Acta Theriologica 53 (4): 365–374, 2008. PL ISSN 0001–7051
Habitat selection and the change in distribution of Asian elephants in Mengyang Protected Area, Yunnan, China Liu LIN, Limin FENG, Wenjing PAN, Xianming GUO, Jianwei ZHAO, Aidong LUO and Li ZHANG*
Lin L., Feng L., Pan W., Guo X., Zhao J., Luo A. and Zhang L. 2008. Habitat selection and the change in distribution of Asian elephants in Mengyang Protected Area, Yunnan, China. Acta Theriologica 53: 365–374. Elephants were confined to Mengyang Protected Area in China and their distribution range had reduced greatly compared to past records. A preliminary study of habitat selection by Asian elephants Elephas maximus Linnaeus, 1758 and their distribution was conducted in Mengyang Protected Area and its surrounds using site visits and transect surveys from July 2003 to December 2006. Although no variable significantly influenced their habitat selection, elephants still showed preference for altitudes between 900 and 1200 m, gradients < 30°, and orientations to the south-east, south and south-west. Human activities, including habitat transformation and degradation, disturbance by large infrastructure and poaching were considered to be the main factors inducing elephant distribution changes. Key Laboratory for Biodiversity Science and Ecological Engineering, Ministry of Education, College of Life Science, Beijing Normal University, Beijing, 100875, China, e-mail:
[email protected] (LL, LF, WP, LZ); Xishuangbanna National Nature Reserve Bureaus, Jinghong, Yunnan, 666100, China (XG, JZ, AL)
Key words: Elephas maximus, logistic regression, human-wildlife conflict, transformation, poaching, habitat alteration, conservation, road ecology
(80%) are confined to Mengyang, Mengla and Shangyong protected areas in the Xishuangbanna National Nature Reserve, and its surrounds (Zhang et al. 2006). Mengyang Protected Area has China’s largest wild elephant population (80–100). Knowledge of animal habitat selection patterns can help avoid needless waste of conservation resources and help prioritize areas for conservation efforts by targeting those habitats
Introduction The Asian elephant Elephas maximus Linnaeus, 1758 is ranked in the Red Data Book as a first class national protected animal in China and as an endangered species by the IUCN (Santiapillai and Jackson 1990, Wang 1998). It is estimated that there are 165–213 wild elephants left in China, and about 136–179
* corresponding author [365]
366
L. Lin et al.
that are most important for species, particularly when incorporated into a conservation landscape that would support viable populations (Li et al. 1999, Ostro et al. 2000, Pradhan and Wegge 2007). Elephant habitat selection, home range and migration patterns can be influenced by geographical variation in vegetation cover, topography characteristics (Codron et al. 2006), and human activities (Ishwaran 1993, Wu et al. 1999). Three surveys of wild elephant distribution in Xishuangbanna have been carried out since the 1970s. The first survey indicated that wild elephants occupied a wide, contiguous area (south to 22°35’N) around Jinghong city and Mengla county (Fig. 1; Research Group 1, Institute of Zoology of Yunnan 1976). At that time, according to the Survey Group of Xishuangbanna Nature Reserve (1985), the elephants were also found in a large number of areas in Jinghong city, Mengla county; and Mengyang Protected Area, where elephants were found at sites scattered inside the protected area and its surrounding areas. However, surveys during 1996–1997 showed that elephants were found only in Mengyang and Shangyong Protected Areas, while they had been extirpated from Mengla Protected Area (Wu et al. 1999). No detailed distribution maps were created following these three surveys due to a lack of authentic field data. Zhang et al. (2006) created a map of elephant distribution in China, yet the distribution in each specific area was coarse as the map was drawn at a large scale. Elephant distribution can be greatly affected by human activities, such as farmland encroachment, logging, infrastructure and illegal poaching (Choudhury 1999, Blom et al. 2004, Buij et al. 2007). In recent years, natural forests in Xishuangbanna, especially tropical rain forest, have conspicuously decreased and been replaced by rubber plantations due to the high price of rubber (Zhu 2008). The spreading farmland had greatly increased the heterogeneity of natural landscapes in Mengyang (Wu et al. 1997). Therefore, it is likely that habitat transformation and reduction has influenced elephant distribution. Habitat selection by Asian elephants has been studied in several range states in Asia (Sukumar
1989, Ishwaran 1993, Zhang and Wang 2003, Feng and Zhang 2005, Pradhan and Wegge 2007). The aims of the present research were to determine what factors affect elephant habitat selection and to provide an updated, fine-scale map of elephant distribution in Mengyang. The results of our study could be used to assess the status of wild elephants and their associated requirements in elephant conservation strategies.
