In tern a tio n a l

Sch o la rs
Jo u rn a ls
International Journal of Agroforestry and Silviculture ISSN 2375-1096 Vol. 5 (4), pp. 294-304, April, 2017.
Available online at www.internationalscholarsjournals.org International Scholars Journals
Author(s) retain the copyright of this article.

Full Length Research Paper

Diversity of shade tree species in smallholder coffee

farms of western Oromia, Ethiopia
1
Ebisa Likassa* and 2Abdela Gure
1
Department of Natural Resources, College of Agriculture and Natural Resource, Wollega University, Nekemte, Ethiopia,
2
School of Forestry, College of Forestry and Natural Resource, Hawassa University, Shashemene Ethiopia.
Accepted 02 August, 2016

Shade grown coffee has been promoted as means of preserving biodiversity in the tropics even though
expansion of coffee cultivation has been seen as a contributing factor to deforestation and erosion of
biodiversity. The study was conducted on diversity of tree species in smallholder coffee farms of Manasibu
district, Western Oromia, Ethiopia. The aim of the study was to assess diversity and structure of tree species in
smallholder coffee farms as well as farmers management practices in the district. To conduct the study,
stratified random sampling method was used. Vegetation data were collected from 74 rectangular plots. The
vegetation data was analyzed for tree diversity, Importance value index, similarity coefficient, density, Basal
Area, shade cover and other structural parameters. A total of 53 tree species were recorded in which 32
indigenous tree species were common to both forest and coffee farms. Species richness was significantly
higher in the adjacent natural forest than in coffee farms. There was significant difference (p<0.05) between
Peasant associations (PAs) interims of evenness, coffee shrub density, and tree density. Coffee shrub density
was significantly correlated with wealth status. Tree management practices in the study area of coffee farms
were more or less the same among the PAs. It was observed that tree species diversity and House Hold (HH)
dependency on coffee production increased with the closeness of the PAs to the adjacent natural forest. It was
concluded that traditional coffee production system is an important land use system in slowing down loss of
biodiversity and should therefore be encouraged.

Keywords: Diversity, adjacent forest, shade trees, smallholder coffee farms, Western (Oromia) Ethiopia.

INTRODUCTION

Previous studies of biodiversity were mainly on surrounding agricultural matrix with successful
undisturbed ecosystems, with less attention given to management strategies (Tejeda-Cruz et al., 2010).
biodiversity in managed or agricultural ecosystems Widespread agricultural expansion with population
(Moguel and Toledo, 1999). However, the recently growth resulted in the deforestation and biodiversity loss
accepted argument is that not only protected areas are due to lack of appropriate measures or unsustainable
enough to ensure biodiversity conservation but also the management of natural resources like forests. Due to the
current rates of deforestation and land degradation in
different regions of the world, many species may be lost
*Corresponding authorE-mail: [email protected] before they are even known to science (Good, 2004).
Tel.: +251913142412; +251917663009 Even though coffee is under story woody shrub,
1
Likassa & Gure 295

