This article was downloaded by: [Dr Subrata Nandy]On: 10 June 2013, At: 03:20Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: MortimerHouse, 37-41 Mortimer Street, London W1T 3JH, UK

International Journal of Biodiversity Science,Ecosystem Services & ManagementPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/tbsm21

Comparing tree diversity and population structurebetween a traditional agroforestry system andnatural forests of Barak valley, Northeast IndiaSubrata Nandy a & Ashesh Kumar Das ba Forestry and Ecology Department, Indian Institute of Remote Sensing , Indian SpaceResearch Organisation , Dehradun , 248001 , Indiab Department of Ecology and Environmental Sciences , Assam University , Silchar ,788011 , Assam , IndiaPublished online: 06 Dec 2012.

To cite this article: Subrata Nandy & Ashesh Kumar Das (2013): Comparing tree diversity and population structurebetween a traditional agroforestry system and natural forests of Barak valley, Northeast India, International Journal ofBiodiversity Science, Ecosystem Services & Management, 9:2, 104-113

To link to this article: http://dx.doi.org/10.1080/21513732.2012.748691

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form toanyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug dosesshould be independently verified with primary sources. The publisher shall not be liable for any loss, actions,claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly orindirectly in connection with or arising out of the use of this material.

International Journal of Biodiversity Science, Ecosystem Services & Management, 2013Vol. 9, No. 2, 104–113, http://dx.doi.org/10.1080/21513732.2012.748691

Comparing tree diversity and population structure between a traditional agroforestry systemand natural forests of Barak valley, Northeast India

Subrata Nandya,* and Ashesh Kumar Dasb

aForestry and Ecology Department, Indian Institute of Remote Sensing, Indian Space Research Organisation, Dehradun 248001, India;bDepartment of Ecology and Environmental Sciences, Assam University, Silchar 788011, Assam, India

We studied population structure, composition and diversity in a traditional Indian agroforestry system, called paan jhum,in comparison to natural forests of the Barak valley, Assam, northeast India. The phytosociological data from these forestswere analysed quantitatively, to determine species richness, diversity, importance value, stand density and the basal area.The analysis showed that species richness and diversity were higher in paan jhum than in natural forests, in all three studysites. A total of 47, 37 and 48 tree species were recorded in paan jhum, compared with 35, 32 and 42 species in naturalforests of the three study sites, respectively. Paan jhum had higher stand density (790, 934 and 763) and basal area (74.05,41.60 and 55.88 m2 ha−1), whereas natural forests had lower stand density (775, 865 and 522) and basal area (68.75,40.50 and 48.04 m2 ha−1) in all the study sites, respectively. An F-test showed significant differences in the variance inspecies richness, basal area and the stand density at 95% confidence level in the two forest categories. Paan jhum mightbecome a component of a forested landscape that is valued for contributing to resource production, other ecosystem servicesand biodiversity conservation.

Keywords: diversity; population structure; Paan jhum; traditional agroforestry; Khasi tribes; conservations

Introduction

Due to the rapid rise of human population and their con-sequent increased demands for more utilization of forestresources, the natural forests in India as well as in manyother countries are shrinking at an alarming rate. Theconversion of forests has led to habitat loss, which isthe greatest threat to biodiversity (Wong 2011). Hence,conservation and judicious use of natural resources are fun-damental to the sustainable development. For sustainabilityof natural resources, i.e. planned utilization of naturalresources taking full account of the needs of the presentand future generations, traditional knowledge plays a vitalrole. Ramakrishnan (1998) emphasized that the way tra-ditional societies are linked to natural ecosystems andlandscapes provide important clues for biodiversity conser-vation. Tribal communities have identified themselves aspart of the ecosystem and have acquired empirical knowl-edge on the basis of their experience while living close tonatural resources. Based on traditional knowledge, socio-cultural practices and religious beliefs, tribal communitieshave maintained the natural resources in a diverse and pro-ductive state. The multifunctional nature of agroforestrysystems plays an important role in biodiversity conserva-tion, providing goods and services to society, augmentationof carbon storage, enhancing the fertility of the soils andproviding social well-being to people (Pandey 2007).

Paan jhum is a traditional agroforestry system prac-ticed by the Khasi tribes of Barak valley in northeast India.Rahman et al. (2009) also described this kind of indigenous

*Corresponding author. Email: subrato.nandy@gmail.com

agroforestry system practiced by the Khasi tribes inBangladesh. In addition, Wiersum (2004) described themas ‘forest gardens in which wild and cultivated plants coex-ist such that the structural characteristics and ecologicalprocesses of natural forests are preserved’. They are char-acterized by their evolution within the local communitiesrather than as a result of formal agroforestry systems.Similar examples of forest gardens forming a compo-nent within a forested landscape have been describedfor Indonesia (Mary and Michon 1987; Aumeeruddy andSansonnens 1994) as well as the Philippines (Olofson1980; Fujisaka and Wollenberg 1991), Thailand (Withrow-Robinson et al. 1998) and the Amazon region (Denevanet al. 1984; Posey 1985). Chase (1989) suggested theterm ‘domiculture’ for the creation of such forested land-scapes with a series of localized areas (domuses). The term‘in domo’ conservation of biodiversity was suggested byWiersum (2003) for the interaction between people andforest vegetation with each area having a specific set ofmanagement practices. Agroforestry represents a varietyof nature–human relations; they reflect multiple forms ofhuman creativity in dealing with forest resources (Wiersum1997a, 1997b).

