IB

Na

b

c

a

ARRA

KBBOWCM

I

tliictCY

lSoFfe

Y

(

1d

Journal for Nature Conservation 20 (2012) 56– 61

Contents lists available at ScienceDirect

Journal for Nature Conservation

j our na l ho mepage: www.elsev ier .de / jnc

ndia’s biodiversity hotspot under anthropogenic pressure: A case study of Nilgiriiosphere Reserve

agarajan Baskarana, Uthirapathi Anbarasana, Govindasamy Agoramoorthyb,c,∗

Asian Nature Conservation Foundation, Indian Institute of Science, Bangalore 560012, IndiaGujarat Institute of Desert Ecology, PO Box 83, Bhuj 370001, IndiaCollege of Environmental Sciences, Tajen University, Yanpu, Pingtung 907, Taiwan

r t i c l e i n f o

rticle history:eceived 8 October 2010eceived in revised form 26 May 2011ccepted 10 August 2011

eywords:iodiversity hotspotiotic pressurevergrazing

a b s t r a c t

This paper presents data on the impact of biotic pressure in terms of grazing by livestock and wood cuttingby humans on the plant community in the Nilgiri Biosphere Reserve of India. Grass, and herbaceous plantbiomass, number of cattle dung piles, number of woody stems available and damaged by human activi-ties and weed biomass were assessed at different proximity along transects radiating from village-forestboundary to forest interior to measure the ecological impact of livestock grazing and fire wood collec-tion. The grass biomass was positively correlated to overgrazing indicating the adverse effect on naturalvegetation by cattle. Woodcutting was intense along the forest boundary and significantly declined asdistance increased. Similarly, weed biomass and number of thorny species declined positively with prox-

oodcuttingonservationanagement

imity from village-forest boundary and the weed biomass was significantly higher in the pastoral sitescompared to residential sites. The results suggest that human impact adversely affects natural vegeta-tion and promotes weed proliferation in forest areas adjoining human settlements in the ecologicallyimportant Nilgiri Biosphere Reserve. Continued anthropogenic pressure could cause reduction in fodderavailability to large herbivores like elephants, which in turn leads to an increase in human–elephant

conflict.

ntroduction

Anthropogenic pressures are known to influence natural ecosys-ems worldwide. Field studies have shown prolonged grazing byivestock on vegetation resulting in profound ecosystem changes,ncluding altered structure and composition as well as changesn the physical and chemical properties of top soil. The obvioushanges were reported in the under-story composition with reduc-ion or removal of shrubs and herbaceous perennials (Hoffman &owling 1990, 1991; Moolman & Cowling 1994; Stuart-Hill 1992;ates et al. 2000a).

Trampling by livestock can also lead to significant reduction initter and soil crust cover, water infiltration, and soil micro-climate.uch changes in soil structure were reported to reduce regenerationf dominant tree species in the woodland forest (Yates et al. 2000b).urthermore, studies have shown weed invasion prevalence in

ragments grazed by livestock for longer periods (Abensperg-Traunt al. 1998).

∗ Corresponding author at: College of Environmental Sciences, Tajen University,anpu, Pingtung 907, Taiwan. Tel.: +886 916752019.

E-mail addresses: agoram@mail.tajen.edu.tw, agoram@mail.nsysu.edu.twG. Agoramoorthy).

617-1381/$ – see front matter © 2011 Elsevier GmbH. All rights reserved.oi:10.1016/j.jnc.2011.08.004

© 2011 Elsevier GmbH. All rights reserved.

India’s tropical forests are threatened by severe overgrazingby livestock, forest fire, woodcutting, encroachment and agricul-ture (Arjunan et al. 2005; Jha & Singh 1990; Kodandapani et al.2008; Madhusudan 2004; Sen et al. 2008; Silori & Mishra 2001).Such impacts have degraded natural forests that are becomingunsuitable for large mammals, especially herbivores. As a result,managing such degraded forests is increasingly becoming a conser-vation nightmare (Agoramoorthy 2010). For example, India’s NilgiriBiosphere Reserve, which is one among the 25 global hotspots ofbiodiversity with diverse endemic fauna and flora, has been facingsevere man-made ecological disturbances (Desai & Baskaran 1996;Silori & Mishra 1995, 2001).

