Content uploaded by Shrinidhi Ambinakudige
Author content
All content in this area was uploaded by Shrinidhi Ambinakudige on Nov 09, 2016
Content may be subject to copyright.
ORIGINAL PAPER
Comparing tree diversity and composition in coffee
farms and sacred forests in the Western Ghats of India
Shrinidhi Ambinakudige ÆB. N. Sathish
Received: 20 December 2007 / Accepted: 8 October 2008 / Published online: 19 October 2008
ÓSpringer Science+Business Media B.V. 2008
Abstract Expansion of coffee cultivation is one of the causes of deforestation and bio-
diversity loss. However, shade grown coffee has been promoted as a means for preserving
biodiversity in the tropics. In this study we compared tree diversity in two types of coffee
management regimes with the sacred groves in the Western Ghats of India. We computed
species accumulation curves, species diversity indices and evenness indices to compare the
different management regimes. Results of diversity indices showed that shade coffee had
less diversity compared to sacred groves. Exotic species dominated the tree diversity in
lands where the tree harvesting rights are with the growers. Native trees dominated the tree
diversity when growers had no ownership rights on trees. A species accumulation curve
suggested that the sacred grove had higher species richness compared to other two habitats.
Lack of incentive to preserve endemic species as shade trees is forcing growers to plant
more exotic species in shade grown coffee plots. If encouraged, shade grown coffee can
preserve some biodiversity, but cannot provide all ecological benefits of a natural forest.
Keywords Biodiversity Coffee India Land tenure Sacred groves
Species dominance
Introduction
Agriculture is a major livelihood for millions of people in the world. At the same time
widespread agricultural development is one of the major causes of deforestation and
biodiversity loss (Ehrlich 1988; Gorenflo and Brandon 2003; Chichilnisky 1994; Benhin
2006). Conservationists are concerned about the impacts of cultivation of commercial
S. Ambinakudige (&)
Department of Geosciences, Mississippi State University, 355 Lee Blvd, Mississippi State,
MS 39762, USA
e-mail: ssa60@msstate.edu
B. N. Sathish
College of Forestry, Ponnampet, Kodagu, Karnataka, India
e-mail: satibn79@rediffmail.com
123
Biodivers Conserv (2009) 18:987–1000
DOI 10.1007/s10531-008-9502-5
crops like coffee in and around biodiversity hotspots (Hardner and Rice 2002). Recently,
much attention has been focused on measuring tree, ants and bird biodiversity loss due to
agricultural intensification, particularly in coffee agro-ecosystems (Bhagwat 2002;
Perfecto and Vandermeere 2002; Perfecto et al. 1996,2007; Moguel and Toledo 1999;
Philpott et al. 2007,2008). Coffee can be grown under sun or under shade. Highly shaded
farms provide important habitat refuges for biodiversity in some regions (Perfecto et al.
1996,2007). Shade grown coffee may more closely resemble the natural environment
(Geist and Lambin 2001; Jime
´nez-Avila and Martı
´nez 1979; Perfecto et al. 2005; Moguel
and Toledo 1999; Soto-Pinto et al. 2001). Furthermore, type and management of shade
strongly affect species richness (Perfecto and Snelling 1995; Perfecto et al. 1996; Moguel
and Toledo 1999; Blackman et al. 2005). A substantial literature on shade grown coffee in
different parts of the world has shown the superior conservation value of shade-grown
coffee (Perfecto et al. 1996; Moguel and Toledo 1999; Blackman et al. 2005; Jime
´nez-
Avila and Martı
´nez 1979; Soto-Pinto et al. 2001) compared to the sun grown coffee.
However, perception of farmers that increasing shade cover in farms will diminish yields
results in reducing shade density and diversity sometimes provokes losses of mammal,
bird, and arthropod diversity (Philpott and Armbrecht 2006; Perfecto et al. 2007). Fur-
thermore, researchers have also argued that species composition in shade-grown coffee
farms is not identical to that of forest (Soto-Pinto et al. 2001; Rappole et al. 2003).
Shade coffee has been ambitiously promoted by conservationists and the scientific
organizations as a means for preserving biodiversity in tropics (Conservation International
2001; National Audubon Society 2000; Philpott and Dietsch 2003). Coffee is a highly
traded commodity in the world next only to oil in value. For some countries it is the most
important source of foreign exchange (Rice and Ward 1997; Ambinakudige 2006). Today
there are about 29 million coffee drinkers in the world (Ambinakudige 2006). Presence of
this large number of coffee consumers encourages conservationists to promote shade
coffee over sun coffee. This generates enormous pressure on growers to produce shade
coffee with the goal of increased biodiversity conservation (Rappole et al. 2003).
Understanding the context of these pressures and impacts of shade coffee on biodi-
versity can inform policy decisions that influence the livelihoods of millions of coffee
growers and workers throughout the developing world. Coffee is an international com-
modity, and land-use decisions by coffee growers depend on the international coffee
market as well as national and local forest policies (Blackman et al. 2005; Ambinakudige
2006; Sathish 2005; Verbist et al. 2005). Livelihood advocators argue that if the interna-
tional niche market for shade coffee is recognized for its role in conservation of
biodiversity, millions of growers and workers will benefit by premium price aimed at
conserving biodiversity (Ambinakudige 2006). To meet the goals of biodiversity conser-
vation and improving the livelihoods of coffee growers and workers, more studies are
needed to document how biodiversity patterns and shade coffee production coincide in
regions where coffee is an important export.