Study area Mengyang Protected Area is one of five sub-reserves in the Xishuangbanna National Nature Reserve. It is located on the banks of the Mekong River to the north of Jinghong City (100°36’45”~101°16’50”E, 22°6’57”~22°23’30”N), and 2 covers 997.6 km (Fig. 1). The Simao-Xiaomengyang Highway and the parallel 213 National Road separate the reserve into two parts: the eastern and western part (Fig. 1). Mengyang Protected Area has a subtropical monsoonal climate and the annual mean temperature is between 16.4 and 22.0°C. The eastern region of the protected area is wet with annual rainfall > 1600 mm, while the western region is drier with annual rainfall < 1200 mm. The area belongs to the upper section of Mekong River system. The Mekong River runs along the west, the Xiaoheijiang River lies in the east, and six large tributaries run through the middle, including the Mengyang River. The elevation of this mountainous area ranges from 550 to 1600 m above sea level, with most mountains ranging in elevation between 900 and 1200 m and covering 62.6% of the total area. Monsoonal evergreen broad-leaved forest contributes the largest proportion (69.21%) of vegetation cover. Tropical rain forest is found mainly in lower valleys. Tropical bamboo forest is usually found on hillsides, especially in drier areas in the west and humid conifer forest occupies the northeast part of the protected area (Ou et al. 1997).
Material and methods Site visits Since July 2003, field studies have been carried out in different parts of Mengyang Protected Area and its adjacent areas to determine the distribution and status of wild elephants. During field studies, the presence or absence of elephants was ascertained by direct observation, indirect evidence such as footprints and dung, and by interviewing local forest department staff, villagers and hunters. For direct observation and indirect evidence, walked transects along existing and newly cut paths and trails, and vehicle transects along roads and accessible tracks were made. Records prior to 2003 were simultaneously collected through interviews and former survey reports. Position information was obtained by GPS (Garmin 72) and marked on map us-
Asian elephants in Mengyang, China
10
0
101 oE
20 km
Dahuangtian
4
5
Dawotang Diaojing
Dadugang
o
22 20'N
367
2 3
Puwen River
Mekong River
Zhongtianba Mengyang River
1
o 22 10'N
Xiaoheijiang River
Longgu
6
Xinlongshan
Mengyang protected area Distribution range in 2006 Distribution range in 2005 Distribution range in 1997
China Mengyang
Village Highway
Jinghong
River Special area
Xishuangbanna
Fig. 1. The distribution ranges of wild Asian elephants in Mengyang Protected Area in 1997, 2005 and 2006. No. 1–6: areas where changes happened that are referred to in the text.
ing Arcview GIS 3.2. The distribution maps of 2005 and 2006 were based on our field survey. The rough map of 1997 was made according to the report of Wu et al. (1999) and our local interviews, whereas no attempt was made to create distribution maps from the 1970s and 1980s. This work ended in December 2006, by which time more than 70 villages had been visited, covering a total trip length of about 2000 km.
Transect survey Thirty transects were driven. An area 20 m wide and 10 to 50 km long was monitored, ensuring that almost all vegetation types were traversed. Plots (20 ´ 20 m) were sampled at intervals of 2 km along the transects. Within the plots, variables such as vegetation type, elevation, location, human
disturbance, slope, orientation, as well as presence evidence of fresh elephant activities, such as footprints, dung piles, grubbing and foraging were determined (following Feng and Zhang 2005). Vegetation was assigned to six types: evergreen broad-leaved forest, tropical rain forest, humid conifer forest, tropical bamboo forest, shrub and farmland. Elevation was assigned to one of three categories: £ 900 m, 901–1200 m, or > 1200 m. Location was assigned to one of three categories: up (mountaintop or upper hillside), middle (middle hillside), or low (valley or lower hillside). Human disturbance was recorded as follows: low (distance to human settlements > 2 km, few human activities), intermediate (distance: 500 m – 2 km, near a footpath or a secondary road), or high (< 500 m to farmlands, settlements or major roads). Slope was measured and assigned to one of four categories: 0–10°, 11–20°, 21–30°, and > 30°. Orientation was assigned to the following categories: North, Northeast,
L. Lin et al.
368
East, Southeast, South, Southwest, West and Northwest. Transect surveys were conducted from December 2004 to January 2005. A total length of about 900 km was covered and 396 plots were sampled. This work also assisted in determining the distribution range by supplying location information.