Data on the impact of anthropogenic factors affecting the natu-ral forests of southern India are limited. In this paper, we presentquantitative data on the impact of overgrazing by livestock onthe biomass of grass and herbaceous plants and woodcutting byhumans on the density of woody stem in Mudumalai Wildlife Sanc-tuary (presently a Tiger Reserve) and adjoining Nilgiri North ForestDivision of Tamil Nadu state, which is part of the Nilgiri BiosphereReserve of India.

Study area

Mudumalai Wildlife Sanctuary (area 321 km2) is one of themajor protected areas within the Nilgiri Biosphere Reserve (area

N. Baskaran et al. / Journal for Nature Conservation 20 (2012) 56– 61 57

Fig. 1. Map of Mudumalai Wildlife Sanctuary and its adjoining forest in the Nilgiri Biosphere Reserve, India.

5 ature Conservation 20 (2012) 56– 61

5aefww1fioNe2aa

D

mN2otcpllto2

fbDit1rotowcaetalwta(rww

o(wte(la

Distance (m)3500300025002000150010005000

no. o

f cat

tle d

ung

pile

s / h

a

0

50

100

150

200

250

300

350

0

50

100

150

200

250

300

350

biomass of grass and herbaceous plants increased with proxim-ity from the village to forest boundary (Fig. 3). For example, thegrass availability was 20 g/m2 at 0–500 m distance from the village

3500300025002000150010005000

Bio

mas

s

0

10

20

30

40

50

60

70

0

10

20

30

40

50

60

70

Grass Herbs

8 N. Baskaran et al. / Journal for N

520 km2) located in the Western Ghats of southern India (11◦32′

nd 11◦45′ N; 76◦20′ and 76◦45′ E; Fig. 1). The region has anxtended wet season from May to December and a short dry seasonrom January to April. The rainfall has a marked east–west gradientith eastern area receiving lesser rainfall (600–800 mm/year) thanestern area (1000–2000 mm/year). Temperature ranges from

6 ◦C in December to 32 ◦C in April. The vegetation follows the rain-all with tropical dry thorn forest in the east, dry deciduous forestn the middle, and moist deciduous forest in the west with patchesf semi evergreen forest. The sanctuary is connected to the Nilgiriorth Forest Division that acts as a corridor between the West-rn and Eastern Ghats and as a crucial elephant habitat (Sukumar003). Biotic pressures through grazing and collection of firewoodnd cattle dung by people pose a major threat in the eastern sidend adjoining forest areas (Silori & Mishra 2001).

ata collection and analysis

Four villages namely Masinagudi (Site 1), Moyar, (Site 2) Chem-anatham (Site 3) and Odagaramarigudi (Site 4) located in theilgiri Biosphere Reserve were selected to collect data. Sites 1 and

are residential villages harbouring 5000 inhabitants and 40%f them obtain their income through various government jobs,ourism and agriculture while the rest engage in the collection ofattle dung and fire wood or labour jobs. Sites 3 and 4 are smallastoral villages occupied by less than 100 people which include

abourers grazing the cattle of Badagas, a tribal community whoive up in Nilgiri hills. People from all four sites rear cattle for dungo meet the manure demand from tea and coffee plantations locatedutside the sanctuary (Desai & Baskaran 1996; Silori & Mishra 1995,001).

