Woody plant diversity in an Afromontane agricultural landscape (Debark District, northern Ethiopia)

This article was downloaded by: [Addis Ababa University]On: 05 August 2014, At: 22:24Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Forests, Trees and LivelihoodsPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/tftl20

Woody plant diversity in anAfromontane agricultural landscape(Debark District, northern Ethiopia)Belay Teferaa, Morgan L. Ruelleb, Zemede Asfawc & BerhanuAbraha Tsegayd

a Department of Biology, Madawalabu University, Bale Robe,Ethiopiab Department of Natural Resources, Cornell University, Ithaca, NY,USAc Department of Plant Biology and Biodiversity Management, AddisAbaba University, Addis Ababa, Ethiopiad Department of Biology, Bahir Dar University, Bahir Dar, EthiopiaPublished online: 01 Aug 2014.

To cite this article: Belay Tefera, Morgan L. Ruelle, Zemede Asfaw & Berhanu Abraha Tsegay(2014): Woody plant diversity in an Afromontane agricultural landscape (Debark District, northernEthiopia), Forests, Trees and Livelihoods, DOI: 10.1080/14728028.2014.942709

To link to this article: http://dx.doi.org/10.1080/14728028.2014.942709

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. Taylor & Francis, our agents,and our licensors make no representations or warranties whatsoever as to the accuracy,completeness, or suitability for any purpose of the Content. Versions of publishedTaylor & Francis and Routledge Open articles and Taylor & Francis and Routledge OpenSelect articles posted to institutional or subject repositories or any other third-partywebsite are without warranty from Taylor & Francis of any kind, either expressedor implied, including, but not limited to, warranties of merchantability, fitness for aparticular purpose, or non-infringement. Any opinions and views expressed in this articleare the opinions and views of the authors, and are not the views of or endorsed byTaylor & Francis. The accuracy of the Content should not be relied upon and should beindependently verified with primary sources of information. Taylor & Francis shall not beliable for any losses, actions, claims, proceedings, demands, costs, expenses, damages,and other liabilities whatsoever or howsoever caused arising directly or indirectly inconnection with, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Terms &Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions It is essential that you check the license status of any given Open and OpenSelect article to confirm conditions of access and use.

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

Woody plant diversity in an Afromontane agricultural landscape(Debark District, northern Ethiopia)

Belay Teferaa, Morgan L. Ruelleb*, Zemede Asfawc and Berhanu Abraha Tsegayd

aDepartment of Biology, Madawalabu University, Bale Robe, Ethiopia; bDepartment of NaturalResources, Cornell University, Ithaca, NY, USA; cDepartment of Plant Biology and BiodiversityManagement, Addis Ababa University, Addis Ababa, Ethiopia; dDepartment of Biology, Bahir DarUniversity, Bahir Dar, Ethiopia

Woody plants serve a wide range of economic, sociocultural, and ecological functionswithin traditional farming systems. Conservation of woody plant diversity withinagricultural landscapes is therefore critical to farmers’ livelihoods. We studied theconservation status of woody plant species and associated indigenous knowledge ofsmall-holder farming communities in the Debark District of northern Ethiopia. Weconducted interviews with 60 informants and ran free-listing, preference ranking anddirect matrix ranking exercises to measure the use of woody plants by farmingcommunities. To compare farmers’ ranking of plants with their abundance in thelandscape, we measured plant frequencies, densities, and diversity by undertaking avegetation survey. Compared with 55 woody plants named by farmers duringinterviews, only 14 species were encountered in the vegetation plot data; most of theremaining species were rare and therefore located by targeted searches. We foundrelatively low indices of species diversity (H0 ¼ 0.58) and evenness (J0 ¼ 0.21),indicating the low conservation status of most woody plants. Trees and shrubsindigenous to the Debark landscape have been recently replaced by an exotic treespecies (Eucalyptus globulus) with a more competitive growth rate, which is valued byfarmers as an economic species. The most preferred indigenous tree species (e.g. Oleaeuropaea ssp. cuspidata and Juniperus procera) have become increasingly rare, asreported in interviews and confirmed by direct field observation and vegetation plotdata. Households have started planting some indigenous woody species; however, thevast majority of new plantings are E. globulus. Swift coordinated actions are necessaryto prevent the rapid replacement of indigenous woody plant diversity by a monocultureof non-native trees. A community-based program for integrated ecological restorationof indigenous woody plant diversity would require support from local governmentagencies and non-governmental organizations.

Keywords: ethnobotanical methods; Eucalyptus globulus; indigenous ecologicalknowledge; traditional agroforestry; woody plant species; vegetation survey

Introduction

Small-holder farmers throughout the world deliberately maintain trees and shrubs on land

that is used for cropping and/or grazing. The broad range of roles woody plants play in

traditional farming systems have been described by many authors (e.g. Mekonnen 2001;

Elias et al. 2006; Garrity et al. 2006; Negash 2007; Neba 2009; Alemayehu & Hager 2010;

Tabuti 2012). Farmers protect, plant, and promote woody species within and around their

home gardens, fields, and communal pasturelands to derive a range of benefits, including

provisions of food, fodder, construction materials, farm equipment, fuel wood, and

medicines. The retention of trees and shrubs in agricultural landscapes depends on local

q 2014 The Author(s). Published by Taylor & Francis.

This is an Open Access article. Non-commercial re-use, distribution, and reproduction in any medium, provided the original work

is properly attributed, cited, and is not altered, transformed, or built upon in any way, is permitted. The moral rights of the named

author(s) have been asserted.

*Corresponding author. Email: [email protected]

Forests, Trees and Livelihoods, 2014

http://dx.doi.org/10.1080/14728028.2014.942709

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

ecological knowledge regarding the use and conservation of species, the values of plants

within subsistence and market economies, land and other resource tenure systems that

determine access, spiritual beliefs and traditions associated with plants, as well as changes

in sociocultural structures (Martin 1995; Neba 2009).

Deforestation is a significant challenge across mountainous regions of sub-Saharan

Africa, particularly in the Ethiopian highlands. In the Amhara region of northern Ethiopia,

average landholdings of farming families have declined to less than 1 ha per family,

driving rapid conversion of forests, woodlands, and shrublands to crop fields and treeless

rangeland (Mengistu & Hager 2009a). Rates of deforestation in Ethiopia have been

difficult to measure (Wøien 1995, McCann 1997), and recent analyses of satellite imagery

indicate that changes in forest cover are highly variable between landscapes (Gebrehiwot

et al. 2014). Nonetheless, loss of woody plant diversity poses a clear threat to the well-

being of subsistence farming communities (Teketay 2001). In order to reduce existing

pressure on remaining forests and maintain the valuable services provided by woody

plants, many local communities are renewing efforts to retain and plant useful trees and

shrubs within croplands, grazing areas, and other parts of their landscapes. These practices

can increase the availability of food, health, and energy resources to farming families,

diversify opportunities for income generation, and thereby optimize the productivity,

stability, and resilience of farming systems (Hoekstra et al. 1990). Furthermore, planting

and protecting trees and shrubs can help reverse deforestation while offsetting the

socioeconomic impacts of land scarcity, because farming families are able to produce

multiple products from their relatively small landholdings.