Data analysis Because the data comprised multiple categorical variables (habitat characteristics) and a binary response (presence/absence), logistic regression modeling was selected as a suitable statistical technique. This technique is widely used for analyzing multivariate data involving a binary response variable (Li et al. 1999, Greenwood et al. 2002, Buij et al. 2007, Hayward et al. 2007). Prior to logistic regression, a chi-square-test for those variables was carried out to find out whether elephants actually discriminate among sites based on the above environmental variables and only those variables having significant difference (p < 0.05) were selected in the next analysis. In addition, to avoid violations caused by variable interactions, principal component analysis (PCA) was used to summarize covarying explanatory variables after the chi-square-test. Principle components (PCs) and the constant were included as independent variables, while presence/absence was used as the dependent variable in the logistic regression model using an enter procedure. All analysis above was done by SPSS 13.0 for MS Windows. The Vanderploeg and Scavia selectivity index was used to assess single character preference (Vanderploeg and Scavia 1979). Selectivity coefficient Wi and selectivity index Ei were defined as: Wi = (ri/pi)/ å(ri/pi) Ei = (Wi – 1/n) / (Wi + 1/n) where: i – class of certain variable, n – class number, ri – the proportion of sample units used by animals in the ith class to total sample units used by animals in the environment, pi – the proportion of sample units in the ith class to total sample units in the environment. Wi is bound between –1 and 1, Ei categories are defined as follows: Ei > 0.1: preferred; Ei < – 0.1: avoided; –0.1 < Ei < 0.1: selected randomly. Excel 2000 was employed to do the analysis.
listed in Table 1. Only elevation and orientation had significant variance and were selected for the PCA. Results of the PCA showed that the first component (PC1) explained 56.75% of the variance and increased in value with elevation, but decreased with orientation. PC2 increased with both of them (Table 2). In the logistic regression model, two coefficients –0.070 and 0.156 responding to the sampled data PC1 and PC2 and a constant –0.256 resulted, whereas the prediction accuracy of the model was only 57.5%. Accordingly, it was not a fitted model and no variable could be chosen as a significant indicator of elephant’s habitat selection. Habitat characteristics associated with the presence of elephants
A great range of sites with a diversity of physical and biotic characteristics had evidence of elephant presence (Table 3). Tropical bamboo forest was most preferred, followed by humid conifer forest, evergreen broad-leaved forest and tropical rain forest, while farmland and shrub were avoided. Elevations were selected randomly, while areas with elevations between 901 and 1200 m were most preferred, followed by those below 900 m and those above 1200 m. Most locations were selected randomly, however those on the upper areas of mountains were most preferred. Areas with low levels of human disturbance were most preferred. Areas with slopes of 11–20° were preferred to those over 30°, while other slopes were selected randomly. Orientations towards the south-east, south, and south-west were preferred to those towards north-east, east and west, whereas orientations towards north-west and north were selected randomly.
Results Distribution range of elephants in Mengyang Analysis of habitat variables
Of the total 396 plots, 173 (43.7%) had evidence of elephant activities. The frequency distribution and chi-square values with significant level of 0.05 for the independent variables are
Wild elephants have been confined to the Mengyang Protected Area and its surrounding areas since 1997 (Fig. 1). This distribution has shrunk greatly since the 1970s (Research Group 1, Institute of Zoology of Yunnan 1976) and
Asian elephants in Mengyang, China
369
Table 1. Distribution frequency and chi-square-test of variables for habitat selection. Frequency Variables
Categories
Results Present
Absent
97 13 6 28 9 20
116 17 6 24 19 41
c = 7.14, df = 5, p > 0.05
55 100 18
89 100 34
c = 6.75, df = 2, p < 0.05
2
Habitat type
Evergreen, broad-leaved forest Tropical rain forest Humid conifer forest Tropical bamboo forest Shrub Farmland
Elevation (m)
£ 900 901–1200 > 1200
Location
Up Middle Low
63 26 84
67 40 116
c = 1.93, df = 2, p > 0.05
Human disturbance
Low Intermediate High
98 37 38
108 47 68
c = 3.92, df = 2, p > 0.05
Slope (°)
0–10 11–20 21–30 > 30
53 72 44 4
73 77 64 9
c = 2.7, df = 3, p > 0.05
Orientation
North Northeast East Southeast South Southwest West Northwest
25 11 10 35 38 28 5 21
33 30 38 25 28 26 20 23
c = 33.02, df = 7, p < 0.05
Table 2. Principal component analysis (PCA) of habitat variables. Component Variables PC1 Elevation Orientation Total eigenvalues Cumulative (%)
PC2
0.753
0.658
–0.753
0.658
1.135 56.747
0.865 100.00
2
2
2
2
2
1980s (Survey Group of Xishuangbanna Nature Reserve 1985) when elephants occurred in a contiguous area from Mengyang to Shangyong. Elephant range in 1997 covered most areas of that in 2005 and 2006, except Area 4 where elephants were first found in 1998 (Fig. 1). It was believed that elephants migrated from further south into the southern part of Area 4 and their range had expanded to Dahuangtian village by 2006 (Fig. 1). In contrast, in Area 3, where elephants were found frequently before 1998, al-