The cattle grazing impacts at different proximities from villages-orest boundaries were quantified through cattle dung count usingelt transects following the methods of Schaich et al. (2010).omestic animals such as cattle, goat, sheep and buffalo graze freely

n the sanctuary. We estimated the grazing impact of cattle sincehey form the major chunk (75%) of livestock diversity (Baskaran998). The dung of the related wild cattle, the Indian gaur (Bos gau-us) was differentiated from the domestic cattle by size and naturef defecation. Transects (2.5–3.5 km long, 4 m wide) were laid usinghe random table near villages on all directions following the meth-ds of Wilson et al. (1996). Dung seen within 2 m on either sideas counted. The density of dung per unit area was computed for

onsecutive 250 m distance to assess grazing impact. The biomassvailability (grass, herbs and shrubs browsed by ungulates) wasstimated at different proximities from the boundaries of villageso assess the impact on grass/herbaceous layers (Skarpe 1991). Inddition, three transects were laid, one each on the direction of vil-ages to forest interior. On every 100-m interval, two 1 m2 quadrats

ere laid on either side of transects at the fifth metre. All vegeta-ion (except woody stem and weed) within quadrats were scrapedt ground level and segregated into grass and herbaceous plantsWilson et al. 1996). After segregation, wet weight was taken sepa-ately using a balance to estimate biomass/unit area. Biomass dataere pooled at every 500 m intervals and mean biomass/m2 areaas computed to know changes in the availability.

Woodcutting was quantified using belt transects (2.5–3.5 km)rienting from villages to forest following the methods of Nagothu2001). All woody stem above 20 cm girth (damaged/undamaged)ithin 2 m on either side was recorded. Data on species name, sta-

us (damaged/undamaged) and cause of damage (human, elephant,

tc.) were recorded for each species. The biomass of invasive plantsLantana camera, Eupatorium odoratum, Opentia delinii and Tripu-us terestries) was estimated following the methods of Nkurunzizand Milberg (2007). Variables such as grass/herbaceous biomass,

Fig. 2. The average density of cattle dung (no/ha ±SD, n = 4) at different distancesfrom villages-forest boundary to forest interior in the Nilgiri Biosphere Reserve,India.

weed abundance, number of stems, number of human-damagedstems, and number of cattle dung piles were pooled at 250–500 mintervals. In order to know the relationship of each variable withdistance, they were tested against distance using Spearman rankcorrelation. Wilcoxon matched pair test was used to test the differ-ence in weed abundance between sites (SAS Institute 2000).

Results

Cattle grazing

The intensity of cattle use evident from dung density steadilydeclined with an increase in distance from villages to forest bound-ary (Fig. 2). The observed relationship between grazing and distancewas negatively correlated (r = −0.978, n = 13, p < 0.01) indicatingthat grazing pressure decreases with an increase in distance fromvillages. However, the presence of dung up to 3.5 km from the vil-lage showed that cattle movement continued inside the forest. The

Distance (m)

Fig. 3. The average biomass (wet wt g/m2) of grass and herbaceous plants (±SD,n = 4) at different distances from village-forest boundary to forest interior in theNilgiri Biosphere Reserve, India.

N. Baskaran et al. / Journal for Nature Conservation 20 (2012) 56– 61 59

Distance (m)3500300025002000150010005000

No.

of s

tem

s/ha

0

50

100

150

200

250

0

50

100

150

200

250

FaB

triwi

oNaheiasw

W

tnitdidsfea51

FpR

Distance (m)3500300025002000150010005000

Bio

mas

s

0

100

200

300

400

500

600

700

800

0

100

200

300

400

500

600

700

800

ig. 4. The average number of human damaged woody stems per ha (±SD, n = 4)t different proximity from village-forest boundary to forest interior in the Nilgiriiosphere Reserve, India.

o forest boundary. But, it increased to 45 g/m2 beyond 2 km. Theepeated cattle grazing near villages reduced the grass productiv-ty since the cattle grazed to the base level of the grasses and again

hen the grass re-grew. Similarly, herbaceous plants showed anncrement in biomass with distance (Fig. 3).

