sign in sign up
<<
Home , Cornus

Diet and feeding habits of Asiatic black bears in the NorthernJapanese Alps

Oscar C. Huygens1'7, Toshiro Miyashita2, Bj0rn Dahle3, Meghan Carr4, Shigeyuki Izumiyama5, Takashi Sugawara6, and Hidetake Hayashi3

1ShinshuUniversity, Graduate School of Science and Technology,Matsumoto, 390-8621, Japan 2ShinshuUniversity, Department of EnvironmentalSciences, Facultyof Science, Matsumoto,390-8621, Japan 3NorwegianUniversity of Science and Technology,Department of Zoology, N-7491 Trondheim,Norway 4Universityof Tennessee, Departmentof Forestry,Wildlife and Fisheries, P.O. Box 1071, Knoxville, TN 37901-1071, USA 5WildlifeManagement Office, Inc., 5-8 Fuda-ku,Kawasaki, 214-0011, Japan 6Shinshu University,Department of Biology,Faculty of Science, Matsumoto,390-8621, Japan

Abstract: Habitatloss and bear-humanconflicts are reportedthroughout Asiatic black bear (Ursus thibetanus)range. Sound understandingof Asiatic black bear food habits and ecology is needed to develop effective conservationpolicies to protect or restore Asiatic black bear habitatand to reduce bear-humanconflicts. We documentedfood habits and aspects of the feeding ecology of Asiatic black bears with access to an alpine zone in the NorthernJapanese Alps from 408 scats, 220 bear feeding platforms in , 469 radiolocations obtained from 21 bears, and field observations between September 1995 and November 1999. Bears ate oak (Quercus spp.) acorns from the previous fall and dwarf bamboo (Sasa spp.) and shoots in spring;succulent and soft mast, especially Japanesecluster cherry (Prunus grayana), in summer;and hard mast, especially oak acorns, in fall. Bears ate insects in all seasons, with a peak in summer,and also ate Japaneseserow (Naemorhedus crispus) on at least 6 occasions. In summer,bears that moved to alpine elevations relied on succulent plants;bears that remained at lower elevations relied on soft mast. In fall, all bearsmoved to hard-mast producing areas in broad-leavedforests at lower elevations in the montane zone. Montane broad- leaved forests seem importantfor Asiatic black bear survivalin the NorthernJapanese Alps. It may be importantto give these forests more protectionthan they currentlyreceive.

Key words: alpine zone, Asiatic black bear, bear diet, bear movements, broadleaf forest, elevational shifts, food habits, Japan,Northern Japanese Alps, Ursus thibetanus

Ursus 14(2):236-245 (2003)

The Asiatic black bear is listed by the International Increasingrates of Asiatic black bear-humanconflict Union for the Conservation of Nature (IUCN) as have been attributedto shrinkage and degradationof vulnerable over most of its range and as critically en- Asiatic black bear habitat (Azuma and Torii 1980; Hazumi dangered in Iran and Pakistan (Servheen et al. 1999). Furubayashiet al. 1980; Watanabe1981; 1994, Habitat loss is cited as a major concern for Asiatic 1999; Sathyakumar 1999). To help prevent further habitats black bears in ,Japan, India, Russia, and , conflicts, it is importantto protect and restore and conflicts between and Asiatic black bears critical for the survival of Asiatic black bears. Habitat meant to benefit Asiatic black are reported in China, India, Iran, Japan, Russia, and conservation policies conflicts will not suc- Taiwan. Reportedconflicts involve predationon crops, bears and to reduce bear- habitsand of these bears orchards,apiaries, fish farms,and livestock;debarking of ceed unless the feeding ecology the of environ- trees;and mauling,killing, and even consuminghumans. are well understoodthroughout variety ments in which they occur. The food habits of Asiatic black bears in Japan are well documented, except at high elevations (e.g. Takada 1979, Nozaki et al. 1983, and were 7email:[email protected] Torii 1989, Naganawa and Koyama 1994)

236 ASIATICBLACK BEAR DIET IN JAPAN * Huygens et al. 237

N Japana n 0

0 area

P 0

o*

350 km

Fig. 1. Study area in Japan's Nagano prefecture (in * 2645 gray) for a 1995-99 study of food habits of Asiatic black bears with access to an alpine zone.

0 10km summarized by Hashimoto and Takatsuki (1997). Asiatic black bear have been reported to use alpine habitats (Novikov 1956, Hazumi 1994), but the food Fig. 2. The 3 major drainages of the study area in habits of such bears have not been described.We report Nagano prefecture, Japan. The central drainage, the on the food habits of Asiatic black bears with access to Karasu Gawa, has the least rugged topography and had been more. The elevation of is an alpine zone in the NorthernJapanese Alps of central logged peaks included (meters). Japan.

study areaitself, precipitationthere was probablyhigher. Study area Rains were concentratedduring the monsoon-like rainy The approximately 259-km2 study area (36?22'N, season from 20 June to 20 July and during the typhoon 137?45'E)was located in Nagano Prefecture,in the cen- season from 20 August 20 to 20 September. Snow ter of Honshu, Japan's largest island (Fig. 1). The study startedto accumulateat the higher elevations as early as area encompassed3 majordrainages (Fig. 2), straddling October.Snow fields usually did not disappearuntil late the southeasternedge of the Chubu Sangaku National July but could last into Septemberin cooler or in Park in the NorthernJapanese Alps, and was bordered years with above average snow precipitation.Most trees by the MatsumotoPlain on the east. Elevations ranged leafed out in May and early June, and the first signs of from approximately600 m on the east to 2,922 m on the fall were evident in late August in alpine meadows and west. Gravel and paved roads, hotels and hot spring re- by mid-Septemberat lower elevations. sorts, vacation homes, mountainlodges, 2 golf courses, Three distinct vegetational zones occurred in the erosion-controldams, retaining walls, and mini hydro- study area:the montanezone from approximately800 m electric dams occurredin the study area. to about 1,500 m; the subalpine zone, between 1,500 Yearly precipitationand snow fall averaged967 mm and 2,500 m; and the alpine zone above 2,500 m. and 100 cm, respectively, in HotakaTown, at an eleva- The montane zone was characterizedby numerous tion of 540 m in the plains just east of the study area. of broad trees. Mongolian oak Although precipitation data were unavailable in the (Quercus crispula), Japanese white oak (Q. serrata),

