Exploration of Woody Species Diversity and Population Structure

UNIVERSITY OF GONDAR COLLEGE OF AGRICULTURE AND RURAL TRANSFORMATION DEPARTMENT OF GENERAL FORESTRY

WOODY SPECIES DIVERSITY, STRUCTURE AND REGENERATION STATUS OF JUNIPERUS DOMINATED DRY AFROMONTANE FOREST IN BEYEDA DISTRICT, NORTH GONDAR ADMINSTRATIVE ZONE AMHARA NATIONAL REGIONAL STATE NORTH WEST HIGHLANDS OF ETHIOPIA

A Thesis Submitted to the Department of General Forestry, In Partial Fulfillment of the Requirements for the Degree of Master of Science in Forest Management and Utilization

By Muhabaw Taju Muhammed

June, 2018 Gondar, Ethiopia

A Thesis Submitted to the Department of General Forestry, In Partial Fulfillment of the Requirements for the Degree of Master of Science in Forest Management and Utilization

By Muhabaw Taju Muhammed

Advisors: Asmamaw Alemu (PhD) Endalkachew Teshome (PhD)

June, 2018 Gondar, Ethiopia UNIVERSITY OF GONDAR POSTGRADUATE DIRECTORATE

APROVAL SHEET

As thesis research advisor, we hereby certify that we have read and evaluated this thesis prepared under our guidance by Muhabaw Taju Muhammed entitled “Woody Species Diversity, Structure and Regeneration Status of Juniperus dominated Dry Afromontane Forest in Beyeda District, North Gondar Administrative Zone, Amhara National Regional State: North West Highlands of Ethiopia”. We recommend that this thesis document can be submitted as fulfilling the thesis requirements.

Submitted by:

Muhabaw Taju ______Name of student Signature Date Approved by: ______Major Advisor Signature Date ______Co-Advisor Signature Date As members of the Board of Examiners of the M.Sc. thesis open defense examination, we certify that we have read and evaluated the thesis prepared by Muhabaw Taju Muhammed and examined the candidate. We recommend that the thesis be accepted as fulfilling the thesis requirements for the degree of “Master of Science in Forest Management and Utilization”. ______Chairperson Signature Date ______Internal Examiner Signature Date ______External Examiner Signature Date Final approval and acceptance of the thesis is contingent upon the submission of final copy of the thesis to postgraduate office (PGO) through the departmental or school graduate committee (DGC or SGC) of the candidate

STATEMENT OF THE AUTHOR

By my signature below, I declare and affirm that this thesis with the title “Woody Species Diversity, Structure and Regeneration Status of Juniperus dominated Dry Afromontane Forest in Beyeda District, North Gondar Administrative Zone, Amhara National Regional State: North West Highlands of Ethiopia” is my own work. I have followed all ethical principles of scholarship in the preparation, data collection, data analysis and completion of this thesis. All scholarly matter that is included in the thesis has been given recognition through citation. I affirm that I have cited and referenced all sources used in this document. Every serious effort has been made to avoid any plagiarism in the preparation of this thesis.

This thesis is submitted in partial fulfillment of the requirement for MSc. Degree in Forest Management and Utilization from the Postgraduate Directorate at University of Gondar. The thesis is deposited in the University of Gondar Library and is made available to borrowers under the rules of the library. I solemnly declare that this thesis has not been submitted to any other institution anywhere for the award of any academic degree, diploma or certificate.

Brief quotations from this thesis may be used without special permission provided that accurate and complete acknowledgement of the source is made. Requests for permission for extended quotations from, or reproduction of, this thesis in whole or in chapter may be granted by the Head of the School or Department or the Dean of the Postgraduate Directorate when in his or her judgment the proposed use of the material is in the interest of scholarship. In all other instances, however, permission must be obtained from the author of the thesis.

Name: Muhabaw Taju Signature: ------Date: ------

College of Agriculture and Rural transformation/ Department of General Forestry

BIOGRAPHICAL SKETCH

Muhabaw Taju was born on June 17, 1984 in North-western Amhara Region, in Tikel Dengay town of Lay Armachiho District. He attended his Elementary School at Limat- ber Primary School and his Secondary School at Limat-ber and Fasiledes Comprehensive High Schools. In 2000 he joined Wondo Genet College of Forestry and by 2002 he received the award for Diploma in General Forestry. After graduation, Muhabaw was immediately employed by Adiarkay district Office of Agriculture as an expert. After serving few years, in 2006 he has got scholarship with the sponsorship of Austrian Development Co-operation programme and joined Bahir Dar University and by 2010 he received the award for Bachelor of science in Natural Resource Management. He served for the last 14 years in different professional as well as leadership positions in the field of Forestry, Natural resource management, and General agriculture in different District of North Gondar. Now he is working in Central Gondar Zone office of Agriculture. Being serving in this zone, in 2016/17 he has got a scholarship with the sponsorship of Central Gondar Zone office of Agriculture in collaboration with the University of Gondar and Joined the University of Gondar, College of Agriculture and Rural Transformation to pursue his MSc. in Forest Management and Utilization.

iii

ACKNOWLEDGMENTS

First and foremost, I would like to thank and praise the Almighty “Allah” helping me to accomplish my work. I feel great pleasure in expressing my deep sense of gratitude to my Advisors Dr. Asmamaw Alemu and Dr. Endalkachew Teshome for their unreserved professional guidance, supervision and assistance starting from proposal writing to the end of my thesis work. Both of them were advising me not only being an academic advisor, but also in a sense of brotherhood and I am very much grateful for that.

I am very grateful to University of Gondar and Central Gondar Zone Office of Agriculture for providing me MSc. scholarship to study in the University of Gondar. Thanks to the Amhara Region Bureau of Finance and Economic Development in providing me GIS- based maps. I am also grateful to my dearest friend Melkamu Kasseye for his un reserved technical as well as professional support. I would like to express my special thanks to my friends Abebe Mulugeta, Muluken Tefera, Ayanaw Tadesse, and Amsalu Abich for their technical support and friendly advice.

My heartfelt thanks go to Beyeda District Administrative Office and the District Office of Agriculture staff members; Especially Kenaw Assefa, Woretaw Demiss, Molla Berihun, Tigabe Alemayehu, Atallo Admasu, Zewditu Mezgebu, Tigabe Alemayehu, Belachew Gedamu, and Negash Adissu for their so much kind and genuine support during reconnaissance survey and data collection. I also owe my grateful thanks to Central Gondar Zone office of Agriculture staff members in supporting me with Courage and initiation. I would like to say thank you for all those unmentioned instructors and classmates from my schooling life till now who directly or indirectly helped me to reach up to this level.

Last but not least my special thanks go to my beloved wife Teyba Abdulkadir for her restless support and courage throughout my schooling life. She was backing me by taking the whole responsibility in caring our children with love and affection. Without her support it was very much difficult to complete this educational journey.

ACRONYMS AND ABBREVIATIONS

AA Afro-Alpine belt

ACB Acacia-Comiphora woodland and bushland

ANOVA Analysis of Variance

ANRS Amhara National Regional State

BA Basal Area

BDoA Beyeda District Office of Agriculture

BoFED Bureau of Finance and Economic Development

CBD Convention on Biological Diversity

CTW Combretum-Terminalia woodland and wooded grassland

DAF Dry evergreen Afromontane forest and grassland complex

DBH Diameter at Breast Height

DSS Desert and semi-desert scrubland

EB Ericaceous belt

EBI Ethiopian Biodiversity Institute

EWCA Ethiopian Wild Life Conservation Authority

FLV Fresh water lakes, lakeshores, marshes, flood plains and vegetation

GPS Geographical Positioning System

HDF Highly Disturbed Forest

IBC Institute of Biodiversity Conservation

ITCZ Intertropical Convergence Zone

IUCN International Union for Conservation of Nature

IVI Importance Value Index

LDF Less Disturbed Forest m.a.s.l: meter above sea level

MAF Moist evergreen Afromontane forest

MANOVA Multi Variate Analysis of Variance

MDF Moderately Disturbed Forest

RD Relative Density

RDO Relative Dominance

RF Relative Frequency

RV Riverine vegetation

SLV Salt-water lakes, lakeshores, salt marshes and pan vegetation

SMNP Semen Mountains National Park

SNNPR Southern Nations Nationalities and Peoples Region

TRF Transitional rainforest

WGG Wooded grassland of the western Gambela region

TABLE OF CONTENTS APROVAL SHEET I STATEMENT OF THE AUTHOR II BIOGRAPHICAL SKETCH III ACKNOWLEDGMENTS IV ACRONYMS AND ABBREVIATIONS V LIST OF TABLES X LIST OF FIGURES XI LISTS OF APENDICES XII ABSTRACT XIII 1. INTRODUCTION 1

1.1. Background of the study 1

1.2. Statement of the Problem 4

1.3. Significance of the Study 5

1.4. Objectives 6 1.4.1. General objective 6 1.4.2. The Specific Objectives 6

1.5. Research questions 6 2. LITERATUREREVIEW 7

2.1. Concept of Biodiversity 7

2.2. Status of Biodiversity in Ethiopia 8

2.3. Vegetation Types in Ethiopia 8 2.3.1. Dry evergreen afromontane Forest of Ethiopia_(DAF) 9 2.3.2. The Juniperus dominated forests of Ethiopia 10

2.4. Population Structure of the Forest Species 11

2.5. Regeneration Status of the Dry Evergreen Montane Forests of Ethiopia 12 2.5.1. Regeneration Status of Juniperus forest 12 2.5.2. Factors affecting regeneration of dry evergreen montane forests 13

2.6. Threats facing Forest Biodiversity in Ethiopia 14

2.7. Disturbances in a Forest Ecosystem 15

2.8. Human Activities Causing Biodiversity Loss in Forest Ecosystems 16

vii

2.9. Species Richness and Diversity along Disturbance Gradient 16

2.10. Measurement of Diversity Using Different Indices 17 2.10.1. Species diversity and richness 18 2.10.2. Measurement of similarity and dissimilarity 19 2.10.3. Measurement of Species density and frequency 19 2.10.4. Measurement of importance value index (IVI) 20

2.11. Conceptual framework of the study 20 3. MATERIALS AND METHODS 22

3.1. Description of the study area 22 3.1.1. Geographic Location 22 3.1.3. Topography, geology and climate 23 3.1.4. Vegetation 24 3.1.5. Land use and Land cover 26

3.2. Vegetation Data Collection and Analysis 26 3.2.1. Reconnaissance Survey 26 3.2.2. Sampling strategy 26 3.2.3. Woody Species Vegetation Data Collection and Identification 29

3.3. Data Processing and Analysis 30 3.3.1. Vegetation data analysis 30 4. RESULT AND DISCUSSION 36

4.1. Woody Species composition, richness, diversity, and similarity along the

gradients of disturbance 36 4.1.1. Composition of woody species 36 4.1.2. Woody Species Richness 39 4.1.3. Woody Species diversity 40 4.1.4. Woody species similarity 41