In Barak valley, the Khasi tribe is associated with thepractice of paan jhum (Piper betle cultivation in naturalforest) that is considered as having significant implica-tions on forestry management (Unlocking . . . 2005). Paan(Piper betle), a perennial creeper, is cultivated for its leaf.In paan jhum, it is planted as a subordinate crop mixed with

© 2013 Taylor & Francis

Dow

nloa

ded

by [

Dr

Subr

ata

Nan

dy]

at 0

3:20

10

June

201

3

International Journal of Biodiversity Science, Ecosystem Services & Management 105

forest trees to get the economic benefit and also the ecolog-ical benefits from the resultant interaction. In this study, acomparative phytosociological analysis of tree species wascarried out in paan jhum and nearby natural forests.

Materials and methods

Study area

Three sites were selected for the present study, namelysite 1 – Badshahitila (24◦20′54.2′′–24◦25′43.7′′N lat-itude and 92◦20′53.3′′–92◦21′29.4′′E longitude), site2 – Loharbond (24◦34′37.5′′–24◦35′9.5′′N latitude and92◦45′10.9′′–92◦46′42.1′′E longitude) and site 3 – Sultani(24◦20′54.6′′–24◦21′46.6′′N latitude and 92◦31′21.4′′–

92◦31′43.3′′E longitude) (Figure 1). These sites arelocated in Barak valley, situated in the southern Assam ofnortheast India. The region shares its borders with NorthCachar Hills district of Assam and Meghalaya in the north;Manipur in the east; Mizoram in the south and Tripuraand Sylhet district, Bangladesh in the West. The valley hasan undulating topography characterized by hills, hillocks(locally known as tillah), wide plains and low-lyingwaterlogged areas (locally called beels). Badshahitila liesin Karimganj district, Loharbond is in Cachar district andSultani is located in Hailakandi district of Assam, north-east India. For the purpose of study, in each of these sitestwo areas were selected, one in paan jhum and the otherin nearby natural forest, to observe the difference in tree

India

Karimganj

Hailakandi

Plot location

Barak valley

Site 1

Site 3

Site 2

Cachar

Assam

Figure 1. Location of study area.

Dow

nloa

ded

by [

Dr

Subr

ata

Nan

dy]

at 0

3:20

10

June

201

3

106 S. Nandy and A.K. Das

diversity and population structure in both the areas. Thestudy sites have a warm humid climate with mean annualrainfall of 2660 mm, most of which is received duringthe south-west monsoon season (May to September). Themean maximum temperature ranges from 25.4◦C (January)to 32.6◦C (August) and the mean minimum temperaturevaries from 11◦C to 25◦C (August). The dry season usuallycorresponds to the period from December to February. Theforest types in these areas are mainly Cachar tropical ever-green forest (1B/C3) and Cachar tropical semi-evergreenforest (2B/C2) (Champion and Seth 1968).

Piper betle and the Barak valley

Piper betle is native to Malaysia and is cultivated through-out south-eastern Asia (The wealth . . . 1969). The leavesform the well-known masticatory with the addition of lime,catechu, areca nuts and other flavouring ingredients. It isaromatic, carminative, stimulant and astringent. The betelleaf occupies a significant place in the everyday life of thepeople of northeast India. Chewing of paan is an ancienthabit among all classes of people of northeast India. To thefarmers, a betel leaf garden provides the much-needed cashfor meeting their day-to-day requirements. The cultivationof the vine under relatively natural conditions is limited totropical forest areas on the south-western coast of peninsu-lar India and the hill areas of Assam, where the cultivationof crop is done in the shade of forest trees or in establishedareca nut gardens. Under artificially created conditions,however, it is grown almost in every part of the country,although in a scattered fashion (Chagule 1960).

In the hilly areas of Barak valley, the cultivation of thevine, known as paan jhum, is practiced in large scale. Thepractice involves pollarding of trees in a forest and subse-quently planting the stem cuttings of the vine at the base ofsuch trees. This is done during June–July. The pollardinghelps sunlight to penetrate through the forest canopy, whichis essential for the growth of paan. The vines are propa-gated by terminal stem cuttings collected from sufficientlymature plants. Within 1 month of planting, new buds startcoming up from the newly planted paan plants. The vine isgrown on all the trees in the paan jhum area. The collec-tion of paan starts after 3 years of plantation and continueall the year round. But collection is more during June–July,i.e. during rainy season, as high rainfall facilitates highproduction of the leaves.

Methodology

For the phytosociological analysis in paan jhum as well asin natural forest, a total of 60 quadrats were laid (10 ran-dom quadrats in each site). The size of the quadrat wasdetermined by species area curve method (Misra 1968;Mueller-Dombois and Ellenberg 1974) and was found tobe 0.04 ha. The sites were selected to represent the majorvegetation. In each quadrat, the girth at breast height (gbhat 1.37 m from the ground) of all the trees with gbh≥30 cm were measured and recorded. The vegetation datawere quantitatively analysed for abundance, density andfrequency (Curtis and McIntosh 1950). The importancevalue index (IVI) for the tree species was determined asthe sum of the relative frequency, relative density and rela-tive dominance (Curtis 1959). The ratio of abundance tofrequency was used to interpret the distribution patternof the species (Whitford 1949). The ratio of abundanceto frequency indicates regular distribution if below 0.025,random distribution between 0.025 and 0.05 and conta-gious if >0.05 (Curtis and Cottam 1956). Species diversityof each site was determined using Shannon–Weaver index(1949):

H = −∑ [

(ni/N) logn (ni/N)]

(1)

where, ni is the total number of individuals of species i andN is the total number of individuals of all species in thatstand; log implies to natural log.

An F-test was carried out to work out whether thepopulation attributes, namely species richness, basal areaand stand density differ significantly in both the forestcategories under study. A questionnaire-based survey wasundertaken in the individual households of the Khasi tribeswho practice paan jhum. During the survey, the infor-mation regarding the details of the paan jhum practiceincluding the yield of paan and the income generated werenoted down.