In this study, we analyzed the impacts of different types of tree ownership rights on tree
diversity and composition in shade coffee and compared it with sacred groves in the Western
Ghats of India. Within shade coffee plots redeemed lands and unredeemed lands are the two
tree ownership regimes analyzed in this study to document the influence of tree ownership on
the decision of the coffee growers to maintain particular types of tree diversity. In redeemed
land, land owner holds the tree right. In unredeemed land, government owns the tree even
though the land is owned by the coffee grower (Shrinidhi and Lele 2001). We compared tree
species diversity and distribution within these two types of shade grown coffee lands with
adjacent sacred groves. Sacred groves are patches of forests where human impact is minimal.
988 Biodivers Conserv (2009) 18:987–1000
123
These lands are set aside for village communal activities (Ambinakudige 2006; Kushalappa
and Bhagwat 2001). This comparative study provides a broader picture of the human
dimensions of biodiversity change due to coffee cultivation. It also informs region-specific
agricultural policy as to the pros and cons utilizing shade grown coffee to ameliorate habitat
decline.
Methods
This study was conducted in the Kodagu district of Karnataka state in the Western Ghats of
the Southern Indian peninsula (Fig. 1). Kodagu is located on the slopes of Western Ghats
Mountains. The Western Ghats region of India is one of the 34 global biodiversity hotspots
in the world dedicated as such due to high levels of endemism and endangered flora and
fauna (Mittermeier et al. 2005). The Western Ghats are located between 8°220N and
20°400N latitude and 73°–77°E longitude (25). Kodagu was selected because it is one of
the major coffee growing regions in India. Both Coffea arabica (Arabica coffee) and
Coffea canephora (Robusta coffee) are grown in Kodagu. Large areas of forests have been
converted to shade grown coffee cultivation in the last century in Kodagu (Ambinakudige
2006; Menon and Bawa 1998; Lele 2001). Traditionally in Kodagu, rice is grown for
subsistence. Apart from coffee, black pepper, cardamom, orange and ginger are the other
commercial crops of the region. Of all these crops, coffee played a major role in socio-
economic development of Kodagu. The State supported and nourished the Indian coffee
sector for many years by providing subsidies and extension services. When the Coffee
Board of India allowed an open market for coffee in early 1990s, growers received higher
prices. This encouraged coffee growers to extend the area under coffee into lands that were
assigned to supplement rice cultivation and into the government-controlled forestlands.
However, the recent slump in the international coffee market has weakened the local
economy of Kodagu (Ambinakudige 2006).
Coffee was introduced to Kodagu on a commercial scale by British planters in 1854.
Later, native farmers began coffee cultivation. This introduced more severe transformations
Fig. 1 Study area in Kodagu region of the Western Ghats
Biodivers Conserv (2009) 18:987–1000 989
123
of the landscape (Elouard 2000; Bhagwat 2002). The French Institute of Pondicherry found
that the increase in the area under coffee cultivation between 1977 and 1997 resulted in a net
loss of forest habitat. Although shaded coffee plantations resemble the original forest they
have a relatively high biodiversity (Moguel and Toledo 1999; Perfecto et al. 1996). Only
about 36% of the land remains covered by forest in Kodagu. This is a drastic reduction from
88% observed in 1920. Around 71% of this forest loss is due to coffee cultivation (Menon and
Bawa 1998). However, because of the shade trees interspersped within the coffee, there is
still a substantial amount of tree cover.
To analyze the biodiversity changes due to coffee cultivation in Kodagu, one would
ideally compare biodiversity found in coffee habitats to that found in habitats where the
biodiversity has not been affected by human impacts. However, it is difficult to find a
‘‘pristine’’ landscape anywhere in the world that has no human impact. In the study area,
the forestlands under state control are highly managed and planted with the commercial
tree species like teak. The only landscapes with minimal impacts are the region’s sacred
groves.
In many parts of Asia, Africa, Europe, America and Australia people preserve a section
of natural environment as sacred groves for various social, cultural, religious and eco-
logical purposes (Rappole et al. 2003). In India, sacred groves are located in varieties of
habitats ranging from heavily forested areas of Western Ghats to the deserts of Rajasthan.
Several studies (Kushalappa and Bhagwat 2001; Verbist et al. 2005; Chandran et al. 1998)
have demonstrated the importance of sacred groves in biodiversity conservation.
In Kodagu, the state forest department’s estimates show that there are 1,214 sacred
groves covering an area of 6,375 ha (i.e., about 1.6% of total geographical area of Kod-
agu). These patches are locally referred as Devarkaadus—which could be literally
translated as God’s forests. Today many of the sacred groves have been encroached for
coffee cultivation and for human settlement. Today out of 97 big sacred groves (more than
2 ha in size), 9 have been fully encroached, 30 partially encroached, 40 are intact and 9 are
improved by the forest department (Kushalappa and Kushalappa 1998). To quantify the
impacts of tree ownership types on tree diversity in the Kodagu region of the Western
Ghats, we compared tree diversity and distribution within shade grown redeemed and
shade grown unredeemed coffee lands with that contained within sacred groves.