L. Lin et al.
370
Table 3. Character preference by wild Asian elephants in Mengyang Protected Area and its surrounding areas. Y – the number of plots having evidence of elephant activities in the ith class, T –the total number of plots in the ith class, P – preferred, NP – not preferred, R – randomly selected, EBF – Evergreen broad-leaved forest. ri – the proportion of sample units used by animals in the ith class to total sample units used by animals in the environment, pi – the proportion of sample units in the environment, Wi – selectivity coefficient, Ei – selectivity index. Characters
Rank (i)
Y
T
ri
pi
Wi
Ei
Use
97 13 6 28 9 20
213 30 12 52 28 61
0.5607 0.0751 0.0347 0.1618 0.052 0.1156
0.5379 0.0758 0.0303 0.1313 0.0707 0.154
0.1768 0.1680 0.1942 0.2090 0.1247 0.1273
0.0294 0.0040 0.0763 0.1126 –0.1439 –0.1339
R R R P NP NP
55 100 18
144 200 52
0.3179 0.578 0.104
0.3636 0.5051 0.1313
0.3111 0.4071 0.2818
–0.0346 0.0997 –0.0838
R R R
Habitat type
EBF Tropical rain forest Humid conifer forest Tropical bamboo forest Shrub Farmland
Elevation (m)
£ 900 901–1200 > 1200
Location
Up Middle Low
63 26 84
130 66 200
0.3642 0.1503 0.4855
0.3283 0.1667 0.5051
0.3732 0.3034 0.3234
0.0565 –0.0471 –0.0151
R R R
Human disturbance
Low Intermediate High
98 37 38
206 84 106
0.5665 0.2139 0.2197
0.5202 0.2121 0.2677
0.3732 0.3456 0.2812
0.0564 0.0180 –0.0848
R R R
Slope (°)
0–10 11–20 21–30 > 30
53 72 44 4
126 149 108 13
0.3064 0.4162 0.2543 0.0231
0.3182 0.3763 0.2727 0.0383
0.2671 0.3068 0.2587 0.1673
0.0331 0.1021 0.0171 –0.1981
R P R NP
Orientation
North Northeast East Southeast South Southwest West Northwest
25 11 10 35 38 28 5 21
58 41 48 60 66 54 25 44
0.1445 0.0636 0.0578 0.2023 0.2197 0.1618 0.0289 0.1214
0.1465 0.1035 0.1212 0.1515 0.1667 0.1364 0.0631 0.1111
0.1321 0.0823 0.0639 0.1788 0.1765 0.1588 0.0613 0.1463
0.0275 –0.2061 –0.3237 0.1771 0.1708 0.1192 –0.3417 0.0786
R NP NP P P P NP R
most no sign of them has been found since 1998. In Area 1, elephants disappeared in the early 1990s. In Area 2, Area 5 and Area 6, elephants disappeared after 2005.