Previous studies have documented local ecological knowledge associated with trees

and shrubs in the Ethiopian highlands and demonstrated that woody plant diversity is

critical to farmers’ livelihoods (e.g. Bekele-Tesemma 2007, Negash 2007, Mengistu &

Hager 2009a, 2009b). In order to support the conservation of useful woody plants, it is

necessary to compare local knowledge and values of plants with their abundance and

distribution in the landscape. Therefore, the goals of this research were to document

knowledge and use of woody plants by farmers, to assess their relative values to local

communities, and to measure abundances and diversity of woody plants in the surrounding

agricultural landscape. Additional objectives of our research included observing

differences in woody plant knowledge among gender and age groups, ascertaining

major threats to woody plants, and identifying local actions necessary for future

conservation efforts. We selected the Debark District of northern Ethiopia as a

representative Afromontane landscape with a long agricultural history that includes recent

depletion of indigenous woody species and replacement by Eucalyptus globulus, as well as

to highlight challenges and opportunities likely shared by other highland communities in

sub-Saharan Africa.

Materials and methods

Description of the study area

Our research was conducted in October and November of 2012 in communities

surrounding the market town of Debark, located in the North Gonder Zone of the Amhara

Region of Ethiopia at an elevation of 2850m above sea level (Figure 1). The Debark

District encompasses 1647 km2 on the western slopes of the Semien Mountains. From

2002 to 2011, the average monthly temperatures in Debark ranged between 13.0 and

16.58C in December and April, respectively; most of the precipitation occurred between

June and September, ranging from 900 to 1400mm/year (NMSA 2013). Based on the

2 B. Tefera et al.

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

amount of annual rainfall and altitudinal range, plant ecologists estimate that the land

around Debark was once covered by Afromontane forests dominated by Juniperus procera

(Friis et al. 2011), but it is likely that much of the area was cleared for farming hundreds if

not thousands of years ago. According to the 2007 national census, the population of the

Debark District was 159,153 individuals; 13% live within Debark town and the remaining

87% reside in rural areas (CSA 2007). About 95% of Debark residents are Ethiopian

Orthodox Christians; the remainder is predominantly Muslim (CSA 2007). The vast

majority of Debark residents speak Amharic as their mother tongue.

The study area is located between 2600 and 3000m above sea level, which falls within

the DEGA(cold highland) traditional agroecological zone (Hurni 1998). The people within

this zone are subsistence farmers practicing mixed crop and livestock agriculture. The

majority of fields are devoted to barley (Hordeum vulgare), wheat (Triticum spp.), fava

bean (Vicia faba), pea (Pisum sativum), and flax (Linum usitatissimum). In home gardens

and irrigated fields lying close to streams and rivers, farmers plant potatoes (Solanum

tuberosum), garlic (Allium sativum), and leafy greens (e.g. Brassica carinata A. Braun).

Most farmers own domestic animals, including cattle, horses, donkeys, mules, sheep,

goats, chickens, and bees.

Sampling and collection of ethnobotanical data

Of the 31 kebeles (administrative sub-districts) of Debark District, we conducted

interviews in six kebeles located within 12 km of Debark town. These kebeles encompass a

rural landscape consisting of small villages, croplands, grazing areas, and adjacent spaces

Figure 1. Map of study area, including Debark city and surrounding kebeles (sub-districts) visitedfor interviews and vegetation surveys.

Afromontane agricultural landscape 3

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

associated with agricultural life. Participants were selected by random sampling to equally

represent kebeles and avoid oversampling easily accessible kebeles and households.

Fifteen random points were generated within each kebele using the ‘Create Random

Points’ tool in a geographic information system (ESRI ArcMap 10.0). The coordinates for

these points were uploaded to a Garmin eTrex Legend geographic positioning system

(GPS), which was used to navigate to each point. Upon arriving at the random point, the

research team identified the closest settlement and solicited household members to

participate in an interview. In cases where no household member was available or in rare

cases when those present did not consent to an interview, the team navigated to the next

random point, until 10 interviews had been conducted in each of the six kebeles, for total

60 informants.

In the first round of interviews, we applied ethnobotanical methods (Martin 1995,

Kindt 2006) to investigate the relationships between woody plants and farming

communities. Informants free-listed the local names of woody plant species including

trees, shrubs, and lianas located within their village and the surrounding landscape. For

each species listed, the interviewer asked a series of questions about its use, cultivation,

and association with specific habitats. Informants were also asked about criteria for

selecting woody species for cultivation, major threats to woody plants, and the

management practices that local communities use for their conservation.

After the first round of interviews was complete, 12 key informants (seven men and

five women) were preferentially sampled from among the original 60 informants,

following previous ethnobotanical research (e.g. Hailu & Asfaw 2009). Key informants

were selected based on their knowledge and enthusiasm regarding woody plants. A second

round of detailed interviews was conducted with the key informants, beginning with a

series of questions about the parts of plants used for purposes discussed earlier and the

abundance of woody plants within the surrounding landscape. Preference ranking and

direct matrix ranking exercises were conducted with each key informant.

The 12 woody plants that were most frequently listed during the first round of

interviews were selected for preference ranking in the manner recommended by Martin

(1995). The names of these plants were printed on cards, and key informants were asked to

arrange the cards according to their general preferences (they placed the most preferred

plant at the top and the least preferred at the bottom). In cases where a key informant was

not able to read Amharic script, the research team read the names of plants aloud and

arranged the cards according to the key informant’s instructions.

The five most frequently mentioned plants during the first round of interviews were

selected for direct matrix ranking following Martin (1995) with key informants. In

addition, the four most frequently mentioned uses and four most commonly cited criteria

for cultivation were included as attributes for matrix ranking. Key informants ranked the

five plants according to each attribute. Ranks were recorded in a matrix (five for the

highest rank, one for the lowest), and a total score for each species was computed for the

12 key informants and the species were each assigned an overall rank.

Vegetation survey

A vegetation survey was conducted in November 2012 to estimate the relative abundance

of woody plants in the landscape. Total 200 random plots were generated using the ‘Create

Random Points’ tool in ESRI ArcMap 10.0. Due to the sparse distribution of woody plants

within the study area, a large proportion of these plots did not contain any woody plant.

Therefore, before vegetation sampling, each point was viewed in high-resolution aerial

4 B. Tefera et al.

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

photography using Google Earth. Points located within 50m of any sign of woody

vegetation were selected for on-the-ground vegetation sampling; all other points were

documented as containing no woody elements. To correct for any errors in observing

aerial photography, 10% of the points in which no woody vegetation was observed in

Google Earth were visited to verify that no woody plants were present. For these and all

points where presence of woody plants had been detected, the research team navigated to

the point with a Garmin eTrex Legend GPS. A circular plot of 10-m radius (0.0314 ha) was

centered on each point. All woody plants within the plot greater than 5 cm in height were

counted and recorded per species.