Vegetation sampling
Sampling plots in coffee plots and the least disturbed sacred groves were selected in the
southeastern part of Kodagu (Fig. 1). The average altitude is 900 m and average rainfall is
about 2,000 mm. The soils in the area are classified as red loamy (FRLHT 2000). Sampling
plots in coffee and sacred groves had relatively similar topography.
Based on the list of sacred groves collected from the forest department records, 15
sacred groves in the study area were initially inspected. Seven sacred groves out of 15
found were less disturbed and eight groves were completely or partially encroached for
coffee cultivation or housing. Seven undisturbed sacred groves, seven redeemed coffee
plots and seven unredeemed coffee plots were selected for vegetation sampling. People in
and around these sacred groves were interviewed using open-ended questions. The ques-
tions included the family structure, land holdings, type of tree ownership (redeemed or
unredeemed), and shade management practices.
Randomly located rectangular quadrants of 20 920 m were deployed in the sacred
forests. Other studies in the Western Ghats and elsewhere in tropics have used sample plots
ranging from 10 910 to 10 950 m (Condit 1994; Gentry 1990; Daniels et al. 1995;
990 Biodivers Conserv (2009) 18:987–1000
123
Chandran 1993). Due to lower density of trees within coffee plantations, larger sample
plots of 25 950 m size were used.
Seven plots in each of the three land tenure categories were sampled. A nylon rope of
50 m length was used to mark the central transect in coffee plots. A rope of 20 m in length
was used in sacred groves. A 10 m rope (sacred groves) and 12.5 m rope (coffee plots)
were used to measure the distance of each tree from the central line of transect. All trees
greater than or equal to 10 cm diameter at breast height (DBH) were enumerated and
identified. DBHs were measured using a measuring tape. Pascal and Ramesh’s (1987) field
key was used for identification of trees (Bhagwat 2002; Daniels et al. 1995; Utkarsh et al.
1998). To measure the structural complexity and heterogeneity, stem density per hectare
and basal area for hectare were calculated. Forest structural heterogeneity was determined
by the variability in basal area of trees in a given forest stand.
To analyze the difference in the composition of the tree species in three land habitats,
species importance, species relative density and species relative dominance were
calculated.
Species relative density ¼Number of trees in each species
Total number of trees 100
Species relative ðbasal dominanceÞdominance ¼Basal area of each species
Total BA of all species 100
Species importance is the average of relative density and relative dominance. Density can
be very high for a species, but if the individuals are small then dominance may be low.
Conversely, a species with a very high dominance may be very low in density. By taking
the average of these two values we get a better idea of the size and density of vegetation
(Iverson and Prasad 2002; Dallmeier and Comiskey 1998).
Tree species found only in a single land tenure type are classified as ‘unique species’,
those shared by any two of the three tenures were classified as ‘shared species’ and those
found in all three types are referred to as ‘ubiquitous species’ (Bhagwat 2002). A series of
v
2
tests were conducted to test the significance of differences in the frequencies of unique
species of trees in three habitats of differing land tenure.
To compare tree richness in redeemed coffee, unredeemed coffee and sacred groves, we
generated sample-based rarefaction curve (MaoTao estimates) using EstimateS (Colwell
and Coddington 1994;http://www./viceroy/eed/uconn/.edu/estimates). To compare the
richness between the habitats we rescaled the sample based rarefaction curves to the
number of individual (Gotelli and Colwell 2001).
To compare the species diversity in terms of species richness and species evenness in
shade grown coffee and the sacred groves, we used Simpson’s reciprocal index (Magurran
2004; Gimaret-Carpentier et al. 1998) and the Shannon Wiener index (Magurran 2004;
Gimaret-Carpentier et al. 1998).
The Shannon diversity index (H) is commonly used index to characterize species
diversity in a community. Shannon’s index accounts for both abundance and evenness of
the species present. The Shannon index is calculated from the equation:
H¼X
S
i¼1
piln pi
The quantity (p
i
) is the proportion of individuals found in the ith species. A higher value
of Hindicates high species diversity in the sample (Magurran 2004).
Biodivers Conserv (2009) 18:987–1000 991
123
Shannon’s evenness is calculated as
ji¼H
Hmax
where H, Shannon index. Equitability assumes a value between 0 and 1 with 1 being
complete evenness (Magurran 2004).
Simpson’s reciprocal index is often used to quantify the biodiversity of a habitat. It
takes into account the number of species present, as well as the abundance of each species
(Magurran 2004). Simpson’s Index (D) measures the probability that two individuals
randomly selected from a sample will belong to the same species.
Simpson’s index is calculated as
D¼Xniðni1Þ
NðN1Þ
where n
i
is the number of individuals in ith species, and Nis the total number of indi-
viduals. As Dincreases, diversity decreases. Simpson index is therefore usually expressed
as 1 -Dor 1/D(Dallmeier and Comiskey 1998). It captures the variance of the species
abundance distribution. Value of the complement (1 -D) or reciprocal (1/D)ofDwill rise
as the assemblage becomes more even (Dallmeier and Comiskey 1998). We used reci-
procal of Din this study. We can measure the evenness by dividing the reciprocal form of
the Simpson’s index by the number of species in the sample (Magurran 2004).
E1=D¼ð1=DÞ
S
where E
1/D
is Simpson’s evenness. S, number of species in the sample and 1/Dis the
reciprocal. The Simpson’s evenness measurement ranges from 0 to 1 and is not sensitive to
species richness (Magurran 2004).