The Spearman rank correlation test revealed that the biomassf grass and herbaceous plants increased positively with distance.onetheless, only the grass had correlation (r = 0.893, n = 7, p < 0.01)nd not herbaceous plants (r = 0.714, n = 7, p < 0.071). Althougherbaceous biomass did not have statistical significance, it hadcological impact on the survival of endemic vegetation. A compar-son was made among the biomass of grass and herbaceous plantsnd cattle grazing pressure at different proximities, and the resultshowed that grass and herbaceous plants availability increasedith decrease in cattle grazing pressure (Figs. 2 and 3).

oodcutting

The intensity of woodcutting per unit area was quantified fromhe village to forest boundary to the interior forest up to 3.5 km. Theumber of stems damaged (lopped/cut) by people decreased with

ncrease in distance (Fig. 4). Spearman rank correlation revealedhat damage to tree species had strong negative correlation withistance (r = −0.956, n = 13, p < 0.01; Fig. 4). The woodcutting was

ntense at closer distance from the village-forest boundary andeclined gradually in the interior of forest. The density of woodypecies showed a gradual increase from village-forest boundary toorest interior due to high wood cutting pressure near villages. For

xample, the density of woody stems was less than 400 stems/hat 0–250 m distance from the boundary and it increased to over00 stems/ha after 1 km and further increased to 600 stems/ha after.75 km (Fig. 5).

Distance (m)3500300025002000150010005000

No.

of s

tem

s/ha

0

200

400

600

800

1000

0

200

400

600

800

1000

ig. 5. The average density of woody stems (trees) (±SD, n = 4) available at differentroximity from village-forest boundary to forest interior in the Nilgiri Biosphereeserve, India.

Fig. 6. The average biomass (wet wt g/m2) of weed (±SD, n = 4) estimated at differentproximity from village-forest boundary to forest interior in the Nilgiri BiosphereReserve, India.

Tree density was tested with distance using rank correlation,which showed positive correlation with distance (r = 0.841, n = 13,p < 0.01). The density of trees was significantly lower in forestcloser to villages than the forest interior suggesting that wood-cutting significantly reduced tree availability near villages. Theresults also show that the density of woody stem decreased signif-icantly with increase in human damage and correlated negativelywith human damage on tree species (r = −0.731, n = 13, p < 0.01).Furthermore, the woodcutting pressure altered woody speciescomposition, for example, thorny stem dominated the forest nearvillages and declined significantly as distance increased (r = −0.686,n = 13, p < 0.01). On the other hand, the non-thorny woody stemshowed a reverse trend with a positive increase with distance(r = 0.842, n = 13, p < 0.01). The intensive removal of woody stemnear villages might have resulted in thorny species dominating for-est near villages. Thus the non-thorny species availability increasedas distance from village-forest boundary increased.

Weed abundance

The biomass estimation of weed abundance in relation to dif-ferent proximities from village to forest boundary showed thatweed decreased with increase in distance (Fig. 6). The relation-ship between weed biomass and distance was tested using rankcorrelation that showed negative correlation, and the abundancedecreased with distance (r = −0.786, n = 7, p < 0.05). Notably, the twosmall pastoral villages (cattle pens) had significantly higher weedabundance (mean 381 g/m2) than other two residential sites withrelatively recent cattle grazing pressure (mean 131 g/m2; Wilcoxonmatched pair test, Z = 2.028, n = 7, p < 0.04) due to prolonged over-grazing and trampling of soil leading to higher weed proliferation.Comparison of overgrazing/woodcutting with weed abundanceshowed higher weed abundance near grazing/woodcutting areasand declined gradually as biotic pressure decreased. The weedabundance was positively correlated with cattle grazing (r = 0.964,n = 7, p < 0.01) and woodcutting (r = 0.964, n = 7, p < 0.01), whichclearly showed weed proliferation through biotic pressures.

