Download pdf - Response of a Group of Sichuan Snub‐Nosed Monkeys to Commercial Logging in the Qinling Mountains, China

Contributed Paper

Response of a Group of Sichuan Snub-NosedMonkeys to Commercial Logging in the QinlingMountains, ChinaSONGTAO GUO,∗† WEIHONG JI,‡ BAOGUO LI,∗†∗∗ AND MING LI§∗Department of Geology, Northwest University, Xi’an 710069, China†Key Laboratory of Resource Biology and Biotechnology in Western China of Ministry of Education, and College of Life Sciences,Northwest University, Xi’an 710069, China‡Institute of Natural Resource, Massey University, Albany, Auckland, New Zealand§Institute of Zoology, Chinese Academy of Sciences, Beijing 100101, China

Abstract: Habitat destruction is one of the greatest threats to primates worldwide. To understand the impact

of forest logging on the habitat use of primates in temperate mixed forest, we compared the range, habitat

used, population size, and diet of a troop (ERT) of Sichuan snub-nosed monkeys (Rhinopithecus roxellana)

in the Qinling Mountains before (1989), during (1997), and after (2002–2003) commercial logging. Logging

significantly changed the composition of the forest and the food supply for the troop. Some areas were heavily

logged and formed patches in the forest that lacked canopy cover. The troop moved 7 km away from their

original range when logging took place and returned to their original range after logging stopped, but they

avoided heavily logged areas that lacked canopy cover. Their movement indicated some degree of site fidelity

in this species. Diet and home range changed after logging, but the population size remained stable, which

suggests that this species has some ability to adapt to habitat changes. Our results may reflect a natural

flexibility in primates to adapt to the changing food resources in temperate areas with marked seasonal

variations in food availability and distribution. This flexibility may have contributed to their higher degree of

resilience to habitat alterations caused by human activities compared with tropical forest primates that have

a more specialized diet. Our findings provide important baseline information that will help decision makers

in their efforts to conserve primates, especially in temperate regions, and to sustainably manage primate

habitat.

Keywords: commercial logging, habitat modification, primate diet, primate range shift, Qinling Mountains,Rhinopithecus roxellana, Sichuan snub-nosed monkey

Respuesta a la Explotacion Forestal Comercial de un Grupo de Monos Rhinopithecus roxellana en las MontanasQinling, China

Resumen: La destruccion del habitat es una de las mayores amenazas a los primates del mundo. Para

entender el impacto de la explotacion forestal comercial sobre el uso de habitat por primates en bosques

templados mixtos, comparamos la distribucion, el habitat utilizado, el tamano poblacional y la dieta de

una tropa (ERT) de monos Rhinopithecus roxellana en las Montanas Qinling antes (1989), durante (1997) y

despues (2002–2003) de la explotacion forestal comercial. La explotacion forestal cambio significativamente

la composicion del bosque y la disponibilidad de alimento para la tropa. Algunas areas fueron explotadas

intensivamente y formaron parches en el bosque que carecıan de cobertura de dosel. La tropa se desplazo

7 km de su distribucion original cuando la explotacion forestal se llevo a cabo y retorno a su distribucion

original cuando ceso, pero los monos evitaron las areas intensamente explotadas que carecıan de dosel. Este

desplazamiento indico algun grado de fidelidad de sitio en esta especie. La dieta y el rango de hogar cambiaron

∗∗Address correspondence to Baoguo Li, email [email protected] submitted September 13, 2007; revised manuscript accepted December 31, 2007.

1055Conservation Biology, Volume 22, No. 4, 1055–1064C©2008 Society for Conservation BiologyDOI: 10.1111/j.1523-1739.2008.00975.x

1056 Response of R. Roxellana to Logging

despues de la explotacion forestal, pero el tamano de la poblacion permanecio estable, lo que sugiere que esta

especie tiene cierta habilidad para adaptarse a cambios en el habitat. Nuestros resultados pueden reflejar

una flexibilidad natural de los primates para adaptarse a los cambios en los recursos alimenticios en areas

templadas con una variacion estacional marcada en la disponibilidad y distribucion de alimento. Esta

flexibilidad puede haber contribuido al mayor grado de resiliencia a las alteraciones del habitat provocadas

por actividades humanas en comparacion con primates de bosques tropicales que tiene una dieta mas

especializada. Nuestros resultados proporcionan importante informacion basica que ayudara a los tomadores

de decisiones en sus esfuerzos por la conservacion de primates, especialmente en regiones templadas, y al

manejo sustentable del habitat de primates.

Palabras Clave: cambio en la distribucion de primates, dieta de primates, explotacion forestal comercial,modificacion del habitat, montanas qinling, Rhinopithecus roxellana

Introduction

Habitat modification is one of the main threats to pri-mate populations around the world. Forest logging of-ten is the most destructive type of habitat modificationsfor primates because it alters the distribution and abun-dance of their resources such as food and shelter (Johns1986a; Marsh et al. 1987; Gilbert 2003). A number of re-searchers have reported the effects of logging on tropicalforest primates. Changes in range and population densityare common following logging (Johns 1986a; Estrada &Coates-Estrada 1996; Chapman et al. 2000, 2003). One ofthe causes of range shifts by primates in response to log-ging is changes in the distribution of food resources, andthe nature of these range shifts depends on the diet ofthe species concerned. Primates with a specialized dietare more vulnerable to extinction than generalists whenalterations in their habitats disrupt the availability of theirkey food items (Yamagiwa 1999), and logging tends tohave more effect on the home ranges of primates withmore specialized diets (Jeffrey 1978; Johns 1986b).

The effects of logging on primate population size variesand appear to be associated with changes in availabil-ity of food resources and the dietary flexibility of thespecies involved. In Kibale, Uganda, a population of Pro-

colobus badius with a specialized diet declined after theirhabitat was logged. On the contrary, Colobus guereza,which has a broad diet, is more abundant in logged areas(Johns & Skorupa 1987). In Malaysia the main character-istic of frugivore and folivore primates that enables themto survive in logged forests is their considerable ability tochange their diet and feeding behavior (Johns 1986a).