A randomization test as described by Solow (1993) was used to compare diversity indices
(both Shannon’s and Simpson’s) in all the three land tenure classes. Species diversity and
richness software was used to conduct randomization test. In this test, three data sets were
combined into a single set. The combined data set was randomly partitioned into three
subsets. Shannon and Simpson diversity indices were calculated for each subset and the
differences in the indices values between land tenure classes were recorded. The procedure
was repeated 10,000 times and the estimated P-value was calculated. The number of sim-
ulations producing a difference in the diversity indices greater than or equal to that observed
estimates the probability that any observed difference was due to chance (Solow 1993).
Qualitative methods
To understand the human dimensions of the tree diversity change in coffee and the sacred
groves a study of archival materials on the issues of coffee cultivation and the sacred
groves in the region was conducted. The intent of this archival work was to reveal the
history of management decisions in coffee and sacred groves. In addition, semi-structured
interviews were conducted to provide detail on the role of forest policies and coffee market
on people’s land use decisions (Gimaret-Carpentier et al. 1998). Five knowledgable
individuals in the study area were interviewed. Knowledgable individuals are those persons
who have lived in the area for long time and are involved in coffee cultivation. We asked
villagers to guide us to most experienced, and or older persons in the community. In
992 Biodivers Conserv (2009) 18:987–1000
123
particular, interviews sought to distil more information about the preferred tree species in
different land tenures. State policies regulating harvests among different tenures and its
influence on the decisions on retaining of any particular species were also recorded. The
interviews focused solely on collecting qualitative information and not quantitative
information. All the interviews were recorded and transcribed. Data were then coded and
analyzed using QSR N6 qualitative data analysis software.
Results
A total of 109 different species were found in all tenures together (Fig. 2). Sacred groves
had the highest number of species (62 species). Redeemed and unredeemed coffee plots
had nearly the same number of species. Thirty-eight species were recorded in redeemed
plots and 39 species were recorded in unredeemed coffee (Fig. 2). Basal area was highest
in sacred groves. Total basal area for sacred groves was 31.54 m
2
ha
-1
. While redeemed
coffee recorded 18.91 m
2
ha
-1
, unredeemed coffee recorded 15.59 m
2
ha
-1
as the total
basal areas (Table 1).
In the sampled plots, 4.5% of all species were ubiquitous, 18% were shared, and 76%
were unique (Fig. 2). The v
2
test revealed that observed number of unique species was
higher than expected in the sacred grove but lower than expected by chance in t
he redeemed coffee and unredeemed coffee (v
2
test, v
2
=22.65, df =2, P=0.000).
Fig. 2 Distribution of ubiquitous, shared and unique species in three habitat types differing in land tenure
classes
Biodivers Conserv (2009) 18:987–1000 993
123
Table 1 Species importance, relative density and dominance in three land tenure classes
Unredeemed coffee lands Redeemed coffee lands Sacred Groves
Species Species count Species Species count Species Species count
Dalbergia latifolia 73 Grevillea robusta 96 Dimocarpus longan 60
Grevillea robusta 28 Acrocarpus fraxinifolius 38 Acronychia pedunculata 40
Citrus reticulata 20 Erythrina suberosa 31 Mimisilon malabarica 37
Species Species importance Species Species importance Species Species importance
Dalbergia latifolia 27.9% Grevillea robusta 26.3% Dimocarpus longan 13.4%
Grevillea robusta 9.2% Acrocarpus fraxinifolius 12.8% Acronychia pedunculata 8.1%
Spathodea companulata 8.3% Erythrina suberosa 8.5% Mimisilon malabarica 7.6%
Species Relative density Species Relative density Species Relative density
Dalbergia latifolia 30.5% Grevillea robusta 29.1% Dimocarpus longan 15.7%
Grevillea robusta 11.7% Acrocarpus fraxinifolius 11.5% Acronychia pedunculata 10.5%
Citrus reticulata 8.4% Erythrina suberosa 9.7% Mimisilon malabarica 9.7%
Species Relative dominance Species Relative dominance Species Relative dominance
Dalbergia latifolia 25.3% Grevillea robusta 23.5% Dimocarpus longan 13.4%
Ficus racemosa 9.9% Acrocarpus fraxinifolius 14.0% Margaritaria Indica 14.2%
Spathodea companulata 9.5% Ficus racemosa 8.3% Mangifera indica 4.0%
Total stand density per ha 273 Total stand density per ha 367 Total stand density per ha 1,361
994 Biodivers Conserv (2009) 18:987–1000
123
The v-squared test also revealed that the difference between number of unique species in
redeemed coffee and sacred groves (v
2
=0.4.27, df =1, P=0.039) and between
unredeemed coffee and sacred groves (v
2
=2.89, df =1, P=0.089) are significant.
Both redeemed and unredeemed coffee plots had nearly the same number of observed and
expected unique species. There is no significant difference between the distribution of
unique species between redeemed and unredeemed coffee (v
2
test, v
2
=0.140, df =1,
P=0.70).
According to species accumulation curve (Fig. 3) and 95% confident intervals, there are
significantly more species of trees in sacred groves compared to redeemed lands. Similarly,
unredeemed coffee lands had significantly higher number of species per number of indi-
viduals compared to redeemed coffee plots. However, between sacred groves and
unredeemed lands species accumulation per number of individual trees did not differ
significantly (Fig. 3).