Many plant species that had been free-listed by farmers were never encountered during

the vegetation survey. Therefore, targeted searches were made in March 2013 based on

recommendations of key informants and other local plant experts. For specimen

collection, 38 of the remaining species were located; six species (each mentioned by a

single informant) could not be located and were therefore excluded from further analysis.

Plant specimens were collected, pressed, and dried using standard methods (Martin 1995),

and transported to Bahir Dar University where identification and confirmation was made

based on the relevant volumes of the Flora of Ethiopia and Eritrea (Hedberg & Edwards

1989; Edwards et al. 1995; Hedberg et al. 2003, 2006, 2009; Tadesse 2004), and

supplementary field guides (Puff & Nemomissa 2005, Bekele-Tesemma 2007).

Data analysis

The total number of woody plants free-listed by all informants were recorded and

tabulated per local and scientific names. Differences in the number of plants listed per age

and gender categories were analyzed using non-parametric statistics, including the Mann–

Whitney U test (McKnight & Najab 2010) to compare the number listed by men and

women, and the Kruskal–Wallis H test (Dalgaard 2008) to detect differences between age

classes. Preference ranking data were analyzed by calculating the mean rank assigned to

each woody plant by key informants. Direct matrix data were analyzed by summing the

ranks obtained for all attributes by each plant for all key informants.

From vegetation surveys, plot frequency was calculated by dividing the number of

plots containing a plant by the total number of plots analyzed, including those that did not

contain any woody plant (n ¼ 200). Density of each woody plant was calculated by

dividing the number of individuals counted within all 200 plots by the total area sampled

(6.28 ha). Relative abundance of each woody plant was calculated by dividing the number

of individuals of a species counted across all plots by the total number of individuals of all

woody plant species counted in all plots.

Diversity of woody plants was analyzed by calculating the Shannon diversity and

evenness indices (Shannon 1949). The diversity index (H0) is the negative sumof all relative

abundances ( p) multiplied by the natural logarithm of the relative abundance (ln p):

H0 ¼ 2XR

i

pi ln pi:

The evenness index (J0) was calculated by dividing the diversity index by the maximum

possible diversity (H0max ) based on the number of species observed across the study area (S):

J 0 ¼ H 0

H 0max

¼ H 0

ln S:

Afromontane agricultural landscape 5

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

Results and discussion

Knowledge of farmers regarding woody plant species

The 60 informants generated a list of names of woody plants, which were later identified

using the voucher specimens and found to consist of 55 species belonging to 49 genera and

34 families (Table 1). These woody plants included 48 indigenous species (including two

endemics) and seven introduced species. The Fabaceae and Rosaceae were each

represented by six species, Euphorbiaceae by four species, and Moraceae by three species.

Asteraceae, Cupressaceae, Lamiaceae, Malvaceae, Myrsinaceae, Oleaceae, and

Solanaceae were each represented by two species. The remaining 22 families were each

represented by a single species. These results are similar to recent findings of Abate et al.

(2012) in arid rangelands of south-eastern Ethiopia (500–1780m above sea level), in

which Fabaceae accounted for the largest proportion of woody plants. Considering the

temperate origins of the Rosaceae, it is not surprising that they also show high diversity in

the northern Ethiopian highlands where the climate is similar to temperate regions.

The woody plant habitats described by informants included home gardens, crop fields,

fence lines, grazing lands, church compounds, riparian areas, roadsides, forest patches,

and hilltops (Figure 2). Farmers’ listing showed that home gardens and yards contained the

greatest woody plant richness (35 species), followed by crop fields (29 species) and forests

(26 species). Other habitats mentioned by farmers include riparian zones (12 species),

grazing areas (11 species), hilltops (11 species), church compounds (10 species), fences

and hedgerows (including living fences and boundary plantings, nine species), and

roadsides (nine species). In addition, 16 species were described as ubiquitous in the

landscape, indicating that their habitat requirements are less specific, being ecologically

more generalist than other species.

Our results align with those of Tolera et al. (2008), who found that the woody plant

species richness was much higher in home gardens than in adjacent crop fields. Our

interviews indicate that home gardens have the highest woody plant diversity in part

because homeowners feel a stronger sense of ownership, and plants in home gardens

benefit from the direct care of household members. In particular, perennial plants in home

gardens provide continued benefits by their long-term presence in close proximity to the

family home.

Uses of woody plants

Woody plants serve a wide range of purposes for farming families in the Debark District.

The most frequently mentioned uses include fencing (34 species), fuel wood (33 species),

and shade (32 species). Woody plants are also used as forage and fodder (25 species),

construction materials (20 species), tools and equipment (18 species), sources of income

(17 species), cleaning supplies (branches and leaves for sweeping the home and washing

dishes, 16 species), traditional medicines (10 species), improving soil fertility (10 species),

furniture (nine species), bee forage (nine species), ornamentals (nine species), food (nine

species), spiritual practices (four species), and toothbrushes (two species). Furthermore,

most woody plants are used for multiple purposes. For example, 29 species are used for

both fencing and fuel wood (cf. Table 1).

Most of the trees, shrubs, and woody climbers named by our informants are used

throughout the DEGA zone of the Ethiopian highlands and are therefore well-documented

in existing ethnobotanical literature (e.g. Bekele-Tesemma 2007, Negash 2007). To our

knowledge, only two previous studies have investigated use of woody plants within the

6 B. Tefera et al.

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

Table

1.

Woodyplantslisted

byinform

ants,includingthepercentofallinform

antswhoincluded

each

plantduringfree-listing(n

¼60).

No.Scientificnam

eaFam

ily

Localnam

e(A

mharic)

Growth

habit

Free-list

(%)

Habitatb

Usesc

1AcaciaabyssinicaBenth.

Fabaceae

Girar

Tree

87

Hg,Cf,Ft,Ub,Ch,Rz,

Ga,Ht1,2,3,5,6,7,8,10,15,14

2AcaciabrevispicaHarms

Fabaceae

Qentefa

Liana

2Hg,Ft

1,13

3AcaciadecurrensWilld.j

Fabaceae

Yeferenjigirar

Tree

10

Hg,Rz

1,2,3,4,6,7,8,10,11,13

4AcaciamelanoxylonR.Br.j

Fabaceae

Yeferenjiweyra

Tree

3Hg

2,3,5,6,8,10,11

5ArundodonaxL.

Poaceae

Shem

beko

Shrub

7Hg

1,5,6,8,13

6Buddleja

polystachya

Fresen.

Scrophulariaceae

Anfar

Tree

8Hg,Ub,Ch,Rs

1,2,3,4

7Calpurnia

aurea(A

iton)Benth.

Fabaceae

Zigita

Shrub

8Cf,Ft

1,2,6

8Carissaspinarum

L.

Apocynaceae

Agam

Tree

5Ft

1,2,12

9ClutiaabyssinicaJaub.&

Spach.Euphorbiaceae

Fiyelefej

Shrub

8Hg,Cf,Ub,Ht

1,2,3,4,5,14

10

Cordia

africanaLam

.Boraginaceae

Wanza

Tree

3Hg

3,10,12

11

CrotonmacrostachyusHochst.

exDelile

Euphorbiaceae

Bisana

Tree

2Cf,Ft

1,3

12

CupressuslusitanicaMill.j

Cupressaceae

Yeferenjitsid

Tree

18

Hg,Cf,Ft,Ch,Rz

1,2,3,6,7,8,15,10

13

CytisusproliferusL.f.j

Fabaceae

Menoqinteba

Shrub

38

Hg,Cf,Ch,Rz,Fn

1,2,3,4,5,6,7,10,11,13

14

Discopodium

penninervum

Hochst.