Ursus 14(2):236-245 (2003) 238 ASIATICBLACK BEAR DIET IN JAPAN * Huygens et al.

Japanesechestnut (Castanea crenata), several species of pine, hinoki cypress, and Japanese cedar (Crypto- cherries (Prunus spp.), 2 species of dogwood (Cornus meria japonica). The central drainageof the study area spp.), walnut (Juglans mandshurica), beech (Fagus had the least rugged topography and consequently crenata), and Sorbus alnifolia were potential bear had experienced the most cutting. This encouragedthe foods. Mesic sites, rocky ridges, and steep slopes were propagation of early successional species, some of characterizedby other communities that included which were good bear-food producers in the summer conifers. Forests dominatedby Japanesered pine (Pinus (such as Prunus spp., Cornus spp., and Vitus spp.). In densiflora)also occurredat lower elevations, usually as contrast, the northern and southern drainages of the an early successional species maintained by periodic study area had steeper topography, had been logged cutting of competing deciduous trees. The layer less, and were characterizedby later successional spe- was diverse and included species of Viburnum,Hydran- cies, among which hard mast producing trees such as gea, Euonymus, Ilex, and Rhus. The shrub layer was Mongolian oak and Japanese chestnut were important further characterizedby widespread thickets of dwarf bear-foodproducers in the fall. bamboo (Sasa spp.) up to 2 m high and thick enough to Japanese macaque (Macaca fuscata), Japanese be serious obstacles to both natural and artificial serow, red fox (Vulpes vulpes), raccoon dog (Nycter- regeneration.These dwarf bamboo thickets, combined eutes procyonoides), Eurasianbadger (Meles meles), 2 with the steep terrainand the lack of trails, also denied species of weasels (Mustela sibirica and M. erminea), practical foot access for research to large parts of the and Japanesemartens (Martes melampus) were the most study area. common mammalsof the study area. The subalpine zone was dominatedby Abies veitchii and A. mariesii, althoughPicea jezoensis var. hondoen- sis was also common. Ridges and steep dry slopes were Methods often covered diversifolia in conjunctionwith by Tsuga Scat analysis Pinus parviflora,P. parvifora var. pentaphylla, Korean From September 1995 to November 1999 we stone pine (P. koraiensis), Thuja standishii, and some- collected scats opportunisticallyas we captured and times hinoki cypress (Chamaecyparis obtusa). Birch radiolocatedbears. From early May to early Novem- (Betula ermanii) was conspicuous in the subalpinezone ber 1999, we also searched for scats in specific areas, and regularlyformed pure stands at the boundarywith mostly of the montane and alpine zones, by walking the alpine zone. Broad-leaved deciduous forests com- along animal trails, ridges, and any other land or veg- prised primarilyof Salix spp. and Alnus spp. occurred etation features we thought might attract bears (2-3 along riversides. The shrub layer was often poorly searches week). We did not collect scats that con- developed, although Viburnumfurcatum and several per tained obvious anthropogenicfoods (near orchards or species of Sorbus were common. mountain sites). In 1999 we The alpine zone was characterizedby alpine scrub, lodge garbage disposal collected scats from bears in barreltraps. snowfield grasslands, and alpine deserts. Impenetrable caught We recordedthe and other thickets of'Japanese stone pine (P. pumila), a creeping drainage,vegetation type, relevant informationfor each scat. We used 1:25,000 pine, was often the climax vegetation in this zone. scale to estimate elevation, and Alpine heaths, grasslands,and deserts occurredeither as topographic maps a U-test to test for differences in soft edaphicor topographicclimaxes inducedby exposureto Mann-Whitney diets of bears at low and elevations wind, late thawing snow, the presenceof serpentinesoil, mast of feeding high in summer. We included old scats (>2 months) in the or as pioneer communities on landslips and scree. In when we were able to their con- concave depressions,Japanese stone pines occurredwith analysis only identify broad-leaveddeciduous such as Alnus maximo- tents, usually seeds or pits that also allowed us to assign to the scat. In fall 1999 we left wiczii, Sorbus spp., Acer tschonoskii, and others. Bog an approximatedate estimated100 scats and meadow vegetation typically developed on slopes approximatelyhalf of an (containing about below late-melting snowfields. Numata (1974) and mostly acor but also walnut) deposited during area. we collected Wada (1985) provide further detail on the vegetation 1.5 months in a 0.25-km2 Otherwise, all scats found in of Nagano Prefecture. all the scats we found, including estimated the Selective cutting of trees had taken place on many of clusters and near day beds. We visually it to the month the steeper slopes within the study area, whereas the freshness of each scat and assigned (i.e., we believed it had been gentler slopes had usually been clearcut and converted days 1-10, 11-20, 21-31) a scat found on 22 to plantationsof larch (Larix kaempferi),Japanese red deposited.For example, we assigned