Differences in species composition in three land tenure systems
Sacred groves had the highest average tree stand density (1,361 ha
-1
±599 stems ha
-1
)
compared to in redeemed (377 ha
-1
±112 stems ha
-1
) and unredeemed (273 ±96 ha
-1
)
lands (Table 2). Different species were dominant in each of the land tenures. In redeemed
coffee lands, (Table 2)Gravellia robusta was the important species (sp. impor-
tance =26.3%, relative density =29.1%, relative dominance =23.5%). In unredeemed
coffee lands, Dalbergia latifolia was important (sp. importance =27.9%, relative den-
sity =30.5%, relative dominance =25.3%) and in sacred groves, Dimocarpus longan had
the highest species importance and density (sp. importance =13.4%, relative den-
sity =15.7%, relative dominance =13.4%). In sacred groves, Mangifera indica species
had the highest relative dominance (14%).
Gravellia robusta is non-indigenous exotic species in the region, and was introduced
mainly as a shade tree in the coffee plantations. D. latifolia and D. longan are endemic
species. The dominant species in unredeemed lands, D. latifolia, is a deciduous species.
D. latifolia (Rosewood) is also one of the reserved trees (can only harvested under special
license) in the state and has high commercial value. G. robusata, the dominant species in
redeemed plots is an exotic species. The dominant species in sacred groves, D. longan and
M. indica are evergreen species. Thirteen percent of all species in redeemed coffee plots
0
20
40
60
80
100
120
140
160
180
200
# Individuals
# Species
Unredeemed Coffee Redeemed Coffee Sacred Groves
1000
600 8004002000
Fig. 3 Species accumulation
curves for trees in redeemed,
unredeemed coffee plots and the
sacred groves. Accumulation
curves values were extracted in
EstimateS and show Mao Tao
estimates. Error bars show 95%
interval, nonoverlapping bars
show statistical differences
Biodivers Conserv (2009) 18:987–1000 995
123
were reserved trees. Unredeemed coffee also had 13% reserved trees, while this value
increased to 16% in sacred groves. Forest department list them as reserved trees because of
their rarity and thus the higher value. These reserved trees are properties of the govern-
ment. Land owners need a special permission from the government to harvest these trees.
Diversity indices
The Shannon Wiener indices (H) for redeemed coffee, unredeemed coffee and sacred
groves were 2.68, 2.71 and 3.29 respectively. The higher Hvalue in sacred grove indicates
higher diversity. The randomization test (Solow 1993) used to compare the Hvalues in all
three land tenures indicated that the sacred groves are more diverse than both redeemed
coffee and unredeemed coffee at 95% confident interval. However, the redeemed and
unredeemed coffees had 70% probability of having the same diversity at 95% confident
interval. Whereas the there is only 0.46% probability that sacred groves are less diverse
than the redeemed and unredeemed coffee combined. The Simpson’s reciprocal index (D)
also showed that the sacred groves are more diverse than both the redeemed and unre-
deemed coffee plots. The Dvalue for sacred groves is 16.05, whereas for redeemed coffee
Dis 8.09 and for unredeemed coffee Dis 7.98.
No difference in evenness was observed among the three land tenures. The Shannon’s
evenness measure (j0) for three land use classes were 0.46, 0.49 and 0.55 for redeemed
coffee, unredeemed coffee and sacred groves respectively. The evenness measure assumes
a value between 0 and 1 with 1 being complete evenness (Magurran 2004). This indicates
Table 2 Results of randomization test using 10,000 random partitions on Shannon and Simpson indices in
three land tenure classes
Simpson’s index Shannon index
Unredeem coffee and scared grove Unredeemed coffee and scared grove
Observed, unredeemed 7.98 Observed, unredeemed 2.71
Observed, sacred grove 16.05 Observed, sacred grove 3.29
Difference (delta) 8.06 Difference (delta) 0.57
Number of simulated |delta| [
observed |delta|
276 Number of simulated |delta| [
observed |delta|
0
P0.03 P0.00
Redeemed coffee and scared grove Redeemed coffee and scared grove
Observed, redeemed 8.09 Observed, redeemed 2.68
Observed, sacred grove 16.05 Observed, sacred grove 3.29
Difference (delta) 7.96 Difference (delta) 0.61
Number of simulated |delta| [
observed |delta|
182 Number of simulated |delta| [
observed |delta|
0.00
P0.02 P0.00
Redeemed coffee and unredeem coffee Redeemed coffee and Unredeem coffee
Observed, unredeemed 7.98 Observed, unredeemed 2.71
Observed, redeemed 8.09 Observed, redeemed 2.68
Difference (delta) 0.11 Difference (delta) 0.03
Number of simulated |delta| [
observed |delta|
9,462 Number of simulated |delta| [
observed |delta|
8,067
P0.94 P0.81
996 Biodivers Conserv (2009) 18:987–1000
123
that there is not much difference in the evenness among the three land tenures. Simpson’s
evenness (E
1/D
) also indicated that the three tenures had similar evenness. Simpson’s
evenness (E
1/D
) for sacred grove was 0.25, for redeemed coffee was 0.21 and for unre-
deemed coffee was also 0.21.