4.2. Density, frequency, dominance and importance value index of woody species 42 4.2.1. Density of species along disturbance gradient 42 4.2.2. Frequency of woody species along disturbance 45 4.2.3. Dominance of woody species along disturbance gradients 45 4.2.4. Importance Value Index (IVI) of woody species in each study sites 48

viii

4.3. Population Structure of the Common Woody Species along Disturbance 49 4.3.1. Diameter and height class distribution of the common woody species 49

4.4. Regeneration Status of Woody Species along Gradients of Disturbance 55

4.5. Investigates of the effect of anthropogenic disturbance 57 4.6.1. Effect of elevation on species parameters 59 4.6.2. Effect of Slope on species parameters 61 4.6.3. The effect of Aspect on plant parameters 61 5. CONCLUSION AND RECOMMENDATION 64

5.1. Conclusion 64

5.2. Recommendations 66 6. REFERENCES 67 7. APPENDICES 74 APPROVAL SHEET 82

LIST OF TABLES

Table 1: Description of the study sites along disturbance levels 27 Table 2:Research Methodology and data analysis Summary 35 Table 3: Lists of woody species identified 36 Table 4: Species diversity, richness and evenness along disturbance gradients 38 Table 5: Similarity of woody species between the three sites 42 Table 6: Density of life stages along disturbance 43 Table 7: Density of life stages of all species along the gradients of disturbance 44 Table 8: Relative Density, frequency, dominance and importance value of woody species across the three sites 47 Table 9: Mean table to show the effect of elevation on woody species diversity, evenness, richness and density 60 Table 10: Mean table to show the effect of slope on plant parameters 62 Table 11:The effect of aspect on species parameters 63

LIST OF FIGURES

Figure 1: Conceptual framework (own sketch) 21 Figure 2: Map of the study area (Extracted from BoFED, 2016_Gis data) 22 Figure 3: Map showing the study forest and sampling plots along disturbance 23 Figure 4: Soil Map of the District (Adapted from BoFED, 2016) 24 Figure 5: Juniperus afromontane forest of Beyeda District (Photo By MT) 25 Figure 6: Picture shows study sites along disturbance levels (Photo MT) 27 Figure 7: Simple sketch of data collection quadrates 28 Figure 8:Species accumulation curve 29 Figure 9: Composition of woody species along the gradients of disturbance 37 Figure 10: Woody species richness along disturbance gradients 40 Figure 11: Density of the forest along the gradients of disturbance 43 Figure 12: Basal area along disturbance levels 46 Figure 13: The proportion of the dominant woody species along disturbance gradient 49 Figure 14: Horizontal structure of the forests 50 Figure 15: population structure of representative species in LDF site 51 Figure 16: population structure of representative species in MDF site 52 Figure 17: population structure of representative species in HDF site 53 Figure 18: Vertical structure of the forests 54 Figure 19: The proportion of seedlings, saplings, and adult trees along disturbance 57 Figure 20: Density of dead stumps/ha in along disturbance 58 Figure 21: Density of coppiced stumps/ha along disturbance 58

LISTS OF APENDICES

Appendix 1: Photo shows data collection team during field work 74 Appendix 2: GPS points of the study forest and sample plots 74 Appendix 3: Data collection sheet for diversity 76 Appendix 4: Regeneration data collection sheet 77 Appendix 5: ANOVA table for species parameter difference in life form 78 Appendix 6: ANOVA Table for species parameter difference across Disturbance 78 Appendix 7: MANOVA Table for difference in species parameters across elevation 79 Appendix 8:MANOVA Table for difference in species parameters across Slope 80 Appendix 9: MANOVA Table for difference in species parameters across Aspect 81

xii

ABSTRACT

Owing to the diversity in climate, topography, and vegetation; Ethiopia is recognized as a major center of Biodiversity. The vegetation types of the country are highly diverse, varying from Afro-alpine to Desert vegetation types. Even if the country is rich in Biodiversity, its forest resources are being destroyed alarmingly by anthropogenic and natural factors. This study was conducted in the Juniperus dominated Afromontane forest in Beyeda district, Amhara Region, North west highlands of Ethiopia. The study aims to determine woody species diversity, population structure and regeneration status along the gradients of disturbances. For vegetation survey the forest was stratified in to three disturbance levels (Less disturbed, Moderately disturbed, and highly disturbed forests). The vegetation assessment followed a Systematic random sampling in which the first sampling plot was located randomly, and the subsequent plots laid systematically at regular interval. Data were collected from 41 square (400m2) main sample plots laid on transects at regular interval of 200m between transects and 100m between plots. Species diversity and evenness were computed using Shannon diversity and Evenness indices. The similarities along disturbances were computed using Sorenson’s similarity index. Further ANOVA and MANOVA were used to test differences among disturbance levels. A total of 24 woody species belonging to 20 families and 24 genera were identified. Species Richness, Evenness, density and basal area of woody species decreased as intensity of disturbance increased. There was no statistical difference in overall Shannon diversity index along disturbance levels. woody species density and Richness along disturbance were differ significantly at 95% confidence intervals. Population structure showed trends of inverted ‘J’ shape pattern along the gradients of disturbance. As a result of the above mentioned facts disturbance has a negative effect on woody species parameters; but mild disturbance has a positive effect in facilitating regeneration. In addition to these environmental variables its own has effect to change species parameters So, attention should be given in formulating forest management plan and strategy to limit the impact of anthropogenic disturbance in the forest so as to sustained the study forest. Key words/Phrases: Juniperus dominated forest, Afromontane forest, woody species diversity, population structure, Less disturbed forest, Moderately disturbed forest, Highly disturbed forest

xiii

1. INTRODUCTION

1.1. Background of the study

Ethiopia is endowed with diverse flora and fauna in tropical Africa due to its great geographical diversity and the resulting vegetation types, soil types and diverse climatic conditions led to the emergence of habitats that are suitable for the evolution and survival of various plant and animal species (Teketay, 2001; Gurmessa et al., 2012). The vegetation of the country is very heterogeneous and has a rich endemic element (Tilahun, 2015). The topographic features that range from 110m below sea level to 4,543 meters above sea level has created quiet diverse ecological conditions in the country (Zegeye et al., 2011). This wide range of ecological variation coupled with the corresponding diverse climate and topography has made the country to be one of the very important diversity rich areas in the world (Worku, 2006).

Vegetation types in Ethiopia are highly diverse, varying from Afro-alpine to desert Vegetation types (Zegeye et al, 2011). Various authors have classified the vegetation of Ethiopia into various types and subtypes; for example based on (EBI, 2009) the vegetation of Ethiopia was classified into ten vegetation types. Friis and Demissew (2001) classifies the Ethiopian vegetation in to nine vegetation types including the Coastal vegetation of Eritrea. More recently, the potential vegetation of Ethiopia is systematically revised to have 12 vegetation types and 12 subtypes (Friis et al., 2010, Wassie, 2017). The recent classification was mainly based on environmental parameters and GIS- methodology. These vegetation types and sub types are (1) Desert and semi-desert scrubland (DSS). (2) Acacia-Commiphora woodland and bushland (ACB) (with the subtypes (2a) Acacia-Commiphora woodland and bushland proper and (2b) Acacia wooded grassland of the Rift Valley). (3) Wooded grassland of the western Gambela region (WGG). (4) Combretum-Terminalia woodland and wooded grassland (CTW). (5) Dry evergreen Afromontane forest and grassland complex (DAF) (with the subtypes (5a) Undifferentiated Afromontane forest, (5b) Dry single-dominant Afromontane forest of the Ethiopian highlands, (5c) Afromontane woodland, wooded grassland and grassland, (5d) Transition between Afromontane vegetation and Acacia-Commiphora bushland on

1 the Eastern escarpment). (6) Moist evergreen Afromontane forest (MAF) (with the subtypes (6a) primary or mature secondary moist evergreen Afromontane forest, and (6b) Edges of moist evergreen Afromontane forest, bushland, woodland and wooded grassland. (7) Transitional rainforest (TRF). (8) Ericaceous belt (EB). (9) Afro alpine belt (AA). (10) Riverine vegetation (RV). (11) Fresh water lakes vegetation (FLV), (with the subtypes (11a) Fresh-water lake vegetation (open water) and (11b) Freshwater marshes and swamps, floodplains and lake shore vegetation). (12) Salt lakes vegetation (SLV) (with the subtypes (12a) Salt lake vegetation (open-water) and (12b) Saltpans, saline/brackish and intermittent wetlands and salt-lake shore vegetation) (Friis et al., 2010).

The Ethiopian highlands comprise more than 50 % of the land area with Afromontane vegetation, from which the dry evergreen Afromontane forests and grass land complexes forming large coverage in different parts of the country (Berhanu et al., 2017). Many studies suggested that the Ethiopian highlands were once covered by forests, with co- dominant Juniperus and Olea species (Bishaw, 2001; Sterck et al., 2010). The current reminant fragmented forests in the Ethiopian highlands are several centuries old, with small and ancient forest patches remaining around churches and in inaccessible valleys (Wassie and Teketay, 2006).

The Dry evergreen montane forests of Ethiopia are distributed in central (East, West and North Shewa, Arsi and Gurage zones), Northern (East and West Gojam, North and South Gondar, South and North Wello, Agew Awi, and South, East and West Tigray Zones) Eastern (East and West Hararghe, Afar and Dire Dawa Zones) and Southern (Bale, Borena, and South and North Omo Zones) parts of Ethiopia ( Sterck et al., 2010; Etefa, 2011). Typical dry evergreen montane forests in Ethiopia are situated on highlands and mountains occurring at altitudinal ranges of 1,500 to 3,200 m. a. s. l (Woldu and Saleem, 2000). This vegetation is characterized by Olea europaea sub species cuspidata, Juniperus procera, Prunus africana, Celtis africana, Euphorbia ampliphylla, Carissa spinarum, Euclea divinorum, Rosa abyssinica, Pittosporum viridiflorum and Ekebergia capensis (Sterck et al., 2010). There are mixed provenances of Juniperus procera, some of which can get very big while others remain small (Bekele, 1993).

Even if the country is rich in biodiversity, its biological resources are being destroyed at an alarming rate for settlement, agriculture and fuelwood as the majority of the Ethiopia’s population is settled in the afromontane areas (Friis, 1986). Recently, construction projects, resettlements, forest fire and urban expansion are damaging the remaining forest patches in northwestern, western and southwestern Ethiopia (Wassie, 2017). Forest resources degradation is intense in the northern highlands of Ethiopia compared to other parts of the country owing to large human and livestock population and long history of settlement (Amare et al., 2016). Studies indicated that few remnant high forests are found in southwestern and western parts and as patches in conservation sites, churches and sacred areas in the country (Senbeta, 2006; Alelign et al., 2007; Wassie et al., 2007 ; Zegeye et al., 2011). Afromontane forests belong to the most fragmented forest ecosystems which should be accorded high conservation priority (Zegeye, 2006).