Solanaceae

Alm

itShrub

7Hg,Cf,Ft

1,2,3,5,9

15

Dodonaea

angustifoliaL.f.

Sapindaceae

Kitkita

Tree

12

Hg,Ft

1,2,3,4,5,6,7,10

16

Dombeyatorrida(J.F.Gmel.)

Bam

ps

Malvaceae

Wulkifa

Tree

12

Hg,Cf,Ft,Ch,Ga

1,2,3,5,6,10,11,13

17

Erica

arboreaL.

Ericaceae

Wuchena

Tree

12

Cf,Ft,Ub,Ga,

Ht

1,2,3,4,5

18

EucalyptusglobulusLabill.j

Myrtaceae

Bahirzaf

Tree

100

Hg,Cf,Ft,Ub,Rz,Ga,

Fn,Fb

1,2,3,6,7,8,10,13,14

19

Euphorbia

abyssinicaJ.F.Gmel.Euphorbiaceae

Qulqual

Tree

12

Hg,Cf,Ch

1,2,6,8,10

20

FicussurForssk.

Moraceae

Shola

Tree

5Cf,Rs

1,2,3,12

21

[FicusthonningiiBlume]

[Moraceae]

Chiviha

Tree

2Hg

3,5,14

22

FicusvastaForssk.

Moraceae

Warka

Tree

2Ft

3,12

23

Hagenia

abyssinica(Bruce

exSteud.)J.F.Gmel.d

Rosaceae

Koso

Tree

23

Hg,Cf,Ft,Ch,Fn

1,2,3,7,8,9,10,14,15

24

Hypericum

revolutum

Vahl

Guttiferae

Amija

Shrub

20

Hg,Cf,Ft,Ub,Rz,Ga,

Ht

1,2,3,4,5,6,10,11,13

25

Jasm

inum

grandiflorum

L.

Oleaceae

Tem

belel

Shrub

2notdiscussed

gnonementioned

h

(Continued

)

Afromontane agricultural landscape 7

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

Table

1–continued

No.Scientificnam

eaFam

ily

Localnam

e(A

mharic)

Growth

habit

Free-list

(%)

Habitatb

Usesc

26

Juniperusprocera

Hochst.ex

Endl.

Cupressaceae

Tsid(habesha)

Tree

57

Hg,Cf,Ft,Ub,Ch,Ga,

Fn,Ht

1,2,3,4,5,6,7,8,10

27

Justicia

schimperiana(H

ochst.

exNees)

T.Anderson

Acanthaceae

Sim

iza

Shrub

5Hg,Cf,Fn

1,2,6,9,14

28

Kalanchoesp.

Crassulaceae

Endaw

ula

Shrub

3Ft,Rz

nonementioned

h

29

Maesalanceolata

Forssk.

Myrsinaceae

Swarya

Tree

2notdiscussed

gnonementioned

h

30

MalussylvestrisL.j

Rosaceae

Apil

Tree

7Hg

731

Malvaverticillata

L.

Malvaceae

Lut(yesew

)Shrub

3notdiscussed

g5,6,9,13

32

Maytenusarbutifolia(H

ochst.ex

A.Rich.)R.Wilczek

dCelastraceae

Atat

Tree

20

Hg,Cf,Ft,Ub,Rz,Ga,

Fn,Rs

1,2,3,5,9

33

MyricasalicifoliaHochst.ex

A.Rich.

Myricaceae

Sinich

Tree

13

Cf,Ft,Ub,Rz,Ht

1,2,3,5,6,10

34

MyrsineafricanaL.

Myrsinaceae

Qechem

oShrub

3Cf,Ft

1,2,6

35

Nuxiacongesta

R.Br.ex

Fresen.Stilbaceae

Atquaro

Tree

5Hg,Ht

1,2,3,4,5

36

Ocimum

lamiifolium

Hochst.Ex

Benth.

Lam

iaceae

Dam

akese

Shrub

2Ht

1,4

37

Oleaeuropaea

L.ssp.cuspidata

Oleaceae

Weyira

Tree

38

Hg,Cf,Ft,Ub,Ch,Ga

1,2,3,4,5,6,7,10,14,15,16

38

Opuntiaficus-indica(L.)Mill.

Cactaceae

Ashew

aqulqual

Shrub

2Hg

3,5,12

39

Osyrislanceolata

Hochst.&

Steud.

Santalaceae

Qeret

Tree

2Ft,Ch

2,5

40

Phytolaccadodecandra

L’H

er.

Phytolaccaceae

Endod

Liana

27

Hg,Cf,Ft,Ub,Rz,Fn

1,2,3,4,5,7,9,10

41

Prunuspersica

(L.)Batschj

Rosaceae

Kok

Tree

18

Hg,Ga

7,12

42

Rhamnusprinoides

L’H

er.

Rham

naceae

Gesho

Shrub

62

Hg,Cf,Rz,

Ht

3,5,7

43

RhusvulgarisMeikled

Anacardiaceae

Embes

Tree

15

Cf,Ft,Ub,Ga

1,2,3,5,6,7

44

RicinuscommunisL.

Euphorbiaceae

Bulqa

Tree

13

Hg,Cf

3,7,11

45

Rosa

abyssinicaR.Br.ex

Lindl.d

Rosaceae

Qaga

Shrub

27

Hg,Cf,Ft,Ub

1,2,3,4,5,10,11,12,14

46

Rosa

x-richardiiRehd.

Rosaceae

Tsigereda

Shrub

2Hg

947

RubussteudneriSchweinf.d

Rosaceae

Enzoryia

Tree

2Ft

2,6,12

48

Rumex

nervosusVahl

Polygonaceae

Embacho

Shrub

23

Hg,Cf,Ft,Ub,Ht

1,2,3,4,5,9,14

8 B. Tefera et al.

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

49

Ruta

chalepensisL.

Rutaceae

Tenaadam

Shrub

2Hg

7,9,12

50

Rydingia

integrifolia(Benth.)

Scheen&

V.A.Alberte

Lam

iaceae

Tinjut

Shrub

2Cf

2,4,7

51

Salixmucronata

Thunb.f

Salicaceae

Haya

Tree

3Hg,Cf,Ga

10,16

52

Solanecio

gigas(V

atke)

C.Jeffrey

iAsteraceae

Mogneqitel

Shrub

43

Hg,Cf,Ub,Fn,Ht

1,3,9,11,13,14

53

Solanum

marginatum

L.f.

Solanaceae

Embuay

Shrub

3Ub,Rz

454

Urera

hypselodendron(H

ochst.

exA.Rich.)Wedd.

Urticaceae

Lanquso

Liana

2notdiscussed

gnonementioned

h

55

Vernonia

rueppelliiSch.Bip.ex

Walp.