Logging generally induces changes in forest composi-tion and structure in the short and long term (Johns 1988;Plumptre 1996). This can have serious negative effectson the food and other resources in primate habitats (Sko-rupa 1988; Davies 1994; Ganzhorn 1995; Chapman et al.2000). Nevertheless, the degree of the effect varies overtime and with the extent of logging (Johns & Skorupa1987). Low-intensity logging may induce an increase inplant diversity and abundance of some tree species. For

example, in Madagascar increased exposure of the re-maining trees to sunlight led to an increase in abundanceof fruit for frugivorous lemurs and improved the qualityof leaves for folivorous species, at least in the short term(Ganzhorn 1995). Nevertheless, heavy logging in Kibaleforest, Uganda, removed about 50% of all trees and thebasal area of food trees for red colobus (P. badius), andthis caused a major decline in their numbers (38%) (Sko-rupa 1986). Understanding the tolerance of primates fordifferent intensities of logging is important for decisionmaking about the future conservation and sustainable useof forest resources.

Much less is known about the responses of primatesto habitat disturbance in temperate forests, but somepredictions can be made on the basis of their diets inthese highly seasonal habitats (Kirkpatrick et al. 1998;Xiao et al. 2003; Ding & Zhao 2004). For example,Yunnan snub-nosed monkeys (Rhinopithecus bieti) areless specialized in their habitat and food choice and, al-though much reduced in numbers, can survive in frag-mented, mixed coniferous–broadleaf forest as well asundisturbed coniferous forest (Xiao et al. 2003; Ding &Zhao 2004). Their persistence seems to be related totheir ability to make seasonal diet changes in responseto extreme seasonal fluctuations in food availability as-sociated with the temperate climate. Likewise, the dietof R. roxellana varies by site and season (Li 2001; Guoet al. 2007), as does the diet of rhesus macaques (Macaca

mulatta) in Himalayan pine forests (Wada 1984) and ofJapanese macaques (Macaca fuscata) on the Kinkazanand Yakushima islands (Nakagawa 1989; Hanya 2004).

One key to the adaptability of these species to a tem-perate climate is their consumption of seeds from dif-ferent trees, including Pinus spp., and of a variety ofhigh-fiber foods, such as mature leaves and bark. Thediet of Rhinopithecus spp. varies significantly with sea-son (Ding & Zhao 2004; Guo et al. 2007). They havesurvived some habitat degradation caused by livestockgrazing (Xiao et al. 2003) and the collection of firewoodand plants and seeds used in medicines. We predict thatsnub-nosed monkeys can tolerate low-intensity logging;

Conservation Biology

Volume 22, No. 4, 2008

Guo et al. 1057

however, they naturally have very large home rangescompared with colobine species in the tropics (Kirk-patrick et al. 1998; Li et al. 2000). Therefore, confine-ment in fragmented habitat could reduce their ability toadapt to habitat modified by logging unless range shiftsare possible.

Sichuan snub-nosed monkeys, R. roxellana, are semi-arboreal (Ren et al. 2001; Zhang et al. 2006) and liveon the northern edge of colobine distribution in mixedforests dominated by the families Betulaceae, Fagaceae,and Pinaceae. They often form unusually large troopsof more than 100 individuals consisting of social groupsmade up of several one-male units (a single adult malewith multiple adult females and juveniles) and all-maleunits (Kirkpatrick 1998; Kirkpatrick et al. 1998; Ren et al.2000; Tan et al. 2007; Zhang et al. 2003). They haveextensive home ranges (28–55 km2) and low popula-tion densities (1.6–11.8 individuals/km2) (Davies 1994;Kirkpatrick 1998). The ranges of neighboring troops donot usually overlap (Li et al. 1999). R. roxellana canchange their diet in the event of unusual climate changewhen food becomes scarce (Li et al. 2003). Nevertheless,their ability to adapt to habitat disturbance, such as forestlogging, has not been documented in detail.

Habitat loss due to human commercial activities, in-cluding logging, has been a major threat to the survival ofR. roxellana (Hu 1998; Li et al. 2001). To date, the effectof logging on 2 groups of R. roxellana in different areasof China has been studied. A troop of monkeys in theYuhuangmiao region of the Qinling Mountains, ShaanxiProvince, immediately moved out of their summer rangein response to logging (Li et al. 1999). In ShennongjiaReserve, Hubei Province, R. roxellana prefer primaryand lightly logged forest to the grasslands formed afterclearcutting (Li 2004). Neither of these studies providedcomplete information on range use, population size, orbehavioral responses before, during, and after logging.Such information is important for conservation and sus-tainable use of forests.

Commercial logging was common in the habitat ofR. roxellana in the Qinling Mountains in the 1960s–1990s, which caused habitat fragmentation and degra-dation. From 1958 to 1998, 5 national logging companiesoperated in the Qinling Mountains (Li et al. 1999). Be-tween 1994 and 1998, commercial logging was carriedout in the Yuhuangmiao region where troops of R. roxel-

lana live. Logging stopped when a new forest policy wasimplemented by the Chinese government in 1998 (Zhanget al. 2000).

We used data collected before (1989–1990), during(1997), and after (2002–2003) logging to investigate theability of R. roxellana to adapt to habitat destruction.Such long-term studies built on this type of data are rare.Our findings should improve understanding of the re-sponse of primates to habitat destruction in temperateregions, and they should have important implications for

conservation and management of this endangered speciesand for conservation of primates in general, particularlythose in temperate regions.