Results

Species richness and diversity

In all study sites, species richness was found to be higher inpaan jhum than the natural forest (Table 1). Highest speciesrichness was observed in paan jhum of site 3. A totalof 48 species was observed in paan jhum compared to

Table 1. Phytosociological analysis of natural forest and paan jhum in three study sites in Barak valley, Assam, Northeast India.

Badshahitila Loharbond Sultani

Variable Natural Forest Paan jhum Natural Forest Paan jhum Natural Forest Paan jhum

Number of species 35 47 32 37 42 48Number of genera 32 41 30 31 36 41Number of families 22 30 23 24 28 32Shannon–Weaver index 3.00 3.30 3.07 3.12 3.27 3.36Basal area (m2 ha−1) 68.75 74.05 40.50 41.60 48.04 55.88Stand density (stems ha−1) 775 790 865 934 522 763

Dow

nloa

ded

by [

Dr

Subr

ata

Nan

dy]

at 0

3:20

10

June

201

3

International Journal of Biodiversity Science, Ecosystem Services & Management 107

42 species in natural forest. Species richness ranged from32 to 42 species in the natural forest, whereas in paan jhumit ranged between 37 and 48. The number of genera as wellas number of families was found to be higher in paan jhumareas than in natural forest in all the three sites. Shannon–Weaver index showed the highest species diversity in site 3,where diversity in paan jhum (3.36) was more than the nat-ural forest (3.27); similarly in other two sites also speciesdiversity in paan jhum was higher than the natural forest(Table 1).

Density, basal area and importance value

Table 1 shows the stand density (stems ha−1) in all thestudy sites. Highest stand density was observed in paanjhum of site 2 (934 stems ha−1), whereas lowest stand den-sity was observed in natural forest of site 3 (522 stemsha−1). It was observed that the stand density was higherin paan jhum than the natural forest of all the sites. Thebasal area was found to be slightly higher in paan jhumthan in the natural forest (Table 1). The basal area rangedfrom 40.50 to 68.75 m2 ha−1 in natural forest and between41.60 and 74.05 m2 ha−1 in paan jhum.

The IVI of top ten species in paan jhum as well as nat-ural forests of all the study sites is shown in Figure 2. Theresult showed that in site 1, the highest IVI in natural forestwas represented by Amoora wallichii (41.56), whereas inpaan jhum area of the same site Dipterocarpus turbina-tus was the dominant tree species having IVI of 43.43.A comparison of the IVI values in natural forest and paanjhum area of this site allowed several noteworthy obser-vations. Dipterocarpus turbinatus, which is an evergreenspecies, has more IVI in paan jhum area (43.43) than inthe natural forest (7.64). One more interesting observationis that Mangifera sylvatica (a rare species), a wild mango,has an IVI of 11.06 in paan jhum area, whereas in thenatural forest of the same site it is absent. In site 2, thehighest IVI in natural forest was represented by Ziziphusrugosa (40.51), whereas in paan jhum area of the same site,the highest IVI was represented by Palaquium polyanthum(67.93), an evergreen species. In the third site, the highestIVI in natural forest was represented by Cynometra polyan-dra (53.46), whereas in the paan jhum area of the samesite, Alseodaphne owdenii has the highest IVI (55.32).Alseodaphne owdenii is the best tree species available inthis region used for making furniture.

Stand density and forest structure

Tree species richness, stand density and Shannon–Weaverindex across girth classes in all the areas showed a decreas-ing trend with increase in stem girth classes (Table 2).Tree size class distribution in both the categories of forestareas followed a reverse J-pattern. The mean tree density inlower and middle girth classes (30–60 cm, 60–90 cm and90–120 cm) of site 3 was 50% greater in paan jhum thanin the natural forest (Table 2). In site 1, stand density inlowest girth class (30–60 cm) was more in natural forest(275 stems ha−1) than in paan jhum (210 stems ha−1),

whereas in middle and higher girth classes stand densitywere higher in paan jhum. In site 2, the stand density inthe lowest girth class (30–60 cm) was more in paan jhum(600 stems ha−1) than in natural forest (495 stems ha−1).The mean basal area contribution in the lowest girth class(30–60 cm) was higher in natural forest compared withpaan jhum, but in the middle as well as in the higher girthclasses, mean basal area was found to be higher in paanjhum of site 1 (Figure 3). In site 2, the basal area of lowestgirth class (30–60 cm) was 73.59% greater in paan jhum(9.24 m2 ha−1) than the natural forest (6.8 m2 ha−1), whilethe other girth classes were represented almost equally inboth the areas. The mean basal areas in lower and middlegirth classes of site 3 were more in paan jhum, whereasthe higher girth classes in natural forest were representedby higher values. The F-test showed that the variance inspecies richness, basal area and stand density differ signifi-cantly at 95% confidence level in both the forest categoriesin all the sites except species richness and basal area ofsite 2.

Distribution pattern

The analysis of distribution pattern of trees is shown inFigure 3. In site 1, most of the species were regularly dis-tributed both in natural forest and in paan jhum. The sametrend was observed in site 2 also. But comparison showsthat the species in paan jhum were more regularly dis-tributed than the natural forest in sites 1 and 2. In site 3,most of the species in both natural forest and paan jhumwere contagiously distributed (Table 3).

Economic benefits

The Khasi tribes get economic benefits by selling paan,which is their main occupation. The annual yield of paanis approximately 2,52,000 individual leaves per hectare.The economic life span of paan plant is about 30 years.The vine starts yielding paan from the age of 3 years.The production is more during June–July (rainy season).During January–March (winter season), paan productionis relatively low, as the leaves fall due to abscission. Newleaves start coming up on the onset of monsoon. Due toless production of paan during February–March, the pricebecomes higher, i.e. Rs.350 per kuri (1 kuri = 6720 leaves)(Rs.69.68 = 1C). But, in other months, the price is less(Rs.40–50 per kuri) because of more production.