expansion of coffee cultivation is one of the causes of mean annual rainfall of the district is 950mm. The
deforestation and biodiversity loss (Ambinakudige and altitudinal variation ranges from 1249 to 1933 masl.
Sathish 2009). However, Agroforestry systems are widely Fluvisols, Regosols, and Vertisols are the three soil types
seen as the means that can reduce the impacts of of the study area.
deforestation (Tengnas 1994) through providing eco- Agriculture with timber and non-timber forest products is
agricultural solutions that successfully combine objectives the basis for the livelihood of most of the inhabitants of
for increased food security and biodiversity conservation Manasibu district. The farming system in the area is
gains (Kindtet al. 2008). For instance, coffee shade characterized by crop and livestock mixed farming
systems host diverse plant species (Ambinakudige and system; though the level of contribution of each
Sathish, 2009). Accordingly, traditional shaded coffee enterprise to the economy of small scale farmers is less
production system have received considerable attention understood. Coffee plantations are the major ones in
from conservation organizations in recent years (Perfecto sustaining the life of most households (HHs). Small
et al. 2005) since the system supports much more fragmented forest cover is restricted to the foot hills of
biodiversity conservation (Perfecto et al. 1996; Soto-Pinto some hills, where coffee is commonly grown and where
et al. 2000; Faminow and Rodringuez 2001; Soto-Pinto et farmers did not convert the forest into other land use
al. 2001; Perfecto et al. 2005; Gordon et al. 2007) and type.
cash income generation from the sale of both timber and The peasant association (PAs) in which the study was
non- timber forest products (Beer 1987; Beer et al. 1988; conducted were Qorke 01, Guyo SachiLaftoSalga
Gordon et al. 2007). (G/S/L/Salga) and BukeHena (B/Hena). Relative
Although coffee ranks among the five most valuable closeness to the nearby adjacent forest and level of
agricultural exports from developing nations (Ricketts et household dependency on coffee production was
al., 2004), there are few studies that describe structural decreasing from Qorke 01 to B/Hena PA, respectively.
characteristics and woody plant species diversity to The adjacent natural forests used for sampling purpose
define shade grown coffee stands. Particularly in Africa, were highly disturbed and not demarcated until 2009.
only few studies of biodiversity in human made These forests are dominated by Carrisaedulis particularly
landscapes such as coffee agroecosystems have been in Bembe forest. Most of the trees and shrubs in these
conducted (Komar, 2006). Likewise, even though coffee forests are encircled by thorny species like Carrisaedulis.
production in Manasibu district of Ethiopia is
predominated by small scale farming activity, there is no Sampling Design
study which documented species diversity, structures,
effect of tree species on coffee shrubs, and traditional Three PAs and two represent villages were selected
coffee farm management practices of the study area. purposively based on their relative proximity to road and
Therefore, the aims of the study were:(i) to assess tree nearby forests, and also the relative dependency on the
species diversity and structural parameters in smallholder farming practice (coffee production), which were
coffee farms, (ii) to evaluate the species diversity and supposed to affect diversity of tree species. Stratified
similarity between the coffee farms and the adjacent random sampling procedure was followed to select target
natural forests, (iii) to assess the effect of species HHs from each of the wealth categories after been
richness, shade tree structural parameters, and ranked with the help of key informants based on amount
household (HH) wealth status on coffee shrub density, of coffee and cereal crops produced in quintal per year,
and (iv) to document farmers management practices in number of cattle, landholding size, and type of house.
the area. Key informant selection was carried out by following the
techniques used by Den-Biggelaar (1996). Subsequently,
three representative HHs from each wealth categories (9
MATERIALS AND METHODS HHs) from each village and a total of 54 HHs were
randomly selected. Selected HHs were those whose
Site description coffee farms were used for data collection. Two disturbed
adjacent natural forests were also selected for the
The study was conducted in Manasibu districts, western reference.
Oromia, Ethiopia (Fig.1). Manasibu is geographically Data was collected by both questionnaire survey and
found at 09 48'N and 35 06'E. From the district, mid vegetation survey. Questionnaire was distributed to each
altitude and lowland areas constitute 68% and 32%, of the selected coffee producing HHs to identify the
respectively. Mean annual temperature varies from 19 to farmers management practices of the shade coffee farms

22 C. April and May are the hottest months of the area from the study area. Due to low density of trees, 54
whereas July and August are the coldest months. The random plots of 35m x 35m were selected at the center of
1
 296 Int. J. Agrofor. Silvicult.

Fig.1. Location of the study area [Manasibu district] from administrative map of Ethiopia showing the main coffee
growing areas.

the farms (one plot per farm) following Lopez-Gomez et vegetation structure was estimated using the density,
al. (2008) and Ambinakudige and Sathish (2009). basal area, per cent shade cover, diameter and height.
Simultaneously, 10 quadrants of 20m x 20m were Species richness, Shannon diversity index, and Shannon
selected randomly in each of the adjacent natural forests. Evenness were analyzed to evaluate the change in tree
Measurements on tree species of diameter at breast height species between the adjacent forest and coffee farms.
(dbh or 1.3m) of 5cm and height of 3m has been These parameters were also compared among PAs and
conducted in each plot. Plant species below 5cm dbh and wealth classes to evaluate the effect of PAs position and
3m height were not considered by this study because they status of household on tree species diversity in the coffee
were continuously removed from the system through farm farms. Similarly, both tree structure and coffee shrub
management activities. Diameter measurement was done density in the coffee farms were compared between the
using caliper or diameter tape. Trees were categorized into PAs, and among the wealth classes. Tree diameter and
10 diameter classes: 5-15, 15.1-25, 25.1-35, 35.1-45, 45.1- height were computed only in the coffee farms. The
55, 55.1- 65, 65.1-75, 75.1-85, 85.1-95 and >95cm. Height
relative importance of each species in the coffee farms
of each tree was measured by hypsometer and classified
was computed using the importance value index (IVI),
into nine different height classes: 3-6, 6.1-9, 9.1-12, 12.1-15,
15.1-18, 18.1-21,21.1-24, 24.1-27 and 27.1-30m. Crown
which is the sum of relative density, relative frequency
diameter was measured by measuring perpendicular canopy and relative dominance (Kent and Coker, 1992) after
rays for each tree by using measuring tape. Coffee shrub pooling the data from each plot for each species to
abundance from each plot was counted. All tree species evaluate the dominance of a species in the shade coffee
5cm dbh were enumerated and identified by referring to the system.
Flora of Ethiopia and Eritrea (Edwards et al. 1995; Edwards Per cent shade cover was calculated after first calculating
et al. 1997; Edwards et al. 2000; Hedberg et al. 2003; crown area and crown diameter of each over story tree
Hedberg et al. 2004; Hedberg et al. 2006). as indicated below. Crown cover of each tree in a plot
was calculated and changed to percentage canopy cover.
Data analysis Per cent shade cover =
s
i=1 CA
100..Equation 1
The species diversity and similarity were estimated using sample area
2
the Shannon diversity index, Shannon Evenness and CA= crown area of individual tree (m ); S= number of
Sorensen similarity index, respectively. Whereas, the individuals in the sample plot
1
Likassa & Gure 297