Methods

Study Site and Subjects

The study site was in the Yuhuangmiao region, ZhouzhiNature Reserve, on the northern slope of the QinlingMountains (108◦14

′– 108◦18

′E, 33◦45

′– 33◦ 50

′N, ele-

vation: 1400–2900 m above sea level) in Shaanxi, China(Fig. 1). The site included steep terrain on West Ridgeand East Ridge which was divided by Middle Ridge andthe Nancha River (Fig. 1). Vegetation included deciduousbroadleaf (1400–2200 m), mixed deciduous broadleafand coniferous (2200–2600 m), and coniferous forest(>2600 m). The climate in the area is strongly seasonal.Average annual temperature is 6.4 ◦C (range: −8.3 ◦Cin January to 21.7 ◦C in July) and annual rainfall is 980mm (Li et al. 2000). We defined the seasons as spring(21 March–22 June), summer (23 June–23 September),autumn (24 September–22 December), and winter (23December–20 March 20).

Figure 1. Ranges of R. roxellana in 1989–1990 (before

logging), 1999 (shortly after logging); and 2002–2003

(5 years after logging) (1, D [Da-ai-gou] Valley; 2, H

[Huang-shi-gou] Valley; 3, S [Shang-he-gu] Valley; 4, L

[Lao-ji-gou] Valley; ERT (East Ridge Troop). Data on

home range of West Ridge Troop (WRT) is from Li

et al. (2000) and Lu (2007).


Volume 22, No. 4, 2008


There were 2 troops of R. roxellana in the region(Li et al. 2001): the free-ranging East Ridge Troop (ERT)and the semi-provisioned (from 2001) West Ridge Troop(WRT). We chose the ERT as our study group becauselogging was conducted within this troop’s range.

Commercial logging was conducted in 3 valleys in thestudy area. The extent of logging varied among thesevalleys from the edge to the center of the ERT homerange. The biggest trees were marked and cut down. InShang-he-gu (S valley), the forest was clearcut, and a 9-km logging road was built. Habitat destruction in S valleywas the heaviest. Both Huang-shi-gou (H valley) and Da-ai-gou (D valley) were selectively logged, but the degreeof logging was higher in H valley (40% of trees >5 cm dbh[diameter breast height] were cut) than in D valley (20%of trees >5 cm dbh were cut) (Zhang 1996).We recordedlocations of the ERT group and collected foraging databefore, during, and after logging. Data were pooled byseason.

Botanical Survey

To understand the change of vegetation compositioncaused by logging, we reanalyzed the data from a sur-vey by Zhang (1996) in the same area before logging andconducted a botanical survey after logging. Sampling ef-fort before logging was higher and covered a bigger areabecause the survey prior to logging was primarily to in-form the logging company. To avoid sampling bias, weused only the data from transects located where postlog-ging vegetation surveys were conducted. The survey areaincluded the range of the ERT before and after logging.

Before logging, at each 50-m elevation interval (n =63), a belt transect of 100 × 10 m was set up on each slopewithin the range of the ERT. The surveyed area covered6.30 ha. Within transects, all trees with >5 cm dbh wereidentified and the diameter was recorded (Zhang 1996).

After logging, in September and October 2002, 75botanical plots of 30 × 15 m covering 3.38 ha were estab-lished systematically on each slope within the range ofthe ERT. Of these 75 plots, 35 were allocated in D, H, andS valleys where commercial logging was conducted. Wesampled the same plots before and after logging. Theother 40 plots were located in primary forest withinthe home range of the ERT. In this temperate moun-tain forest, vegetation types varied with elevation. Weset the vegetation survey plots at 100-m elevation inter-vals from 1600 to 2650 m so as to cover all forest typeswithin the ERT range. All tree species in each plot wereidentified, and trees with >5 cm dbh were measured.Subsequently, we measured the diversity of tree speciesand the diversity of food plants used by the monkeysin these plots with the Shannon–Wiener diversity index(H ′) : H ′ = −∑s

i=l pi log pi , where s is the number ofspecies, and pi is the proportion of the total number ofindividuals represented by the ith species. Tree biomass

was estimated by calculating total basal area (BA); BA =∑(π r2) of trees in a species, genus, or family.

We tested for differences in diversity of tree species,diversity and density of species of food trees, and rangeuse by the ERT in 2002 between logged forest and areasof primary forest with a t test. To understand the changein abundance of food trees before and after logging, wecompared biomass of the main species of food trees be-fore and after logging with a t test.

Diet

We observed monkeys before (1989–1990) and after(2002–2003) logging. We used binoculars (8 × 40) andspotting scopes (25–60×) to watch them from a distanceof 50–200 m, and we recorded the plant species they ate.Monkeys were clearly visible in the crowns of conifertrees, the open crowns of deciduous trees, and when trav-eling on the ground. In the course of tracking the monkeytroops if individuals could be seen clearly, we conductedscan surveys over 5 min at 15-min intervals and recordedthe diet of all visible animals (Altmann 1974). The namesof plant species, plant parts eaten, and the frequencies ofconsumption of each plant species were recorded. Wecategorized food items as leaf, seed, or other (e.g., bark,buds, twigs, lichens). We ranked the top 15 species offood trees on the basis of the time R. roxellana spentfeeding on these species in the 3 study periods.

Range Use

Before, during, and after logging (2002–2003), we fol-lowed the monkeys on foot and monitored movement ofthe troop during an average of 8 h of observation per dayand for at least 10 days/month to ensure a good samplingin different seasons. We recorded the central location ofthe troop on a topographic map (scale 1:25,000) everyhour, or noted when the troop traveled more than 200m. We also recorded the frequency of the ERT enteringeach vegetation plots.

We calculated range use by estimating area coveredby quadrants (250 × 250 m) used by monkeys in a year.Range size was calculated with the following formula:0.25 × 0.25 × the number of quadrants/cos (45◦). Wedesignated the core area (most frequently used cells) asthe activity center of the group and identified locationsof the centers of seasonal ranges before and after log-ging. The effect of logging on range use was measuredas distance between the activity centers before and afterlogging.