Discussion

Different tribal communities around the world practice var-ious types of agroforestry systems for their livelihood.Khasi tribes, like other tribal communities, practice shift-ing cultivation for many years (Bareh 1967). These tribeshave also developed their own market-oriented sustainabletree crop production system (Nath et al. 2003). Paan jhumareas are conserved and protected imbued with the reli-gious beliefs and taboos of the local people. Felling of trees

Dow

nloa

ded

by [

Dr

Subr

ata

Nan

dy]

at 0

3:20

10

June

201

3

108 S. Nandy and A.K. Das

Site 2

Site 3

Name of species

Paan jhum

0

15

30

45 41.76 39.76

6.629.4311.0612.7613.76

20.5221.68

43.43

Am

oora

wal

lich

ii

Dip

tero

carp

us tu

rbin

atus

Cyn

omet

ra p

olya

ndra

Art

ocar

pus

cham

a

Hol

igar

na lo

ngif

olia

Bac

aure

a ra

mif

lora

Man

gife

ra s

ylva

tica

Mag

noli

a ho

dgso

nii

Syzi

gium

cum

inii

Pal

aqui

um p

olya

nthu

m

Site 1

0

15

30

60

75

45

6.517.098.679.5810.6010.6611.57

29.9331.45

67.93

Cyn

omet

ra p

olya

ndra

Zan

thox

ylum

bud

rung

a

Sara

ca a

soca

Schl

eich

era

oleo

sa

Sem

ecar

pus

anac

ardi

um

Ziz

phus

rug

osa

Gar

cini

a xa

ntho

chym

us

Syzi

gium

cum

inii

Gyn

ocar

dia

odor

ata

Pal

aqui

um p

olya

nthu

m

Impo

rtan

ce v

alue

inde

x (I

VI)

0

15

30

45 40.51

17.66

10.8111.2512.3312.70

19.2724.0524.62

31.16

Dio

spyr

os m

elan

oxyl

on

Ziz

phus

rug

osa

Syzi

gium

cum

inii

Cas

tano

psis

indi

ca

Mes

ua fl

orib

unda

Chr

ysop

hyll

umla

nceo

latu

m

Vat

ica

lanc

eaef

olia

Ster

culi

a fo

etid

a

Pod

ocar

pus

neri

ifol

ius

Pal

aqui

um p

olya

nthu

m

Impo

rtan

ce v

alue

inde

x (I

VI)

0

15

30

60

45

12.35

53.46

32.5425.64

22.1116.18 14.76

9.77 9.426.21

Cyn

omet

ra p

olya

ndra

Dio

spyr

os m

elan

oxyl

on

Bis

chof

ia ja

vani

ca

Art

ocar

pus

cham

a

Sara

ca a

soca

Syzi

gium

cum

inii

Mes

ua fl

orib

unda

Dip

tero

carp

us tu

rbin

atus

Pal

aqui

um p

olya

nthu

m

Hol

igar

na lo

ngif

olia

Natural forest

Impo

rtan

ce v

alue

inde

x (I

VI)

0

15

30

45 41.56

34.81

20.29

Am

oora

wal

lich

ii

Chu

kras

ia ta

bula

ris

Cin

nam

omum

cac

hare

nsis

Mes

ua fe

rrea

Gar

cini

a xa

ntho

chym

us

Dip

tero

carp

us tu

rbin

atus

Cyn

omet

ra p

olya

ndra

Mag

noli

a ho

dgso

nii

Syzi

gium

cum

inii

Pal

aqui

um p

olya

nthu

m

14.58

6.65 6.35 5.707.648.21

41.45

0

15

30

60

45

8.73

55.32

39.49

21.1714.18 12.37 12.09 7.56 8.55 8.14

Am

oora

wal

lich

ii

Alp

hons

ea v

entr

icos

a

Chu

kras

ia ta

bula

ris

Sara

ca a

soca

Als

eoda

phne

ow

deni

i

Syzi

gium

cum

inii

Kay

ea fl

orib

unda

Too

na c

ilia

ta

Gyn

ocar

dia

odor

ata

Pal

aqui

um p

olya

nthu

m

Figure 2. Ten dominant species encountered in natural forest and paan jhum of the three study sites in Barak valley, Assam, NortheastIndia.

is not allowed inside paan jhum areas. Rather fast grow-ing species like Amra (Spondias pinnata), Simul (Bombaxceiba), Bajrang (Zanthoxylum rhetsa), Udal (Sterculia vil-losa) and Tula (Tetrameles nudiflora) are planted in gaps,if required. Hence, paan jhum practice helps to conservebiological diversity along with income generation.

Rahman et al. (2009) investigated the cultural andfinancial management technique and the conservation roleof paan (betel leaf) based agroforestry practiced by theKhasi community of Sylhet district of Bangladesh. Theyfound that the mean annual income from one hectare ofbetel leaf plantation was Tk.80979 (Tk.105.19 = 1C). The

Dow

nloa

ded

by [

Dr

Subr

ata

Nan

dy]

at 0

3:20

10

June

201

3

International Journal of Biodiversity Science, Ecosystem Services & Management 109

Table 2. Mean tree species richness, stand density (stems ha−1) and Shannon–Weaver index for different girth classes in natural forestand paan jhum of the study sites in Barak valley, Assam, Northeast India.