Table 1. Diversity indices, species richness, and tree density in the coffee farms and the adjacent natural forests (mean
SD).
Land Density Species richness Diversity indices
-1
use ha
Total mean H' E
b b b b
Coffee 123.36(60.96) 36 6.5(2.33) 1.63(0.42) 0.91(0.07)
farms
a a a a
Natural 182.45(24.74) 49 14.8(3.14) 2.56(0.24) 0.96(0.02)
forests
F-value 35.16 152.5 86.8 7.2
P <0.000 <0.000 <0.000 <0.008
Means with the same letter in the column were not significantly different at P=0.05

Sorensen similarity index (Ss) measures how the floristic forest. Similarly, Shannon diversity index and Shannon
compositions of different coffee farms or land use evenness were also higher in the adjacent forests. This
systems are similar. Species similarity analysis was difference may be explained in terms of the difference in
complemented by calculating the floristic similarities the management practices in the two land uses. Coffee
between the coffee farms and the adjacent forests; and farms were generally characterized by selective retention
among PAs using Sorensens similarity index (Magurran of some amount of over story trees as shade tree while
1988): there may not be such intentional management in the
2C adjacent natural forests. The dominance of some species
Ss = ..Equation2
A+B
in the coffee farms can be explained by the importance
C= the number of species common to both sites; A= the
attached to those species by the farmers for additional
number of species present in one of the sites to be
purposes like timber extraction, medicinal value, honey
compared; B= the number of species present in the other
production, fodder for their cattle, fuel wood and organic
site.
matter production.
Statistical analysis The adjacent natural forest is generally expected to have
higher number of tree species than coffee farms since
Descriptive statistics were used to present the results. coffee farms need continuous management that
The results were subjected to one-way ANOVA Tukeys eliminates seedling, sapling and shrubs to create free
test to compare whether there was significant mean space for coffee shrubs so as to reduce competition.
differencein tree species diversity betweenthe land uses, Tadesse (2003); Schmitt (2006); Feyera and Denich
among PAs and wealth categories. The result was also (2006) also reported higher plant species diversity in
subjected to Pearson correlation using SAS version 9. forest coffee than in the semi forest coffee system due to
the shade reduction. However, coffee production, which
RESULTS AND DISCUSSION is often considered a threat to natural forest biodiversity
have important contribution to tree species diversity
Tree species diversity (Ambinakudige and Sathish, 2009).
In the coffee farms of the current study, although average
A total of 53 tree species were recorded from the coffee number of tree species and Shannon diversity index were
farms and the adjacent forests, of which 49 were native not significantly different among the PAs, it was slightly
and 4 were exotic species. Therewas higher tree species higher in Qorke 01 PA than the others (Table 2). The
richness recorded in the adjacent forests (49) than in the lowest species diversity in the B/Hena coffee farms may
coffee farms (36) (Table1) which is comparable with the be explained in terms of uneven distribution of shade tree
results of the studies by Hylander and Sileshi, (2009) and species across the sampled coffee farms, the relatively
Ambinakudige and Sathish (2009) where forest and less dependency of the resident farming HHs on coffee
coffee home garden; sacred forest and coffee farms production since cereal production was equally important
(redeemed and unredeemed coffee plots) were and perhaps also due to the relatively longer distance
compared, respectively. Contrary to the current study, from the adjacent forest. Whereas, the highest tree
Lopez-Gomez et al. (2008) recorded higher number of species diversity in Qorke 01 coffee farms may be
tree species in the coffee farms than in the adjacent explained in terms of more dependency of dweller
1
 298 Int. J. Agrofor. Silvicult.

Table 2. Species richness, diversity indices and structural variables of trees and coffee shrub density (mean SD) in the coffee
farms at PA level n=54.
Variables G/S/L/Salga B/Hena Qorke 01 F-value P
Total S 29 26 25
a a a
Mean S 6.61(2.87) 5.89(2.25) 7.00(1.75) 1.04 0.361
a a a
H' 1.59(0.52) 1.50(0.40) 1.81(0.26) 2.71 0.076
ba b a
E 0.90(0.07) 0.88(0.10) 0.95(0.03) 3.76 0.029
a a a
BA 16.13(8.44) 13.52(7.83) 10.54(4.59) 2.75 0.073
b b a
Coffee shrub 3541(976.35) 3965(1061.26) 4752(621.33) 8.28 0.001
-1
density ha
- ba a b
Tree density ha 120.18(42.07) 155.10(84.28) 94.78(27.74) 5.14 0.009
1
a a a
Shade % 66.01(26.01) 48.99(19.78) 54.63(19.19) 2.48 0.093
S- Species richness, BA- basal area
Means with the same letter in the row were not significantly different at P=0.05