Ursus 14(2):236-245 (2003) ASIATICBLACK BEAR DIET IN JAPAN * Huygens et al. 239

July and estimatedto be 3-4 weeks old to the last third Telemetry of June. We used scats that were easy to age to acquire We captured bears with barrel traps set in the the experience necessary to age other scats. For montanezone in 2 drainagesof the study area and near example, a scat found in early August and containing 2 lodges in the Alpine zone. We attemptedto triangulate pits of a cherry that had ripened in early July was most radiocollaredbears twice a week from the groundwith 2 likely about 4 weeks old, giving us an indication as to or 3-element Yagi antennas from August 1996 to what other 4 week-old scats looked like. Up to summer October 1997. In most cases a minimum of 3 azimuths 1998 we froze scats for storage and thawed them be- obtained within 30 minutes were transcribed onto fore analysis. Beginning in summer 1998, we analyzed 1:25,000 topographicmaps and used for each location. unfrozen scats within 1 week of finding them. We soft- Less than 3 azimuths were used for some locations ened all scats in waterand washed them through3 sieves where the lay of the land either made it difficult to get (aperturesof 4, 2.36, and 0.5 mm) to separateindividual more or gave us reasonableassurance that 2 or 1 were food items. All items were identified to the finest sufficient. The extremely complex topography, by al- taxonomic resolution possible. We visually estimated lowing us to quickly move behind land features where the relative volume of each food item and either the signal should dim or die if the location was assigned it to 1 of 6 volume categories: trace, 1-25%, reasonably correct, also often allowed us to verify our 25-50%, 50-75%, 75-100%, and 100% (Maehr and radiolocations. We estimated the elevation of radio- Brady 1984), or, when possible, assigned an exact locations from 1:25,000 topographicmaps. volume (for food items in scats that were composed of equal volumes of 2-3 food items). Items found in trace amountswere given an arbitraryvolume of 1%, but we Results excluded bear hairs, stones, and particles. We Scat analysis calculated percent frequency of occurrence (FO) and We found a total of 408 scats (56, 53, and 299 in percent volume (PV) of items as follows: spring, summer, and fall, respectively) between Sep- tember 1995 and November 1999. Of these, we found of item = numberof scats same Frequency containing 215 opportunisticallyfrom fall 1995 to fall 1998 (12 and item 203 in summer and fall, respectively), and 193 during Percentfrequency of occurrence(FO) = (frequencyof systematic searches in 1999 (56, 41, and 96 in spring, item/totalnumber of scats) X 100 summer,and fall, respectively). Percentvolume (PV) = total volume of item/ percent This sample allowed us to identify 3 main bear food total numberof scats seasons (Table 1). During spring (den emergence-30 Jun) bears fed on oak acorns from the previous fall as For PV calculations, we used the midpoint of each long as they were available. this volume category (we gave an item in the 1-25% volume They complemented with dwarf bamboo leaves (PV = category a value of 12.5)(Dahle et al. 1998); we used during early spring 5.95 before 20 and 1.84 and exact amounts when we estimatedthe exact volume. May after) with dwarf bamboo shoots during late spring (PV = 0 before 20 May and 20.6 after). Kuma dana During summer (1 Jul-31 Aug), bears fed on We collected additionalinformation on bear foraging succulent plants and consumed soft mast when avail- behavior from direct observation of bears and field able. However, scats found at higher elevations con- signs. Asiatic black bears break branchesin treetops to tained less soft mast (x = 9.2%, SE = 4.1) than scats reachhard and soft mast throughouttheir range (Bromlei found at lower elevations (x = 59.7%, SE = 8.5; U = 1965, Schaller 1969, Reid et al. 1991, Hazumi 1999, 149.5, P < 0.001, Table 2). The mean elevation of scats Hwang et al. 2002). This feeding behavior results in found at higher elevations (n = 28) was 2,496 m (SE = conspicuous crude feeding platformsin treetops, called 13), whereas the mean elevation of scats found at lower kumadana (bearshelf) or enza (roundseat) in Japanese. elevations (n = 23) was 1,082 m (SE = 38). We found no We walked roads in the study area in late fall and early scats at the lower elevations from 10 to 31 July and no winter of 1995, after deciduous leaves had fallen, and scats between the elevations of 1,760 and 2,450 m scanned the slopes with binocularsto count kumadana. during the study. Where possible, we walked to these kuma dana to Bears fed on oak acors and walnuts as early as 21 identify species. August, but soft mast comprisedthe bulk of the diet until

Ursus 14(2):236-245 (2003) 240 ASIATICBLACK BEAR DIET IN JAPAN * Huygens et al.