Asteraceae

Shuteni

Shrub

3Fn

1,2,5,11

Notes:In

additionto

55specieslisted

here,threeplantsnam

edbyinform

ants(Leliha,Ambel,Qererto)could

notbelocatedforspecim

encollection.Allspeciesareindigenousunless

indicated

otherwise.

aAllscientificnam

eshavebeenverified

ontheInternational

plantlist(w

ww.plants.org).

bHabitatswithin

thestudyarea

aslisted

byfarm

ers:Hg!

Homegardens;Cf!

Cropfields;Ft!

Forest;Ch!

Church;Rz!

Riparianzones;Ga!

Grazingareas;Fn!

Fences;

Ht!

Hilltops;Rs!

Road

sides;Bd!

Boundaries;Ub!

Ubiquitous.

cUseswithin

thestudyarea

aslisted

byfarm

ers:

1!

Fencing;2!

Fuel

wood;3!

Shade;

4!

Cleaningsupplies;5!

Fodder;6!

Construction;7!

Income;

8!

Furniture;

9!

Medicine;

10!

Tools;11!

Bee

forage;

12!

Food;13!

Ornam

ental;14!

Soilfertility;15!

Spiritual;16!

Dentalcare.

dThesenam

esarecurrentlyunresolved

ontheInternational

plantlist.

eSynonym

ofOtostegia

integrifoliaBenth.

fSynonym

ofSalixsubserrata

Willd.

gThesingle

inform

antwholisted

thisplantdid

notreportitshabitat.

hThesingle

inform

antwholisted

thisplantdid

notreportanyuses.

iEndem

icto

Ethiopia.

jIntroduced(notnativeto

Ethiopia).

Afromontane agricultural landscape 9

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

Debark District; the first focused on indigenous edible fruits (Mengistu & Hager 2009a,

2009b) and the second on medicinal plants (Abebe 2011). With regard to edible fruits, our

informants identified only six species as compared to 11 identified by Mengistu and Hager

(2009b) in Debir kebele alone. Several of the species free-listed in the latter study (e.g.

Coffea arabica) are unlikely to grow in the DEGA zone, and may reflect farmers’

knowledge and use of surrounding lowlands. With regard to medicinal plants, 74.5% of the

species named by our informants were described as medicinal plants by Abebe (2011),

whereas only 10 (18%) of the species were identified as medicines by our informants. It is

likely that the medicinal uses of many plants were known to the traditional healers and

medicinal plant vendors interviewed by Abebe (2011), but not to our informants.

Special values of some woody plants

Preference ranking indicated that E. globulus was the most preferred woody plant by

almost all farmers, while Rhamnus prinoides, Olea europaea ssp. cuspidata, Acacia

abyssinica and J. procera were ranked second, third, fourth, and fifth, respectively

(Table 2). As in other regions where it has been introduced, E. globulus is favored for its

rapid growth rates, minimal management requirements, and coppicing ability (Turnbull &

Pryor 1978; FAO 1981). By comparison, the slow growth rates of most indigenous tree

species were mentioned by farmers as a critical factor discouraging their cultivation.

Confirming the preference ranking results, direct matrix ranking also indicated that

E. globulus is the most highly valued tree based on multiple attributes (Table 3). Key

informants ranked it highest for its use in fencing and construction, as well as a source of

cash income. Only A. abyssinica was ranked higher than Eucalyptus as a source of fuel

wood. Eucalyptus was also ranked higher than all other trees based on its growth

performance, the availability of saplings at local extension offices, and its overall

multifunctionality (variety of uses). The only attribute for which Eucalyptus was not the

first or second ranked tree was as a source of fodder, because its leaves are not palatable to

livestock. These findings are confirmed by the report of FAO (2002), which characterized

Figure 2. Conservation of Acacia abyssinica (trees in foreground) within a crop field south ofDebark town. Note the abundance of E. globulus in the background.

10 B. Tefera et al.

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

Table

2.

Preference

rankingresultsforkey

inform

antsformostfrequentlylisted

woodyplants.

Key

inform

ant

code

Species

Eucalyptus

globulus

Rhamnus

prinoides

Olea

europaea

ssp.

cuspidata

Acacia

abyssinica

Juniperus

procera

Phytolacca

dodecandra

Hagenia

abyssinica

Cytisus

proliferus

Rosa

abyssinica

Maytenus

arbutifolia

Rumex

nervosus

Solanecio

gigas

DB1

12

10

11

98

67

53

41

2DB2

12

111

10

97

86

45

32

DB3

11

12

87

610

45

91

23

DB4

10

12

11

85

29

76

43

1DB5

12

11

97

810

65

42

13

DB6

11

12

710

89

14

62

53

DB7

12

11

10

94

81

27

56

3DB8

11

12

89

10

67

52

34

1DB9

10

12

11

79

62

85

31

4DB10

10

712

56

811

39

42

1DB11

912

10

811

75

36

24

1DB12

12

11

97

86

510

13

24

Total

132

123

117

96

92

85

66

63

62

38

34

28

Rank

12

34

56

78

910

11

12

Afromontane agricultural landscape 11

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

Eucalyptus as the most preferred multipurpose tree for small-holder farmers throughout

the Ethiopian highlands.

Variation in woody plant knowledge

Analysis of the number of plants listed by men and women indicated that there is some

difference in their knowledge about woody plants (Table 4). Men listed more woody plants

than women, and the difference is statistically significant as shown by the results of the

Mann-Whitney U test ( p , 0.05). The difference in knowledge observed between genders

can be explained by the notion that woody plant knowledge emerges from interactions

between people and plants. Women tended to list woody plants that have household uses

(e.g. those used for food preparation and storage) while men mentioned woody plants that

are used to fashion farming tools, in home construction, and as fodder for livestock. In

following traditional gender roles, the variety of woody plants used by men may be greater

than those used by women. Furthermore, men have a higher probability of mentioning

woody plants that are found around homesteads, farmlands, and in uncultivated areas

within the agricultural landscape because they are more likely to encounter them on a

regular basis.

On the other hand, there is a possibility that our results were slightly influenced by the

nature of interactions between researchers (all men) and informants. Some female

informants may have been uncomfortable speaking at length with researchers resulting in

the shorter lists of plants. Women in rural areas are sometimes wary of talking with men

from outside the community, and several cases were noted in which interviews with

women had to be cut short.

At the a ¼ 0.05 level of significance, there was not sufficient evidence to draw any

conclusion about differences in the numbers of woody plants listed between the four age

groups (Table 4). This result is supported by informants, who often stated that all age

groups interact with woody plants and are equally involved in their management as well.

This finding indicates that each age group is about equally knowledgeable about woody

plants in the study area.

Frequency and abundance of woody plants

Woody plants are sparsely distributed in the agricultural landscapes of Debark District;

therefore, relatively few (39 plots) of the random circular plots (n ¼ 200) contained any

Table 3. Direct matrix ranking results from key informants, including five most frequently listedplants and eight most frequently discussed attributes.