Table 3. Species richness, diversity indices and structural variables of tree species, and coffee shrub density (mean SD) in
the coffee farms among wealth classes n=54.
Parameters Poor Medium Rich F-value P
Total S 22 30 26
b a ba
Mean S 5.44(2.15) 7.28(2.16) 6.78(2.46) 3.38 0.046
a a a
H' 1.57(0.45) 1.76(0.37) 1.57(0.41) 1.27 0.289
a a a
E 0.94(0.05) 0.91(0.08) 0.88(0.09) 2.42 0.099
a a a
Coffee shrub 3689(1022) 4204(1008) 4366(969) 2.24 0.117
-1
density ha
a a a
Tree 97.51(34.69) 131.97(59.35) 140.59(75.89) 2.68 0.078
-1
density ha
Means with the same letter in the row were not significantly different at P=0.05

farming HHs on coffee production and more closeness of coffee home gardens in that most of the shade trees in
the PA to the adjacent natural forest. Furthermore, the the coffee home gardens were trees commonly found in
results of the current study indicated that farmers species forests. In the contrary, Mendez et al. (2007) reported
selection, preservation and management of coffee shade lower similarity in species composition between shade
trees were more or less similar across the study sites. coffee cooperatives and forest sites in El Salvador. Out of
However, tree species diversityin the coffee farms of a total of 53 tree species, 32 (60.4%) were common,
medium HH wealth class was higher than that recorded while 17 (32.1%) and 4 (7.5%) were unique to the
from the poor and rich wealth class (Table 3). The result adjacent natural forest and coffee farms, respectively. In
implies that tree species composition and diversity in the the adjacent natural forests, all of the unique tree species
coffee farms were not linear with HH wealth status for this were indigenous as might be expected, while the unique
particular study. The finding of this study was in line with species in the coffee farms were exotics. Those unique
that of Motuma et al. (2008) and contrary to that of tree species restricted to the natural forests were the
Zebene (2003). least preferred by the farmers primarily because of their
The results of the current study indicated that there was a unsuitability for shade and generation of income;
high species overlap between the adjacent natural forests whereas those species unique to the coffee farms were
and the coffee farms. Sorensens similarity index purposefully introduced fruit tree species. Ambinakudige
between the coffee farms and the adjacent natural forests and Sathish (2009) also reported similar results from
was 0.75 indicating high overlap between the two land sacred grove and coffee farms in India.
use types since most of the tree species in the coffee Regardless of the difference in tree species diversity,
farms were native and remnants from the conversion of there was high species overlap between PAs.
natural forest to coffee farms. Hylander and Sileshi Accordingly, it ranged from 69 to 81% between sites
(2009) also reported similar results from forests and (PAs). The highest similarity (0.81) in species

1
Likassa & Gure 299

Table 4. IVI of total tree species recorded in the surveyed coffee farms (n= 54).
Species scientific name Abundance RF % RDe% RDo% IVI
Acacia abyssinica 28 5.556 3.074 4.093 13
Albiziagrandibracteata 26 4.094 2.854 2.269 9
Albiziagummifera 109 10.527 11.965 20.519 43
Apodytesdimidiata 12 3.217 1.317 1.254 6
Bersamaabyssinica 32 4.094 3.513 2.789 10
Brideliamicrantha 6 1.462 0.659 0.239 2
Bruceaantidysenterica 1 0.877 0.11 0.142 1
Capparisfascicularis 6 1.17 0.439 0.157 2
Citrus sinesis 7 1.17 0.768 0.055 2
Cordiaafricana 168 14.036 18.441 24.887 57
Croton macrostachyus 163 12.866 17.892 13.397 44
Dovyalisabyssinica 8 2.047 0.878 0.482 3
Ekebergiacapensis 2 0.585 0.22 0.022 1
Elaeodendronbuchananii 3 0.585 0.329 0.17 1
Entadaabyssinica 2 0.585 0.22 0.007 1
Fagaropsisangolensis 1 0.292 0.11 0.134 1
Faureaspeciosa 2 0.585 0.22 0.01 1
Ficussur 9 2.339 0.988 1.159 4
Ficusthonningii 6 1.754 0.659 0.689 3
Ficusvasta 17 4.094 1.866 10.42 16
Maesalanceolata 8 1.754 0.878 0.672 3
Magniferaindica 15 2.047 1.647 2.393 6
Millettiaferruginea 29 5.263 3.183 2.931 11
Oleacapensis 1 0.292 0.11 0.011 0
Perseaamericana 1 0.292 0.11 0.02 0
Piliostigmathonningii 1 0.292 0.11 0.167 1
Psychotriaorophila 3 0.877 0.329 0.33 2
Sapiumellipticum 18 4.386 1.976 3.943 10
Sennapetersiana 3 0.585 0.329 0.279 1
Sesbaniasesban 4 0.877 0.439 0.028 1
Steganotaeniaaralaceae 3 0.292 0.439 0.297 1
Stereospermumkunthianum 1 0.292 0.11 0.009 0
Syzygiumguineense 7 1.17 0.768 1.217 3
Tecleanobilis 6 0.877 0.659 0.155 2
Vangueriaapiculata 5 1.17 0.549 0.203 2
Vernoniaamygdalina 199 7.603 21.844 4.452 34