Table 1. Percent frequency of occurrence (FO) and Table 2. Percent frequency of occurrence (FO) and percent fecal volume (PV) of food items in 408 percent fecal volume (PV) of food items found in Asiatic black bear scats in the Northern Japanese Asiatic black bear scats collected during summer at Alps, 1995-99. low elevations (925-1,760 m) and high elevations (2,450-2,650 m) in the Northern Japanese Alps, Springa Summerb Fallc 1995-99. (n = 56) (n = 53) (n = 317) Food item FO PV FO PV FO PV Summera = = Hardmast low (n 25) high (n 28) Quercusspp. 80 63 6 5 78 73 Food item FO PV FO PV Castanea crenata 2 0.7 8 6 Hardmast 15 5 11 3 Juglans 11 mandshurica Quercusspp. 12 4 Pinus koraiensis 4 0.04 6 5 Castanea crenata 1.5 32 Pinus 8 0.5 Juglans mandshurica 9.6 pumila Pinus koraiensis 8 0.1 Soft mast Pinus pumila 14 1 Prunusgrayana 25 23 Soft mast Rubus vernus 9 3 Prunus 52 48 Cornus 6 2 grayana 4 1.5 controversa Prunusmaximowiczii 12 3.5 Vitis 4 1 Corus controversa coignetiae 3 kaki 3 2 Vitiscoignetiae 8 Diospyros 4 1.5 Sorbus commixta 2 2 Acanthopanaxspinosus 18 6 Linderaumbellata 2 1 Rubus vernus Rubus 7 2 Other 43 4 13 4 pseudojaponicus Other 36 2.2 36 2 Graminoids Graminoids Sasa shoots 29 21 4 3 spp. Sasa shoots 8 6 Sasa leaves 32 8 4 2 spp. spp. Sasa leaves 7 3 leaves 11 6 11 6 spp. Other, Other 21 11 Forbs Forbs (stems & leaves) Rubus 8 5 Angelicaspp. 32 19 pseudojaponicus Rubus 14 9 Elatostema 4 4 pseudojaponicus Cardiocrinumspp. 11 8 japonicum Petasites spp. 11 7 Angelicaspp. 17 10 Cirsiumspp. 14 6 Cardiocrinumspp. 6 4 Urticaspp. 4 2 Petasites spp. 6 4 Cirsiumspp. 8 3 Otherplant material 28 5.2 32 23 Otherplant material 54 4 30 15 16 2 Insects Insects Formicidae 40 5.6 36 3 8 0.5 4 0.5 9 0.1 38 4 1 0.3 Vespidae Formicidae Other 4 0.5 Vespidae 2 0.02 6 0.5 3 0.1 Bibionidae 7 0.5 aFrom1 Jul until30 Aug. Mammals Naemorhedus 11 2 1 0.4 eggs (Formicidae); and bees and wasps (Vespidae) in all crispus Other 13 0.1 19 2 23 1 seasons, but peaking during summer. Hair and meat in scats also indicated bears ate serows on at least 6 aFromden until30 Jun. emergence and 3 in total of 9 bFrom1 Jul until31 Aug. occasions (3 in spring, fall; scats). CFrom1 Sep untilden entrance. We foundonly 19 scats (4.7%)in the subalpinezone, of which 16 were of Korean stone pine seeds, 2 were of dwarf bamboo shoots, and 1, from a bear capturedin consistedof 31 August (Table 1). During fall (1 Sep-den entrance), a barreltrap set in the subalpinezone, stems, forbs and berries almost disappearedfrom the diet and leaves, ants, and Actinidia arguta . We spent for scats in the were replaced by walnuts, Korean stone pine seeds, comparatively less time searching bears chestnuts, and especially Mongolian oak and Japanese subalpinezone, as radiotelemetry results indicated white oak acoms. Bears ate small quantitiesof ants, ant spent little time there (Huygens 1998, Carret al. 2002).

Ursus 14(2):236-245 (2003) ASIATICBLACK BEAR DIET IN JAPAN * Huygens et al. 241 . ~~m ~ ~~ m m m m m

Table 3. Species of trees containing kuma dana determination. Of these, 162 (75%) were in either (bear feeding plafforms) observed in the Northern Mongolian or Japanese white oak. Over 92% of the fall 1995. Japanese Alps, kuma dana were in trees that produce hard mast in the Scientific name Occurrence % fall (Table 3). We found 2 kuma dana in trees that had broken Quercuscrispula 112 51.9 obviously while the bears where in them, Quercusserrata 37 17.1 causing the bearsto fall to the ground,one from a height Quercusspp.8 13 6.0 >10 m. Castanea crenata 34 15.7 Linderaobtusiloba 10 4.6 Diospyroskakib 5 2.3 Telemetry Juglans mandshurica 3 1.4 We radiocollared21 bears (11 adult and 3 juvenile Cornusmacrophylla 2 0.9 males; 5 adult and 2 juvenile females) between August Total 216 100.00 1995 and October 1997 and obtained 469 usable aQ. crispulaor Q. serrata. radiolocations on them by December 1997 (Huygens bFromplanted but abandoned trees in an area near a village 1998). It was difficult to obtain bear locations in some where the forest had reclaimedfarm land on a steep slope. roadless areas, but this could often be circumventedby hiking up to trails on the high ridges. Even so, we were We found only 7 scats (1.7%) in tree plantations, unable to obtain radiolocationsof bears in some areas, even such though plantations made up 19.7% of the especially of the subalpinezone. study area. Telemetry locations (Huygens 1998) indi- During summer,5 bears (4 males, 1 female) remained cated bears occasionally used tree plantations,either in at lower elevations (< 1,500 m) and 10 bears (6 males, 4 the spring when they presumablyfed on dwarf bamboo females) either stayed above 2,300 m or moved back leaves or shoots, or in the fall when kumadana in either and forth between these high elevations and elevations Japanesechestnut trees or Mongolian oaks, which occur <1,500 m. Up to 3 unknown adult bears fed concur- sporadically in larch plantations, indicated bears had rently in the same meadow above timberlinein summer. been feeding in them. After August, all radiocollared bears moved to the We found scats on hiking trailsonly in the alpinezone. montane zone at elevations <1,500 m to feed on hard Especially during spring and summer, there was a pre- mast, especially oak acorns. ponderance of scats along small, steep ridge-like Two bears (1 male, 1 female) remainedin the central formations formed when a steep slope meanders back drainage (which had experienced more timber cutting) and forth.This may have been because these steep mini- throughoutthe , but 5 bears (4 males, 1 female) left ridges were easierfor us to search.However, 19 of the 25 that drainagefor the more pristinedrainages in the fall. day beds we found were located on these ridge-like The fact that of 299 fall scats only 41 (13.7%) were from formations,often on the flatterarea immediatelyabove the central, more heavily cut drainage, probably also a large tree, further suggesting that bears used them reflects this exodus. No radiocollaredbear moved to the frequently.Bears may like such vantage points because more disturbeddrainage in the fall (Hugyens 1998). allow they surveillanceof the surroundings,offer easy Field signs, direct observationof bears, and depreda- escape routes, are breezier and therefore cooler, and tion reports indicated bears fed on other foods besides possibly also offer an occasional chance to prey upon those revealed in scats. Although we never found unsuspectingapproaching ungulates using such ridges as beechnuts or beech buds in scats, we found beeches travel routes. Predation by Asiatic black bears in our with numerousolder claw marksleading up theirtrunks, study was suggested when we saw a young, radio- suggesting that in previous years bears eitherhad fed on collaredmale drag the carcass of a serow across a creek beech buds in the spring or on beechnuts in the fall. In on 1 November 1995. Upon inspectionthe next day the spring 1999 we occasionally found evergreentrees (e.g. serow was to have been judged healthy, with no obvious hinoki cypress and Japanesered pine) with peeled bark, and with clean a leg injuries, hooves, good dentition;it indicating bears had fed on the cambium. On 19 May was 10-15 as determined years old, by ring counts on its 1996, we saw an adult bear feed on the budding leaves horns. of a Salix sachalinensis. A minimum of 13 bears were killed on the southeasternedge of the study area where they were raiding orchardsand apiaries,indicating that Kuma dana the diet of some bears in the study area included these We found 264 kumadana in the study areaduring fall anthropogenicfoods. We observed a bear feed on the of 1995, which we accessed 216 (81.8%) for species nest of the wasp Vespula lewisi located in the hollow