Discussion Habitat selection by elephants
According to the chi-square-test and logistic regression model, no variable regression signifi-
cantly influenced elephant habitat selection. Elephants in this area were found in areas with an elevation of 1500 m or with slopes exceeding 50°, and they even risked entering villages for food occasionally. Similar discoveries have been found in Simao and Shangyong (Zhang et al. 2003, Feng and Zhang 2005). Diet composition and foraging ecology of Asian elephant were studied in Shangyong and their diet was composed of 106 plant species, which were distributed in diverse vegetation habitats (Chen et al. 2006). Therefore, as an intelligent species with high-levels of
Asian elephants in Mengyang, China
tolerance of human disturbance and as a generalist feeder, elephants can acclimatize themselves to various habitat types. Although they did not select habitat significantly, elephants still showed preference to certain habitat characteristics. Preference for different habitats can enhance an animal’s survival and reproduction rates, and it is also a reflection of the balance between physiological requirements and predation risk (Stephens and Krebs 1986). An adult Asian elephant consumes 135–300 kg (fresh weight) plants per day (Sukumar 1992), so food resources associated with vegetation types are an important factor influencing their habitat preference. Bamboo constitutes one of the major food resources for wild elephants (Chen et al. 2006), and elephants depend on bamboo leaves and shoots mainly during June and September each year (Guo et al. 2006). Thus areas with abundant bamboo may attract elephants to visit them frequently. Crop raiding by elephants has been reported frequently in Xishuangbanna (Bureau of Statistics, Xishuangbanna 2002) and in Simao (Zhang et al. 2003). Therefore, based on simple food requirements (Sukumar 1990), farmlands should be a preferred habitat type for elephants. Yet our results showed they avoid farmlands, probably due to the frequent human interference. Elephants usually use nearby natural forests for concealment before raiding (Zhang et al. 2003). As the area of conifer forests within elephants’ 2 distribution range is small (3.09 km , 0.28%, Ou et al. 1997) and adjacent to farmlands, elephants frequently occurred there. Elephants preferred areas with lower elevations and milder slopes, for mountaineering is an energy-expensive activity for the largest mega-herbivore (Wall et al. 2007). Their preference for elevations 901–1200 m is expected as 62.6% of the areas in Mengyang talls within this range (Wu et al. 1997) and most of areas below 900 m is occupied by human settlements and farmland (Wu 1997). Preference in orientation was much more significant than other variables and the topography characteristics may explain it. Xishuangbanna is located in the southern part of Hengduan
371
Mountain Range which runs north to south, and the principle orientation of the range runs northwest-southeast through Xishuangbanna (Survey Group of Xishuangbanna Nature Reserve 1985). Therefore, we hypothesize that the principle migration routes of elephants may run northwest-southeast or north-south, although a further survey is needed to certify it. Elephant tolerance of relatively high levels of disturbance and human activities was not unexpected as they raid crops throughout their distribution (Santiapillai 1991, Nyhus et al. 2000, O’Connell-Rodwell et al. 2000, Willams et al. 2001). However, human disturbance still had a negative influence on their habitat preferences. Fifty-seven percent (n = 98) of the plots with presence evidence occurred in areas with low-level human disturbance and elephants preferred to conduct their activities at night, such as crop-raiding (Zhang et al. 2003) and highway crossing (W. J. Pan et al., unpubl.), in order to avoid human conflict. Therefore, habitat preference of elephants illustrate the need for large intact habitats with low human disturbance levels for the conservation of wild elephants. Possible causes of elephant decline
The distribution of elephants in Mengyang keeps shrinking, except in Area 4, and the expansion in Area 4 is probably a response of range shrinkage in other areas. Human activities are thought to be the major cause of this decline. The area covered by natural forest in Xishuangbanna decreased from 1.05 million ha in 1952 to 0.3 million ha in 1994, whereas the area of land represented by farmland increased from 36 242 ha in 1949 to 114 774 ha in 1998 (Guo et al. 2002). In response to this, the current habitat of elephants in China is divided into isolated patches situated among extensive stretches of rubber and tea plantations, and croplands (Zhang et al. 2006). Owing to the increase of the local human population and immigration from other prefectures, the number of settlements inside Mengyang increased from 16 in 1950 (Wu et al. 1997) to 56 in 2006. This casts grave doubts about the value of Mengyang to act as a pro-