Population Size and Composition

The number of individuals in the ERT was counted beforeand after logging under optimal visibility during winterwhen monkeys could be observed clearly among decidu-ous trees. We conducted censuses when the group was


Volume 22, No. 4, 2008

Guo et al. 1059

at rest during the day. We made observations throughbinoculars and spotting scopes at distances varying fromabout 10 to 200 m. We categorized individuals into 4 ageand sex classes (Zhang et al. 2003): adult males, adultfemales, juveniles (1–4 years old), and infants (<1 yearold). Individuals that could not be placed in any of the 4classes were categorized as unidentified.

Results

Change in Forest Composition

Forest composition and structure changed in the loggedarea after logging. The most dominant trees were loggedand replaced by understory species. Some species thatwere present in the understory before logging, such asPrunus kansuensis, Salix kansuensis, and Lindera ob-

tusiloba, grew up after canopy trees were removed andtheir density and basal area increased. The overall densityof trees >5 cm dbh increased, but many of them werestill small after logging (Table 1).

In the selectively logged valleys (H and D), some impor-tant food trees that had previously dominated these val-leys, such as Populus purdomii, Fraxinus mandshurica,

Table 1. The mean density, diameter, and basal area of tree species in 3 logged valleys in the range of R. roxellana before and after logging inShang-he-gu (S), Huang-shi-gou (H), and Da-ai-gou (D).

Total density Overall mean Total basalDominant tree species∗ (stems/ha) diameter (cm) area (cm2/ha)

Before logging (1994)S Valley 715 20.5 300,478.75Populus purdomiiQuercus spp.Pinus armandiiAcer mono

H Valley 940 20.7 402,780.60Quercus spp.Betula albosinensisPop. purdomiiPin. armandii

D Valley 788 18.20 261,017.12Quercus spp.Tilia oliveriPop. purdomiiA. mono

After logging (2002)H Valley 1068 15.06 242,226.26Quercus spp.Abies fargesiiTsuga chinensisCarpinus turczaninowii

D Valley 862 15.65 211,200.54Quercus spp.Platycarya strobilaceaPrunus kansuensisB. albosinensis

∗Dominant trees are the biggest trees with the most coverage and biomass and the greatest height. Forest was clearcut in D Valley.

Tilia oliveri, Betula sp., were logged, although Quercus

spp. still dominated the south-facing slopes in this region.L. obtusiloba individuals grew up and dispersed rapidlyafter the canopy trees were removed. Monkeys ate thenew leaves and twigs of this species in summer. Nev-ertheless, the diameters of their trunks were too small(<5 cm) to be measured during the survey. In S Val-ley, the big trees were clearcut to make a road. Trees inthis area were too small to be measured after logging.The overall mean diameter of food trees was significantlysmaller after logging, but their density increased signif-icantly after selective logging (Table 2). The basal areaof main food trees decreased after logging, except forPru. kansuensis and S. kansuensis, which showed theopposite trend (Table 3). Compared with unlogged areas,diversity and density of trees, including species includedin the diet of the ERT, were significantly higher in theselectively logged area, but the basal area of food treeswas not (Table 4).

Range Use

The ERT used almost the entire area of D, H, and S valleysbefore commercial logging in 1990. The troop had en-tered 227 grid squares when the cumulative number of


Volume 22, No. 4, 2008


Table 2. Changes in total basal area, overall density, and diameter (>5 cm dbh) of food tree species in the range of a R. roxellana group (ERT)before and after logging in the Qinling Mountains.

Before logging After logging Difference (t, df)

Total mean basal area (cm2/ha) (SD) 47,187.47 (36,570.29) 402,45.27 (23324.63) 1.8473,a 59Overall mean density (individuals/ha) (SD) 805.29 (388.48) 930.07 (383.96) 3.1291,b 337Mean diameter (cm) (SD) 19.83 (13.54) 15.33 (10.46) 61.8452,b 1781

ap < 0.05.bp < 0.01.

squares entered approached the asymptote and annualrange size was 20.1 km2. Within 1 year after logging,the troop had entered 207 squares when the cumulativenumber of squares entered approached the asymptoteand the annual range was 18.3 km2. After logging, therange of the ERT extended to the area of West Ridge,and part of it (5.83 km2, 66 squares) overlapped with therange of a neighboring troop, the WRT, in spring 2003(Fig. 1). The distance between the original range centerbefore logging and the new range center (4 in Fig. 1) dur-ing logging was 7.0 km. This range shift is significantlylarger than the average seasonal range shift of 0.97 km(SD = 0.39) (t = − 65.6541; p < 0.01).

Habitat Use

During logging in 1997, areas in the S, H, and D valleyswhere logging was conducted were not visited by theERT. Their annual range size during logging contractedto 15.4 km2. In 2003, 5 years after logging stopped, thetroop returned to D and H valleys. Shang-he-gu Valley,where trees were clearcut and a road was built was notused after logging (Fig. 1). Despite their avoidance of

Table 3. The change of basal area of the top 15 species of food treesof R. roxellana before and after logging in the Qinling Mountains.

Basal area (cm2/ha)

before afterSpecies∗ Family logging logging

Populus purdomii Salicaceae 98,897 10,370Quercus aliena Fagaceae 96,736 70,413Fraxinus Oleaceae 61,911 5,169

mandshuricaTilia oliveri Tiliaceae 44,439 17,658Betula sp. Betulaceae 44,164 5,710Corylus chinensis Fagaceae 22,393 4,119Pinus armandii Pinaceae 15,924 614Acer mono Aceraceae 12,739 3,440Lindera obtusiloba Lauraceae 5,374 2,313Sorbus koehneana Maloideae 4,877 2,797Helwingia japonica Cornaceae 3,583 0Salix spathulifolia Salicaceae 2,677 1,475S. kansuensis Salicaceae 2,229 3,651Prunus kansuensis Prunoideae 806 8,070Viburnum Caprifoliaceae 488 121

betulifolium

∗Ranked according to time spent feeding on each prior to logging.

the S Valley, the mean frequency of visits by the ERT tosample plots was significantly higher in logged areas thanunlogged areas (Table 4).