Ursus 14(2):236-245 (2003) 242 ASIATICBLACK BEAR DIET IN JAPAN * Huygens et al. branchof a windblown tree. Anotherbear dug up a nest availability,related in turnto successional stage, caused of V. flaviceps (a ground dwelling wasp), and another this movement. Early successional species (e.g. Prunus unsuccessfullyattempted to enlargethe entranceof a nest spp., Cornus spp., Vitus spp.), which were more com- of wild honey bees (Apis mellifera) in a hollow tree. mon in the central drainage, appearedto be adequate Bears in the study area peeled acorns and chestnuts producers of soft mast during summer and thus of before eating them, and we found numerousacorn peels interestto the bears. However, we hypothesize that late below kuma dana in oak trees and no peels, or only successional species (particularlyMongolian oak, which traces, in scats. was more common and maturein the pristinedrainages), We found 3 day-beds of interwoven, uncut dwarf were better producers of hard mast during fall, thus bamboostems and leaves in dwarfbamboo thickets (2 in attractingbears. early spring on relatively flat areas and 1 in late autumn Asiatic black bears peeled acorns and chestnuts and on a steep slope). These somewhat oval-shaped beds scats consequently lacked these peels. This made it formed a springy mattress approximately90 cm long difficult to differentiate between acorn and chestnut and 30 to 40 cm high. scats, especially when they were older. Although we observed kuma dana in chestnut trees throughoutthe study, we noticed chestnuts in scats only in 1999, the Discussion only year the oak acorn crop was poor. We may have Except for their use of the alpine zone, the behavior missed chestnutsin scats in previous years due to a lack and food habits of Asiatic black bears in the Northern of experience. Japanese Alps were consistent with those reported in Hwang et al. (2002) also found ground nests the literaturethroughout Asiatic black bear range (e.g. intricately built with live vegetation in Taiwan, and Bromlei 1965, Schaller 1969, Reid et al. 1991, these nests resulted in a bowl-like structurein which Hashimoto and Takatsuki 1997, Hwang et al. 2002). a bear might hide. The nests we found resembled an Virtuallyall Asiatic black bear diet studies reportshifts insulatingspringy cushion on which a bear might more from succulent forbs and grasses in spring, to soft mast likely rest than hide, possibly during cold periods. in summer, and hard mast, either oak acors, Korean Personal observations indicated that Japanese stone stone pine seeds, or walnuts, in fall. When available, pines, which are abundantin the alpine zone of the study Asiatic black bears ate hard mast from the previous area,produced poor seed crops duringthe 3 summersof fall in spring (Bromlei 1965, Yudin 1993, Hashimoto 1997 to 1999. Throughoutthe world, bears depend on and Takatsuki1997, Goto 2000). Eating dwarf bamboo the seeds of stone pines whereverbears and stone pines The seeds of in spring has been reported in Japan (Takada 1979, are sympatric(Mattson and Jonkel 1990). than the seeds of Naganawa and Koyama 1994) and China (Wu 1983, Japanesestone pines are much smaller be of less Wang 1988, Schaller et al. 1989, Reid et al. 1991). Korean stone pines and may consequently Asiatic black bears in other studies also moved to sea- interest to bears. It is possible, however, that in good on the seeds of sonal feeding areas and made little use of the subalpine crop years bears in our study area feed zone (Wu 1983, Hazumiand Maruyama1987, Reid et al. Japanese stone pines more extensively than reported in mid- 1991, Yudin 1993, Hwang et al. 2002). The paucity of here, especially because these seeds ripen Asiatic black bear scats found on trails, the peeling of summer(Hayashida 1994) when other hard mast is not acors, the eating of the cambium of trees, yet available. even in beech buds, and beechnuts, and the possible ambush- Oak acorns dominatedin the scats each fall, when the acorn was ing of ungulates have also been previously reported 1999, the only year crop relatively This a (Furubayashiet al 1980, Watanabe 1980, Nozaki et al. poor (personal observations). suggests longer- our fall 1983, Schaller et al. 1989, Hwang et al. 2002). Finally, term study would not greatly affect results, American black bears (U. americanus) were also except in years of total acorn crop failure. Our spring results are based on 1999 scats, reportedto have different concurrentdiets at low and and summer mostly too short to document bear diet. high elevations (Graber and White 1983, Raine and a period adequately that studies Kansas 1990). Mattson et al. (1991) suggested long-term document bear food habits We observed a partial exodus of bears from the are necessary to adequately substantial differences centraldrainage, which had been logged more intensely, because they noted among-year scat content in Yellowstone towardthe 2 more pristinedrainages during autumn. We in grizzly bears (U. arctos) and that estimates of scat hypothesize that differences