AttributesEucalyptusglobulus

Acaciaabyssinica

Juniperusprocera

Rhamnusprinoides

Solaneciogigas

Construction 57 30 51 0 0Fodder 20 56 31 35 7Fencing 55 46 42 2 23Fuel wood 47 54 42 2 0Growth performance 53 25 28 27 47Seedling availability 58 28 25 35 34Source of income 57 29 33 49 0Variety of uses 57 38 40 33 12Total 404 306 292 183 123Overall rank 1 2 3 4 5

12 B. Tefera et al.

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

tree, shrub, or liana. The total density of woody plants in the study area was 246

individuals/ha, which greatly varied between species (Table 5). More than 85% of the

individual trees encountered during the vegetation survey belonged to a single exotic tree

species, E. globulus. E. globulus had by far the greatest overall density (192 individuals/

Table 4. Non-parametric tests for number of plants listed by gender and age groups. (a) Mann–Whitney U test output for gender (b) Kruskal–Wallis H test for age groups.

(a) Number of plants listed by gender

Gender N Mean rank Sum of ranks

Men 21 37.1 779Women 39 27.0 1053Total 60

Test outputsMann–Whitney U 271Z-score 22.16p-value 0.031

(b) Number of plants listed by age group

Age group (n ¼ 4) N Mean rank

19–29 years old 20 24.930–40 years old 16 31.941–61 years old 16 35.362–82 years old 8 32.3Total 60

Test outputsChi-square 3.52Degrees of freedom (n 2 1) 3p-value 0.318

Table 5. Frequency, densities, and relative abundances of woody plants encountered duringvegetation surveys (n ¼ 200).

Scientific nameGrowthhabit

Plotspresent

Frequencyacross

plots (%)Total

individualsDensity(ind./ha)

Relativeabundance

(%)

Eucalyptus globulus Tree 24 12.0 1204 192.72 88.5Rumex nervosus Shrub 3 1.5 60 9.55 4.4Cupressus lusitanica Tree 1 0.5 33 5.25 2.4Euphorbia abyssinica Tree 2 1.0 20 3.18 1.5Discopodium penninervum Tree 1 0.5 11 1.75 0.8Vernonia purpurea Shrub 1 0.5 7 1.11 0.5Solanum marginatum Shrub 3 1.5 7 1.11 0.5Acacia abyssinica Tree 5 2.5 6 0.96 0.4Clutia abyssinica Shrub 2 1.0 4 0.64 0.3Solanecio gigas Shrub 1 0.5 3 0.48 0.2Phytolacca dodecandra Liana 1 0.5 2 0.32 0.1Myrsine africana Shrub 1 0.5 2 0.32 0.1Rhamnus prinoides Shrub 1 0.5 1 0.16 0.1Rosa abyssinica Shrub 1 0.5 1 0.16 0.1Total 39 19.5 1361 216.72 100

Afromontane agricultural landscape 13

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

ha), while other species were much less dense, ranging from 0.2 to 10 individuals/ha.

Most of the woody plant species mentioned during the interviews were not found during

the vegetation survey because their densities were too low to detect by a randomized

survey design.

The Shannon diversity index for woody plants within the study area was 0.58. The

low value of this index (which usually ranges from 0 to 3.0) confirms our observation

that woody plant diversity is low despite the relatively high number of species occurring

within the study area. This indicates the extremely low densities of most woody species

and the high relative abundance of E. globulus. Similarly, the Shannon evenness index

(0.21) indicates low evenness, i.e. dominance by a single or few species. Indeed,

E. globulus is the dominant tree and accounts for most of the woody plants within the

study area.

Comparison of data from interviews and the vegetation survey

Species richness observed during the vegetation survey (14 species) was slightly more

than one-fourth of the total listed by informants (55 plant names). Of the 55 woody plant

species listed by informants, 13 were encountered during the vegetation survey (Tables 1

and 5). One additional woody plant species that had not been mentioned by informants

was also encountered (Vernonia purpurea Sch. Bip. ex Walp.). Informants’ knowledge

about a greater number of species may reflect a time in the recent past when woody

plants were more abundant and diverse in the surrounding landscape. This theory was

supported by some informants who said that the woody plants were disappearing from

their surroundings and are now found only in the Semien Mountains National Park,

which borders the study area to the north and may serve as a refuge for many woody

plants.

The high density of E. globulus relative to other woody plants in the study area

(Table 5) reflects the special attention given by households to its planting and

management, which reflects an increasing need for cash income. In addition, for the past

several decades, local extension offices have ensured high availability of Eucalyptus

saplings; only recently have there been efforts to increase the availability of indigenous

woody species. Informants asserted that widespread and large-scale cultivation of

E. globulus is driven by its fast growth rate, multiple uses, suitability for local

environmental conditions, and high market value. As observed in the field, E. globulus is

cultivated in most land use types, and is therefore seen throughout the landscape, despite

the fact that it cannot reseed on its own. The majority of eucalyptus plantations are

densely planted woodlots, which often replace crop fields as soil fertility and crop

production decline, and single-row plantings that line roads and field boundaries.

Eucalyptus is also observed at the periphery of church compounds, along streams and

rivers, and as hilltop groves. Within Debark villages, eucalyptus poles are clearly the

primary construction material, and its branches are the most common fencing material

and fuel wood. Although there are many other multipurpose trees, including J. procera

and O. europaea ssp. cuspidata, they are said to be not as fast-growing or as easy to

manipulate as E. globulus. Furthermore, although indigenous saplings require protection

from livestock, Eucalyptus is not palatable to most livestock and therefore requires little

protection. Because of these differences, a considerable proportion of the local

community plant only Eucalyptus. After they are harvested for use, indigenous species

are increasingly replaced by Eucalyptus, resulting in its rising ecological dominance over

indigenous woody plants.

14 B. Tefera et al.

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

Future prospects for woody plants in Debark

Threats identified by farmers

According to informants, indigenous woody plants within the study area are threatened by

overexploitation, lack of natural regeneration, lack of management, low availability of

saplings for planting, climate change, planting of introduced species (specifically

E. globulus), and expansion of croplands (Figure 3). Each of these factors contributes

differently to the decline in abundance and diversity of woody plants. Around 41% (33) of

the informants reported that overexploitation is driving the decline of woody plant

diversity and density. Some of the factors reported by a small proportion of informants

in Debark are seen as primary threats to woody plant diversity in other parts of Ethiopia.

For example, Lulekal et al. (2008) reported that deforestation due to agricultural expansion

is the greatest threat to woody plants in Mana Angetu District, southeast Ethiopia.

Likewise, Hundera (2007) also showed that agricultural expansion is the major threat to

woody plants in the Jimma Zone of southwestern Ethiopia. By contrast, only two (3%)

informants in Debark mentioned expansion of croplands as a major threat to woody plant

diversity.

One female key informant in the study area commented that both the least and most

useful woody plants have the best chance of being conserved in agricultural landscapes.

Her observation seems to align with our field measurements of woody plant abundance

and diversity. The least useful woody plants and those with fewer uses are conserved

because they are left alone and not harvested. Some of these species (e.g. Rumex nervosus)

are largely ignored by farmers because they are rarely used for fuel wood or fencing

material, and can only be browsed by goats or sheep, and are therefore less threatened by

overexploitation.