Discussion

India harbours one of the largest domesticated bovine popula-tions (294 million) in the world (Agoramoorthy & Hsu 2008). Thedomestic cattle in rural India regularly travel inside the forest forgrazing and they impact directly by grazing and indirectly by tram-

pling. The negative correlation observed in this study betweencattle dung density and distance provides evidence supportingthe pressure. This study illustrates that overgrazing significantlyreduced grass and herbaceous cover as evidenced from the negative

6 ature

rbcRrstcibbsrplv

t1afsfMtocf2

stethRdf1c(towrci(

iiupscif(

gobSbop

0 N. Baskaran et al. / Journal for N

elationship between cattle density and grass, and herbaceousiomass. A similar trend in cattle grazing effect on grass and herba-eous plant community was reported from India (Baskaran 1998;ameshkumar 1994). Furthermore, Painter and Belsky (1993)eported overgrazing by livestock in USA held the grasses andedges at lower densities. Experimental studies in forested ecosys-ems have found that livestock substantially reduced the vegetationover, especially perennial bunchgrasses (Bock & Bock 1993). Sim-larly, recovery of vegetation cover, especially grass and decline ofare ground in the absence of cattle grazing have been documentedy Dobkin et al. (1998) in the USA. Studies in Australia have alsohowed that prolonged use of native vegetation by livestock haveesulted in reduction and removal of native shrubs and herbaceouserennials (Yates et al. 2000a). Hence it is clear that over grazing by

ivestock is a major cause leading to the decline of the understoryegetation.

In our study area, the vegetation was not only affected by cat-le grazing but also by the collection of dung by people. About5,000 livestock have been reared in the study area and 90% of themre of an unproductive low yielding variety and are mainly rearedor dung production (Baskaran, unpublished data). Estimates havehown 6400–6800 tonnes of dry cattle dung collected annuallyrom seven forest villages of Mudumalai Wildlife Sanctuary (Silori &

ishra 2001). Therefore, the cattle dung that is supposed to enrichhe soil of overgrazed areas is not at all recycled. Impact by livestockvergrazing and dung collection has been reported to alter physi-al and chemical properties of soil as well as composition of microauna and flora (Hobbs 2001; Kinnear & Tongway 2004; Yates et al.000b).

The negative correlation observed between stem damage inten-ity and woody stem availability shows that woodcutting decreasedhe availability of woody stem in adjoining areas. Similarly, Sagart al. (2003) reported highest woody stem density at the least dis-urbed site and lowest at the highly disturbed site of Vindhyanill ranges of west-central India. Elephants in the Nilgiri Biosphereeserve are known to browse on plants such as Tectona gran-is, Dichrostachys cinearea, Acacia sp. and Albizzia sp. and theseood trees are being exploited by humans for firewood (Baskaran998; Sukumar 2003). About 2000 tonnes of firewood annuallyame out of seven villages where the present study was conductedSilori & Mishra 2001). Woody stem composition was altered byhe wood removal pressure evident from the positive correlationf non-thorny species and negative correlation of thorny speciesith distance. As a matter of fact, extensive use of firewood in

ural areas was known to cause forest decline in many developingountries. According to recent estimates, consumption of firewoodn rural India accounts for 87% with demand exceeding supplyAgoramoorthy & Hsu 2008).

The positive correlation among weed abundance, cattle graz-ng and woodcutting revealed that biotic pressure significantlyncreased weed abundance. Similarly, Rameshkumar (1994) doc-mented positive correlation between weed abundance and bioticressure. The weed abundance per unit area was higher in pastoralettlements. The higher weed abundance recorded in pastoral sitesould be the reflection of the impact of intensive past overgraz-ng. Besides, weed invasion tends to be more prevalent in forestragments that have been grazed by livestock for a long periodAbensperg-Traun et al. 1998; Hobbs 2001).

This study shows that human disturbances through livestockrazing and woodcutting could reduce the diversity and populationf grass, herbaceous and woody plants and increase weed out-urst in the ecologically sensitive Nilgiri Biosphere Reserve of India.

evere biotic pressure in the tropical forest not only affect flora,ut also large herbivores such as elephants and gaur that dependn native vegetation for survival. Field reports indicate that ele-hants have strayed or moved out of their traditional ranges due to

Conservation 20 (2012) 56– 61

severe habitat degradation around villages located in forest areas(Baskaran et al. 1995; Baskaran 1998). If appropriate measures arenot taken by the government of Tamil Nadu State to protect thenatural forest vegetation from cattle grazing and wood cutting,lethal human–elephant conflicts cannot be avoided in the NilgiriBiosphere Reserve.