372
L. Lin et al.
tected area. Moreover, most villages moved to lower elevations from mountain areas. In 1950, 69% of the villages were located in areas at elevations between 1200 and 1400 m, but now more than 83% of them are scattered across the more fertile lowlands or river valleys below 1100 m, which used to be covered by tropical rainforest that served as refuges for many large herbivores, such as elephant and gaur Bos gaurus (Wu et al. 1997). Forest degradation and cultivation in these areas caused the loss and fragmentation of elephant habitat. Furthermore, contiguous natural habitat was cut off by villages along rivers or roads. Area 1 is separated from other parts of Mengyang Protected Area by Mengyang River (Fig. 1). This area had long been utilized by elephants, but in the 1970s humans began to settle along the Mengyang River (Wu et al. 1997). Currently there are 7 villages along the river and most of the natural forest alongside has been cleared and replaced by farmland and rubber trees (Wu 1997). Elephants now appear to have been completely extirpated from this region, undoubtedly in response to the increase in human disturbance and large-scale landscape changes. In Area 2, the bamboo forest was completely cleared for rubber plantations in 2005 and in Area 5, rubber trees have been widely planted since 2005. It is unquestionable that the habitat loss drove elephants extinct here. The cause of disappearance of elephants in Area 3 is still uncertain, but human disturbance is thought to have played a role. Owing to habitat fragmentation and increased landscape heterogeneity, the distance between natural forest and farmlands has decreased and the potential for encountering crops by elephants increased near their natural habitats or along their migration routes, leading to increased crop-raiding (Sukumar 1992, 2003) and threatening the continuation of the migration (Berger 2004). Conversely, gun confiscation and hunting bans in the region have reduced local resident’s abilities to drive wild elephants away from their crops, through sound, light, fire, trenches and electric fences, which proved ineffective (Zhang and Wang 2003). In Area 4, elephants began to settle in the southern part of
this region in 1998 and consumed sugar cane, maize and rice heavily. By 2006, they ranged over the whole of Area 4. Although lacking experimental tests, correlations between continuous range expansion of elephants and their crop-raiding are likely. Large infrastructure is generally excluded from protected areas because it invariably has detrimental impacts on wildlife habitats, particularly when they bisect protected areas and hamper the migration of wildlife (Saunders et al. 2002, Rico et al. 2007). They might also increase the mortality of wildlife when passing through them (Clarke et al. 1998, Alexander and Waters 2000). The Simao-Xiaomengyang highway, which traverses Mengyang, has caused inevitable impacts on elephant migration because it diminished the cross-road corridors used by elephants. Moreover, elephants entered the freeway occasionally and have caused several traffic accidents (W. J. Pan et al., unpubl.). In Area 6, there was tropical bamboo forest that elephants had utilized frequently in the past, but during June 2003 and April 2006 when the construction of the highway occurred, elephants were seldom found in the area (W. J. Pan et al. unpubl.). Furthermore, in 2006 the bamboo forest was cleared and replaced by rubber plantations and elephants disappeared. Steps should therefore be taken to reduce the adverse influence on wildlife due to the building of large infrastructures. Additions to these infrastructures, such as wildlife overpasses and underpasses (Cleavenger and Waltho 2000), should be planned or modified to maintain the connectivity of habitat and migration routes of animals (McDonald and Clair 2004, Ng et al. 2004). Such measures should be introduced to all large Chinese infrastructure projects that unavoidably cross important habitat of threatened wildlife species. Indeed, this should be part of an appropriate environmental assessment process. Poaching has always been a major threat to the survival of elephants in China (Feng and Zhang 2005, Zhang et al. 2006) and the only objective is tusks. Elephant-poaching was serious in the 1980s and early 1990s. Seventeen elephants were killed between 1979 and 1983, and 16 were killed between 1994 and 1995 (Xishuang-
Asian elephants in Mengyang, China
banna National Nature Reserve unpubl.). Although no systematic records are available, wild elephants are still threatened by poaching in Mengyang and the latest accident occurred in October 2005. Some killings go unreported and undetected, especially in remote areas. Poaching is likely to be an important factor determining elephant distribution (Douglas-Hamilton et al. 2005, Buij et al. 2007). Douglas-Hamilton and colleagues (2005) noted that elephants habitually and rapidly vacated areas where people were perceived as hostile. During our visits, we also have heard several similar anecdotal stories which told that elephants usually avoided areas where poaching had occurred. Therefore, more effective anti-poaching and appropriate models of law enforcement are urgently needed and should be adopted for better conservation of wild elephants and their natural habitats in China. Acknowledgements: We are grateful to the International Fund for Animal Welfare (IFAW), Asian Elephant Conservation Grant of the USFWS and Xishuangbanna National Nature Reserve Management Bureau for both financial and logistic support. We greatly appreciated J. G. Huang, Y. H. Dong, Y. L. Guo, Z. Q. Yuan, W. Q. Zhu for their guidance and advice, and H. D. Zhao, D. F. Cao, Z. H. Li, K. X. Xiewen for their hard work and field assistance.