The Juniperus dominated Natural forest of Beyeda District is one of the few remnant patches of single species dominated dry evergreen Afromontane forest in the Northern highlands of Ethiopia adjacent to the North eastern part of Semien Mountains National Park. The Juniperus forest is a very important woody species to produces timber of high economic value. The wood is used for manufacturing of lead pencil, construction and lining of buildings, as well as for a variety of outdoor works owing to its fine texture, straight grain, medium hardness, resistance to termite attack and workability (Sterck et al., 2010). Despite these values, the forest is getting deteriorated by several anthropogenic and natural factors that lead the forest to endangered status in some provenance (Sertse et al., 2011).

Although the study forest is covered by such forest, we lack documented scientific information uncovering the resource status and possible potential ecological and productive benefits from its sustainable management. Due to lack of such information appropriate management and utilization of the forest is not well designed and because of this the coverage of the forest is getting decreased from time to time by several factors as the forest lacks attention. The existence of such reminant forest and its present situation is envisaged as important case for scientific investigation in linking the species diversity and population status of major single species dominated dry Afromontane and sub afro-

3 alpine species with disturbance levels. Hence, this thesis tried to show a result of study conducted on species diversity and population structure along the gradient of disturbance at Juniperus dominated forest in Beyeda district, North west highlands of Ethiopia.

1.2. Statement of the Problem

The Juniperus dominated forest which is part of the dry evergreen afromontane forest of Ethiopia once covered large area for the past 100 years (Pohjonen and Pukkala, 1992). But, these forests being surrounded by agricultural lands and settlement areas; the size and quality of remnant forest patches have been largely degraded at an alarming rate due to various factors such as encroachment, disturbance and selective logging, leading to the gradual loss of key species like Juniperus procera (Teketay, 2005; Sertse et al., 2011; Wassie et al., 2017).

Adequate understanding of the structure and composition of a given vegetation formation and the ecological processes such as the response of the constituting species to the different levels of disturbance is essential for designing appropriate management practices. The key research question is thus ‘how does the species composition, diversity, structure and regeneration status vary along the gradients of disturbance. Such scientific investigations can contribute to uncover the ecological processes in such remaining vegetation that were preserved due to inaccessibility. These are important galleries and possible potential germplasm sources for ecological studies, conservation initiatives and plantation development of threatened single species dominated dry Afromontane and sub Afro alpine species.

Various studies have been conducted in relation to forest diversity, population structure and regeneration status of dry ever green Afromontane forests at different parts of the country like Denkoro forest (Ayalew et al., 2006); Ericaceous forest of Semien Mountains (Teshome, 2007); Bale Mountains National Park Forest (Yineger et al., 2008); Menagesha suba state forest (Benti, 2011); Menagesha Amba Mariam forest (Tilahun, 2015); Hugumburda Forest (Aynekulu et al., 2016). But, limited scientific knowledge are available for the responses of single dominated dry evergreen Afromontane and sub afro- alpine vegetation species to the gradients of disturbance like (Mamo et al., 2015).

Moreover, reminant gallery vegetation such as the Juniperus dominated Afromontane forest in Beyeda are not yet investigated and even their existence is not recognized by the scientific community. Hence due to lack of sufficient baseline information about the existence and distribution of such type of forest in other areas like Beyeda district; most of the researchers concentrated on few National forests priority areas like the above- mentioned areas.

The Juniperus dominated afromontane natural forest of Beyeda district is situated adjacent to the North Eastern part of Semien Mountains National park. This forest, which once covered a very large area is currently getting deteriorated because of several factors mainly by anthropogenic factors and becomes restricted around the mountain escarpment of southern and central part of the District. Appropriate scientific information is required to design proper managerial conservation measures before the forest is completely destroyed. Therefore, this study is initiated to investigate the woody species diversity, population structure and regeneration status along the gradients of disturbance to generate scientific information so as to design proper measures for conservation and sustainable utilization of the forest resources.

1.3. Significance of the Study

Research on the effect of disturbance on species composition, diversity and regeneration of the forests has been lacking. But, currently, there is a need for information that would contribute to conservation and sustainable utilization of forests as disturbance may be one of the management tool to favor or inhibit the regeneration of target species. To have such kind of information, documentation and identification of woody species diversity, structural analysis and description of the forest is important. This is particularly vital for prioritizing threatened woody species conservation. So, the results of this research can provide crucial information on the status of the Juniperus dominated natural forests for local research and development institutes in particular and serve as baseline information for research and policy makers for further development, sustainable management and utilization of the reminant forest resource. Moreover, the communication of the results from this investigation is expected to attract the attention of research community and policy makers for sustainable utilization of this gallery vegetation. Hence, the forest may

5 gain special attention and it may be used as an alternative provenance for the conservation as well as restoration of the most valuable woody species in the area.

1.4. Objectives

1.4.1.General objective

The general objective of the study was to determine woody species diversity, structure, and regeneration status of Juniperus dominated afromontane forest of Beyeda district along disturbance level; so as to contribute ecological knowledge for the formulation of appropriate conservation measures, sustainable forest management and sustainable utilization plan of the respective reminant forest resource.

1.4.2.The Specific Objectives

The Specific objectives of the study were to:

 Assess woody species composition along the gradients of disturbance  Determine woody species diversity, Evenness and Richness along the gradients of disturbance  Determine the vertical and horizontal structure of the forest along gradients of disturbance  Determine the regeneration status of the forest along gradients of disturbance  Examine the effect of anthropogenic disturbance along disturbance gradient  Examine the effect of environmental variables (elevation, slope, and aspect) on species parameters (Species diversity, Richness, Evenness and Density)

1.5. Research questions

Through detailed ecological investigation of the Juniperus dominated reminant Dry Afromontane forest in Beyeda district, this study addressed a key research question ‘how does the single dominated Afro montane vegetation species composition, diversity, density, population structure, and regeneration status varies along the gradients of disturbance?

2. LITERATUREREVIEW

2.1. Concept of Biodiversity

Biodiversity is a very complex and multifaceted term and many scientists in various ways have defined the term and each definition reveals a new dimension and emphasizes a different aspect to suit the work at hand (Afua, 2011). Biological diversity is the variability among living organisms from all sources including, inter alia, terrestrial, marine and other aquatic ecosystems and the ecological complexes of which they are a part; this includes diversity within species, between species and of ecosystems (Magurran, 2004 and EBI, 2005). Species diversity and genetic diversity are the most important concept of Biodiversity (Chandrakar, 2012). The former refers to the number of species found within a given area. Whereas, genetic diversity refers to the variety of genes within a particular species, variety or breed. Plant biodiversity refers to the totality and variability of all plants and their ecosystems (Melese and Ayele, 2017).

Biodiversity plays a key role in ecosystem functioning and has been widely used as an indicator of ecosystem health (Betemariyam, 2011). According to EBI (2005) biodiversity provides four main services: (1) supporting (nutrient cycling, soil formation, primary production, etc.), (2) provisioning (food, fresh water, wood and fiber, fuel, etc.), (3) regulating (climate regulation, flood regulation, disease regulation, water purification, etc.), and (4) cultural (aesthetic, spiritual, educational, recreational, etc.) services.

Ecologists investigating terrestrial systems often focus on species diversity of plant communities since green plants usually account for a large proportion of the biomass in a given system. In forests, biological diversity allows species to evolve and adapt the changing environmental conditions and to support their ecosystem functions (Alelign et al., 2007). Plant biodiversity is one of the major groups of biological diversity. It can be affected by different biotic and abiotic factors. Globally, patterns of plant species diversity are influenced by altitudinal and soil gradients apart from other factors like, topography and aspect (Kebede et al., 2013).

2.2. Status of Biodiversity in Ethiopia

Ethiopia is internationally recognized as a major center for biodiversity (Zegeye et al., 2006).The major physiographic features of the country are massive highland mountains and plateau divided by the Great Rift Valley and surrounded by lowlands along the periphery (Bekele, 1993). There is great altitudinal variation from 110 meters below sea level to 4543 meters above sea level (Teketay, 2001). Although Ethiopia is a tropical country with typically hot and dry lowland areas, it has varied macro and micro climatic conditions. The rainfall is variable in different parts, highest (2200 mm) in highlands (>1500 meters) and lowest, (250 mm) in low lands (<1500 meters). All these factors have contributed to richness of Ethiopian biodiversity with great level of endemism and genetic diversity (Azamal et al., 2012).

Ethiopia is one of the world's rich biodiversity countries and it deserves attention regionally and globally (Gebretsadik, 2016). It is estimated to contain about 7025 species of higher plants, >10% are endemic (Tilahun, 2015). Endemism is particularly high in the high mountains and in the dry forests of Ogaden and Borena lowlands (Etefa, 2011). There is also great diversity of fauna in Ethiopia, owing to the diversity in climate, vegetation, and terrain (Zegeye et al., 2006). It is estimated that there are 281 species of mammals, 861 species of birds of which 29 species of mammals and 15 species of birds are endemic. There are about 201 species of reptiles of which 87snakes, 101 lizards, one species of crocodile, and 13 species of tortoises and turtles and 9 of them are endemic and a total of 63 species of amphibians have also been recorded in Ethiopia of these23 species are listed as endemic (Gebretsadik, 2016). The country also own high genetic diversity of four of the world’s widely grown food crops like wheat, barley, sorghum, peas, and industrial crops like linseed, castor bean and cotton and also cash crops like coffee are found in Ethiopia (Etefa, 2011).

2.3. Vegetation Types in Ethiopia

The vegetation of Ethiopia is one of the richest in floristic diversity in Africa and/or in the world. Most recently, the country is recognized as one of the top 20 Megadiverse countries in the world (Wassie, 2017). Ethiopia is also known to be one of the 12

Vavilovian centers of origin and diversity for many cultivated crops and higher plants in the world (EBI, 2009). Ethiopia (including Eritrea) has been divided into 16 Floristic Regions, namely Gojam, Gondar, Bale, upland Shewa, Sidamo, Afar, upland Wello, Tigray, Welega, Illubaboor, Keffa, Gamo-Gofa, Arsi, Hararghe, East and West Eritrea (Friis et al., 2010).

According to EBI (2005) there are 92 high forests in Ethiopia out of which 56 are dry evergreen montane forests, 29 moist montane forest, 5 transitional dry moist evergreen montane forests and 2 lowland semi-evergreen forests. Furthermore, the afromontane vegetation of Ethiopia are one of the richest in species composition and endemism (EBI, 2009). The recent vegetation classification of Ethiopia categorized in to 12 type and 12 sub types based on environmental variables and GIS-technologies (Friis, 2010). Based on such classification, forest patches of the study area have been classified as dry afromontane and sub classification of single dominated dry evergreen Afromontane forest.

2.3.1.Dry evergreen afromontane Forest of Ethiopia_(DAF)

The dry evergreen afro montane forests are among the least known and unique vegetation types confined to a few localities in the world (Bekele, 2000). The Ethiopian highlands comprise more than 50 % of the land area with Afromontane vegetation, from these the dry evergreen Afromontane forests forming large coverage in different parts of the country (Berhanu et al., 2017). The current fragmented forest in the Ethiopian highlands is several centuries old, with small and ancient forest patches remaining around churches and in inaccessible valleys (Wassie and Teketay, 2006).