Diet

The number of tree species used by the ERT as food in-creased after logging. The annual diet survey of the troopbefore logging revealed that it fed on 33 tree species be-longing to 15 families. After logging the ERT fed on 84 treespecies, including trees and shrubs of 29 families. Of the33 species recorded as food before logging, only 18 werestill eaten after logging, but not all of them had been extir-pated. Sixty-six new species of food trees were recordedin their diet after logging. Most of the food items wereplant parts, including leaves, seeds, buds, twigs, bark,and lichens (Ramalina sinensis and Parmelia spp.). Theproportion of these food items in the annual diet did notchange significantly after logging (χ2 = 6.22 × 10−10;p > 0.05).

Population Size and Composition

Before logging, in 1991, 5 population censuses were con-ducted. The largest troop encountered contained 84 in-dividuals, including 10 adult males, 26 adult females, 18juveniles, 6 infants, and 24 unidentified individuals. Adultsex ratio (male vs. female) was 1:3.1, the ratio of adultsto immatures (juveniles and infants) was 1:0.8, and theratio of infants to adult females was 1:4.

After logging, we conducted 8 censuses during 2002.The largest troop contained 112 individuals, including 11adult males, 41 adult females, 14 juveniles, 21 infants, and25 unidentified individuals. The adult sex ratio (male vs.female) was 1:3.7, the ratio of adults to immatures (ju-veniles and infants) was 1:0.7, and the ratio of infants toadult females was 1:2. The population structure was sig-nificantly different before and after logging (χ2 = 7.912;p < 0.05).

Discussion

Habitat Changes

When food becomes scarce, Colobines either rangemore widely and eat a wider diversity of species


Volume 22, No. 4, 2008

Guo et al. 1061

Table 4. Changes in forest structure and habitat use for an R. roxellana group (ERT) in logged and unlogged plots in the Qinling Mountains.a

Unlogged Logged Difference (t, df)b

Overall diversity of tree species, H′(SD) 2.15 (0.83) 2.50 (0.66) 1.9432,∗ 69

Diversity of food tree species, H′(SD) 0.31 (0.22) 0.40 (0.03) 1.9895,∗ 73

Density of food tree species (stems/ha) (SD) 480 (372) 685 (397) 2.1473,∗ 73Basal area of food tree (cm2/ha) (SD) 41,981.11 (25,188.66) 37,841.66 (18920.83) 1.9841, 73Frequency of the ERT entering the plots 1.08 (2.03) 4.18 (4.06) 4.4231,∗∗ 74

(times/plot) (SD)c

aData from 2002 to 2003.b∗p < 0.05; ∗∗p < 0.01.cNumber of times a plot was entered by the troop during the survey period.

(Presbytis melalophos) (Johns 1986a, 1986b) or try toconserve energy by ranging less widely and narrow-ing their diet to basic staples (C. satanas) (McKey &Waterman 1982). Changes in forest composition and pri-mate food resources often occur after logging (Johns1986a; Lovejoy et al. 1986; Estrada & Coates-Estrada1996; Onderdonk & Chapman 2000), but the extentof change depends on the extent of damage and thespecies of plants affected. In West Malaysia loggingcaused the food resources to became more clumped,and P. melalophos groups were forced to range morewidely than before (Johns 1986a). In the Yuhuangmiaoregion trees preferred by the ERT as food before log-ging decreased in biomass and abundance after logging,and several species including Helwingia japonica disap-peared. To compensate for such loss, the ERT feed onmore varieties of plants that were distributed widely onboth the east and west ridges. New food trees also grewup after logging, and the overall diversity and density offood trees increased in selectively logged areas. Thesechanges influenced the ranging pattern, habitat use, diet,and population size of the ERT.

Range Shifts

Range shifts could be seasonal or could result from log-ging. The range pattern of R. roxellana showed signif-icant seasonality, and the species seasonal ranges over-lapped with each other. Ordinarily (i.e., in habitat thathas not been logged), they do not leave the core area oftheir range except in spring, when they follow widelydistributed foods (Li et al. 2000; Tan et al. 2007). Theshift in home range by the ERT following logging was asignificantly larger shift than the seasonal range shift. Likemany primates, the immediate response of R. roxellana

was to escape from the disturbance of logging. The groupmoved out of their original home range during logging.Nevertheless, the ERT returned to some areas of its orig-inal home range after logging. This indicates that theyhave the ability to explore new areas and can toleratesome degree of habitat disturbance.

After logging the ERT expanded their home range fromEast Ridge to West Ridge and into the home range ofthe neighboring troop (WRT). This shift in home range

was mainly due to the change in food distribution anddecrease of food resources in their original range. TheERT also lost part of its original range area in S valleyto clearcutting, which forced the troop to shift and ex-pand its home range to overlap with that of the neigh-boring group. The size of their range after logging alsodecreased. Such changes resulted in an increase of pop-ulation density in this region, which could in turn resultin an increase in the chance of conflict between the ERTand neighboring groups; a decrease in foraging efficiencybecause of their unfamiliarity with the area; an increasein the chance of predation (Marsh et al. 1987; Gebo &Chapman 1995; Li 2004;) and an increase in the chanceof contracting a disease (Gillespie & Chapman 2006).

Habitat Use and Dietary Adaptation to Habitat Change

In response to logging R. roxellana showed significanthabitat preference. The ERT preferred selectively loggedareas to unlogged and clearcut areas. This may be becausethe density and diversity of food trees were higher in se-lectively logged areas. Nevertheless, 2 groups of R. roxel-

lana in Shennongjia Reserve showed a different habitatpreference (Li 2004): they preferred unlogged forest tolightly logged forest. This was because unlogged forestwas rich in lichen, which was an especially importantfood for these groups. Although R. roxellana in the Qin-ling Mountains also feed on lichen, it is not as importantas it is for monkeys in Shennongjia (Guo et al. 2007).