Woody plants that are highly preferred by the local community are also relatively

abundant in the study area, particularly E. globulus. The high abundance of this and other

species can be attributed to the communities’ enthusiasm for cultivation to meet their own

household needs and local market demand. By comparison, woody plants such as

Figure 3. Major threats to woody plants reported by informants (n ¼ 60) in Debark District.

Afromontane agricultural landscape 15

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

O. europaea ssp. cuspidata and J. procera are dwindling according to the informants. The

few remaining trees are used by the community without replanting. The persistence of

many woody plant species is doubtful unless farmers and other community members

increase the rate of replanting to match the rate of removal.

Efforts to promote woody plant diversity

Informants mentioned many efforts to increase the abundance of woody plants within the

Debark District (Figure 4). Management and conservation practices include fencing,

watering, hoeing, adding fertilizers, preventing illegal cuttings, weeding, pruning

and creating shade. Fencing and watering were reported by 45 (75%) and 38 (63%)

informants, respectively. Because these management practices are more easily applied to

cultivated plants, it may be that non-cultivated woody plants are under greater threat. In

general, we noted a desire to conserve and increase woody plant diversity among Debark

residents, but pressures to exploit indigenous trees and replace them with Eucalyptus

surpass efforts to conserve indigenous woody plants, which clearly need actions in support

of households.

An effort to conserve woody plant diversity in the Ethiopian highlands could be

facilitated by spiritual beliefs and religious institutions. Ethiopia is home to a diversity of

spiritual systems, religious traditions, and associated cultural taboos, including many that

are related to woody plants, particularly large charismatic and emblematic trees. For

example, some farmers in the study area said that croplands can become unproductive due

to a curse brought on by the evil eye. To remedy this problem, farmers will till their land

with a plough made from the wood of Hagenia abyssinica. This species is also said to

attract lightning during the rainy season, thereby preventing lightning strikes of people,

livestock, or houses. The value of Hagenia and other species related to these and similar

beliefs and practices may contribute to their conservation in the landscape.

Figure 4. Management practices reported by informants (n ¼ 60) to promote woody plants.

16 B. Tefera et al.

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

As in other parts of Ethiopia, we observed that trees located close to Ethiopian

Orthodox churches are protected from cutting. Although we excluded church yards from

this study because we focused on the areas managed by individual households, we

observed that woody plants such as J. procera, O. europaea ssp. cuspidata, Cupressus

lusitanica, Rosa abyssinica, Euphorbia abyssinica, H. abyssinica, and A. abyssinica are

more abundant around Ethiopian Orthodox churches than in other parts of the landscape.

Many churches in Debark District are surrounded by rather large patches of trees

dominated by J. procera. Throughout northern Ethiopia, the trees and shrubs surrounding

churches and monasteries may be representative examples of remnant forest patches

(Wassie et al. 2005). These forests are protected by spiritual/religious prohibitions against

cutting trees and shrubs from these areas. As in other parts of the world, taboos and social

sanctions protect remnant forests from deterioration (King-Oliver et al. 1997). In Debark,

church yards may be among the few remaining habitats for indigenous woody species that

require the understory conditions provided by mature trees. Churches and monasteries are

examples of in situ conservation sites that have chances of survival into the foreseeable

future. These patches not only showcase the beauty and benefits of indigenous woody

species but may also serve as important sources of seeds as a living source of germ plasm

for raising saplings for planting.

Conclusion and recommendations

Our study on woody plant diversity in the environs of Debark revealed an alarmingly low

diversity of trees, shrubs, and woody climbers. It is important to remember that although

the study area was once primarily Afromontane forest and woodland, it was transformed

into an agricultural landscape a long time ago. Native woody plants are retained within this

landscape because they have been important to farmers, and local community members

underline that trees and shrubs continue to play a wide range of roles in their day-to-day

lives. However, most valuable indigenous trees and shrubs are being threatened with

overexploitation and habitat modification.

Among the land use types we investigated, home gardens are characterized by a higher

number of woody plant species than most other land use types. This is likely because home

gardens – besides being refuges for indigenous woody plants that become rare in the

environment – are under the consistent supervision and care of the household, which has

clear tenure rights for perennial plants grown around individual homesteads.

Selection of woody species for planting by households is based primarily on variety

and intensity of uses. The major uses of woody plants are for fencing and fuel wood. In the

absence of the preferred multipurpose indigenous woody plant species, E. globulus has

become the most widely planted woody plant largely on account of its fast growth, sapling

availability, and high market value that offers opportunities for income generation.

Local and regional governments as well as non-governmental organizations need to

support the conservation of ecologically, socially, and economically valuable indigenous

woody plants in agricultural landscapes. Cutting of trees for use and to make space for

farming are major threats to woody plant diversity that can be reduced by stricter

enforcement of existing laws that prohibit such activities. Self-recruitment of indigenous

species (particularly valuable woody plants such as O. europaea ssp. cuspidata,

J. procera, H. abyssinica, and A. abyssinica) should be protected from trampling and

consumption by livestock with adequate fencing. We observed that local nurseries are

raising key indigenous woody species, including the species cited above, and promotion of

their planting is an encouraging recent development. In addition, local extension workers,

Afromontane agricultural landscape 17

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

the regional forestry department, and non-governmental organizations need to collaborate

with communities to develop effective programs and policies to facilitate the protection

and care of indigenous trees and shrubs after they are planted. Farmers clearly value

woody plant diversity, but local communities need support in order for collective,

voluntary ecological restoration to gain momentum.

Acknowledgements

The authors would like to extend their gratitude to the farmers of Koha Zebzeba, Kino Libanos,Debir, Gomya, Zebena, and Mikara kebeles of the Debark District for their invaluable contributionsto this research. This research would not have been possible without the energetic field assistance ofAmanuel Birhanie, Yohannes Desalegn, and Fekadu Alem. We also thank the Debark Ketema (city)and Wereda (district) administrations for facilitating our work. Finally, we would like to emphasizeour appreciation for the editor of this journal and two anonymous reviewers whose thoughtfulcomments strengthened our work.

References

Abate T, Ebroand A, Nigatu L. 2012. Evaluation of woody vegetation in the rangeland of southeastEthiopia. Int Res J Agric Sci Soil Sci. 2:113–126.

Abebe E. 2011. Ethnobotanical study on medicinal plants used by local communities in DebarkWereda, North Gondar Zone, Amhara Regional State, Ethiopia [MSc thesis]. Addis AbabaUniversity.

Alemayehu L, Hager H. 2010. Woody plants diversity and possession, and their future prospects insmall-scale tree and shrub growing in agricultural landscapes in central highlands of Ethiopia.J Small-scale For. 9:153–174.

Bekele-Tesemma A. 2007. Useful trees and shrubs for Ethiopia: Identification, propagation andmanagement for 17 agroclimatic zones. Nairobi: Regional Soil Conservation Unit, SwedishDevelopment Authority.