In order to reduce the dependency of people on forest, we sug-gest the following recommendations to the government of TamilNadu, India:

1. People who own a large number of low milk-yielding cowscan replace them with a few high milk-yielding cow varietiesthrough government assistance.

2. Fodder for livestock can be supplied through people’s partic-ipation from forest areas that come under the Joint ForestManagement System.

3. The Forest Department can increase the temporary/regular jobopportunities for people who depend on livestock dung and fuelwood for livelihood.

4. The government can provide free household biogas plants orsupply liquefied petroleum gas at subsidized rate through theclimate change scheme.

5. The Forest Department must restrict cattle rearing by those whoare employed by the government.

6. The Forest Department must ban dung collection and sale fromvillages.

7. The government must implement alternate livelihood measuresfor villagers to enhance biodiversity conservation in the NilgiriBiosphere Reserve.

Acknowledgements

We thank the Tamil Nadu State Forest Department for providingpermission for this study. We also thank Dr. R. Sukumar (IndianInstitute of Science-IISc) for providing logistics at Masinagudi FieldStation, and at IISc extended to UA.

References

Abensperg-Traun, M., Smith, G. T., Arnold, G. W., & Steven, D. E. (1998). Exotic plantinvasion and understorey species richness: A comparison of two types of euca-lypt woodland in agricultural Western Australia. Pacific Conservation Biology, 4,21–32.

Agoramoorthy, G. (2010). Global wildlife conservation. An ecologist’s jungle journal.Berlin: Lambert Academic Publishing.

Agoramoorthy, G., & Hsu, M. J. (2008). Biogas plants ease ecological stress in India’sremote villages. Human Ecology, 36, 435–441.

Arjunan, M., Puyravaud, J. P., & Davidar, P. (2005). The impact of resource collec-tion by local communities on the dry forests of Kalakad—Mundanthurai TigerReserve. Tropical Ecology, 46(2), 135–143.

Baskaran, N. (1998). Ranging and resource utilization by Asian Elephant (Elephasmaximus Linnaeus) in Nilgiri Biosphere Reserve, South India. Ph.D., Thesis. India:Bharathidasan University.

Baskaran, N., Balasubramanian, M., Swaminathan, S., & Desai, A. A. (1995). Homerange of elephants and its implications for the management of Nilgiri BiosphereReserve. In J. C. Daniel, & H. S. Datye (Eds.), A week with elephants (pp. 296–313).Bombay Natural History Society, Bombay.

Bock, C. E., & Bock, J. H. (1993). Cover of perennial grasses in southeastern Arizonain relation to livestock grazing. Conservation Biology, 7, 371–377.

Desai, A. A., & Baskaran, N. (1996). Impact of human activities on the rangingbehaviour of elephants in the Nilgiri Biosphere Reserve, South India. Journalof the Bombay natural History Society, 93(3), 559–569.

Dobkin, D. S., Rich, A. C., & Pyle, W. H. (1998). Habitat and avifaunal recovery fromlivestock grazing in a Riparian meadow system of the Northeastern Great Basin.Conservation Biology, 12, 209–221.

Hobbs, R. H. (2001). Synergisms among habitat fragmentation Livestock grazing, andbiotic invasions in Southwestern Australia. Conservation Biology, 15, 1522–1528.

Hoffman, M. T., & Cowling, R. M. (1990). Desertification in the lower Sundays RiverValley, South Africa. Journal of Arid Environment, 19, 105–117.

Hoffman, M. T., & Cowling, R. M. (1991). Phytochorology and endemism along aridityand grazing gradients in the lower Sundays River Valley, South Africa: Implica-tions for vegetation history. Journal of Biogeography, 18, 189–201.