The ERT avoided clearcut areas, even though therewas no hunting in the area. This was probably due tolack of food resources and the change of forest struc-ture. An immediate effect of clearcutting was the lossof food trees, and R. roxellana initially allocated partof their time to feeding on seeds on the ground (Renet al. 2001; Guo et al. 2007), a behavior also exhibitedby other primate species exposed to deteriorating habi-tat in temperate forests (Wada 1984; Hanya 2004). Theunderstory vegetation may initially supply some food forthe ERT, but the understory shrubs and grass grew uprapidly without the tree canopy and became very dense.The ERT was no longer sighted in this valley, perhaps be-cause of the dense understory and loss of canopy trees.Lack of canopy increases the risk of predation from both


Volume 22, No. 4, 2008


aerial (e.g., eagles) and ground predators (e.g., snakes).R. roxellana in Shennongjia also avoided the clear-cutareas (Li 2004). Therefore, clearcutting not only affectedtheir food supply, but also destroyed the structure of themicrohabitat for R. roxellana and caused total loss ofhabitat.

Flexibility of diet is generally thought to influencethe ability of primate species to live in forest fragments(Struhsaker & Oates 1975; Estrada & Coates-Estrada 1996;Onderdonk & Chapman 2000), although some colobineshave relatively monotonous diets and yet adapt well tologging and forest fragmentation (Johns & Skorupa 1987;Onderdonk & Chapman 2000). R. roxellana can shifttheir diet seasonally (Guo et al. 2007) and even occasion-ally the composition of their diet to cope with changes infood availability caused by climate change in the QinlingMountains (Li et al. 2003). This dietary flexibility mighthave been the key to their survival of logging and is likelyto be important for their long-term survival in the QinlingMountains.

Population Responses to Logging

Because leaf material is the main component of theirdiet, the population size of colobines is directly corre-lated with the biomass of food trees (Davies 1994). Log-ging may cause population decline because of reductionin food sources (Plumptre & Reynolds 1994; Ganzhorn1995; Chapman et al. 2000). For example, heavy loggingcaused a 49% decline of overall trees and a 42% reduc-tion in basal area of food trees that supplied more than80% of the red colobus’ diet in Kibale Forest, Uganda. Asa result, the population of red colobus declined by 38%(Skorupa 1986). Nevertheless, logging in the Yuhuang-miao region did not lead to a decline in the population ofthe ERT; rather, their population increased slightly. Com-pared with the ERT population, the population size ofthe neighboring group (WRT), whose habitat was neverlogged, was stable over the same period: 90 individu-als in 1990, 93–95 in 1998 (Ren 1999), and about 90in 2001 (B. G. Li & K. Wada, unpublished data). There-fore, the small increase in population size of the ERT wasmore likely a response to the increase in food diversity re-sulting from secondary growth created through selectivelogging. Other researchers who study colobine monkeys(P. melalophos and P. obscura) also report increases inpopulation size after logging because of increases in foodresources (Johns 1986a).

Reduction in food availability can lead to increased in-fant and juvenile mortality in primates (Struhsaker 1976;Altmann et al. 1977; Dittus 1977; Skorupa 1988). In ourstudy the ratio of adults to immature individuals was sta-ble several years after logging.

Although our results agree with the results of someprevious studies in that an increase in food diversity dueto selective logging resulted in increased group sizes of

some colobine monkeys, longer-term study is essentialfor determining whether such an increase in populationsize is maintained over the long term. Longer-term studyis also required to investigate whether such an increasein a local primate groups where logging is carried out isbeneficial for the primate populations in the region.

Conservation Implications

The immediate and long-term response of monkeys toforest disturbance is a key issue for their conservationbecause the response will influence policy making foranimal protection and sustainable use of forest. TheR. roxellana in the Qinling Mountains were able to ad-just their foraging behavior to adapt to habitat change.Nevertheless, adjustment is not unlimited, and they can-not live in heavily disturbed areas such as the S Valleyafter clearcutting. R. roxellana exhibit little or no rangeoverlap between neighboring troops naturally (Hu et al.1980; Kirkpatrick 1998; Li et al. 1999) or before andduring logging (Li et al. 1999, 2000, 2001). The rangeoverlap between the ERT and WRT after logging and thedecrease of the home range of the ERT led to a higherdensity of R. roxellana in this region (Li et al. 2000;Tan et al. 2007) as a result of compression of groupsinto smaller area. Although no negative effects of higherdensities have yet been observed, crowding can poten-tially limit the growth of colobine populations (Davies1994). For example, the population of C. guereza at Boledeclined when a large group was compressed within asmall territory because aggression between group mem-bers often interfered with reproduction (Dunbar 1987).This could be a potential threat for R. roxellana in theQinling Mountains, especially because the carrying ca-pacity of this forest for R. roxellana is still unknown.

Although our results, as well as results from some pre-vious studies, show some short-term positive responsesof primate to selective logging, such as increased groupsizes, it is premature to conclude that selective loggingmay be carried out in the habitat of this primate becausethe immediate response of primates to logging may notpredict the long-term impact of such disturbances on pri-mate populations (Chapman et al. 2000). It is importantthat longer-term effect of selective logging on primatepopulations and on other animals in the area be investi-gated before a conclusion is made. Information on otherfactors related to logging, such as infant mortality, dis-ease, and predation, is urgently needed. Monitoring ofthe ERT will be continued in Qinling Mountains.

Besides commercial logging, other less destructive hu-man activities, such as collection of medicinal plants andfirewood by local people and grazing of livestock, alsomay have caused R. roxellana groups to move awayfrom the area (Li et al. 1999). These activities can alsolead to forest fragmentation (Chapman et al. 2003) andshould be limited in monkey habitat, especially within


Volume 22, No. 4, 2008

Guo et al. 1063

the reserve. More research is needed to investigate thelong-term effects of such human activities and the abil-ity of R. roxellana to adapt to varying degrees of humandisturbance.