References Alexander S. M. and Waters N. M. 2000. The effects of highway transportation corridors on wildlife: a case study of Banff National Park. Transportation Research P C 8: 307–320. Berger J. 2004. The last mile: how to sustain long-distance migration in mammals. Conservation Biology 18: 320–331. Blom A., Zalinge R. V., Mbea E., Heitkönig I. M. A. and Prins H. H. T. 2004. Human impact on wildlife population within a protected Central Africa forest. African Journal of Ecology 42: 23–31. Buij R., Mcshea W. J., Campbell P., Lee M. E., Dallmeier F., Guimondou S., Mackaga L., Guisseougou N., Mboumba S., Hines J. E., Nichols J. D. and Alonso A. 2007. Patch-occupancy models indicate human activity as major determinant of forest elephant Loxodonta cyclotis seasonal distribution in an industrial corridor in Gabon. Biological Conservation 135: 189–201. Bureau of Statistics, Xishuangbanna. 2002. Report on economic and life loss caused by wild animals. The Bureau of Statistics, Xishuangbanna: 1–200. Chen J., Deng X. B., Zhang L. and Bai Z. L. 2006. Diet composition and foraging ecology of Asian elephants in
373
Shangyong, Xishuangbanna, China. Acta Ecologica Sinica 26: 309–316. Choudhury A. U. 1999. Status and conservation of the Asian elephant Elephas maximus in north-eastern India. Mammal Review 29: 141–173. Clarke G. P., White P. C. L. and Harris S. 1998. Effects of roads on badger Meles meles populations in south-west England. Biological Conservation 86: 117–124. Clevenger A. P. and Waltho N. 2000. Factors influencing the effectiveness of wildlife underpasses in Banff National Park, Alberta, Canada. Conservation Biology 14: 47–56. Codron J., Lee-Thorp J. A., Sponheimer M., Codron D., Grant R. C. and de Ruiter D. J. 2006. Elephant (Loxodonta africana) diets in Kruger National Park, South Africa: spatial and landscape differences. Journal of Mammalogy 87: 27–34. Douglas-Hamilton I., Krink T. and Vollrath F., 2005. Movements and corridors of African elephants in relation to protected areas. Naturwissenschaften 92: 158–163. Feng L. M. and Zhang L. 2005. Habitat selection by Asian elephant (Elephas maximus) in Xishuangbanna, Yunnan, China. Acta Theriologica Sinica 25: 229–236. [In Chinese with English summary] Greenwood A., Churchfield S. and Hickey C. 2002. Geographical distribution and habitat occurrence of the water shrew (Neomys fodiens) in the Weald of South-East England. Mammal Review 32: 40–50. Guo H., Padoch C., Chen A. and Fu Y. 2002. Economic development, land use and biodiversity change in the tropical mountains of Xishuangbanna, Yunnan, Southwest China. Environmental Science and Policy 5: 471–479. [In Chinese with English summary] Guo Y. L., Zhang L. and Dong Y. H. 2006. Foraging behavior of Asian elephants in Xishuangbanna, Yunnan. Acta Theriologica Sinica 26: 54–58. [In Chinese with English summary] Hayward M. W., de Tores P. J., Dillon M. J. and Banks P. B. 2007. Predicting the occurrence of the quokka, Setonix brachyurus (Macropodidae: Marsupialia), in Western Australia’s northern jarrah forest. Wildlife Research 34: 194–199. Ishwaran N. 1993. Ecology of the Asian elephant in lowland dry zone habitats of the Mahaweli River Basin, Sir Lanka. Journal of Tropical Ecology 9: 169–182. Li W. J., Wang Z. J., Ma Z. J. and Tang H. X. 1999. Designing the core zone in a biosphere reserve based on suitable habitats: Yancheng Biosphere Reserve and the red crowned crane (Grus japonensis). Biological Conservation 90: 167–173. McDonald W. and St. Clair C. C. 2004. Elements that promote highway crossing structures use by small mammals in Banff National Park. Journal of Applied Ecology 41: 82–93. Ng S. J., Dole J. W., Sauvajot R. M., Riley S. P. D. and Valone T. J. 2004. Use of highway undercrossings by wildlife in southern California. Biological Conservation 115: 499–507.