The dry evergreen montane forests in Ethiopia are multistoried forest vegetation situated on highlands and mountains occurring at altitudinal ranges of 1,500 to 3,200 m. a. s. l (Woldu et al., 1999). The first stratum also known as the top storey/canopy layer consists of a non-uniform, non-compact layer of tall trees. These trees are known as "emergents" because they project above the vegetation mass. Characteristic plant species at the canopy layer include Juniperus procera, Podocarpus falcatus, Apodytes dimidiata, Prunus africana and Acacia abyssinica (Bekele, 1993). The sub-canopy layer, which is below

9 the canopy layer, is a mass of shorter trees of various heights such as Allophylus abyssinicus, Euphorbia ampliphylla, Myrsine melanophloes and Olinia rochetiana. The third storey usually known as shrub layer consists of species such as Myrsine africana, Calpurnia aurea and Dovialis abyssinica. The ground layer is covered with grasses, ferns and mosses(Woldu et al., 1999). But, the forests in this ecosystem have greatly diminished due to expansion of agriculture and other interference by people and domestic animals and have been replaced by bush land and scrub in most areas (Etefa, 2011).

2.3.2.The Juniperus dominated forests of Ethiopia

The Juniperus dominated forest is classified under DAF and sub-classification of the single-dominant dry Afromontane ever green forest of the Ethiopian highlands (Wassie et al., 2017). In the classification of Ethiopian highland forest Juniperus procera grows in the zones of ‘Woina Dega’ (1800-2500 m) and ‘Dega’ (2500-3200 m) above sea level. The optimum altitude is from 2200-2500m where Juniperus procera forms either pure stands or is the dominant species with Hagenia abyssinica and Olea eropeana. Between 1800m and 2200m above sea level Juniperus procera and Podocarpus gracilior often grow mixed (Pohjonen and Pukkala, 1992). There are only few notable natural, older Juniperus procera forest areas in the country: one in Menagesha National Park (central highlands), one in Gara Ades (Eastern highlands), and another in the Bale Mountains (southern highlands) and small patches in many mountainous parts of northern Amhara, Tigray, and Oromia regional states (Sterck et al., 2010). The largest remaining occurrence is in the Menagesha State forest, 20 km west of Addis Ababa. Originally the whole range of Wochecha Mountain surrounding Menagesha was well stocked with Juniperus procera. But, now a day’s these Juniperus dominated forests are cleared for arable crops and grazing, or was harvested for fuel wood (Pohjonen and Pukkala, 1992). Juniper forest populations are extremely small and fragmented in its natural habitat due to anthropogenic disturbance (mainly logging). Thus, it is considered as endangered species that should be given the highest priority for conservation (FAO, 1997 cited in Teketay et al., 2006)

2.4. Population Structure of the Forest Species

Population structure is the numerical distribution of individuals of differing size or age classes within a population of a given species at a given moment of time as populations are dynamic. It can also be defined as the distribution of individuals of each species within a population in arbitrarily defined diameter height classes (Afua, 2011). Population structure data have long been used by foresters and ecologists to investigate the regeneration profile of a species of a given population in the study quadrats (Peters, 1996). In broad sense, computing the population structure of a given population can tell us two sets of information. First, the distribution of plants among age, or size class categories which again tell us something about the history and the future of the population and the second information provided by age/size distribution is the immediate means of identifying the poorly represented stages of life history of a given population (Worku, 2006).

Tree stem diameter distribution often used to infer past disturbances, regeneration patterns and successional trends in tree populations (Gebrehiwot, 2003). Therefore, the analysis of population structure is a basic tool for orienting management activities, for identifying species susceptibility to local extinction, to reveal compositional changes of the vegetation, and for assessing the impact of resource exploitation. Based on the data from constructed population structure, it has been depicted that, tree species with population structure skewed to an inverted ‘J’ (Silvertown, 1982), or an ‘L-shaped’ distribution i.e. with many small individuals and few larger ones, are considered to have a favorable status for regeneration and, hence, stable populations. If the population structure constructed from both diameter and height class frequency distributions has very few individuals at lowest classes, then that particular species has serious problems in regeneration (Worku, 2006).

Human-induced disturbances are the major causes for changes in forest structure and composition and the extent of these effects are dependent on the type and severities of the disturbances (Mamo et al., 2015). As described in the works of Afua (2011) human disturbance causes disruption of forest structure and changes composition of species which ultimately leads to disruption of tree population structure. The effects on forest

11 structure and composition caused by heavy forest disturbance (such as heavy mechanical logging) results in the destruction of both mature trees and young stock and the alteration of certain environmental conditions which hinder forest regeneration.

2.5. Regeneration Status of the Dry Evergreen Montane Forests of Ethiopia

Natural regeneration is a site-specific ecological process, and it is usually difficult to characterize the factors that control the regeneration processes (Aynekulu et al., 2016). Regeneration is a central component of tropical forest ecosystem dynamics and restoration of degraded forest lands (Tesfaye et al., 2009). Tropical forest plants regenerate from one or more pathways, i.e. 1) seed rain: recently dispersed seeds; 2) soil seed bank: dormant seeds in the soil; 3) seedling bank or advance regeneration: established, suppressed seedlings in the understory; and 4) coppice: root or shoot sprouts of damaged individuals (Teketay, 2005). The formation of a seedling-sapling bank under the forest canopy is the major regeneration route(Teketay et al., 2006). Important factors for regeneration are micro site, canopy openness; treatment, and species (Wassie et al., 2009). The vascular plants of the Afromontane forests of Ethiopia maintain their population through natural regeneration (Aynekulu et al.,2009). However, large-scale degradation of natural forest in Ethiopia has created a major challenge with respect to the regeneration of key native tree species (Wassie et al., 2009).

The study of regeneration of forest trees has important implications for the management of natural forests and is one of the thrust areas of forestry. Understanding of natural regeneration processes and the distribution of recruits is of paramount importance to examine the build-up of future forest structure and composition (Vishal, 2011). Forest resources are renewable only because they can regenerate. The regeneration dynamics or pace, at which older trees of a forest are replaced by younger ones in time and space, is determined by the regeneration potential of the component tree species and the factors that influence the actual regeneration (Neelo et al., 2013).

2.5.1.Regeneration Status of Juniperus forest

Regeneration of Juniperus procera is reportedly very poor. The possible causes related to biological stress resulted from an insect damages of the cones, climatic change as a

12 possible effect of the poor regeneration of Juniperus (El-juhany et al., 2009). Juniperus is reported to be a strong light demander and does not regenerate under natural conditions of closed canopy; its seeds do not germinate on areas of thick humus while its seedlings are unable to penetrate a deep layer of litter and humus. The quantity and spectral distribution of light reaching the ground is critical for the regeneration of Juniperus procera (Bekele, 1993). Hence, its regeneration seems to be dependent on the disturbance of both the canopy and humus layer for a full exposure of light and exposed mineral soil (Sharew et al., 1997). In open patches, seeds germinate better as can be found more frequently in semi-natural forests. Controlled local thinning may be considered to create some open patches to encourage natural regeneration. In the secondary forests, nursery beds and enrichment planting may be considered in such open patches to promote regeneration and to avoid a future age-class gap in the tree population (Sterck et al., 2010).

2.5.2.Factors affecting regeneration of dry evergreen montane forests

Dry Afromontane forests are among the most exploited forest ecosystems and have been increasingly fragmented. Several forests and forest tree species in the country have shown decline in their population structure and regeneration due to past and present disturbances, conversion of land uses and ongoing destructive forest management practices (Tesfaye et al., 2009). As investigated by Wassie et al. (2010) factors that potentially influence regeneration at the early stage are those that determine the probability of seedling establishment and those that affect seedling survival and growth. Site conditions, like nutrient and water availability, overstorey conditions, e.g. tree species, and canopy density, are important determinants of regeneration (Wagner et al., 2010). In another study coined by Teshome (2007) showed that grazing, microsite condition like; slope, exposure to sun light, night frost and soil fertility status are important factors for effective establishment of seedlings. Canopy openness has also a great effect on the regeneration and survival of seedlings inside a certain natural forest especially for those light demanding species (Wassie et al., 2009). Temperature also play an important role for enhancing regeneration in a disturbed forest ecosystem. It also helps seeds to detect the

13 formation of gaps in the canopy of vegetation since vegetation canopy acts as effective insulator of temperature changes (Teketay, 2005).

2.6. Threats facing Forest Biodiversity in Ethiopia

Ethiopia is one of the most diverse and important sources of biodiversity in the world for wild, cultivated, or domestic organisms (Azamal et al., 2012). However, the trees and forests of Ethiopia are under tremendous pressure because of the drastic decline in mature forest cover and the continual pressures of population increase, rudimentary farming techniques, land use competition, land tenure, and forest degradation and conversion, and the status of the forest resources should be considered at risk (Etefa, 2011).

According to Betemariyam (2011) and Gebretsadik (2016) the major causes of biodiversity decline are natural land use changes, pollution, changes in atmospheric CO2 concentrations, changes in the nitrogen cycle and acid rain, climate alterations, and the introduction of exotic species. The causes of human-induced loss on biodiversity are the habitat destruction and fragmentation, over-exploitation of natural resources; pollution of air and water (by several activities such as agriculture); introduction of non-native (alien, or exotic) species and climate change induced biodiversity loss. These factors being inextricably linked with some or all the loss of biodiversity is expected to continue at an unchanged increasing pace in the coming decades (Morris, 2010; Afua, 2011; Azamal et al., 2012)

Despite conservational efforts, the Ethiopian biodiversity is being increasingly threatened and reduced, making Ethiopia one of the most degraded biodiversity hotspots in the world and the threats are multiple and interconnected (Mckee, 2007; Azamal et al., 2012). Human population densities have increased considerably in recent times due to various reasons, at a rate faster than the ability of the land to support. As a result, in order to maintain basic living standards, natural resources have been used up faster than can be naturally replenished or before new sources have been found (Bishaw, 2001). The consequences of this unplanned and unsustainable use of natural resources, together with alterations to the climate and natural ecological processes, have been extensive land degradation, and loss of habitat together with the loss of valuable genetic reserves. It is

14 therefore now imperative that the degradation processes be halted, and even reserved, in order to ensure the sustainable utilization of the numerous ecosystems for the Ethiopian people, both for present and future generations (Etefa, 2011).