The randomization test results to compare the significance of Shannon and Simpson
indices in three land tenure classes are in Table 2. This test indicated that in both Shannon
and Simpson indices, sacred groves are significantly more diverse than redeemed and
unredeemed coffee. The test also indicated that there is no significant difference between
tree diversities in redeemed and unredeemed coffee lands (Table 2).
Results of archival records and the interviews
Historical data collected from the administrative reports of various government department
since British annexation of the region (Haller 1910) revealed information on the expansion
pattern of coffee in the region. Records indicated that initially when the coffee was
introduced, growers completely cleared the forest to plant coffee. However, coffee plants
could not survive heavy monsoon rains, so shade grown coffee became the cultivation
practice in this area. Gravillea robusta was introduced to Kodagu coffee plots around
1860s to protect coffee from the attack of leaf rust which had already devastated the coffee
cultivation in Sri Lanka (Haller 1910). Gravillea robusta plants were also promoted by the
Coffee Board of India as shade trees in the coffee growing region. Over the years G.
robusta became a dominant shade tree in redeemed coffee. Since the tree ownership in
redeemed lands rests with the land owner, once the trees are harvested, fast growing and
high timber value exotic species like G. robusta are preferred and planted. Coffee growers
in Kodagu intercrop black pepper among shade trees. Gravillea robusta grows tall and
straight and acts as stand for pepper vines.
In unredeemed coffee lands tree ownership rests with the government, so tree harvesting
is not permitted. However, farmers still plant G. robusta in the areas where there are fewer
shade trees. As a result G. robusta is the second dominant species in unredeemed coffee. In
sacred groves, G. robusta was not recorded. We also noticed some sacred groves not
included in the study, that were completely or partially subsumed by coffee plantations and
housing suggesting that G. robusta may be in these border zones.
The interviews of the knowledgeable individuals also revealed to us the possible
relationship between the coffee crisis in mid 1990s and the preference for exotic species.
Global coffee market witnessed lowest coffee prices in 30 years. Coffee producers though
out the world, unable to make a living with coffee, are changing to other crops or aban-
doning their plantations (Perfecto et al. 2005). A similar trend was observed in the study
area also. Some producers started growing ginger a short term commercial crops to cope
with the coffee crisis. Though we need a separate study to decisively make the claim about
the lower coffee price and the producer’s preference for shade trees in coffee, the
knowledgeable individuals are in the opinion that the people are planting more exotic trees
hoping it will insurance them during the future coffee crisis. Indian coffee market once
regulated by the Coffee Board of India now is an open market. Coffee producers especially
the small producers who have little or no experience in the international market are looking
alternative to survive during the period of lower coffee price. The one alternative seems to
be planting more exotic shade tree species in coffee plots. Premium prices for these exotic
timber species and the less restriction in harvesting and marketing encourages the pro-
ducers to plant more exotic species like G. robusta.
Biodivers Conserv (2009) 18:987–1000 997
123
Based on the interviews it was clear that the gradually G. robusta will dominate the
coffee shade trees. The reason for this observation is that it is easier to acquire forest
department’s permission to harvest and transport exotic trees than the native trees. Also,
these exotic species grow faster and straighter, so they will be more profitable.
Discussions
Sacred groves have higher tree density and diversity compared to unredeemed and
redeemed coffee. Species evenness measurements indicated that all the three habitats
differing in land tenures had similar evenness. Redeemed and unredeemed coffee plots
were similar in species richness abundance but were different compared to sacred groves.
The interviews with the land owners of the coffee lands revealed that the government
forest policies affecting tree rights have played a major role in the current documented
species composition in the coffee lands. The government has a stringent law on harvesting,
transporting and marketing of the wild tree species (Shrinidhi and Lele 2001). However,
the law is less stringent on the harvesting exotic species like G. robusta (Shrinidhi and Lele
2001). In Mexico in shade grown coffee plots, Inga edulis higher dominance. This plant is
promoted by Instituto Mexicano del Cafe
´(INMECAFE) as a better shade tree for coffee in
Mexico (Ambinakudige 2006). The government policies in both cases have led to dif-
ferences in tree species composition—in India because farmers can cut exotic species, and
in Mexico because the government agency actively supported planting of one species. It is
interesting to note that the forest department’s restrictions designed to protect native trees
are having the reverse effect in Kodagu. The lack of restrictions on harvesting exotic trees
is leading to widespread planting of these species and further dominance of these species
over native species. This illustrates how state policies have a major influence on the change
and potential trends in tree diversity in the coffee growing regions.
In well-protected sacred groves, biodiversity is well preserved. At the same time, coffee,
which is often considered as a threat to biodiversity, had significant tree diversity.
Although diversity was less in coffee plots than sacred groves, replacing endemic species
with more profitable tree species like G. robusta may further erode biodiversity within
coffee plantations. At present, in unredeemed coffee plots, the native species D. latifolia
has a high dominance. In the redeemed coffee lands 74% of the basal area (dominance) is
generated by native species. In unredeemed lands, native species have a similar dominance
(72%). This exemplifies how shade grown coffee has the propensity to conserve pockets of
biodiversity even though it has less diversity than natural forests with sacred groves.