Natural forest Paan jhum

Site girthclass (cm)

Mean speciesrichness

Mean standdensity

Shannon–Weaverindex

Mean speciesrichness

Mean standdensity

Shannon–Weaverindex

1 30−60 190.00 275.00 2.68 140.00 210.00 2.8160−90 160.00 190.00 2.68 180.00 255.00 2.9090−120 120.00 130.00 2.62 130.00 150.00 2.78120−150 75.00 75.00 1.89 40.00 45.00 2.62150−180 45.00 45.00 1.98 50.00 60.00 2.04180−210 30.00 30.00 1.10 15.00 15.00 1.79210−240 35.00 35.00 1.79 35.00 35.00 1.95

>240 10.00 10.00 1.39 20.00 20.00 0.69

2 30−60 295.00 495.00 3.00 308.33 600.00 2.7960−90 115.00 195.00 2.43 133.33 166.67 2.2990−120 90.00 100.00 2.44 66.67 83.33 1.89120−150 30.00 30.00 1.56 41.67 41.67 1.61150−180 20.00 20.00 1.39 33.33 33.33 1.04180−210 10.00 10.00 0.69 16.67 16.67 0.69210−240 5.00 5.00 0.32 0.00 0.00 0.00

>240 10.00 10.00 0.69 0.00 0.00 0.00

3 30−60 93.75 125.00 2.80 175.00 279.17 2.9360−90 100.00 118.75 2.75 170.83 208.33 3.2090−120 93.75 115.63 2.81 141.67 187.50 3.07120−150 40.63 50.00 1.99 20.83 20.83 1.61150−180 43.75 43.75 2.35 29.17 29.17 1.75180−210 12.50 12.50 1.39 4.17 4.17 0.69210−240 34.38 40.63 2.10 16.67 20.83 1.05

>240 15.63 15.63 0.95 12.50 12.50 1.10

practice was proved to be a profitable business where thebenefit cost ratio was calculated to be 4.47. Moreover, thespecies composition in the betel leaf plantation area wasfound to be very promising to play a significant role inconservation of biological diversity, making the practice asustainable agroforestry system.

In the present study, the numbers of tree species inthe natural forest were 35, 32 and 42 belonging to 32,30 and 36 genera and 22, 23 and 28 families, respec-tively, in the three study sites. In paan jhum areas ofthe three study sites, it was observed that there were 47,37 and 48 tree species belonging to 41, 31 and 41 gen-era and 30, 24 and 32 families, respectively. The numberof species, genera and families were found to be higherin paan jhum areas compared with natural forest in allthe three sites. The native primary tropical forests, sec-ondary forests and sal plantations of the Garo hills inwestern Meghalaya had higher number of species (162,132 and 87 tree species belonging to 54, 53 and 37 families,respectively) than the present study (Kumar et al. 2006).Khumbongmayum et al. (2006) observed 96 woody speciesin four sacred groves dominated by subtropical forests inthe Imphal east and Imphal west districts of Manipur. Thespecies richness in undisturbed wet evergreen forest inArunachal Pradesh, eastern Himalayas, India, was found tobe 47 belonging to 28 families and 42 genera (Bhuyan et al.2003). Number of tree species in undisturbed, moderately

disturbed and highly disturbed tropical wet evergreen for-est in and around Namdapha National Park, ArunachalPradesh, India were found to be 50, 22 and 18 belongingto 41, 22 and 16 genera and 25, 16 and 16 families, respec-tively (Nath et al. 2005). A total of 82 species belongingto 59 genera and 39 families and 80 species belonging to62 genera of 41 families were recorded in 0.5 ha plots inthe Ialong and Raliang sacred groves of the Jaintia hills inMeghalaya, northeast India (Upadhaya et al. 2003). Hence,it was observed that in all the protected forests as well as inthe undisturbed forests, the number of species, genera andfamily are higher compared with unprotected forests.

The Shannon–Weaver index of all the paan jhumareas is higher (3.12 to 3.36) compared with the naturalforest (3.00 to 3.27) in the study sites. It shows that thistraditional agroforestry system helps in conserving thebiodiversity of this region. The Shannon–Weaver index ofpaan jhum areas correspond to the study done by Upadhayaet al. (2003) in Ialong and Raliang sacred groves, repre-sented by subtropical broad-leaved forest, of the Jaintiahills in Meghalaya, northeast India, where they founddiversity value of 3.42 and 3.55, respectively. Shannon–Weaver index for tree species in the Garo hills of westernMeghalaya were found to be 4.27, 3.78 and 2.47 forprimary forests, secondary forests and sal plantations,respectively (Kumar et al. 2006). Tree species diversityranged from 0.70 to 2.02 in highly disturbed, moderately

Dow

nloa

ded

by [

Dr

Subr

ata

Nan

dy]

at 0

3:20

10

June

201

3

110 S. Nandy and A.K. Das

Site 1

Site 2

Site 3Girth class (cm)

30–60 60–90 120–150 150–180 >240210–240180–21090–120

700

600

500

400

300

200

100

0

10

9

7

8

6

3

4

5

2

0

1

Tre

e de

nsity

(st

ems

ha–1

)Natural forest

300

250

200

150

100

50

030–60 60–90 120–150 150–180 >240210–240180–21090–120

Tre

e de

nsity

(st

ems

ha–1

)

14

12

10

8

6

4

2

0

Paan jhum

300

250

200

150

100

50

030–60 60–90 120–150 150–180 >240210–240180–21090–120

14

12

10

8

6

4

2

0

Bas

al a

rea

(m2 h

a–1)

-1)

30–60 60–90 120–150 150–180 >240210–240180–21090–120

700

600

500

400

300

200

100

0

10

9

7

8

6

3

4

5

2

0

1

Bas

al a

rea

(m2 h

a–1)

300

250

200

150

100

50

030–60 60–90 120–150 150–180 >240210–240180–21090–120

Tre

e de

nsity

(st

ems

ha–1

)

18

16

12

14

10

6

8

4

0

2

300

250

200

150

100

50

030–60 60–90 120–150 150–180 >240210–240180–21090–120

Bas

al a

rea

(m2 h

a–1)

18

16

12

14

10

6

8

4

0

2

Figure 3. Stand structure based on tree density (solid line) and basal area (dotted line) in natural forest and paan jhum of the study sitesin Barak valley, Assam, Northeast India.