2.7. Disturbances in a Forest Ecosystem

Disturbance is any relatively discrete event in time that disrupts ecosystem, community or population structure and changes resources, substrate availability or the physical environment (Hill and Hill, 2009). Disturbance could be a result of human activities or natural causes. These forms of disturbances severely threatened biodiversity (Uniyal et al., 2010). Natural disturbance varies in spatial extent, temporal frequency and magnitude (Hill and Hill, 2009; Shrestha et al., 2013). Natural disturbance can be in the form of fire, wind, etc. among these, wind is the major agent of natural disturbance that renews and modifies forests. Collection of leaf litter for animal bedding and compost making, grazing and collection of fuel wood, harvesting of trees for timber, collection of raw materials for local industries etc. are all forms of human disturbances (Uniyal et al., 2010). Comparing natural disturbances with human disturbances, repetitive process of the natural disturbance could cause more effect on diversity, structure and species richness of plants in forests than the human disturbance (Addo-Fordjour et al., 2009). Ecologists think that if human effects were limited, the forest would return to its natural state (Afua, 2011).

Forest resources are the main source of livelihood of the people living in forest communities. Especially in developing countries like Ethiopia where the livelihood of their nations is directly or indirectly linked to the natural resources; degradation and deforestation of tropical forests due to anthropogenic activities are the major causes of decline in global biodiversity (Mamo et al., 2015). This coupled with the increasing population trend over the past few years, has led to the massive exploitation of natural forests (Uniyal et al., 2010). In recent times many species are threatened due to land- cover and land-use changes all over the world understanding the impact of disturbance on vegetation and the resilience of plant communities to disturbance is imperative to environmental management (Afua, 2011).

2.8. Human Activities Causing Biodiversity Loss in Forest Ecosystems

Habitat destruction has been one of the primary threats to the world’s biodiversity (Afua, 2011). This is the process in which natural habitat is rendered functionally unable to support the species present. In this process, the organisms that previously used the site are displaced or destroyed, reducing biodiversity (Gebretsadik, 2016). Some of the factors responsible for habitat loss include mining, logging, slash-and-burn agriculture, etc. (Afua, 2011). Forest clearance destroys the habitat and generally causes a decline in forest species abundance and diversity. Apart from destroying habitat, forest clearance can fragment the remaining forest, leaving areas of forest that are too small for some species to persist, or too far apart for other species to move between, resulting in a long process of decay in residual diversity from the remaining habitat (Morris, 2010).

Overexploitation can also be the other human induced factor that lead to resource destruction, including extinctions (Morris, 2010). Logging, slash and burn for agricultural expansion can be a cause for over exploitation of forest biodiversity (Azamal et al., 2012). Invasive species are non-native species that have established outside their natural range by human action, has also a tendency to spread, which is believed to cause damage to the environment (Morris, 2010; Gebretsadik, 2016). Invasive species often coexist with native species for an extended time, and gradually the superior competitive ability of an invasive species becomes apparent as its population grows larger and denser and it adapts to its new location; because of this the biodiversity of the forest became very much affected (Gebretsadik, 2016).

2.9. Species Richness and Diversity along Disturbance Gradient

Species richness, is a count of the number of plant species in a quadrat, area or community (Kent, 2012). Species richness often varies from one forest to another depending on a number of factors. Maximum number of species richness has been obtained in intermediately disturbed forest type (Uniyal et al., 2010). On the other hand, other studies have reported of higher species richness in undisturbed stands in relation to other forest types (Addo-Fordjour et al., 2009). Whatever the case, it is clear that species richness is negatively affected by intensive disturbance as observed by (Afua, 2011). Higher species

16 richness at intermediate disturbance level may occur due to the presence of intermediate resource levels present in which many species make use of it. It was also explained by Uniyal et al. (2010) that mild disturbance provides greater opportunity for species turnover, colonization and persistence of high species richness.

The pattern of species richness along a disturbance gradient may vary from one life form to another whereas tree species richness peaked at undisturbed level, that of shrubs peaked at disturbed forest (Pitchairamu et al., 2008). Lalfakawma et al. (2009) find out that higher tree and shrub species richness was recorded in undisturbed stand than in a disturbed stand. However, the number of shrub species richness remained higher in the disturbed stand compared to the undisturbed stand. Higher species richness of trees in the undisturbed forest was attributed to absence of human disturbance (harvesting of trees) that of shrubs in the disturbed forest has been explained by their ability to resist and adapt any form of disturbance (Hill and Hill, 2001). Whatever the case frequency and magnitude of disturbance are key factors for changes in species richness and diversity (Shrestha et al., 2013).

Species richness and diversity also varied between life forms. The difference that exists between the various life forms does not only represent habitat diversity of different sites; it also reflects the adaptation of species to different environments and the large-scale influences on plant (Zhang, 2005). As stated in the works of Afua (2011), tropical rain forest species diversity is not open to any single or simple explanation, but it is due to a complex interplay of factors that must be resolved individually for any particular forest.

2.10. Measurement of Diversity Using Different Indices

A biological community has an attribute that we call species diversity and many different ways have been suggested for measuring it. Recent interest in conservation has generated a strong focus on how to measure species diversity in both plants and animals using different diversity indices (Fisaha et al., 2013). The two main factors taken into account when measuring diversity are richness and evenness. So that, diversity index must be sensitive to both factors, thus must also be sensitive to the different number of species in two or more communities (Ellenberg et al., 1974). A large number of indices of diversity

17 have been devised, each of which seeks to express the diversity of a sample or quadrat by a single number. Diversity is thus measured by recording the number of species and their relative abundances (Kent, 2012).

2.10.1. Species diversity and richness

Species diversity can be viewed from different perspectives: Whittaker (1965, 1972, 1975 cited in Kent, 2012) made a distinction between three types of diversity; Alpha diversity refers to the number of species within the sample area or community, such as a certain type of woodland. Beta diversity is the difference in species diversity between samples or communities that correspond to ‘pieces’ in the landscape mosaic. Beta diversity is thus sometimes called habitat diversity because it represents differences in species composition between very different areas or environments and the rapidity of change of those habitats. The smaller and more numerous the ‘pieces’ of the mosaic of different plant communities, the higher the beta diversity. Alpha diversity remains the number of species within each individual piece of the mosaic. Gamma diversity represents the product of alpha and beta diversity within a landscape the number and frequency of ‘pieces’ within the landscape mosaic (beta diversity) combined with the number of species within each ‘piece’ (alpha diversity). These diversity indices provide information about community composition (Kent, 2012).

Different types of indices of diversity have been devised, each of which seeks to express the diversity of a sample or quadrat by a single number. Some indices are also known by more than one name and are presented in different ways. Of the various indices, the most frequently used is the simple totaling of species numbers to give species richness (Magurran, 2004). However, of the indices combining species richness with relative abundance, probably the Shannon diversity index (H’) sometimes correctly called the s Shannon-Wiener index (퐻′ = − ∑ Pi ln(Pi) is the most commonly used diversity i=1 index which is done based on information theory and the concept that the diversity or information in a sample or community can be measured in a similar way to the information contained within a message or code. The index makes the assumption that individuals are randomly sampled from an ‘infinitely large’ population and also assumes

18 that all the species from a community are included in the sample. Values of the index usually lie between 1.5 and 3.5, although in exceptional cases, the value can exceed 4.5. The Shannon index increases as both the richness and the evenness of the community increase (Kent, 2012).

Species richness refers to the total number of species in a community; this can be measured by counting the number of species in the forest while evenness is the relative abundance of species within the sample or community. The two components can be examined independently or combined in some form of index. Richness can be expressed in a mathematical index to compare diversity between sites. Species richness index is of great importance in assessing taxonomic and ecological values of habitats (Magurran, 2004).

2.10.2. Measurement of similarity and dissimilarity

Many measures exist for the assessment of similarity between vegetation samples or quadrats. Some are qualitative and based on presence/absence data, while others are quantitative and will work on abundance data. Similarity indices measure the degree to which the species composition in quadrats is alike; whereas dissimilarity coefficient assesses which of the two quadrats differ in composition. Another term that is widely used to describe similarity is resemblance, and similarity coefficients are sometimes called resemblance functions. Sorenson’s coefficient of similarity Ss=2a/(2a+b+c) is one of the most common binary similarity coefficient, because it relies on presence or absence data and also gives more weight to species that are present in both quadrats and less weight to species that are present in only one quadrat (Kent, 2012).

2.10.3. Measurement of Species density and frequency

Density is the number of individual plants of a given species per unit area. It can be used to show spatial distribution and ranges over time (spatio-temporal variation of species). Frequency is the proportion of quadrats in which a species occurs. It is a measure of the occurrence of a given species in a given area. Frequency indicates how the species are dispersed and is an ecologically meaningful parameter. According to Kent (2012), it can give an approximate indication of the homogeneity of the quadrats under consideration.

2.10.4. Measurement of importance value index (IVI)

Importance value index (IVI) indicates the relative ecological importance of a woody species in the forest (Gurmessa et al., 2012). IVI was calculated as the sum of relative abundance (%), relative dominance (%), and relative frequency (%) of the species following (Neelo et al., 2013). Plant species vary in their responses to environmental factors. A given species will have a unique set of tolerances to environmental variables, such as disturbance, pH, altitude, light, temperature, moisture, and nutrients. At the community level, these differences in tolerances will cause various species to have competitive advantages depending on the nature of those environmental factors. Therefore, species importance value index permits a comparison of species in a given location and reflects the dominance, occurrence and abundance of a given species in relation to other associated species in an area (Kent, 2012). Generally, to develop conservation strategy and plan, species importance value index is a good index for summarizing vegetation characteristics and to rank species for management and conservation practices and to prioritize them (Gurmessa et al., 2012).

2.11. Conceptual framework of the study

Based on the review of theoretical background a conceptual framework was developed to guide the study (Figure 1). Studies on woody species diversity and population structure in a forest are instrumental in the sustainability of forests since it plays a major role in the conservation and management of forest resource (Addo-Fordjour et al., 2009). As described through section two of the theoretical framework, so many factors affect woody species diversity and population structure of the study forests. As coined by Ordonez et al., (2014) the responsible factors that brings change in woody species diversity and population structures are natural as well as anthropogenic (Afua, 2011). Natural disturbance can be in the form of fire, wind, land degradation and others. In addition, environmental variables like climate, topography, elevation, aspect and others have also the capacity to change woody species diversity, population structure and regeneration status of the study forests (Gebretsadik, 2016).

By conducting a forest survey, one can have indicators and variables in order to decide the status of the study forest for future decision. So, the conceptual frame work (Figure 1) tries to integrate those mentioned factors that will guide towards realizing the objectives or intent of the present study and the casual relationships. The part of the frameworks sketched in bold lines are those concepts which were tried to be covered by the present study; while those sketched in broken lines are socio economic factors that are actually affect the woody species diversity; but, needs further socioeconomic survey’ which was not addressed in this study yet.