Acknowledgments

We thank Zhouzhi National Nature Reserve for permis-sion to conduct this study. We are most grateful to Q.Wang and Y. Liang for their arduous efforts in trackingthe monkey troop. We thank P. Zhang, X. G. Qi, H. P.Wang, and a number of Northwest University studentsfor their assistance in the field work. We also thank Q. K.Zhao, M. E. Rogers, K. Wada, C. L. Tan, and S. Huo for pro-viding helpful comments on the manuscript. We thank2 reviewers for their invaluable comments on the earlierdrafts. This study was supported by National Basic Re-search Program of China (973 Program; 2007CB411600)and the National Nature Science Foundation of China(30630016, 30570312, and 30770375), Primate Conser-vation, Inc., COSMO Oil Eco Card Foundation, and theZoological Society of San Diego.

Literature Cited

Altmann, J. 1974. Observational study of behavior: sampling methods.Behaviour 49:227–267.

Altmann, J., S. A. Altmann, G. Hausfater, and S. A. McCluskey. 1977.Life history of yellow baboons: physical development, reproductiveparameters, and infant mortality. Primates 18:315–330.

Chapman, C. A., M. J. Lawes, L. Naughton-Treves, and T. Gillespie.2003. Primate survival in community-owned forest fragments: aremetapopulation models useful amidst intensive use? Pages 63–78 inL. K. Marsh, editor. Primates in fragments: ecology and conservation.Kluwer Academic/Plenum Publisher, New York.

Chapman, C. A., S. R. Balcomb, T. R. Gillespie, J. P. Skorupa, and T. T.Struhsaker. 2000. Long-term effects of logging on primates in KibaleNational Park, Uganda: a 28-year comparison. Conservation Biology14:207–217.

Davies, A. G. 1994. Colobine populations. Pages 286–310 in A. G.Davies and J. F. Oates, editors. Colobine monkeys: their ecology,behaviour and evolution. Cambridge University Press, Cambridge,United Kingdom.

Ding, W., and Q. K. Zhao. 2004. Rhinopithecus bieti at Tacheng, Yun-nan: diet and daytime activities. International Journal of Primatology25:583–598.

Dittus, W. P. J. 1977. The social regulation of population density andage-sex distribution in the toque monkey. Behaviour 63:281–322.

Dunbar, R. I. M. 1987. Habitat quality, population dynamics, and groupcomposition in colobus monkeys (Colobus guereza). InternationalJournal of Primatology 8:299–330.

Estrada, A., and R. Coates-Estrada. 1996. Tropical rain forest fragmen-tation and wild populations of primates at Los Tuxtlas, Mexico.International Journal of Primatology 17:759–783.

Ganzhorn, J. U. 1995. Low-level forest disturbance effects on pri-mary production, leaf chemistry, and lemur populations. Ecology76:2048–2096.

Gebo, D. L., and C. A. Chapman. 1995. Habitat, annual, and seasonaleffects on positional behavior in red colobus monkeys. AmericanJournal of Physical Anthropology 96:73–2.

Gilbert, K. A. 2003. Primates and fragmentation of the Amazonforest. Pages 145–157 in L. K. Marsh, editor. Primates in frag-ments: ecology and conservation. Kluwer Academic/Plenum,New York.

Gillespie, T. R., and C. A. Chapman. 2006. Prediction of parasite infec-tion dynamics in primate metapopulations based on attributes offorest fragmentation. Conservation Biology 20:441–448.

Guo, S. T., B. G. Li, and K. Watanabe. 2007. Diet and activity budget ofRhinopithecus roxellana in the Qinling Mountains, China. Primates48:268–276.

Hanya, G. 2004. Diet of a Japanese macaque troop in the coniferousforest of Yakushima. International Journal of Primatology 25:55–71.

Hu, J. 1998. Rhinopithecus roxellanae. Pages 65–68 in S. Wang, editor.China red data book of endangered animals: mammalian. SciencePress, Beijing.

Hu, J. C., Q. X. Deng, Z. W. Yu, S. D. Zhou, and Z. X. Tian. 1980.Research on the ecology and biology of the giant panda, goldenmonkey, and other rare animals. Journal of Nanchong Teacher’sUniversity 2:1–29 (in Chinese).

Jeffrey, S. M. 1978. The effects of logging on rainforest fauna of WestAfrica and East Kaliman-tan. Pages 393–400 in Proceedings of theEighth World Forestry Congress. Volume VII. U.N. Food and Agri-culture Organization, Rome.

Johns, A. D. 1986a. Effects of selective logging on the behavioral ecol-ogy of West Malaysian primates. Ecology 67:684–694.

Johns, A. D. 1986b. The effects of commercial logging on a WestMalaysian primate community. Pages 206–209 in D. M. Taub andF. A. King, editors. Current perspectives in primate social dynam-ics. Van Nostrand Reinhold, New York.

Johns, A. D. 1988. Effects of selective timber extraction on rain foreststructure and composition and some consequences for frugivoresand folivores. Biotropiea 20:31–36.

Johns, A. D., and J. P. Skorupa. 1987. Responses of rain forest primates tohabitat disturbance: a review. International Journal of Primatology8:157–191.

Kirkpatrick, R. C. 1998. Ecology and behavior in snub-nosed monkeyand douc langurs. Pages 155–190 in N. G. Jablonski, editor. The nat-ural history of the doucs and snub-nosed monkeys. World ScientificPublishing, Singapore.

Kirkpatrick, R. C., Y. C. Long, T. Zhong, and L. Xiao. 1998. Social orga-nization and range use in the Yunnan snub-nosed monkey Rhinop-

ithecus bieti. International Journal of Primatology 19:13–51.Li, B. G., P. Zhang, K. Watanabe, C. L. Tan, F. Fukuda, and K. Wada. 2003.

A dietary shift in Sichuan snub-nosed monkeys. Acta TheriologicaSinica 23:358–360.

Li, B. G., B. P. Ren, and Y. F. Gao. 1999. A change in the summer homerange of Sichuan snub-nosed monkeys in Yuhuangmiao, QinlingMountains. Folia Primatologica 70:269–273.