Jha, C. S., & Singh, J. S. (1990). Composition and dynamics of dry tropical forest inrelation to soil texture. Journal of Vegetation Science, 1, 609–614.

ature

K

K

M

M

N

N

P

R

S

S

N. Baskaran et al. / Journal for N

innear, A., & Tongway, D. (2004). Grazing impact on soil mites of semi-arid cheno-pod shrublands in Western Australia. Journal of Arid Environments, 56, 63–82.

odandapani, N., Cochrane, M. A., & Sukumar, R. (2008). A comparative analysis ofspatial, temporal, and ecological characteristics of forest fires in seasonally drytropical ecosystems in the Western Ghats. Forest Ecology and Management, 256,607–617.

adhusudan, M. (2000). Recovery of wild large herbivores following livestockdecline in tropical Indian wildlife reserve. Journal of Applied Ecology, 41,858–869.

oolman, H. J., & Cowling, R. M. (1994). The impact of elephant and coat grazing onthe endemic flora of South African succulent thicket. Biological Conservation, 68,53–61.

agothu, U. S. (2001). Fuelwood and fodder extraction and deforestation: Main-stream views in India discussed on the basis of data from the semi-arid regionof Rajasthan. Geoforum, 32, 319–332.

kurunziza, L., & Milberg, P. (2007). Repeated grading of weed abundance and mul-tivariate methods to improve the efficacy of on-farm weed control trials. WeedBiology and Management, 7, 132–139.

ainter, E. L., & Belsky, A. J. (1993). Application of herbivore optimization theory torange lands of the western United States. Ecological Applications, 3, 2–9.

ameshkumar, S. (1994). Ecology of Asian elephant and its interaction with man insouth India. Unpublished Ph.D. Thesis. India: Bharathidasan University.

agar, R., Raghubanshi, A. S., & Singh, J. S. (2003). Tree species composition, dis-persion and diversity in a dry tropical forest region of India. Forest Ecology andManagement, 186, 61–71.

AS Institute. (2000). SAS/ETS software: Changes and enhancements. Cary: SAS Insti-tute.

Conservation 20 (2012) 56– 61 61

Schaich, H., Szabo, I., & Kaphegyi, T. A. M. (2010). Grazing with Galloway cattle forflood plain restoration in the Syr Valley Luxembourg. Journal for Nature Conser-vation, 18, 266–277.

Sen, A., Johri, T., & Bisht, N. S. (2008). Analysis of the effect of anthropogenic inter-ferences on tree species composition in the tropical forests of Dadra and NagarHaveli, India. Current Science, 95(1), 50–58.

Skarpe, C. (1991). Impact of grazing in savanna ecosystems. Ambio, 20, 351–356.Silori, C. S., & Mishra, B. K. (1995). Pressure and resource dependency of Masinagudi

group of villages on the surrounding elephant habitat. In J. C. Daniel, & H. S. Datye(Eds.), A week with elephants (pp. 270–278). Bombay: Bombay Natural HistorySociety.

Silori, C. S., & Mishra, B. K. (2001). Assessment of livestock grazing pressure in andaround the elephant corridors in Mudumalai Wildlife Sanctuary, South India.Biodiversity and Conservation, 10, 2181–2195.

Stuart-Hill, G. C. (1992). Effects of elephants and goats on the Kaffrarian succulentthicket of the eastern Cape, South Africa. Journal of Applied Ecology, 29, 699–710.

Sukumar, R. (2003). The living elephants: Evolutionary ecology, behavior and conser-vation. New York: Oxford University Press.

Wilson, D. E., Cole, F. R., Nichols, J. D., Rudran, R., & Foster, M. S. (1996). Measuringand monitoring biological diversity: Standard methods for mammals. Washington:Smithsonian Institution Press.

Yates, R. J., Hobbs, R. J., & Atkins, L. (2000). Establishment of perennial shrubs and

tree species in degraded Eucalyptus salmonophloia remnant woodlands: Effectof restoration treatments. Restoration Ecology, 8, 135–143.

Yates, C. J., Norton, D. A., & Hobbs, R. J. (2000). Grazing effects on soil and micro-climate in fragmented woodlands in southwestern Australia: Implication forrestoration. Austral Ecology, 25, 36–47.