Natural factors Indicators and (Environmental variables factors) Existing -Species type & -Climate Vegetation abundance -Altitude types -Diameter and height -Topography -High forest class -Aspect -bush land -Number of seedling, -Woodland and saplings -Count of dead and coppice stumps Anthropogenic factors (disturbance) -Socio economic factors like rise of Woody Regeneration population species ecology -Agricultural land diversity -Soil seed bank expansion & -Seed dispersion -land use/cover change Population -Growth habit of -Deforestation and status seedling -Gap creation forest degradation -availability of seed -Free grazing in the forest and -Fire (energy source) soil seed bank -Increasing demand of -presence of Sustainable forest resources dispersal agents management of Juniperus dominated Afromontane forest

Figure 1: Conceptual framework (own sketch)

3. MATERIALS AND METHODS

3.1. Description of the study area

3.1.1.Geographic Location

The study was conducted in Beyeda District (Figure 2), North Gondar Zone of Amhara National Regional State (ANRS), located adjacent to the Semien Mountain National Park (SMNP). It harbors an extensive area of Juniperus dominated dry evergreen montane forest which might be preserved due to in accessibility of the area; and hence not investigated and even its existence not that much known by the scientific community. The District is situated at about 230 km of Gondar city. Geographically the District lies between 12056’30’’- 13024’00’’N latitude and 38021’00’’-38037’30’’E longitude geographical grids. The District is bordered on the North by Telemt District, on the South by Sahala Seyemit District of Waghimira Zone, to the west by Janamora, and to the East by Abergele District of Waghimira Zone. The district has 17 kebeles and covers an estimated area of 87,459.79 ha. The study forest has an area of 1015.33 ha which was geographically located between 1304’40’’- 1306’40’’N latitude and 38024’00’’- 38026’40’’E longitude geographical grids. Specifically the forest is found and stretched in three kebeles of the district namely Wati, Ayiga, and Janbelew Kebeles administrative having significant coverage of Juniperus dominated natural forest which is conserved and managed by the district as protected forests (BoFED, 2016 ; BDoA, 2017).

Figure 2: Map of the study area (Extracted from BoFED, 2016_Gis data)

Figure 3: Map showing the study forest and sampling plots along disturbance 3.1.2.Demography According to the Amhara Bureau of Finance and Economic Development, (2017) annual report the District has projected total population of 115,853; out of this 57,468 are males and 58,385 are females, from the total population 6.7 % live in urban areas, the population density is 132 person/km² that makes Beyeda one of the sparsely populated district of Amhara region. Since population of the area is sparse more than 90% of the population is living in rural areas and most of the settlement areas are steep mountain escarpments and marginal areas. As a result, the livelihood of the people were based on scarce natural resources (BoFED, 2017, Adugna, 2017).

3.1.3.Topography, geology and climate

As the information gained from Beyeda District Office of Agriculture, (2017) annual report; the topography of Beyeda district is characterized by rugged mountains, cliffs, hills, plateaus, valleys and gorges with an altitudinal range of 1232 - 4543 m above sea level. According to Beyeda District Agricultural Office Report, the topography covers 10% of flat, 70% mountainous, 6% valleys and the rest 14% is hills. Hence, the area has varied landscapes composed of steep escarpments as it is situated to the north eastern part

23 of Semien Mountains National Park where Ras Dejen the highest mountain in Ethiopia is situated in this District. Such undulating land form of the area aggravates soil erosion and low soil fertility which leads the area to be food insecure for several decades. The major soil types of the district as shown in (Figure 4) are umbric leptosol, eutric leptosol, umbric andosol and eutric cambisol (BoFED, 2016; BDoA, 2017; Adugna, 2017).

3.19%

24.38% 1.10% Eutric Cambsols 63.84% 0.19% Eutric Leptosols Haplic Alisols 7.30% Lithic Leptosols Umbric Andosols Umbric Leptosols

Figure 4: Soil Map of the District (Adapted from BoFED, 2016) According to the report from the District Office of Agriculture, the district has four Agro- ecological Zones; which are Wurch (29%), Dega (53%), Woina Dega (12%) and kola (6%). The rainfall is characterized by a uni-modal distribution with mean annual rainfall of 1172 mm. Because of the mountainous nature of the area the rainfall pattern is Orographic type and lasts for 3 months stretching from June till August. The rainfall of this area mainly depends on the prevailing winds which are governed by the movement of the intertropical convergence zone (ITCZ), but much modified by local relief (Friis, 1992 cited in Teshome, 2007). The mean annual average temperature ranges between 130C to 29oC (BDoA, 2017, Adugna, 2017). The District have one long wet season from June to September. During this time, at altitudes above 3,600 m, July and August are characterized by infrequent hailstorms and heavy mists that reduce visibility to 30 m or less for much of the day until the end of September (Teshome, 2007).

3.1.4.Vegetation

As the study area is situated on the North eastern part of Semien Mountains National Park it shares similar vegetation characteristics. Due to the altitudinal differences of the area

24 there are three main vegetation zones. These are Montane forest (1,900 to 3,000m. a.s.l), Ericaceous belt (Sub-Afro alpine) (2,700 to 3,700m a.s.l) and finally the Afro alpine zone (3,700 to 4,533m a.s.l) (Teshome, 2007). Most of the study area is covered by the dry ever green Afromontane vegetation. The Natural Forest of Beyeda District is dominated by Juniperus procera, Myrica salicifolia, Olea europaea subsp. Cuspidate, Erica arborea, Osyris quadripartita, Allophylus abyssinicus, Dodonaea angustifolia, Myrsine africana, and Maytenus arbutifolia (BDoA, 2017). One of the unique vegetation features of the natural forest of Beyeda is; the existence of extensive coverage of Juniperus forest which is uncommon to the rest of the neighboring districts of Northern Amhara. Juniperus is found in both tree and shrub form as depicted in (Figure 5). About 15,742.76 ha of the district is covered by plantation and natural forests including bush lands. Among these the Juniperus dominated forest (study area) accounts 1015.33 ha. This forest is the source of wood for the local people as a construction material as well as for fuel wood as there is a little alternative wood source like Eucalyptus in the area (EWCA, 2015 and BDoA, 2017).

Figure 5: Juniperus afromontane forest of Beyeda District (Photo By MT)

3.1.5.Land use and Land cover

According to the (2017), annual report from the District Office of Agriculture; Beyeda District is one of agriculturally unsuitable and food insecure district of Amhara region due to the climate and topography of the area. But for the sake of survival, the community surrounding the current study area is mainly agrarian. And cereal farming and livestock production is the most dominant land use practice in the area. Barley, wheat, bean, field pea and lentil were the principal crops, and livestock comprised mainly cattle, sheep, goats, donkeys, and horses. The land use and land cover of the district is Crops cover 56% the area, pasture 16 %, forests and shrubs 18%, settlements 3% and the rest 7% is a miscellaneous land. As most land is used for cereal growing, the relative proportion of land allocated to trees and forests is low. According to the comment from elders and professionals in the District; since there is no alternative fuel wood source like Eucalyptus the natural forest is becoming heavily exploited for local uses (BDoA, 2017).

3.2. Vegetation Data Collection and Analysis

3.2.1.Reconnaissance Survey

Initially reconnaissance survey was conducted in the third week of December, (2017) to collect baseline information, to gain a general insight on the study area and the vegetation. It helps to have mental picture and visual information on the study area in relation to its ecological attributes and to identify the possible sampling sites and the number of transect lines to be laid down across the Forest. This preliminary survey was aimed to have a better planning and organizations of the stratifications of the study forests, layout of transects and sample plots and hence the data collection.

3.2.2.Sampling strategy

After the reconnaissance visit the entire forest area was stratified into three homogeneous strata following disturbance gradients based on practical visual observation and following the methodology used by (Afua 2011, and Mamo et al., (2015). These are 1) Less disturbed forest hereafter (LDF), 2) Moderately disturbed forest (MDF) and 3) Highly disturbed forest (HDF) (Table 1 and Figure 5).

Table 1: Description of the study sites along disturbance levels Forest strata/site along Description/practical definition disturbance gradient Less Disturbed Forest This site is situated far from settlement and crop (LDF) cultivation areas, and with relatively low level of logging and grazing, surrounded by mountain cliffs and rivers and availability of permanent forest guard Moderately Disturbed This site is situated in which there were no crop cultivation Forest (MDF) and settlement, but there is some considerable levels of selective logging and grazing; Highly Disturbed Forest This site is situated nearer to settlement and crop (HDF) cultivation areas, there is relatively high intensity of grazing, fuel wood collection, logging, charcoal making and high number of dead stumps.

LDF MDF HDF

Figure 6: Picture shows study sites along disturbance levels (Photo MT) Systematic random sampling was employed for the vegetation sampling in which the first plots in the transects are selected randomly and the subsequent plots are laid out at a specified interval (Magurran, 1988). Systematic sampling was selected to include sufficient representative samples of vegetation from all gradient levels (Krebs, 1999). Following the procedures used in Teketay (2001) and Fisaha et al., (2013) transect lines

27 were laid in the forest following the disturbance gradient of the site starting from the top of ridge to bottom valley and the sampling plots were situated along disturbance gradients (LDF, MDF, and HDF).

As illustrated in (Figure 7), main and sub sample plots were used for data collection on adult trees, seedlings and saplings. Totally 41 square main sample plots (15 for LDF, 16 for MDF, and 10 for HDF) with a size of 20 m  20 m for the sampling of adult trees were laid. Two 10 m  10 m sub plots in each corners of the main plot (totally 82 sub plots) were laid for the sampling of saplings and five 5 m  5 m sub plots in four corners and one at the center of the main plot (totally 205 sub plots) for the inventory of seedlings were laid down at a distance between transect lines and sampling plots of 200 m and 100 m respectively using Garmin 60X GPS and Compass. The first sampling plot was located randomly, and the subsequent plots were established at 100 m intervals systematically.

Figure 7: Simple sketch of data collection quadrates

Species accumulation curve was used to determine minimal sample plots as species accumulation curves illustrate the rate at which new species are found (Magurran, 2004). The adequacy of sample size was estimated by stopping sampling at the point at which additional quadrats did not significantly affect the mean of species (Ellenberg, 1974).

10 Less disturbed SPECIES RICHNESS SPECIES 5 Moderately distuebed

Highly disturbed 0 1 2 3 4 5 6 7 8 9 1011 1213141516 NUMBER OF SAMPLING PLOTS

Figure 8:Species accumulation curve 3.2.3.Woody Species Vegetation Data Collection and Identification

To analyze species diversity and population structure of the study forest along disturbance, complete lists of woody species were recorded from each sample plot throughout the entire area under the study. Species were identified using published volumes of flora of Ethiopia and Eritrea (Hedberg et al., 1989; Hedberg et al., 1995; Hedberg et al., 2003; Demissew, 2014). Data on diameter at breast height and height were recorded in all sample quadrants following the procedures used in the works of (Tilahun, 2015).

All adult tree individuals of the woody species encountered within the main plot and saplings and seedlings of woody species within the subplots were recorded (Figure 7). In each sampling plot, the diameter at breast height (DBH) (1.3 m above the ground) of each tree was measured using Caliper and height were measured using hypsometer/marked stick. For this study, plants were categorized as seedling (height ≤ 1 m), sapling (height between 1 m and 2 m) and tree (height ≥2 m and DBH≥2 cm) following (Senbeta and Teketay, 2001; Fisaha et al., 2013; Kebede and Lemenih et al., 2013; Melese and Ayele, 2017). In cases where a bole branched tree found measurement at breast height or below,

29 the diameters were measured separately and considered as two trees and in cases where tree boles buttressed, DBH were measured from the point just above the buttresses.