among drainagesin forage National Park, USA, average Ursus 14(2):236-245 (2003) ASIATICBLACK BEAR DIET IN JAPAN * Huygens et al. 243 composition took 4-6 years to stabilize. Furtherstudies consumption in the fall allows bears to rapidly gain are needed in our study area to document bear diet in weight before the crucial hibernationperiod. Bear re- years of total acorn crop failure and to complete our productive success and levels of bear-human conflicts understandingof bear diet in spring and summer. are directly and indirectly related to food availability Several factors affected the numberof scats we found during summer and especially fall (e.g. Jonkel and each season. Bears did not congregateduring spring and Cowan 1971, Bunnell and Tait 1981, Rogers 1987, Eiler summer because food items were scattered, and scat et al. 1989, Elowe and Dodge 1989, Mattson et al. visibility was lowest duringsummer due to the luxurious 1992). Thus, degradationof the montane zone, where plant development which covered the ground and hid soft and hardmast producingplants are concentrated,is scats. Consequently,scats were most difficult to find in likely to reduce bear reproductivefitness and increase summer,followed by springand fall. Bears also possibly levels of bear-humanconflicts in the NorthernJapanese increasedtheir ingestion and defecationrates duringlate Alps. General habitat degradationcontinues to occur summer and fall, as has been reportedfor brown bears throughoutJapan (Hazumi 1994, 1999) and occurredin (Roth 1980, Mattson et al. 1991), possibly furtherex- the montanezone of the study area duringthis study in plaining our larger fall sample. the form of evergreen plantations and constructionof We could not search for scats in thick dwarf bamboo golf courses, roads, dams, hotels, and second homes. stands or on extremely steep slopes, even though telem- Chubu Sangaku National Park covers mostly scenic etry data indicatedbears used these areas. We possibly but biologically less diverse areas in the subalpine and biasedthe samplefurther by collectingall scats regardless alpine zones of the study area and includes little of the of cluster size, mainly because scats in a cluster often more biologically diverse montane zone where bear containeddifferent food items. The senior authorfound fall foods are concentrated.Strategies aimed at reducing all but 33 (8.1%) of the 408 scats we found, possibly levels of bear-human conflict in Japan should include further biasing the sample because his searches were protectionof remainingbroadleaf forests and conversion based on subjectivecriteria or were often concentratedon of evergreen tree plantationsto broadleaf forests. Our areas where he found scats in previous years. However, recommendationsare similarto those of Takada(1979), searches were also oriented toward areas where radio- Hazumi and Maruyama(1987), and Hazumi (1999) in telemetrydata indicatedbear presence. Additionally,we Japan,of Yudin (1993) and Chestinand Yudin (1999) in did not reach some kumadana for identificationbecause the Russian Far East, of Reid et al. (1991) in China, and of a combination of deep snow, steep slopes, dense of Sathyakumar(1999) in India. vegetation,and distance,biasing our sample. Scat analysis may bias estimates of food habits because of differential disappearanceof foods due to chewing and digestion. This bias can be reduced with Acknowledgments We thank the Zen Rosai Foundationfor its financial correction factors that relate the amount of residue in contributions, J. Moll for his comments on the first feces to the amount of a diet item ingested (Hewitt and version of the manuscript,and DLDH, who provided Robbins 1996). We did not use digestibility correction other invaluableforms of support. factors because we lacked appropriateequations for many of the food items we encountered.Therefore, we probably underestimatedthe importance of highly di- gestible or easily fragmentedfoods and overestimated Literaturecited the importanceof less digestible items relative to their AZUMA, S., ANDH. TORII.1980. Impact of humanactivities on actualproportion in the diet (Hewitt and Robbins 1996). survivalof the Japaneseblack bear. International Confer- ence on BearResearch and Management 4:71-79. BROMLEI,F.G. 1965. Bears of the south far-easternUSSR Managementimplications (Medvediyugo dal'nego vostoka SSSR). Datel'stvo Akad- emiiNauk USSR. During late summer and fall, Asiatic black bears in SSR,Moskva, (Translatedfrom Russian, IndianNational Scientific Documentation Center, New the NorthernJapanese Alps concentratedin the montane Delhi,India, 1973.) zone where fed on and hard mast. they soft, especially, BUNNELL,F.L., ANDD.E.N. TAIT.1981. Hard mast is in fat et al. Populationdynamics high (Landers 1979, Eagle of bears-implications.Pages 75-98 in C.W.Fowler and and Pelton 1983, Kasbohm et al. Hashimoto and 1995, T.D. Smith,editors. Dynamics of largemammal popula- Takatsuki 1997, Kirkpatrickand Pekins 2002), and its tions.John Wiley & Sons,New York,New York,USA.