composition was observed between G/S/L/Salga and the country as coffee shade trees. The finding of the
Qorke 01 PAs, which were relatively farther apart than current study was thus in line with previous study of
between B/Hena and Qorke 01 (0.78) those were Feyera (2006) and Yitebitu (2009). Furthermore, although
relatively closer. The results suggest that distance the total tree species diversity in coffee farms could be
between the PAs did not influence the variation in high, some species may be preferred more by the
species composition among the PAs because the highest farmers to the rest of the species retained in their coffee
similarity was observed from those are apart. farms (Soto-Pinto et al. 2001) due to that shade trees
The IVI of a species increases with the preference the provide several economic and ecological benefits
farmers give to the species. Use value, source of the (Yadessa et al. 2008)
seedling and position of the coffee farms resulted in the Generally, the most important tree species are those
difference in importance of tree species in the coffee most common and abundantly retained in agroforestry
farms. C.africana(57), C.macrostachyus (44), system (Tesfaye 2005; Motuma et al. 2008). Farmers
A.gummifera (43), and V.amygdalina (34), respectively knowledge about the relative advantages of different tree
were found to be the four most important species based species such as provision of balanced shade for coffee
on their IVI (Table 4). The above mentioned tree species shrubs, rapid rate of decomposition of leaf litter to
are generally common in most coffee growing regions of improve soil fertility and availability of seedlings in their
1
 300 Int. J. Agrofor. Silvicult.

vicinity plus additional values (cash, medicinal and relationship. Finding of this was inline with previous
fodder) may have been the major causes of the studies of Mendez et al. (2007); Asteggiano (2008);
difference in the importance of the tree species in the Mendez et al. (2009); Yitebitu (2009). As far as the
study area. relationship between shade cover and tree density is
concerned, it was observed in the current study (Table 2)
Vegetation structure and also from other study (Soto-Pinto et al., 2000) that
shade cover does not necessarily correspond to tree
Tree density per hectare in the coffee farms was density.
significantly lower than in the adjacent natural forests Farmers manage the shade tree canopy to balance
(Table 2) as also reported from Mexico (Lopez-Gomez et optimum shade for coffee production and obtain tree
al. 2008) and India (Ambinakudige and Sathish 2009). products. This involves a yearly pruning of the shade tree
The mean tree density in the coffee farms in the current canopy, aiming to leave 4050% shade cover (Mendez et
study was much higher than the one reported by Tadesse al. 2007) for maximum coffee yield (Soto-Pinto et al.
-1
et al. (2001) which was 60 trees ha (10cm dbh) in the 2000; Yitebitu 2009). The minimum recommended shade
Ethiopian traditional garden coffee farms, while much cover over coffee shrubs was 40% of the land (Faminow
lower than those of reported by Soto-Pinto et al. (2000) and Rodriguez, 2001; SAN, 2005); however, it may differ
and Soto-Pinto et al. (2001) inMexicowhich were 464 and from country to country. Themean per cent shade cover
-1
371.4 ha , respectively. The recommended tree density in the coffee farms of the current study sites was 58.63%
per hectare in coffee farm is to the minimum 70 individual which is higher than the minimum recommended shade
trees with 12 native tree species (SAN 2005). However, cover for the coffee land.
the minimum tree density in the coffee farms was 49 The largest tree diameter and height in the coffee farms
individuals/ha from the current study. were found to be 167cm and 30m, respectively.The
There was significant difference in mean tree density and average dbh and height of trees in the surveyed coffee
coffee density among the PAs, the highest tree density farms was 31.5cm and 12.36m, respectively. Most of the
and coffee density was observed in B/Hena and Qorke trees in the coffee farms had dbh of 45cm (Fig. 2)
01 PA, respectively (Table 2). With respect to the highest which accounted for 79.7% of all the trees in the farms.
tree density in B/Henawas due to the comparatively less F.vasta, A.gummifera, and C.africana trees had higher
intensive management of coffee farms since the high dbh than the others. According to Soto-Pinto et al. (2001)
income of B/Hena farming HHs is from cereal production. most of shade components were in the range of <20cm
Feyera (2006) also discussed continued management dbh. Most of the trees had height of 15m which
suppresses tree regeneration and reduces tree density, accounted for 73.2% of the trees in coffee farms (Fig.2).
while promoting a high number of bigger shade trees Therefore, there was higher tree abundance at both the
over coffee. The mean density of coffee shrub of the lower diameter and height classes.
-1
study area was 4086 ha which is higher than what was
-1
reported by Tadesse et al. 2001 (1000 ha in Hararghe to Effect of shade tree species richness, tree structural
-1
3500 ha in south and southwestern parts of Ethiopia). parameters, and HH wealth status on coffee shrub
The result of the current study was also higher than those density
of Soto-Pinto et al. (2000) from Mexico (800 to 3500
-1
coffee shrubs ha ). Previous studiesof Lopez-Gomez et Coffee shrub formed the under story stratum in the coffee
al. (2008) and Asteggiano, (2008) reported that farms. From this study, there was no significant
asmanagement of shade coffee farms increase,coffee correlation between coffee shrub density and tree
shrub density also increase. In the study area, coffee species richness (r= -0.022; P = 0.437). This indicates
shrub is traditionally planted in an irregular pattern that, density of coffee shrubs was not significantly
instead of row planting; and maintaining such a high influenced by tree species richness in the coffee farms.
coffee shrub density is perhaps a strategy for countering Rather influenced by different interactions between shade
the effects of the widespread coffee blight disease and trees and coffee shrubs in addition to management
termite infestation in the area. practices (Beer et al. 1998). Similarly, there were no
Among the PAs, the highest shade cover was recorded significant positive relationship between coffee shrub
from G/S/L/Salga (66.01%) followed by Qorke 01 density and shade cover (r= 0.17; P = 0.110) and
(54.63%) (Table 2).G/S/L/Salga was characterized by the between coffee shrub density and tree density (r= 0.09; P
dominance of relatively larger trees whose crowns = 0.256), but according to Yitebitu (2009), there was
provide large shades. This was also supported by the strong negative correlation (r = -0.828, P<0.0001)
highest basal area. Correlation between shade cover and between tree density and coffee density. According to
basal area (r = 0.59; p = 0.000) also confirms this Diriba et al. (2007) degree of shading had effect on
1
Likassa & Gure 301