Table 3. Species distribution pattern in natural forest and paanjhum of the study sites in Barak valley, Assam, Northeast India.

Distribution pattern (%)

Site Forest type Random Regular Contagious

1 Natural forest 38.30 51.06 10.64Paan jhum 42.86 51.43 5.71

2 Natural forest 28.13 62.50 9.38Paan jhum 16.22 67.57 16.21

3 Natural forest 9.52 26.19 64.29Paan jhum 4.17 31.25 64.58

disturbed, mildly disturbed and undisturbed stands of trop-ical wet evergreen forests of Arunachal Pradesh (Bhuyanet al. 2003). The diversity in undisturbed, moderately dis-turbed and highly disturbed tropical wet evergreen forest inand around Namdapha National Park, Arunachal Pradesh,India was found to be 1.46, 0.96 and 1.17, respectively(Nath et al. 2005). In the same region, where the presentstudy was carried out, Borah and Garkoti (2011) foundout that the Shannon–Weaver index was higher in undis-turbed forests (1.69 and 1.77) compared with disturbedforests (1.46 and 1.52) of Barak valley, south Assam.

Dow

nloa

ded

by [

Dr

Subr

ata

Nan

dy]

at 0

3:20

10

June

201

3

International Journal of Biodiversity Science, Ecosystem Services & Management 111

Shannon–Weaver index in undisturbed forest and protectedforests, like sacred groves and paan jhum, of this region ismuch higher than the unprotected forest, which shows therole of local tribes in conservation of biodiversity.

The basal area was also found to be higher in paan jhum(41.6 to 74.05 m2 ha−1) than the natural forest (40.5 to68.75 m2 ha−1) in the present study. The basal area ofpaan jhum areas are close to the basal cover, 57.46 and77.44 m2 ha−1, in Ialong and Raliang sacred groves,respectively, of the Jaintia hills of the Meghalaya state ofIndia (Upadhaya et al. 2003). Kumar et al. (2006) observedthat tree basal area (m2 ha−1) ranges from 29–162 in nativeprimary tropical forests, 12–151 in secondary forests and39–74 in sal plantations of the Garo hills in westernMeghalaya. Basal area (m2 ha−1) in undisturbed, moder-ately disturbed and highly disturbed tropical wet evergreenforest in and around Namdapha National Park, ArunachalPradesh, India was found to be 98.58, 21.38 and 7.81,respectively (Nath et al. 2005). The basal area (m2 ha−1)of undisturbed forests (36.88 and 42.12) was found to behigher than that of disturbed forests (9.47 and 16.96) ofBarak valley, south Assam (Borah and Garkoti 2011).Basal area also seems to be higher in sacred groves andpaan jhum compared with unprotected natural forests. Theprotection provided by the local tribes facilitates higherbasal cover in protected forests.

In the present study, the stand density in natural for-est ranged from 522–865 stems ha−1 and 763–934 stemsha−1 in paan jhum. Density of trees (30 cm gbh) perhectare in tropical forests ranged from a low value of245 (Ashton 1964), to intermediate values of 420–617 inBrazilian Amazon (Campbell et al. 1992) and to a highvalue of 639–713 in Central Amazonian upland forest(Ferreira and Prance 1998). The tree density (stems ha−1)in undisturbed forests (846 and 1110) was found to behigher than the disturbed forests (572 and 396) of the Barakvalley (Borah and Garkoti 2011). Tree density (stems ha−1)varies from 417–1023 in native primary tropical forests,620–1111 in secondary forests and 724–980 in sal plan-tations in the Garo hills of western Meghalaya (Kumaret al. 2006). It was also observed that in the present study,paan jhum has higher tree density per hectare than naturalforest in all the sites. Tree density (stems ha−1) in undis-turbed, moderately disturbed and highly disturbed tropicalwet evergreen forest in and around Namdapha NationalPark, Arunachal Pradesh, India were 610, 251 and 34,respectively (Nath et al. 2005). Khumbongmayum et al.(2006) observed tree density (stems ha−1) of 359–1218 infour sacred groves dominated by subtropical forests in theImphal east and Imphal west districts of Manipur. Treedensity (stems ha−1) was found to be 1476 and 938 intwo sacred groves of Ialong and Raliang, respectively, ofthe Jaintia hills of Meghalaya (Upadhaya et al. 2003).In the present study, species richness, stand density andShannon–Weaver index across girth classes in all the studysites showed a decreasing trend with increasing stem girthclasses. The same trend was observed by other studiesconducted in different parts of northeast India (Bhuyan

et al. 2003; Upadhaya et al. 2003; Khumbongmayumet al. 2006). Tree size class distribution in both thecategories of forest areas, namely paan jhum and naturalforests, followed a reverse J-pattern. The same trend wasobserved by Nath et al. (2005) in undisturbed tropical wetevergreen forests in Namdapha National Park, northeastIndia, which indicates good regeneration of the constituentspecies.

Comparative study of IVI in natural forest and paanjhum shows that Palaquium polyanthum and Dipterocarpusturbinatus has more IVI in paan jhum than the natural for-est. Both the species are evergreen and yields good timberused for making furniture. Because of the economic impor-tance, both the species have been exploited in the naturalforest, whereas being a protected area it is safe in paanjhum. One interesting observation was the occurrence ofMangifera sylvatica in paan jhum and its absence in natu-ral forest. The species is utilized as timber and fuelwood,and fruits are edible. It is medicinal also (Barbhuiya et al.2009). Due to its multiple uses, it is exploited very much inthe wild. Alseodaphne owdenii, which yields good qualitytimber for furniture, has higher IVI in paan jhum comparedwith natural forest. This tree species is in great demandin this locality and has been exploited heavily in its natu-ral condition. However, paan jhum acts as a conservationplot to protect the species. Most of the paan jhum areasshowed high species diversity than the natural forest, whichis possible because of the protection provided by the localtribes. As the people living nearby are not allowed to enterin paan jhum, the anthropogenic pressure on the speciesreduces. This facilitates in better conservation of naturalresources.