Introduction of coffee in Kodagu reorganized the patterns of tree diversity. The main
way in which patterns of tree diversity and composition have changed has been due to the
introduction and encouragement of non-indigenous tree species. Another disadvantage of
coffee cultivation is that it discourages the regeneration of tree species because the un-
derstorey is dominated by coffee plants. In this situation, it may be wiser to encourage
shade grown coffee than sun grown coffee as shade coffee retains some amount of bio-
diversity. However, in the regions like Western Ghats where the biodiversity is threatened
by human activities, no more new coffee plantations should be encouraged. Even though
coffee retains some biodiversity, it cannot substitute for natural forest. Existing coffee
plantations should be encouraged to preserve endemic species. The encouragement may be
in the form of niche market for the shade grown coffee where growers receive premium
prices for shade grown coffee. Some conservationists such as Conservation International
and National Audubon Society (Conservation International 2001; National Audubon
998 Biodivers Conserv (2009) 18:987–1000
123
Society 2000; Philpott and Dietsch 2003) already advocating for premium price for shade
grown coffee. Shade coffee can conserve tree biodiversity and could help improve the
livelihoods of the local people if conservation practices and coffee markets are linked.
Acknowledgments We would like to thank Dr. Klooster, Dr. R. K. Hegde, Dr. Kushalappa, Tony Stallins,
and Shea for helping us in various stages of this study. We would also like to thank all the coffee growers
and the villagers in Kodagu for allowing us to do the study in their coffee plots and sacred groves. Finally
we would also like to thank anonymous reviewers for all the comments and suggestions.
References
Ambinakudige S (2006) Differential impacts of commodification of agriculture in the Western Ghats of
India: an extended environmental entitlement analysis. Florida State University, Tallahassee
Benhin J (2006) Agriculture and deforestation in the tropics: a critical theoretical and empirical review.
Ambio 35(1):9–16. doi:10.1639/0044-7447(2006)035[0009:AADITT]2.0.CO;2
Bhagwat S (2002) Biodiversity and conservation of a cultural landscape in the Western Ghats of India.
Oxford Forestry Institute, Department of Plant Sciences, University of Oxford, Oxford
Blackman A et al (2005) Deforestation and shade coffee in Oaxaca, Mexico: key research findings.
Resources for the future, Washington, DC
Chandran MDS (1993) Vegetation changes in the evergreen forest tract of Uttara Kannada District of
Karnataka State. Department of Botany, Karnataka University, Dharwad
Chandran MDS et al (1998) Sacred groves of the Western Ghats of India. In: Ramakrishnan PS, Saxena KG,
Chandrashekar UM (eds) Conserving the sacred for biodiversity management. Oxford and IBH Pub-
lishing Co.Pvt Ltd, New Delhi
Chichilnisky G (1994) North–south trade and the global environment. Am Econ Rev 84:851–874
Colwell RK, Coddington JA (1994) Estimating terrestrial biodiversity through exploration. Philos Trans R
Soc Lond B Biol Sci 345:101–118. doi:10.1098/rstb.1994.0091
Condit A (1994) A field guide to large forest censuses in the tropics: the Bci 50 ha plot and other large-scale
studies. Centre for Tropical Forest Science, Panama
Conservation International (2001) Conservation principles for coffee production. Conservational Interna-
tional, 2007
Dallmeier F, Comiskey JA (eds) (1998) Forest biodiversity in North, Central and South America, and the
Caribbean: research and monitoring. UNESCO, Paris
Daniels RJR et al (1995) Impact of human extraction on tropical humid forests in the Western Ghats in
Uttara Kannada, South India. J Appl Ecol 32:866–874
Ehrlich PR (ed) (1988) The loss of biodiversity: causes and consequences. Biodiversity. National Academy
Press, Washington
Elouard C (2000) Vegetation features in relation to biogeography. In: Ramakrishnan PS, Chandrashekar
UM, Elouard C et al (eds) Mountain biodiversity, land use dynamics and traditional knowledge.
Oxford and IBH Publishing Co.Pvt Ltd, New Delhi, pp 25–42
FRLHT (2000) Kodagu Maps. Foundation for revitalisation of local health traditions, Bangalore
Geist H, Lambin E (2001) What drives tropical deforestation. LUCC International Project Office, University
of Lou vain
Gentry AH (1990) Floristic similarities and differences between Southern Central America and Upper and
Central Amazonia. In: Gentry AH (ed) Four neotropical rainforests. Yale University Press, New Haven
CT, pp 141–157
Gimaret-Carpentier C et al (1998) Sampling strategies for the assessment of tree species diversity. J Veg Sci
9:161–172. doi:10.2307/3237115
Gorenflo LJ, Brandon K (2003) Agricultural capacity and conservation in high biodiversity forest ecosys-
tems. Conserv Biol 17(6):1844. doi:10.1111/j.1523-1739.2003.00150.x
Gotelli NJ, Colwell RK (2001) Quantifying biodiversity: procedures and pitfalls in the measurement and
comparison of species richness. Ecol Lett 4:379–391. doi:10.1046/j.1461-0248.2001.00230.x
Haller G (1910) Revenue settlement of Coorg. Madikeri, Shakthi Printers, Madikeri
Hardner J, Rice R (2002) Rethinking green consumerism. Sci Am 286:88–95
Iverson L, Prasad A (2002) Potential redistribution of tree species habitat under five climate change sce-
narios in the eastern US. For Ecol Manage 155:205–222. doi:10.1016/S0378-1127(01)00559-X
Biodivers Conserv (2009) 18:987–1000 999
123
Jime
´nez-Avila E, Martı
´nez VP (1979) Estudios Ecolo
´gicos Del Agroecosistema Cafetalero. Ii. Produccio
´n
De Materia Orga
´nica En Diferentes Tipos De Estructura. Biotica 4:109–126
Kushalappa CG, Bhagwat S (2001) Sacred groves: biodiversity, threats and conservation. In: Uma Shaanker
R, Ganeshaiah KN, Bawa KS (eds) Forest genetic resources: status, threats and conservation strategies.
