape. tions 21:236–244. Western North American Naturalist 75:339–347. OBBARD, M.E., AND E.J. HOWE. 2008. Demography of black LAYNE, J.N., AND W.G. ABRAHAMSON. 2010. Spatiotemporal bears in hunted and unhunted areas of the boreal forest of variation of fruit digestible-nutrient production in Florida’s Ontario. Journal of Wildlife Management 72:869–880. uplands. Acta Oecologia 36:675–683. ORLANDO, M.A. 2003. The ecology and behavior of an iso- LEGER, D.W., AND I.A. DIDRICHSONS. 1994. An assessment lated black bear population in west central Florida. Thesis, of data pooling and some alternatives. Animal Behaviour University of Kentucky, Lexington, Kentucky, USA. 48:823–832. ORO,D.,M.GENOVART,G.TAVECCHIA,M.S.FOWLER, AND A. LESMERISES,R.,L.REBOUILLAT,C.DUSSAULT, AND M.-H. ST- MART´INEZ-ABRA´IN. 2013. Ecological and evolutionary im- LAURENT. 2015. Linking GPS telemetry surveys and scat plications of food subsidies from humans. Ecology Letters analyses helps explain variability in black bear foraging 16:1501–1514. strategies. PLoS ONE 10:e0129857. PEARLSTINE, L.G., L.A. BRANDT,F.J.MAZZOTTI, AND W.M. LOPEZ´ -ALFARO, C., S.C.P. COOGAN,C.T.ROBBINS,J.K. KITCHENS. 1997. Fragmentation of pine flatwood and marsh FORTIN, AND S.E. NIELSEN. 2013. Assessing nutritional communities converted for ranching and citrus. Landscape parameters of brown bear diets among ecosystems gives and Urban Planning 38:159–169.

Ursus 28(1):92–104 (2017)

Downloaded From: https://www.bioone.org/journals/Ursus on 1/2/2019 Terms of Use: https://www.bioone.org/terms-of-use Access provided by Norwegian University Library of Life Sciences r 104 BEAR FOOD HABITS IN AN ENDANGERED ECOSYSTEM Murphy et al.

PELTON, M.R. 2001. American black bear. Pages 224–233 in J. STENSET, N.E., P.N. LUTNÆS,V.BJARNADOTTIR´ ,B.DAHLE, G. Dickson, editor. Wildlife of southern forests: Habitat and K.H. FOSSUM,P.JIGSVED,T.JOHANSEN,W.NEUMANN,O. management. Hancock House, Blaine, Washington, USA. OPSETH,O.RØNNING, S.M.J.G. STEYAERT,A.ZEDROSSER, PENTERIANI,V.,M.DEL MAR DELGADO,F.PINCHERA,J. S. BRUNBERG, AND J.E. SWENSON. 2016. Seasonal and an- NAV E S ,A.FERNANDEZ´ -GIL,I.KOJOLA,S.HARK¨ ONEN¨ ,H. nual variation in the diet of brown bears Ursus arctos in NORBERG,J.FRANK,J.M.FEDRIANI,V.SAHLEN´ ,O.-G. the boreal forest of southcentral Sweden. Wildlife Biology STØEN,J.E.SWENSON,P.WABAKKEN,M.PELLEGRINI,S. 22:107–116. HERRERO, AND J.V. LOPEZ´ -BAO. 2016. Human behavior STEYAERT, S.M.J.G., J. KINDBERG,K.JERINA,M.KROFEL,M. can trigger large carnivore attacks in developed countries. STERGAR,J.E.SWENSON, AND A. ZEDROSSER. 2014. Behav- Scientific Reports 6:20552. ioral correlates of supplementary feeding of wildlife: Can PETERSON, L.A. 1999. A field guide to edible wild plants: general conclusions be drawn? Basic and Applied Ecology Eastern and central North America. Houghton Mifflin, New 15:669–676. York, New York, USA. STRATMAN, M.R., AND M.R. PELTON. 1999. Feeding ecology PIENAAR, E.F., D. TELESCO, AND S. BARRETT. 2015. Under- of black bears in northwest Florida. Florida Field Naturalist standing people’s willingness to implement measures to 27:95–102. manage human–bear conflict in Florida. Journal of Wildlife SVOBODA, N.J., J.L. BELANT,D.E.BEYER,J.F.DUQUETTE, Management 79:798–806. H.K. STRICKER, AND C.A. ALBRIGHT. 2011. American black RCORE TEAM. 2016. R: A language and environment for statis- bear predation of an adult white-tailed deer. Ursus 22:91–94. tical computing. R Foundation for Statistical Computing, Vi- TEDDER, D.S., J.J. COX,P.H.CROWLEY, AND D.S. MAEHR. enna, Austria. http://www.cran.r-project.org/. Accessed 15 2012. Black bears, palms, and giant palm weevils: An in- Sep 2016. traguild mutualism. Open Ecology Journal 5:18–24. ROBBINS, C.T., C.C. SCHWARTZ, AND L.A. FELICETTI. 2004. ULREY, W.A. 2008. Home range, habitat use, and food habits Nutritional ecology of ursids: A review of newer methods of the black bear in south-central Florida. Thesis, University and management implications. Ursus 15:161–171. of Kentucky, Lexington, Kentucky, USA. ROGERS, L. 1976. Effects of mast and berry crop failures on UNITED STATES FISH AND WILDLIFE SERVICE. 1999. South survival, growth, and reproductive success of black bears. Florida multi-species recovery plan. United States Fish and Transactions of the North American Wildlife and Natural Wildlife Service, Atlanta, Georgia, USA. Resources Conference 41:431–438. WEEKLEY, C.W., E.S. MENGES, AND R.L. PICKERT. 2008. An ROMAIN, D.A., M.E. OBBARD, AND J.L. ATKINSON. 2013. Tem- ecological map of Florida’s Lake Wales Ridge: A new poral variation in food habits of the American black bear boundary delineation and an assessment of post-Columbian (Ursus americanus) in the boreal forest of northern Ontario. habitat loss. Florida Scientist 71:45–64. Canadian Field-Naturalist 127:118–130. WHITTLE, A.J. 2009. Florida panther and black bear: A road ROOF, J.C. 1997. Black bear food habits in the lower Wekiva and urban avoidance/utilization analysis and impacts of land River Basin of central Florida. Florida Field Naturalist use and climate change on large carnivore habitat in Florida. 25:92–97. Thesis, University of Kentucky, Lexington, Kentucky, RYAN, C.W., J.C. PACK,W.K.IGO, AND A. BILLINGS. 2007. USA. Influence of mast production on black bear non-hunting ZAGER,P.,AND J. BEECHAM. 2006. The role of American black mortalities in West Virginia. Ursus 18:46–53. bears and brown bears as predators on ungulates in North SIMEK, S.L., S.A. JONKER,B.K.SCHEICK,M.J.ENDRIES, AND America. Ursus 17:95–108. T.H. EASON. 2005. Statewide assessment of road impacts on bears in six study areas in Florida from May 2001– Received: October 22, 2016 September 2003. Florida Fish and Wildlife Conservation Accepted: March 29, 2017 Commission, Tallahassee, Florida, USA. Associate Editor: Ciucci

Ursus 28(1):92–104 (2017)

Downloaded From: https://www.bioone.org/journals/Ursus on 1/2/2019 Terms of Use: https://www.bioone.org/terms-of-use Access provided by Norwegian University Library of Life Sciences