Li, B. G., C. Chen, W. H. Ji, and B. P. Ren. 2000. Seasonal homerange changes of the Sichuan snub-nosed monkey (Rhinopithecus

roxellana) in the Qinling Mountains of China. Folia Primatologica71:375–386.

Li, B. G., P. J. He, X. Z. Yang, W. K. Wei, B. P. Ren, J. Y. Yang, K. C. Si,and Y. P. Liu. 2001. The present status of the Sichuan snub-nosedmonkey in the Qinling Mountains of China, and a proposed con-servation strategy for the species. Biosphere Conservation 3:107–114.

Li, Y. M. 2001. The seasonal diet of the Sichuan snub-nosed monkey(Pygathrix roxellana) in Shennongjia Nature Reserve, China. FoliaPrimatologica 72:40–43.

Li, Y. M. 2004. The effect of forest clear-cutting on habitat use in Sichuansnub-nosed monkey (Rhinopithecus roxellana) in Shennongjia Na-ture Reserve, China. Primates 45:69–72.

Lovejoy, T. E., et al. 1986. Edge and other effects of isolation on Amazonforest fragments. Pages 257–285 in M. E. Soule, editor. Conservationbiology: the science of scarcity and diversity. Sinauer Associates,Sunderland, Massachusetts.


Volume 22, No. 4, 2008


Lu, J. Q. 2007. Daily activity budget and social structure of Rhino-

pithecus roxellana in Qinling Mountains. PhD thesis. NorthwestUniversity, Xi’an (in Chinese).

McKey, D., and P. G. Waterman. 1982. Ranging behaviour of a groupof black Colobus (Colobus satanas) in the Douala-Edea Reserve,Cameroon. Folia Primatologica 39:264–304.

Marsh, C. W., A. D. Johns, and J. M. Ayres. 1987. Effects of habitatdisturbance on rain forest primates. Pages 83–107 in C. W. Marshand R. A. Mittermeier, editors. Primate conservation in the tropicalrain forest. Alan R Liss, New York.

Nakagawa, N. 1989. Feeding strategies of Japanese monkeys againstdeterioration of habitat quality. Primates 30:1–16.

Onderdonk, D. A., and C. A. Chapman. 2000. Coping with forest frag-mentation: the primates of Kibale National Park, Uganda. Interna-tional Journal of Primatology 21:157–191.

Plumptre, A. J. 1996. Changes following 60 years of selective timberharvesting in the Budongo Forest Reserve, Uganda. Forest Ecologyand Management 89:101–113.

Plumptre, A. J., and V. Reynolds. 1994. The effect of selective loggingon the primate populations in the Budongo Forest Reserve, Uganda.Journal of Applied Ecology 31:631–641.

Ren, B. P. 1999. Spatial distribution patterns in a group of the Sichuansnub-nosed monkeys (Rhinopithecus roxellana) in Qinling Region.MS thesis. Northwest University, Xi’an (in Chinese).

Ren, B. P., S. Y. Zhang, L. X. Wang, B. Liang, and B. G. Li. 2001. Ver-tical distribution of different age-sex classes in a foraging group ofSichuan golden monkeys (Rhinopithecus roxellana). Folia Primato-logica 72:96–99.

Skorupa, J. P. 1986. Responses of rainforest primates to selective loggingin Kibale Forest, Uganda: a summary report. Pages 57–70 in K.Benirschke, editor. Primates, the road to self-sustaining populations.Springer-Verlag, New York.

Skorupa, J. P. 1988. The effect of selective timber harvesting on rainfor-est primates in Kibale Forest, Uganda. PhD dissertation. Universityof California, Davis, California.

Struhsaker, T. T. 1976. A further decline in numbers of Amboseli vervetmonkeys. Biotropica 8:211–214.

Struhsaker, T. T., and J. F. Oates. 1975. Comparison of the be-havior and ecology of red colobus and black-and-white colobusmonkeys in Uganda: a summary. Pages 103–123 in R. H. Tuttle,editor. Socio-ecology and psychology of primates. Mouton, TheHague.

Tan, C. L., S. T. Guo, and B. G. Li. 2007. Ranging behav-ior and population structure of Sichuan snub-nosed monkeys(Rhinopithecus roxellana) in Zhouzhi National Nature Reserve,Shaanxi, China. International Journal of Primatology 28:577–591.

Wada, K. 1984. Ecological adaptation in Rhesus monkeys at the Ku-maon Himalaya. The Journal of the Bombay Natural History Society80:469–498.

Xiao, W., W. Ding, L. W. Cui, R. L. Zhou, and Q. K. Zhao. 2003. Habitatdegradation of Rhinopithecus bieti in Yunnan, China. InternationalJournal of Primatology 24:389–398.

Yamagiwa, J., N. Mwanza, T. Yumoto, and T. Maruhashi. 1994. Seasonalchange in the composition of the diet of eastern lowland gorillas.Primates 35:1–14.

Zhang, P., B. G. Li, K. Wada, C. L. Tan, and K. Watanabe. 2003. Socialstructure of a group of Sichuan snub-nosed monkeys (Rhinopithecus

roxellana) in the Qinling Mountains of China. Acta Zoologica Sinica49:727–735 (in Chinese).

Zhang, P. C., G. F. Shao, G. Zhao, D. C. Le Master, G. R. Parker, J. B.Dunning Jr., and Q. L. Li. 2000. China’s forest policy for the 21stcentury. Science 288:2135–2136.

Zhang P., K. Watanabe, B. G. Li, and C. L. Tan. 2006. Social organizationof Sichuan snub-nosed monkeys (Rhinopithecus roxellana) in theQinling mountains, central China. Primates 47:374–382.

Zhang, T. 1996. Reports of vegetation survey of logging forest in Shang-he-gu, Huang-shi-go, Da-ai-gou for the design of commercial loggingof the Xiao-wang-jian logging company. Louguan Forest Bureau,Xi’an (in Chinese).


Volume 22, No. 4, 2008