374
L. Lin et al.
Nyhus P. J., Tilson R. L. and Sumianto. 2000. Crop-raiding elephants and conservation implications at Way Kambas National Park, Sumatra, Indonesia. Oryx 34: 262–274. O’Connell-Rodwell C. E., Rodwell T., Rice M. and Hart L. A. 2000. Living with the modern conservation paradigm: can agricultural communities co-exist with elephants? A five-year case study in East Caprivi, Namibia. Biological Conservation 93: 381–391. Ostro L. E. T., Silver S. C., Koontz F. W. and Young T. P. 2000. Habitat selection by translocated black howler monkeys in Belize. Animal Conservation 3: 175–181. Ou X. K., Jin Z. Z., Peng M. C., Fang B. and Fang J. M. 1997. Distribution of vegetation in Mengyang Nature Reserve of Xishuangbanna and their ecological characteristics. Chinese Journal of Applied Ecology 8: 8–19. [In Chinese with English summary] Pradhan N. M. B. and Wegge P. 2007. Dry season habitat selection by a recolonizing population of Asian elephants Elephas maximus in lowland Nepal. Acta Theriologica 52: 205–214. Research Group 1, Institute of Zoology of Yunnan, 1976. [The distribution and natural conservation of wild elephant in Yunnan]. Chinese Journal of Zoology 2: 38. [In Chinese] Rico A., Kindlmann P. and Sedláèek F. 2007. Road crossing in bank voles and yellow-necked mice. Acta Theriologica 52: 85–94. Santiapillai C. 1991. Management of elephants in the Xishuangbanna Nature Reserve, PR China. Tiger Paper 18: 1–5. Santiapillai C. and Jackson P. 1990. The Asian elephant – an action plan for its conservation. IUCN/SSC Asian Elephant Specialist Group, IUCN. Gland, Switzerland: 1–80. Saunders S. C., Mislivets M. R., Chen J. and Cleland D. T. 2002. Effects of roads on landscape structure within nested ecological units of the Northern Great Lakes Region, USA. Biological Conservation 103: 209–225. Stephens D. W. and Krebs J. R. 1986. Foraging theory. Princeton University Press, Princeton: 1–247. Sukumar R. 1989. Ecology of the Asian elephant in southern India. I. Movement and habitat utilization patterns. Journal of Tropical Ecology 5: 1–18. Sukumar R. 1990. Ecology of the Asian elephant in southern India: II. Feeding habits and crop raiding patterns. Journal of Tropical Ecology 6: 33–53. Sukumar R. 1992. The Asian elephant: ecology and management. Cambridge University Press, Cambridge: 1–255. Sukumar R. 2003. The living elephants: evolutionary ecology, behavior, and conservation. Oxford University Press, London: 310–321.
Survey Group of Xishuangbanna Nature Reserve, 1985. [Reports on a comprehensive survey of the Xishuangbanna Nature Reserve]. Science and Technology Press, Yunnan: 1–478. [In Chinese] Vanderploeg H. A. and Scavia 1979. Calculation and use of selectivity coefficients of feeding: zooplankton grazing. Ecological Modelling 7: 135–149. Wall J., Douglas-Hamilton I. and Vollrath F. 2007. Elephants avoid costly mountaineering. Current Biology 16: 527–529. Wang S. 1998. [China Red Data Book of Endangered Animals]. Science Press, Beijing: 1–251. [In Chinese] Williams A., Johnsingh A. J. T. and Krausman P. R. 2001. Elephant-human conflicts in Rajaji National Park, northwestern India. Wildlife Society Bulletin 29: 1097–1104. Wu J. L., Jiang W. G., Hu J. S., Li Z. Q. and Li C. H. 1999. [Changes of distribution range of Asian elephant in Xishuangbanna in last 40 years]. Chinese Wildlife 20: 8–9. [In Chinese] Wu Z. L. 1997. Impact of land-use on natural landscapes in Mengyang Nature Reserve of Xishuangbanna. Chinese Journal of Applied Ecology 8: 32–37. [In Chinese with English summary] Wu Z. L., Peng M. C., Liu L. Y. and Dong Y. H. 1997. Settlement patterns in Mengyang Nature Reserve of Xishuangbanna. II: Causes and Effects. Chinese Journal of Applied Ecology 8: 25–31. [In Chinese with English summary] Zhang L., Ma L. C. and Feng L. M. 2006. New challenges facing traditional nature reserves: Asian elephant (Elephas maximus) conservation in China. Integrative Zoology 1: 179–187. Zhang L. and Wang N. 2003. An initial study on habitat conservation of Asian elephant (Elephas maximus), with a focus on human elephant conflict in Simao, China. Biological Conservation 112: 453–459. Zhang L., Wang N., Wang Y. N. and Ma L. C. 2003. A preliminary study on the habitat and behaviors of Asian elephant in Simao, Yunnan of China. Acta Theriologica Sinica 23: 185–192. [In Chinese with English summary] Zhu H. 2008. Advances in biogeography of the tropical rain forest in southern Yunnan, southwestern China. Tropical Conservation Science 1: 34–42.
Received 31 October 2007, accepted 14 April 2008. Associate editor was Matt Hayward.