Oxford and IBH publishing Co Pvt Ltd, New Delhi
Kushalappa CG, Kushalappa KA (1998) Preliminary report: impact assessment of working in the Western
Ghats Forests. College of Forestry, University of Agricultural Sciences, Ponnampet, Bangalore
Lele S (2001) Linking ecology, economics and institutions of forest use in the Western Ghats of Karnataka,
India, Institute for Social & Economic Change, Bangalore and Pacific Institute for SIDES, Oakland
Magurran A (2004) Measuring biological diversity. Blackwell Publishing Company, Malden
Menon S, Bawa K (1998) Tropical deforestation: reconciling disparities in estimates for India. Ambio
27(7):576–577
Mittermeier RA et al (2005) Hotspots revisited: Earth’s biologically richest and most endangered terrestrial
ecoregions. CEMEX—Conservation International, Mexico City
Moguel P, Toledo V (1999) Biodiversity conservation in traditional coffee systems in Mexico. Conserv Biol
13:11–21. doi:10.1046/j.1523-1739.1999.97153.x
National Audubon Society (2000) Coffee and conservation of migratory birds. NAS 2007
Pascal JP, Ramesh BR (1987). A field key to the trees and lianas of the evergreen forests of the Western
Ghats (India). Institut Franc¸ais de Pondiche
´ry, Pondicherry
Perfecto I, Snelling R (1995) Biodiversity and the transformation of a tropical agroecosystem: ants in coffee
plantations. Ecol Appl 5:1084–1097. doi:10.2307/2269356
Perfecto I, Vandermeere J (2002) Quality of agroecological matrix in a tropical montane landscape: ants in
coffee plantations in Southern Mexico. Conserv Biol 16:174–182. doi:10.1046/j.1523-1739.2002.
99536.x
Perfecto I et al (1996) Shade coffee: a disappearing refuge for biodiversity. Bioscience 46:598–608. doi:
10.2307/1312989
Perfecto I et al (2005) Biodiversity, yield and shade coffee certification. Ecol Econ 54:435–446. doi:
10.1016/j.ecolecon.2004.10.009
Perfecto I et al (2007) Shaded coffee and the stability of rainforest margins in Latin America. In: Tscharntke
T, Leuschner C, Zeller M, Guhudja H, Bidin A (eds) The stability of tropical rainforest margins:
linking ecological, economic, and social constraints of land use and conservation. Springer, Envi-
ronmental Science Series, Heidelberg, Germany
Philpott SM, Armbrecht I (2006) Tropical agroforests and the ecological role of ants. Ecol Entomol 31:369–
377. doi:10.1111/j.1365-2311.2006.00793.x
Philpott S, Dietsch T (2003) Coffee and conservation: a global context and the value of farmer involvement.
Conserv Biol 17(6):1844–1846. doi:10.1111/j.1523-1739.2003.00150.x
Philpott SM et al (2007) Field testing ecological and economic benefits of coffee certification programs.
Conserv Biol 21(4):975–985. doi:10.1111/j.1523-1739.2007.00728.x
Philpott SM et al (2008) Biodiversity conservation, yield, and alternative products in coffee agroecosystem
in Sumatra, Indonesia. Biodivers Conserv 17(8):1805–1820. doi:10.1007/s10531-007-9267-2
Rappole JH et al (2003) Coffee and conservation. Conserv Biol 17:334–336. doi:10.1046/j.1523-
1739.2003.01548.x
Rice RA, Ward JR (1997) Coffee, conservation and commerce N the Western Hemisphere. Smithsonian
Migratory Bird Center and Natural Resources Defense Council, Washington, DC
Sathish BN (2005) Assessment of tree diversity in coffee plantations under different land tenure systems in
Virajpet Taluk, Kodagu. College of Forestry, University of Agricultural Sciences, Ponnampet,
Bangalore
Shrinidhi AS, Lele S (2001) Forest tenure regimes in the Western Ghats: a compendium. Bangalore. Inst Soc
Econ Change 90:56
Solow AR (1993) A simple test for change in community structure. J Animal Ecol 62:191–193
Soto-Pinto L et al (2001) Woody plant diversity and structure of shade-grown-coffee plantations in northern
Chiapas, Mexico. Rev Biol Trop 49(3–4):977–987
Utkarsh G et al (1998) On the patterns of tree diversity in the Western Ghats of India. Curr Sci 75:594–603
Verbist B et al (2005) Factors driving land use change: effects on watershed functions in a coffee agro-
forestry system in Lampung, Sumatra. Agric Syst 85:254–270. doi:10.1016/j.agsy.2005.06.010
1000 Biodivers Conserv (2009) 18:987–1000
123