Fig. 2. Height and diameter class distribution of trees in the coffee farms.

counts of coffee shrubs. However, HH wealth status had shade into the coffee farms. Therefore, most farmers
significant positive correlation with coffee shrub density (r maintained trees in the coffee farms not only for shade
= 0.27; p = 0.023) and with tree density (r = 0.29; p = but also for their additional values such as economic
0.016) in coffee farms; with the highest tree and coffee importance (sell of timber and fruits), source of food
shrub density corresponding to rich HHs and the lowest (fruits), fuel wood, fodder, soil and water conservation,
to Poor HHs. andmedicinal value.Earlier study of Soto-Pintoet al.
(2000) also indicated similar benefit of shade trees and
Management practices in coffee farms shrubs.Almost all coffee farmers in the study area
employed similar management practices such as thinning
Farmers management practices of coffee farms shade trees, pruning both shade trees and coffee shrubs,
influenced tree species both positively and negatively. weeding, fertilizing, and hoeing to control the interaction
Shade trees were planted and retained in coffee farms for between coffee shrub and shade trees, and also to make
dual purpose they have. Therefore, shade trees are coffee production sustainable through reducing
treated positively than other tree species, those excluded competition among the components, and controlling
from coffee farms due to their negative effect on coffee shade cover per cent (Table 5). Similar management
production. Few shade tree species were preserved in practices were recorded by Mendez et al. (2009) from
abundantly, whereas more tree species were retained cooperatives coffee plantations in western El Salvador.
rarely. The preference of shade trees by farmers was Coffee shrubs and shade trees are actively pruned in
variable among the most common trees in the coffee most of the plantations where these activities are carried
farms of the study area. Accordingly, the most preferred out in irregular cycles (Hernandez-Martinez et al. 2009).
tree species were C. africana (100%), A. abyssinica However, management approaches related to shade tree
(66.67%), A.gummifera (52.78%), and V. amygdalina density and diversity, pruning of trees and coffee bushes,
(16.67) while Sesbania sesban (5.56%), M.indica and weeding have been shown to affect soil chemical
(5.56%), Senna petersiana (2.78%), Albizia and physical characteristics (Mendez et al. 2009).
grandibracteata (2.78%), and Sapium ellipticum (2.78%) Frequency of these management practices ranged from 1
were the least preferred shade trees. to 3 times per year. Majority of the HHs (63.8%) carried
There were different management activities that are out these activities twice a year. About a third of the HHs
commonly applied in coffee farms to maintain continuous (33.33%) carried out the activities three times a year,
coffee production. The long time tradition in the area while only 2.78% of HHs carried out once a year.
favors incorporation of selected native tree species for Weeding (slashing) and hoeing performed on average
1
 302 Int. J. Agrofor. Silvicult.