To investigated of anthropogenic disturbance on woody species in the forest, all types of disturbances like, the total number of dead standing trees, stumps and logs in the sample plots were counted by involving knowledgeable persons from the local communities. In addition, environmental variables such as altitude; longitude, latitudes, and aspect of each quadrate were measured using “Garmin 60X GPS”. Slope of each quadrate were measured using clinometers.

3.3. Data Processing and Analysis

3.3.1.Vegetation data analysis

Based on the collected ecological data the vegetation structure of the forest at the three disturbance levels (LDF, MDF, and HDF) were evaluated by computing the abundance, frequency, dominance, importance value index (IVI), and by constructing the population structure in terms of DBH class distribution. Heterogeneity of the forest were also determined using Shannon-Weiner diversity and Evenness indices. Comparisons were also made between the vegetation of the three disturbance levels.

Density: density of species is defined as the number of plants of a certain species per unit area. So the mean density of woody species was determined by converting the number of individuals of each woody species encountered in all the quadrats and all transects of the study areas in each of the three sites to equivalent number per hectare as described in Neelo et al., (2013) and Mueller- Dombois (2016). Density of species along disturbance levels was calculated by the following formula.

Number of individuals of species A Total density of species = × 100 Area sampled

Relative density is the proportion of a given species in the study sites to the total density of all species throughout the quadrates under the study. Relative density was computed using the following formula,

Density of species A Relative density of species = × 100 Total density of all species

Frequency: the frequency was calculated as the proportion (%) of the number of quadrats in which each woody species was recorded from the total number of quadrats in each of the sites. It is a measure of occurrence of a given species in a given area. The frequency value obtained reflects the patterns of distribution of species and also it measures the uniformity of the distribution of the species in the study area (Gebrehiwot, 2003). The higher the frequency the more important the plant is in the community. The high frequency value of a given plant species in the given forest indicates that it is widely distributed in the area under the study (Etefa, 2011). In this study Frequency of a species was computed as the proportion of samples within which a species is found, and density was computed by converting the count from the total quadrats into a hectare basis as indicated in (Fisaha et al., 2013). The following formula was used to compute frequency of species, Number of plots with species A Frequency of species = × 100 Total number of plots

The relative frequency was computed as the ratio of the absolute frequency of the species to the sum total of the frequency of all species, and the following formula was used to compute (RF)

Frequency of species A Relative frequency = × 100 Total frequency of all species

Dominance: It refers to the degree of coverage of a species as an expression of the space it occupied in a given area. Usually, dominance is expressed in terms of basal area of the species Kent (2012) and two set of dominance were calculated in this case: absolute dominance (the sum of basal areas of the individuals in m2/ha), and relative dominance, which is the percentage of the total basal area of a given species out of the total measured stem basal areas of all species. Basal area (BA) was calculated for all woody species with DBH > 2cm. Dominance, relative dominance and basal area were calculated using the following formulas,

Total basal area Dominance = Area sampled Dominance of species A Relative dominance = × 100 Total dominance of all species

DBH2 Basal area (m)2 = π 4 Where: BA = basal area per tree in m2 π = 3.1416 DBH = diameter at breast height in cm (1.3m above the ground) Importance value index (IVI): indicates the relative ecological importance of a woody species in the study sites. IVI was calculated as the sum of relative abundance (%), relative dominance (%), and relative frequency (%) of the species following (Neelo et al., 2013). It was calculated by the following formula, Importance Value Index (IVI) = Relative density+ Relative frequency+ Relative dominance

퐵푖 푛푖 푓푖 퐼푉퐼푖 = [ × 100] + [ × 100] + [ × 100] ∑ 퐵푗 ∑ 푛푗 ∑ 푓푗 Where: IVIi = the Importance Value Index (IVI) of the ith species ni = the number of individuals of the ith species; nj = the sum of individual trees of all species Bi = the basal area of the ith species; Bj = the total basal area (m2) of all specie fi = the absolute frequency of the ith species; Species richness (S): is the total number of different woody species in the study sites. It does not take in to account the proportion and distribution of each species at the study sites.

Species diversity of woody species was analyzed by using the Shannon Diversity Index (H’) (also known as the Shannon Weiner Diversity Index in the ecological literature)

(Krebs, 1989; Magurran, 2004). The index takes into account the species richness and proportion of each species in all sampled plots of the study site. The following formula was used to analyze woody species diversity

s Species Diversity (H′) = − ∑ Pi ln(Pi) i=1

Where, H’= Shannon index, S= Species richness, Pi= proportion of S made up of the ith species (relative abundance).

Species Evenness or equitability index: is a measure of similarity of the abundance of the different woody species in the study sites, was analyzed by using Shannon’s Evenness or Equitability Index (E) (Krebs, 1989; Magurran, 2004). Equitability assumes a value between 0 and 1 with 1 being complete evenness. The higher the value of evenness index, the more even the species is in their distribution within the given area (Etefa, 2011).The following formula was used to calculate evenness.

S  piln( pi) i1 Eveness(J)  ln(s)

Where, J’= Evenness and S= Species richness

The similarity in woody species along disturbance (LDF, MDF, HDF) was computed by using Sorensen’s similarity index (Kent, 2012). The values range between 0 and 1: where 0 indicates complete dissimilarity and 1 indicates complete similarity in species. It was calculated using the following formula.

2푎 푆푠 = (2푎 + 푏 + 푐)

Where Ss’= Sorensen’s similarity coefficient, a= number of species common to the samples b=Number of species in sample 1 only and c= Number of species in sample 2 only. Often the coefficient is multiplied by 100 to give a percentage similarity index.

Regeneration status analyses: the regeneration status of the major study species in along disturbance were summarized based on the total count of seedling and sapling of each species across all quadrats and presented in tables, graphs and frequency histograms following Senbeta and Teketay (2001). Individuals were counted as seedlings with a height of ≤ 1m and DBH of ≤ 2 cm) and saplings (with a height > 1m and with a DBH ≤ 2 cm) following Peters (1996).

The regeneration status of the forest was assessed using the following categories used by (Vishal, 2011 and Fisaha et al., 2013). 1. ‘Good’, if presence of seedling > sapling > adult trees; 2. ‘Fair’, if presence of seedling > sapling < adult trees; 3. ‘Poor’, if a species survives only in the sapling stage, but not as seedlings (even though saplings may be <, >, or = mature trees); 4. ‘None’, if a species is absent both in sapling and seedling stages, but present as mature; and 5. ‘New’, if a species has no mature, but only sapling and/ or seedling stages.

Besides the aforementioned ecological analytical procedures both descriptive and inferential statistics were used for further analysis of data in this study. One-way ANOVA and MANOVA at a significance level of 0.05% was used to test the differences in woody species parameters (species diversity, species richness, species evenness and density) along the gradients of disturbances. In addition to this, descriptive statistics was also used to show proportion of stand parameters. The analysis was performed by using statistical package SPSS version 25.0 software. In addition to this Microsoft office Excel were used for data entry and for the analysis of descriptive statistics.

Summary of the data collection and analytical procedures under the respective specific objectives are presented in (Table 2).

Table 2:Research Methodology and data analysis Summary

Objectives Indicators &Variables Data sources and data Sampling procedures Data analysis tools collection methodologies  Assess woody species  Species name  The entire forests  The study area is  Shannon’s diversity and composition along gradients of  Number of species  Vegetation survey: selected and evenness index, and disturbance  Number of all types and number stratified in to LDF,  To analyze species diversity individuals of woody species MDF, and HDF Sorensen’s similarity index along the gradients of (abundance) were recorded disturbance levels.  The IVI were computed by disturbance  Reconnaissance using Microsoft Excel soft survey were made to stratify the forest wares following gradients of disturbance  Determine the vertical and  DBH-class  In the entire forest  Then transect lines  Population structure were horizontal structure of the  Height DBH (>2cm) were were laid with computed using graphs forest along gradients of  Number of trees, measured, defined distance based on DBH and height disturbance saplings and seedlings  systematic random class and number of sampling was individuals employed  Determine the regeneration  Number of trees,  The entire forest  Regeneration status were Square sample quadrats status of the forest along saplings and seedlings  Vegetation survey analyzed based on with a size of 20 m  20 gradients of disturbance were counted (direct count) population size of m, 10 m  10 m and 5 m seedlings, saplings and  5 m for mature matured trees  Investigate of anthropogenic  Number of dead and  The entire forests tree/shrub, sapling and  Counted data of dead and disturbance on woody species coppiced stumps were seedling, respectively, coppiced stumps were and examine the effect of counted were laid down analyzed using MS-Excel alternatively along the environmental variables on  Elevation, slope and  The entire forest  ANOVA and MANOVA line transects then all species parameters (Species Aspect were made using SPSS v-25 diversity, Richness, Evenness variables were recorded and Density)

4. RESULT AND DISCUSSION

4.1. Woody Species composition, richness, diversity, and similarity along the gradients of disturbance

4.1.1.Composition of woody species

From the established sample plots a total of 24 woody species which were belongs to 24 genera and 20 families were identified (Table 3). Among the families, Sapindaceae, Loganiaceae, Rosaceae, and Fabaceae were the most diverse families each represented with 2 species and constituting 40% of the species composition along disturbance gradients; the rest of the 12 families were represented by a single species.

Table 3: Lists of woody species identified

Family name Species name Local Life Disturbance levels name form LDF MDF HDF Cupressaceae Juniperus procera Hochst. Tid Tree * * * ex Endl. Myrtaceae Myrica salicifolia A. Rich. Shinet Tree * * * Oleaceae Olea europaea subsp. Weyira Tree * * * cuspidata Ericaceae Erica arborea L. Wuchena Tree * * * Santalaceae Osyris quadripartita Decn. Keret Shrub * * * Sapindaceae Allophylus abyssinicus Embis Tree * * - (Hochst.) Sapindaceae Dodonaea angustifolia L. f. Kitkita Shrub * * * Myrsinaceae Myrsine africana L. Kechemo Shrub * * * Euphorbiaceae Clutia abyssinica Jaub. & Feyele fejj shrub * * * Spach. Celastraceae Maytenus arbutifolia (A. Atatt Shrub * * * Rich.) Wilczek Rhamnaceae Rhamnus staddo A. Rich. Teddo Shrub - * * Pittosporaceae Pittosporum abyssinicum Weyl Tree * * - Del. Hypericaceae Hypericum revolutum Vahl Amija Shrub * * * Loganiaceae Buddleja polystachya Fresen. Anfar Tree * * * Apocynaceae Carissa spinarum L. (c. Agam Shrub * * * edulis) Rosaceae Rosa abyssincia Lindley. Kega Shrub * * * Polygonaceae Rumex nervosus Vahl Imbwacho Shrub - * *

Fabaceae Calpurnia aurea (Ait.) Digta Shrub * - - Benth. Rosaceae Hagenia abyssinica (Bruce) Kosso Tree * - - J.F. Gmelin Araliaceae Schefflera abyssinica Getem Tree * - - (Hochst. ex A. Rich.) Loganiaceae Nuxia congeta R.Br. ex Askwaro Tree * * - Fresen. Sterculiaceae Dombeya torrida (J. F. Wulkifa Tree * * - Gmel.) P. Bamps Salicacea Salix subserrata Willd. Haya Tree * * - Fabaceae Acacia abyssinica Hochst. B/Girar Tree - * - Key * present - absent

Out of the total species composition, trees share 53%, 42% and 20% in the LDF, MDF and HDF sites, respectively. Similarly, Shrubs share 37%, 58% and 80% in the LDF, MDF and HDF sites respectively (Figure 9). This implies that future fate of the forests may alter to bushy and shrub land as human interference continued.