Ursus 14(2):236-245 (2003) 244 ASIATICBLACK BEAR DIET IN JAPAN * Huygens et al.

CARR,M.M., J. YOSHIZAKI,F.T. VANMANEN, M.R. PELTON, Nature and Natural Resources, Gland, Switzerland, and O.C. HUYGENS,H. HAYASHI,AND M. MAEKAWA.2002. A Cambridge,United Kingdom. multi-scaleassessment of habitatuse by Asiatic black bears -, ANDN. MARUYAMA.1987. Movementsand habitatuse of in centralJapan. Ursus 13:1-9. Japaneseblack bears in Nikko. InternationalConference on CHESTIN,I., ANDV. YUDIN.1999. Status and management Bear Research and Management7:275-279. of the Asiatic black bear in Russia. Pages 211-213 in HEITrr, D.G., ANDC.T. ROBBINS.1996. Estimating grizzly C. Servheen, S. Herrero,and B. Peyton, compilers. Bears. bear food habits from fecal analysis. Wildlife Society Status survey and conservation action plan. International Bulletin 24:547-550. Union for Conservationof Nature and NaturalResources, HUYGENS,O.C. 1998. Asiatic black bear in Nagano prefecture, Gland, Switzerland,and Cambridge,United Kingdom. central Japan: conservation approach and ecological DAHLE,B., O.J. S0RENSEN,E.H. WEDUL,J.E. SWENSON,AND aspects. Thesis, Shinshu University, Matsumoto, Nagano F. SANDEGREN.1998. The diet of brown bears Ursus arctos prefecture,Japan. in central Scandinavia: effect of access to free-ranging HWANG,M.-H., D.L. GARSHELIS,AND Y. WANG.2002. Diets of domestic sheep Ovis aries. Wildlife Biology 4:147-158. Asiatic black bears in Taiwan, with methodological and EAGLET.C., ANDM.R. PELTON.1983. Seasonal nutrition of geographicalcomparisons. Ursus 13:111-125. black bears in the Great Smoky MountainsNational Park. JONKEL,C.J., AND I.M. COWAN.1971. The black bear in the InternationalConference on Bear Research and Manage- spruce-fir forest. Wildlife Monograph27. ment 5:94-101. KASBOHM,J.W., M.R. VAUGHAN,AND J.G. KRAUS.1995. Food EILER,J.H., W.G. WATHEN,AND M.R. PELTON.1989. Re- habits and nutritionof black bears during a gypsy moth production in black bears in the Southern Appalachian infestation.Canadian Journal of Zoology 73:1771-1775. Mountains.Journal of Wildlife Management53:353-360. KIRKPATRICK,R.L., ANDP.D. PEKINS.2002. Nutrition value of ELOWE,K.D., ANDW.E. DODGE.1989. Factors affecting black acorns for wildlife. Pages 173-181 in W.J. McShea and bear reproductive success and cub survival. Journal of W.M. Healy, editors. Oak forest ecosystems: ecology and Wildlife Management53:962-968. managementfor wildlife. Johns Hopkins University Press, FURUBAYASHIK., K. HIRAI,K. IKEDA,AND T. MIZUGUCHI.1980. Baltimore,Maryland, USA. Relationships between occurrence of bear damage and LANDERS,J.L., R.J. HAMILTON,A.S. JOHNSON,AND R.L. clearcuttingin central Honshu, Japan. Interational Con- MARCHINTON.1979. Food and habitat of black bears in ference on Bear Research and Management4:81-84. southeasternNorth Carolina.Journal of Wildlife Manage- GOTO,M. 2000. Asiatic black bear in Asahi village, Nagano ment 43:143-153. prefecture(Nagano-ken Asahi-mura shuhen chiiki ni okeru MAEHR,D.S., ANDJ.R. BRADY.1984. Food habits of Florida Tsukinowaguma no seitai to hogo-kanri jirei ni tsuite). black bears. Journalof Wildlife Management48:230-235. M.S. Thesis, Shinshu University, Minami Minowa, Japan. MATrSON,D.J., ANDC. JONKEL.1990. Stone pines and bears. (In Japanese.) Pages 223-236 in W.C. Schmidt and K.J. McDonald, GRABER,D.M., ANDM. WHITE.1983. Black bear food habits in compilers. Proceedings-Symposium on whitebark pine Yosemite National Park. InternationalConference on Bear ecosystems: ecology and management of a high-mountain Research and Management5:1-10. resource. U.S. Forest Service General Technical Report HASHIMOTO,Y., AND S. TAKATSUKI.1997. Food habits of INT-270. Ogden, Utah, USA. Japanese black bear: A review (Tsukinowaguma no , B.M. BLANCHARD,AND R.R. KNIGHT.1991. Food shokusei: sousetsu). Journalof the MammalogicalSociety habits of Yellowstone grizzly bears, 1977-1987. Canadian of Japan37:1-19. (In Japanesewith English abstract.) Journalof Zoology 69:1619-1629. HAYASHIDA,M. 1994. Role of nutcrackerson seed dispersal , ANDR.R. KNIGHT.1992. Yellowstone grizzly and establishment of Pinus pumila and P. pentaphylla. bear mortality, human habituation, and whitebark pine seed 56:432-442. Pages 159-162 in W.C. Schmidt and F.K. Holtmeier, crops. Journalof Wildlife Management 1994. Food habits of compilers. Proceedings-International workshop on sub- NAGANAWA, K., AND M. KOYAMA. bear in the Tanzawamountains. alpine stone pines and their environment:the status of our Japaneseblack (Tanzawa- no Transactions knowledge. U.S. Forest Service GeneralTechnical Report Sanchi ni okerutsukinowaguma shokusei). 105:539-542. INT-GTR-309.Ogden, Utah, USA. of the Japanese Forestry Society (In HAZUMI,T. 1994. Status of Japaneseblack bear. International Japanese.) the fauna of Conferenceon Bear Research and Management9(1):145- NovIKov, G.A. 1956. Carivorous mammals of 148. the USSR. Keys to the fauna of the USSR. Zoological . 1999. Status and managementof the Asiatic black bear Instituteof the Academy of Sciences of the USSR. No. 62. USSR. in Japan. Pages 207-211 in C. Servheen, S. Herrero,and Izdatel'stvoAkademii Nauk SSR, Moskva, (Trans- for Scientific Trans- B. Peyton, compilers. Bears. Status survey and conserva- lated from Russian, Israel Program tion action plan. InternationalUnion for Conservationof lations, ,Israel, 1962.)