CSA: Census Tabular Reports, National Population and Housing Census [Internet]. 2007. AddisAbaba: Central Statistical Agency of Ethiopia; [cited 2007 Jan 25]. Available from: http://www.csa.gov.et/index.php/2013-02-20-14-51-51/census-tables

Dalgaard P. 2008. Analysis of variance and the Kruskal–Wallis test. In: Introductory Statistics. NewYork: Springer; p. 127–143.

Edwards S, Tadesse M, Hedberg I, editors. 1995. Flora of Ethiopia and Eritrea: Vol. 2 Pt. 2,Canellaceae to Euphorbiaceae. Addis Ababa & Uppsala: The National Herbarium at AddisAbaba University.

Elias E, Jama B, Mogotsi K. 2006. Role of agroforestry in improving food security and naturalresource management in the drylands: A regional overview. J Dryland. 1:206–211.

FAO. 1981. Eucalypts for planting. FAO Forestry and Forest Products Studies 11. Rome.FAO. 2002. Biodrainage: principles, experiences & applications. International Program for

Technology & Research in Irrigation & Drainage, Knowledge Synthesis Report 6. Rome.Friis I, Demissew S, van Breugel P. 2011. Atlas of the potential vegetation of Ethiopia. Addis Ababa:

Addis Ababa University Press and Shama Books.Garrity DA, Okono M, Grayson K, Parrott S, editors. 2006. World agroforestry into the future.

Nairobi: World Agroforesty Centre.Gebrehiwot SG, Bewket W, Gardenas AI, Bishop K. 2014. Forest cover change over four decades in

the Blue Nile Basin, Ethiopia: comparison of three watersheds. Reg Environ Change.14:253–266.

Hailu H, Asfaw Z. 2009. The diversity of food and medicinal plants in the home gardens of SabataTown, Oromia National Regional State, Ethiopia. Ethiop J Biol Sci. 8:31–51.

Hedberg I, Edwards S, editors. 1989. Flora of Ethiopia and Eritrea: Vol. 3, Pittosporaceae toAraliaceae. Addis Ababa & Uppsala: The National Herbarium, Addis Ababa University.

Hedberg I, Edwards S, Nemomissa S, editors. 2003. Flora of Ethiopia and Eritrea: Vol. 4 Pt. 1,Apiaceae to Dipsaceae. Addis Ababa & Uppsala: The National Herbarium, Addis AbabaUniversity.

18 B. Tefera et al.

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014

Hedberg I, Friis I, Persson E, editors. 2009. Flora of Ethiopia and Eritrea: Vol. 8, General Part andIndex to Volumes 1–7. Addis Ababa & Uppsala: The National Herbarium, Addis AbabaUniversity.

Hedberg I, Kelbessa E, Edwards S, Demissew S, Persson E, editors. 2006. Flora of Ethiopia andEritrea: Vol. 5, Gentianaceae to Cyclochilaceae. Addis Ababa & Uppsala: The NationalHerbarium at Addis Ababa University.

Hoekstra D, Torquebiau E, Badege B, editors. 1990. Agroforestry: potentials and research needs forthe Ethiop highlands. Nairobi: ICRAF.

Hundera K. 2007. Traditional forest management practices in Jimma Zone, South West Ethiopia.Ethiop J Educ Sci. 2:2–9.

Hurni H. 1998. Agroecological belts of Ethiopia. Bern: Centre for Development and Environment,University of Bern in association with the Ministry of Agriculture, Ethiopia.

King-Oliver IED, Chitra V, Narasimha D. 1997. Sacred groves: traditional ecological heritage. Int JEcol Environ Sci. 23:463–470.

Kindt R. 2006. Ethnobotanical surveys [Internet]. Nairobi: World Agroforestry Centre; [cited 2014Feb 01]. Available from: http://www.worldagroforestry.org/

Lulekal E, Kelbessa E, Bekele T, Yineger H. 2008. An ethnobotanical study of medicinal plants inMana Angetu District, southeastern Ethiopia. J Ethnobiol Ethnomed. 4:1–10.

Martin GJ. 1995. Ethnobotany: a method manual. London: Chapman & Hall.McCann JC. 1997. The plow and the forest: narratives of deforestation in Ethiopia, 1840–1992.

Environ Hist. 2:138–159.McKnight PE, Najab J. 2010. Mann-Whitney U test. In: Corsini encyclopedia of psychology. New

York: Springer.Mekonnen K. 2001. Practices, constraints and agroforestry interventions in Yeku watershed

northeastern Ethiopia. Ethiop J Nat Resour. 3:161–178.Mengistu F, Hager H. 2009a. Integration of indigenous wild woody perennial edible fruit bearing

species in the agricultural landscapes of Amhara region, Ethiopia. Agrofor Syst. 78:79–95.Mengistu F, Hager H. 2009b. Wild edible fruit species cultural domain, informant species

competence and preference in three districts of Amhara region, Ethiopia. Ethnobot Res Appl.6:487–502.

NMSA. 2013. Climate data for Debark, Ethiopia. National Meteorological Service Agency ofEthiopia, Bahir Dar Branch Office, Ethiopia.

Neba EN. 2009. Management of woody plants in indigenous land use systems of the Sahel: exampleof north Cameroon. Int NGO J. 4:480–490.

Negash M. 2007. Trees Management and Livelihoods in Gedeo’s Agroforests, Ethiopia. Forests,Trees Livelihoods. 17:157–168.

Puff C, Nemomissa S. 2005. Plants of the Simen. A flora of the Simen Mountains and surroundings,northern Ethiopia. Volume 37 of the Scripta Botanica Belgica. Meise (Belgium): NationalBotanic Garden of Belgium.

Shannon CE. 1949. The mathematical theory of communication. Urbana: University of IllinoisPress.

Tabuti JRS. 2012. Important woody plant species, their management and conservation status inBalawoli Sub-county, Uganda. Ethnobot Res Appl. 10:269–286.

Tadesse M. 2004. Asteraceae (Compositae). In: Hedberg I, Friis I, Edwards S, editors. Flora ofEthiopia and Eritrea: Vol. 4, Pt. 2. Addis Ababa & Uppsala: The National Herbarium at AddisAbaba University.

Teketay D. 2001. Deforestation, wood famine, and environmental degradation in Ethiopia’shighland ecosystems: urgent need for action. NorthEast Afr Stud. 8:53–76.

Tolera M, Asfaw Z, Lemenih M, Karltun E. 2008. Woody species diversity in a changing landscapein the south-central highlands of Ethiopia. Agric, Ecosyst Environ. 128:52–58.

Turnbull JW, Pryor LD. 1978. Choice of species and seed source. In: Hillis WE, Brown AG, editors.Eucalyptus for wood production. London: CSIRO.

Wassie A, Teketay D, Powell N. 2005. Church forests in North Gonder Administrative Zone,northern Ethiopia. Forests, Trees Livelihoods. 15:349–373.

Wøien H. 1995. Deforestation, information and citations: a comment on environmental degradationin highland Ethiopia. GeoJournal. 37:501–511.

Afromontane agricultural landscape 19

Dow

nloa

ded

by [

Add

is A

baba

Uni

vers

ity]

at 2

2:24

05

Aug

ust 2

014