Agroforestry systems provide four major ecosystemservices: carbon sequestration, biodiversity conservation,soil enrichment and air and soil quality improvement(Jose 2009). The traditional agroforestry systems play avery significant role in conserving biodiversity (Khumaloet al. 2012). These agroforestry systems are complex sys-tems that have been developed to mimic natural systems.These practices benefit biodiversity through in situ con-servation of native tree species (de Souza et al. 2012).The positive impact of paan jhum agroforestry system onbiodiversity conservation can mainly be attributed to thereduced pressure on the natural forest due to the abilityof this agroforestry system to sustain the daily liveli-hood of the tribal community. Hence, paan jhum areasare socially, economically and ecologically viable systems,as this agroforestry practice provides economic supportto the Khasi tribes in addition to its ecological role forbiodiversity conservation.

Conclusions

Paan jhum forms a unique example of traditional farmingpractice which was adopted by the traditional society asa livelihood security and biodiversity conservation. Thistraditional agroforestry system can be a viable land-useoption which, in addition to alleviating poverty, offers a

Dow

nloa

ded

by [

Dr

Subr

ata

Nan

dy]

at 0

3:20

10

June

201

3

112 S. Nandy and A.K. Das

number of ecosystem services and environmental bene-fits. In the present study, it was found that the speciesrichness and diversity in paan jhum were higher than thenatural forests. The stand density and basal area were alsofound to be higher in paan jhum compared with the naturalforests. The species composition in the paan jhum area wasfound to play a significant role in conservation of biologi-cal diversity. Several indigenous tree species are protectedand managed through this traditional agroforestry sys-tem. Evergreen species like Dipterocarpus turbinatus andPalaquium polyanthum had a higher IVI in paan jhum com-pared with the natural forests. The occurrence of Mangiferasylvatica, a rare wild mango species, in paan jhum and itsabsence in natural forests indicates the role of paan jhumin species conservation.

Paan jhum offers viable options for combiningbiodiversity conservation and management for human ben-efits. Consequently, this agroforestry system might increas-ingly become a component of a forested landscape thatis valued for contributing towards both resource produc-tion and biodiversity conservation. It is also importantto realize the ecological services provided by such tra-ditional agroforestry systems through soil conservation,carbon sequestration and so on, which needs further study.It can be concluded that the paan jhum agroforestry sys-tems practiced in northeast India benefit from conserva-tion, management and maintenance of biodiversity. Thepresent study for the first time reported the paan jhumagroforestry system practiced by the Khasi tribes of Barakvalley, northeast India and its role in conservation.

AcknowledgementsThe authors are thankful to the tribal people of the study areafor their cooperation and help during the study. The authors alsothank two anonymous referees for their critical review and helpfulcomments on the manuscript.

ReferencesAnonymous. 1969. The wealth of India: a dictionary of Indian raw

materials and industrial products Vol III. New Delhi (India):Publications and Information Directorate, CSIR.

Anonymous. 2005. Unlocking opportunities for forest-dependentpeople in India Vol II. Washington (DC): World Bank. WorldBank Report No. 34481-IN.

Ashton PS. 1964. A quantitative phytosociological techniqueapplied to tropical mixed rainforest vegetation. Malays For.27:304–307.

Aumeeruddy Y, Sansonnens B. 1994. Shifting from the simpleto complex agroforestry systems: an example for buffer zonemanagement from Kerinci (Sumatra, Indonesia). AgroforestSyst. 28(2):113–141.

Barbhuiya AR, Sharma GD, Arunachalam A, Deb S. 2009.Diversity and conservation of medicinal plants in Barakvalley, northeast India. Indian J Tradit Knowl. 8(2):169–175.

Bareh H. 1967. The history and culture of the Khasi people. NewDelhi (India): Kluwer Academic Publishers.

Bhuyan P, Khan ML, Tripathi RS. 2003. Tree diversity and pop-ulation structure in undisturbed and human-impacted standsof tropical wet evergreen forest in Arunachal Pradesh, easternHimalayas, India. Biodivers Conserv. 12(8):1753–1773.

Borah N, Garkoti SC. 2011. Tree species composition, diver-sity, and regeneration patterns in undisturbed and disturbedforests of Barak valley, south Assam, India. Int J Ecol EnvSci. 37(3):131–141.

Campbell DG, Stone JL, Roses A. 1992. A comparison of thephytosociology of three floodplain (varzea) forests of knownages, Rio Jurua, Western Brazilian Amazon. Bot J Linn Soc.108(3):213–237.

Chagule BA. 1960. Betelvine cultivation in India. New Delhi(India): Ministry of Food and Agriculture.

Champion HG, Seth SK. 1968. A revised survey of the for-est types of India. New Delhi (India): Government of IndiaPress.

Chase AK. 1989. Domestication and domiculture in north-ern Australia: a social perspective. In: Hillman GC, editor.Foraging and farming, the evolution of plant extraction.London (UK): Unwin Hyman. p. 42–54.

Curtis JT. 1959. The vegetation of Wisconsin Univ. Madison(WI): Wisconsin Press.

Curtis JT, Cottam G. 1956. Plant ecology work book. Laboratoryfield reference manual. Minneapolis (MN): Burgess Publ. Co.