Ursus 14(2):236-245 (2003) ASIATICBLACK BEAR DIET IN JAPAN * Huygens et al. 245

NOZAKI,E., S. AZUMA,T. AOI, H. TORII,T. ITO,AND K. MAEDA. TAKADA,Y. 1979. The food habits of the Japaneseblack bears 1983. Food habits of Japanese black bear. International in the central mountain area, Nagano prefecture(Nagano- Conferenceon Bear Researchand Management5:106-109. Ken Chuou Sanchi ni okeru Nihon-tsukinowagumano NUMATA,M., EDITOR.1974. The flora and vegetation of Japan. shokusei). Journalof the MammalogicalSociety of Japan Kodansha Limited and Elsevier Scientific Publishing 8:40-53. (In Japanesewith English abstract.) Company, American Elsevier Publishing Company, St. TORII,H. 1989. Food habits of the Japanese black bear in Louis, , USA. the headwatersof the Ohwi River (Ohwi-gawa Joryu-iki RAINE,R.M., ANDJ.L. KANSAS.1990. Black bear seasonal food ni okeru tsukinowaguma no shokusei). Journal of the and distribution elevation in Banff National habits by Park, Japanese Forestry Society 71:417-420. (In Japanese with Alberta. InternationalConference on Bear Research and English abstract.) 8:297-304. Management WADA, K. 1985. Vegetation der PrafekturNagano (Nagano- M.D. Z.S. ANDJ.C. Hu. 1991. REID,D., JIANG,Q.T. TENG, QIN, ken no shokusei). Pages 473-482 in A Miyawaki, editor. of the Asiatic black bear (Ursus thibetanus) in Ecology Vegetation of Japan,Chubu. ShibundoPublishers, Tokyo, Sichuan, China. Mammalia55:221-237. Japan.(In Japanese.) ROGERS,L. 1987. Effects of food supply and kinship on social WANG, X.M. 1988. Ecology of black bear (Heixiong de behavior, movements, and population growth of black shengtai).Chinese wildlife (Yesheng dongwu) 2:16-18. (In bears in NortheasternMinnesota. Wildlife Monograph97. Chinese.) ROTH,H.U. 1980. Defecation rates of brown bears. captive H. 1980. to conifers the InternationalConference on Bear Research and WATANABE, Damage by Japanese Manage- black bear. InternationalConference on Bear Researchand ment 4:249-253. Management4:67-70. SATHYAKUMAR,S. 1999. Status and management of the Asiatic . 1981. Black bear damage to artificial in black bear in Japan. Pages 202-207 in C. Servheen, regeneration between control and reservation.Proceed- S. Herrero,and B. Peyton, compilers. Bears. Status survey Japan-conflict of the International Union of Forest Research and conservation action plan. International Union for ings Conservation of Nature and Natural Resources, Gland, Organizations17:581-586. J.Y. 1983. observations on food habits of Switzerland,and Cambridge,United Kingdom. Wu, Preliminary black bear from SCHALLER,G.B. 1969. Food habitsof the Himalayanblack bear Qinling. (Qinling heixiong shixing de chubu (Selenarctos thibetanus) in the Dachigam Sanctuary, guancha).Chinese Journalof Zoology (Dongwuxue Kashmir. Journalof the Bombay Natural History Society Zazhi) 4:48-51. (In Chinese.) 66:156-159. YUDIN,V.G. 1993. The . Pages 479-491 in , Q.T. TENG,K.G. JOHNSON,X.M. WANG,H.M. SHEN, M.A. Vaisfeld and I.E. Chestin, editors. Game animals of AND J.C. Hu. 1989. The feeding ecology of giant pandas Russia and adjacent countries and their environments- and Asiatic black bears in the Tangjiahe reserve, China. bears, brown bear, polar bear, Asian black bear, distribu- Pages 212-241 in J.L. Gittleman, editor. Carnivore tion, ecology, use and protection.Institute of Geographyof behavior,ecology, and evolution. Corell UniversityPress, the Russian Academy of Sciences. Moscow "Nauka", Ithaca,New York, USA. Russia. (In Russian with English abstract.) SERVHEEN,C., S. HERRERO,AND B. PEYTON.1999. Bears. Status survey and conservation action plan. InternationalUnion Received: 9 August 2003 for Conservationof Nature and NaturalResources, Gland, Accepted: 16 July 2003 Switzerland,and Cambridge,United Kingdom. Associate Editor: R.T. Shideler

Ursus 14(2):236-245 (2003)