Table 5. Common tree management practices in the coffee farms of the study area (n = 54).
Management Responded Reasons for practicing Responded HHs (%)
practices HHs (%)
Thinning 88.89% For better growth 97.22%
Pruning 83.33% To reduce competition 91.67%
Weeding 58.33% To reduce shade 80.56%
Fertilizing 5.56%
Hoeing 97.22%

twice a year. When a new coffee farm is established, the coffee farms of lower tree density with very few native
density of shade trees is generally high but with time species should retain or planted to the minimum
shade trees are thinned out to reduce competition. requirement. Similarly, per cent shade cover in coffee
Pruning coffee shrubs occurs when they get old in order farms with lower (<40%) should be increased at least up
to encourage coppicing and thus to improve production. to the moderate shade levels (40-50%) for better shade
coffee production. to provide optimal shade environment
for coffee shrubs, tree density or shade cover should be
CONCLUSION managed.

Shade coffee production system play important role in

harboring various tree species. The highestindigenous ACKNOWLEDGEMENT
tree species composition similarity between coffee farms
and adjacent natural forests indicates the conservation I am highly thankful to all people who contributed to the
role being played by shade coffee production systems. successful completion of my work in one way or another.
As the proximity of the PAs to the adjacent natural forest I am grateful to Ethiopian Environment and Coffee Forum
increased, tree species diversity in the coffee farms also (EECF) and Ethiopian Ministry of education for financial
increased. Similarly, as level of HH dependency on support to cover part of research expenses.
coffee production increased, diversity of tree species also
increased. However, HH wealth status could not be the
cause for tree species diversity in the coffee farms. In the REFERENCES
study area, tree density and shade cover had no
significant positive correlation with coffee shrub density in Ambinakudige S, Sathish BN (2009) Comparing tree
the coffee farms. The relationship among BA, tree diversity and composition in coffee farms and sacred
density, coffee shrub density and shade cover was not forests in the Western Ghats of India. Biodivers.
always linear. There was no significant negative Conserv. 18:987-1000
correlation between tree species richness and coffee Beer J (1987) Advantages, disadvantage and desirable
shrubs density. However, adding tree species without characteristics shade tree for coffee, cacao and tea.
knowing its suitability may affect the coffee shrub density Agroforestry systems 5:3-13
in the coffee farms negatively due to their hindrance Beer J, Muschler R, Kass D, Somarriba, E (1998) Shade
behavior. management in coffee and cacao plantations.
Agroforestry systems 38:139-194
Diriba M, Fassil A, Sileshi N, Granhall U (2007)
RECOMMENDATIONS Composition of coffee shade tree species and density
of indigenous arbuscularmycorrhizal fungi (AMF)
In the face of the alarmingly high and unabated spores in Bonga Natural coffee forest, southwestern
deforestation, and from biodiversity conservation point of Ethiopia. Forest Ecology and Management 241:145
view, converting degraded and highly disturbed natural 154
forests to shade coffee system appears to be a better Edwards S, Mesfin T, Hedberg I (1995) Flora of Ethiopia
option than converting them into a cereal production and Eritrea volume 2, Part 2. Addis Ababa and Uppsala
system when it comes to choosing between the two University, Uppsala
options. From biodiversity and environmental points of Edwards S, Mesfin T, Sebsebe D, Hedberg I (2000) Flora
view, smallholder shaded coffee farms should be of Ethiopia and Eritrea volume2, Part1. Addis Ababa
encouraged through a certification programme. Shade and Uppsala University, Uppsala.
1
Likassa & Gure 303