Curtis JT, McIntosh RP. 1950. The interrelations of certain ana-lytic and synthetic phytosociological characters. Ecology.31(3):434–455.

de Souza HN, de Goede RGM, Brussaard L, Cardoso IM, DuarteEMG, Fernandes RBA, Gomes LC, Pulleman MM. 2012.Protective shade, tree diversity and soil properties in coffeeagroforestry systems in the Atlantic Rainforest biome. AgricEcosyst Environ. 146(1):179–196.

Denevan WM, Treacy JM, Alcorn JB, Padoch C, Denslow J,Paitan SF. 1984. Indigenous agroforestry in the PeruvianAmazon: Bora Indian management of swidden fallows.Interciencia. 9(6):346–357.

Ferreira LV, Prance GT. 1998. Species richness and floristiccomposition in four hectares in the Jau National Park inupland forests in Central Amazonia. Biodivers Conserv.7(10):1349–1364.

Fujisaka S, Wollenberg E. 1991. From forest to agroforestsand logger to agroforester: a case study. Agroforest Syst.14(2):113–129.

Jose S. 2009. Agroforestry for ecosystem services and environ-mental benefits: an overview. Agroforest Syst. 76(1):1–10.

Khumalo S, Chirwa PW, Moyo BH, Syampungani S. 2012. Thestatus of agrobiodiversity management and conservation inmajor agroecosystems of Southern Africa. Agric EcosystEnviron. 157:17–23.

Khumbongmayum AD, Khan ML, Tripathi RS. 2006.Biodiversity conservation in sacred groves of Manipur,northeast India: population structure and regeneration statusof woody species. Biodivers Conserv. 15(8):2439–2456.

Kumar A, Marcot BG, Saxena A. 2006. Tree species diversity anddistribution patterns in tropical forests of Garo hills. Curr Sci.91(10):1370–1381.

Mary F, Michon G. 1987. When agroforests drive back natu-ral forests: a socio-economic analysis of a rice-agroforestrysystem in Sumatra. Agroforest Syst. 5(1):27–55.

Misra R. 1968. Ecology work book. New Delhi (India): Oxfordand IBH Publishing Co. Pvt. Ltd.

Mueller-Dombois D, Ellenberg H. 1974. Aims and methods ofvegetation ecology. New York (NY): John Wiley and Sons.

Nath PC, Arunachalam A, Khan ML, Arunachalam K, BarbhuiyaAR. 2005. Vegetation analysis and tree population struc-ture of tropical wet evergreen forests in and aroundNamdapha National Park, northeast India. Biodivers Conserv.14(9):2109–2136.

Nath TK, Makoto I, Islam MJ, Kabir MA. 2003. The Khasi tribeof north-eastern Bangladesh: their socio-economic status, hillfarming practices and impacts on forest conservation. ForTrees Livelihoods. 13(4):297–311.

Dow

nloa

ded

by [

Dr

Subr

ata

Nan

dy]

at 0

3:20

10

June

201

3

International Journal of Biodiversity Science, Ecosystem Services & Management 113

Olofson H. 1980. An ancient social forestry. Sylvatrop,Philippines For Res J. 5(4):255–262.

Pandey DN. 2007. Multifunctional agroforestry systems in India.Curr Sci. 92(4):455–463.

Posey DA. 1985. Indigenous management of tropical forestecosystems: the case of the Kayapo Indians of the BrazilianAmazon. Agroforest Syst. 3(2):139–158.

Rahman M, Rahman MM, Islam M. 2009. Financial viabil-ity and conservation role of betel leaf based agroforestry:an indigenous hill farming system of Khasia community inBangladesh. J For Res. 20(2):131–136.

Ramakrishnan PS. 1998. Conserving the sacred: where do westand? In: Ramakrishnan PS, Saxena KG, ChandrashekaraUM, editors. Conserving the sacred: for biodiversity man-agement. New Delhi (India): Oxford and IBH Publishing Co.Pvt. Ltd. p. 439–455.

Shannon CE, Weaver W. 1949. The mathematical theory ofcommunication. Urbana (IL): Urbana University Press.

Upadhaya K, Pandey HN, Law PS, Tripathi RS. 2003. Treediversity in sacred groves of the Jaintia hills in Meghalaya,northeast India. Biodivers Conserv. 12(3):583–597.

Whitford WG. 1949. Distribution of woodland plants in relationto succession and clonal growth. Ecology. 30(2):199–208.

Wiersum KF. 1997a. Indigenous exploitation and managementof tropical forest resources: an evolutionary continuum inforest-people interaction. Agri Ecosyst Environ. 63(1):1–16.

Wiersum KF. 1997b. From natural forest to tree crops, co-domestication of forests and tree species, an overview. NethJ Agric Sci. 45:425–438.

Wiersum KF. 2003. Use and conservation of biodiversity inEast African forested landscapes. In: Zuidema P, editor.Tropical forests in multi-functional landscapes. Utrecht (TheNetherlands): Seminar Proceedings. Prins Bernard Centre,Utrecht University.

Wiersum KF. 2004. Forest gardens as an ‘intermediate’ land-usesystem in the nature-culture continuum: characteristics andfuture potential. Agroforest Syst. 61(1–3):123–134.

Withrow-Robinson B, Hibbs DE, Gypmantasiri P, Thomas D.1998. A preliminary classification of fruit-based agroforestryin highland area of northern Thailand. Agroforest Syst.42(2):195–205.

Wong C. 2011. Guidance for the preparation of ESTR products –classifying threats to biodiversity. Canadian biodiversity:ecosystem status and trends 2010. Technical ThematicReport No. 2. Ottawa (ON): Canadian Councils of ResourceMinisters. iii + 30 p.

Dow

nloa

ded

by [

Dr

Subr

ata

Nan

dy]

at 0

3:20

10

June

201

3