Edwards S, Sebsebe D, Hedberg I (1997) Flora of Mendez V E, Shapiro E N, Gilbet GS (2009) Cooperative
Ethiopia and Eritrea volume 6. Addis Ababa and management and its effects on shade tree diversity,
Uppsala University, Uppsala soil properties and ecosystem services of coffee
Feyera S (2006) Biodiversity and ecology of Afromontane plantations in western El Salvador. AgroforestSyst
rainforests with wild Coffeaarabica L. populations in 76:111126
Ethiopia. Dissertation, University of Bonn Moguel P, Toledo VM (1999) Review: Biodiversity
Feyera S, Denich M (2006). Effects of wild coffee conservation in traditional coffee systems of Mexico.
management on species diversity in the Afromontane Conservation biology 13:11-21
rainforests of Ethiopia. Forest Ecology and Management Motuma T, Zebene A, Mulugeta L, Karltun E (2008)
232:6874 Woody species diversity in a changing landscape in the
Gordon C, Manson R, Sundberg J, Angon AC (2007) south-central highlands of Ethiopia. Agriculture,
Biodiversity, profitability, and vegetation structure in Ecosystems and Environment 128:5258
Mexican Coffee agroecosystem. Agriculture, Ecosystems Perfecto I, Rice R A, Greenberg R, Van DerVoort M E
and Environment 118:256266 (1996) Shade coffee: A disappearing refuge for
Hedberg I, Edwards S, Sileshi N (2003) Flora of Ethiopia
biodiversity. BioScience 46:598608
and Eritrea (Apiaceae to Dipsaceae) Volume 4 Part 1.
Perfecto I, Vandermeer J, Mas A, Pinto S L (2005)
Addis Ababa and Uppsala University, Uppsala
Analysis biodiversity, yield, and shade coffee
Hedberg I, Friis I, Edwards S (2004) Flora of Ethiopia and
certification. Ecological Economics 54:435-446
Eritrea (Asteraceae), Volume 4 Part 2. Addis Ababa
Ricketts T H, Daily G C, Ehrlich P R, Michener C D
University, Addis Ababa and Uppsala University,
(2004) Economic value of tropical forest to coffee
Uppsala. 408 p
Hedberg I, Kelbessa E, Edwards S, Demissew S, Persson E
production. PNAS101:12579-12582
(2006) Flora of Ethiopia and Eritrea (Gentianaceae to SAN (Sustainable Agriculture Network), 2005.Additional
Cyclocheilaceae) Volume 5. Addis Ababa and Uppsala criteria and indicators for coffee production.Rainforest
University, Uppsala Alliance, San Jos, Costa Rica.
Hernandez-Martinez G, Manson RH, Hernandez AC (2009) Schmitt C B (2006) Montane rainforest with wild Coffea
Quantitative classification of coffee agroecosystems Arabica in the Bonga region (SW Ethiopia): plant
spanning a range of production intensities in central diversity,wild coffee management and implications for
Veracruz, Mexico. Agriculture, Ecosystems and conservation. Dissertation, University of Bonn
Environment 134:89 98 Schmitt C, Grote U (2006) Wild coffee production in
Hylander K, Sileshi N (2009)Complementary Roles of Home Ethiopia: the role of coffee certification for forest
Gardens and Exotic Tree Plantations as Alternative conservation. German Federal Ministry for Education and
Habitats for Plants of the Ethiopian Montane Rainforest. Research (BMBF).German. www.coffee.uni-bonn.de
Conservation Biology 23:400409 Soto-Pinto L, Alvarado Y R, Caballero-Nieto J, Warnholtz G
Kassahun T (2006) Genetic diversity of wild S (2001) Woody plant diversity and structure of shade-
Coffeaarabicapopulations in Ethiopia as a contribution to grown-coffee plantations in Northern Chiapas, Mexico.
conservation and use planning. Dissertation, University of Rev. Biol. Trop. 49:977-987
Bonn Soto-Pinto L, Perfecto I, Castillo-Hernandez J, Caballero-
Kent M, Coker P (1992) Vegetation Description and Nieto J (2000) Shade effect on coffee production at the
Analysis: A Practical Approach. Belhaven Press, London northern Tzeltal zone of the state of Chiapas, Mexico.
Kindt R, Boffa JM, Lengkeek AG, Degrande A, Mbosso C, Agriculture, Ecosystems and Environment 80:6169
Van O D (2008) Tree Species Diversity in Cultivated Tadesse W (2003) Vegetation of the Yayu Forest in SW
Landscapes: Investigation of Biodiversity Patterns through Ethiopia: Impacts of Human Use and Implications for In
Training of African Researchers. ICRAF, Nairobi Situ Conservation of Wild Coffeaarabica L. Populations.
Komar O (2006) Ecology and conservation of birds in coffee Dissertation, University of Bonn
plantations: a critical review. Bird Conservation Tadesse W G, Denich M, Demel T, Vlek P L G (2002)
International 16:123 Human impacts on Coffeaarabica genetic pool in
Lopez-Gomez A M, Williams-Linera G, Manson R H (2008) Ethiopia and the need for its in situ conservation. In:
Tree species diversity and vegetation structure in shade Engels J, RamanathaRao V, Brown A H D, Jackson M
coffee farms in Veracruz, Mexico. Agriculture, Ecosystems (Eds) Managing plant genetic diversity. CAB
and Environment 124:160172 International / IPGRI, pp 237-247
Magurran A E (1988) Ecological diversity and its Tejeda-Cruz C, Silva-Rivera E, Barton J R, Sutherland W
measurement.Croom Helm Ltd, London J (2010) Why shade coffee does not guarantee
Mendez V E, Gliessman S R, Gilbert G S (2007) Tree biodiversity conservation. Ecology and Society15:1-13
biodiversity in farmer cooperatives of a shade coffee Tengnas B (1994) Agroforestry extension manual for
landscape in western El Salvador. Agriculture, Kenya. ICRAF, Nairobi, Kenya
Ecosystems and Environment 119:145159
1
 304 Int. J. Agrofor. Silvicult.

Tesfaye A (2005) Diversity in Homegarden Agroforestry ZebeneA (2003) Tree species diversity, Top soil
Systems of Southern Ethiopia. Dissertation, University conditions and Arbuscular mycorrhizal association in
of Wageningen the Sidama traditional agroforestry land use, southern
Yitebitu M (2009) The impact of over storey trees on Ethiopia. Dissertation, University of Acta.
sustainable coffee (Coffeaarabica) production in
southern Ethiopia. Dissertation, University of Leibniz