90 Shrub Tree 80 70 60 50 40 30 20 10 Number od trees and and % in shrubs Number trees od 0 LDF MDF HDF Disturbance level Figure 9: Composition of woody species along the gradients of disturbance Six trees and nine shrubs species were common to the three study sites. Juniperus procera, Myrica salicifolia, Olea europaea, Erica arborea, Allophylus abyssinicus, and Buddleja polystachya were among the tree species common to all forest sites. Osyris quadripartita, Dodonaea angustifolia, Myrsine africana, Clutia abyssinica, Maytenus arbutifolia, Rhamnus

Staddo, Hypericum revolutum, Carissa spinarum, and Rosa abyssinica were common shrubs in the three disturbance gradients.

The presence of such common species may be due to similar disturbance tolerance habitat of those woody species and the close distance between sampled plots. Shrubs constituted the major life form of woody species in the HDF, MDF, and LDF sites as compared to trees. This may be due to the ongoing anthropogenic disturbance like selective logging seriously decreased the number of tree species. Specially in highly disturbed site shrubs constitute 80% of the whole population, which may be because of the presence of disturbance tolerant and resistant species. The present findings were agreed with similar study by (Betemariyam, 2011) in Desa Forest.

Comparing the species composition of the study forests with other similar forests in Ethiopia; the species composition of Beyeda Afromontane forest resembles more with Denkoro Forest (Ayalew et al., 2006), Ericaceous Forest of Semien Mountains (Teshome, 2007); Yegof Forest (Mohammed and Abraha, 2013); Menagesha Suba Forest (Benti, 2011); Desa Forest (Betemariyam, 2011); Menagesha Amba Mariam Forest (Tilahun, 2015) and Hugumburda Forest (Aynekulu et al., 2016).

Table 4: Species diversity, richness and evenness of the study forest along disturbance levels (LDF, MDF, and HDF) Forest sites Life Diversity Evenness Woody Species stages (H') species Richness Density/ha Seedlings 2.14 0.64 48565 16 Saplings 2.72 0.83 6666 16 Less disturbed forest Trees 1.18 0.41 3103 18 Seedlings 1.67 0.67 37495 14 Saplings 1.81 0.77 5300 15 Moderately disturbed forest Trees 1.32 0.47 1878 17 Seedlings 1.43 0.72 9312 14 Saplings 1.83 0.76 1885 11 Highly disturbed forest Trees 0.94 0.52 528 11 Mean 1.67 0.64 12748 14.67 CV 22.88 7 23 18 LSD 0.605 0.04 678.82 2.51 Significance [0.05] * ** * *

4.1.2.Woody Species Richness

The overall Species richness of the LDF, MDF, and HDF were 22, 21, and 15 respectively. Hence, Species richness decreases along the gradients of disturbance. The highest woody species richness was recorded in LDF and MDF whereas the woody species richness in HDF was lower. Again, the number of families recoded in the three strata follows similar decreasing trend with the increase in disturbance level. In the LDF, MDF, and HDF the number of families recorded were 20, 19, and 11, respectively. Furthermore, the species richness of life stages along disturbance significantly differ at (p=0.021) showing a general decreasing trend as intensity of disturbance increases (Figure 10). The decrease in the species richness along the increase in intensity of disturbance gradient may reflect high utilization pressure of the forests. That means, due to the ongoing encroachment some of the woody species might be eliminated either by selective logging or by browsing herbivores animals.

The species richness of this forest was lower than other similar forests like; Bale Mountain Forest (230 species) (Yineger et al., 2008); Denkoro Forest (174 species) (Ayalew et al., 2006); Menagesha Suba Forest (112 species) (Benti, 2011); and Wof Washa Forest (62 species) (Fisaha et al., 2013);. This might be due to geographical location, habitat differentiation and other anthropogenic impacts. Moreover altitude may also be another causes for changes in species richness; as the study forest is situated at an altitudinal ranges between 3060m-3248 m a.s.l. which is the upper limit of Juniperus forest (Sharew, 1994). The decrease in species richness at higher elevations may be due to the decrease in temperature, which may reduce productivity (Betemariyam, 2011). The extent of these effects are dependent on the type and severities of the disturbances (Mamo et al., 2015). As expected, the present study demonstrated that woody species richness varies along disturbance gradients in the studied forest. Overall woody plant species richness of Juniperus dominated forest of Beyeda were generally decreasing with increasing disturbance (Figure 10). This pattern of species distribution is supported by other related studies like Addo-Fordjour et al., (2009) in Tinte Bempo Forest of Ghana; and Afua (2011) in Atiwa range Forest. All these studies have related this trend to the influence of human disturbances.

Less disturbed forest Moderately disturbed forest Highly disturbed forest 20 18 16 14 12 10 8 6 4 2 0 Seedling Sapling Trees

Figure 10: Woody species richness along disturbance gradients

4.1.3.Woody Species diversity

The overall diversity index (H’) of the forest along disturbance gradients (LDF, MDF, and HDF) were 1.95, 1.88, and 2.01, respectively (Table 4 in section 4.1.2). The diversity was quantitatively higher in highly disturbed forest; that could be due to the proportional representation of shrubby and stress resistant species like Dodonaea angustifolia, Clutia abyssinica, Rumex nervosus, and Myrsine africana. Likewise, Species evenness index (E’) along disturbance were 0.64, 0.62, and 0.72 respectively. There was no significant variation in overall woody species diversity along the gradients of disturbance. But, the woody species diversity in between life stages (Seedling, sapling, and trees) quantitatively varied as the intensity of disturbance increases showing a decreasing trend as the intensity of disturbance increases (Table 4 in section 4.1.1).

The possible reason for the variation of species diversity between life stages along the increased disturbance level may be due to the growth habit of some species and the intensity of grazing and selective logging that eliminates some species along the gradients of disturbance. Based on the species diversity standards used by Melese and Ayele (2017), the overall species diversity of LDF and MDF were low; whereas it was medium in HDF. With respect to life stages; species diversity of the forest at LDF is medium for seedlings and saplings; low for adult trees. In MDF

40 the species diversity is low for all life stages; and the species diversity in HDF is low for seedlings and saplings; very low for adult trees (Table 4, in section 4.1.1).

Comparing the result to other forests, the species diversity of these study forest along disturbance were lower than Bale Mountain forest (Yineger et al., 2008), Denkoro forest (Ayalew et al., 2006); Wof Washa forest (Fisaha et al., 2013); and Menagesha Amba Mariam forest (Tilahun, 2015). The possible reason may be due to low altitudinal range between those forest sites (3060-3248m a.s.l) as altitude plays paramount role in plant distribution and richness/diversity (Wassie, 2017).

The low diversity of the study forest indicates that the ecological condition of the study forest is unstable and there is unbalanced species abundance along disturbance gradients. There is a concept that, greater diversity leads to greater productivity in plant communities, greater nutrient retention in ecosystems and greater ecosystem stability Tilman (2000) hence, the study forest has a great problem in ecosystem stability as a result of the continuous harvesting of trees for construction wood, fuel wood and timber and clearing of shrubs/trees for other routine purposes by the local community since there are no alternative wood sources like the exotic Eucalyptus in the study area; and therefore, there is a great human pressure on the natural forest. The mean evenness index of the study forest is 0.64 showing more or less an even distribution of the different disturbance tolerant species in the forest. The finding of the present result were in agreement with the finding of Betemariyam (2011) and Kebede et al. (2013).

4.1.4.Woody species similarity

The distribution of woody species along disturbance indicates different similarity patterns. The overall similarity coefficient ranges from 73-88% along disturbance. The highest similarity was observed between LDF and MDF (88%) followed by HDF and MDF (80%). whereas the similarity between HDF and LDF was 73% (Table 5). Therefore, there is high species similarity along disturbance gradients. The similarity may be contributed by the similarity of environmental gradients such as elevation ranges and the rainfall and temperature patterns (similar environmental characteristics) of the area. The most dominant canopy species, namely Juniperus procera, Myrica salicifolia, Olea europaea, Erica arborea, and Allophylus abyssinicus are similarly found in all disturbance levels. Moreover, the sample plots are situated

41 in close proximity. Thus, gene flow is expected along disturbance gradients through seed dispersal agents such as birds and domestic and wild animals (Wassie, 2017).

Table 5: Similarity of woody species between the three sites

Forest sites LDF MDF HDF Less disturbed forest 1.00 Moderately disturbed forest 0.88 1.00 Highly disturbed forest 0.73 0.80 1.00

4.2. Density, frequency, dominance and importance value index of woody species 4.2.1.Density of species along disturbance gradient

The total density (expressed as the number of stems per hectare) of all woody species in LDF, MDF, and HDF were 58334/ha, 44673ha, and 11725/ha respectively (Figure 10 and table 6). The density along disturbance gradient shows a decreasing trend as the intensity of disturbance increases. The total density of life stages also follows similar trend as shown in. As test result shows, the density of life stages (seedlings, saplings, and adult trees) along the gradients of disturbance differ significantly at (p=0.03).

Few species of seedlings, saplings and adult trees were found to predominate the density of the vegetation of the study forest. For instance, five species: Myrsine africana, Juniperus procera, Maytenus arbutifolia, Clutia abyssinica, and Myrica salicifolia contributed to 65 % of the total density at LDF (Table 7). Similarly, five species such as Myrsine africana, Juniperus procera, Clutia abyssinica, Erica arborea, and Myrica salicifolia contributed to 83.5% of the total density ha- 1 in MDF (Table 6). In HDF Myrsine africana, Dodonaea angustifolia, Clutia abyssinica, Maytenus arbutifolia, and Rumex nervosus contributed to 77% of the total density (Table 7). In HDF Dodonaea angustifolia species alone contributed about 30% of the total density; which indicates how much the area is affected by human disturbance; since such species is a common species of disturbed sites (Hedberg et al., 1989).

80000 70000 60000 50000 40000 30000 20000

Density/ha 10000 0 LDF MDF HDF

Figure 11: Density of the forest along the gradients of disturbance The density of seedlings, saplings and trees decreased when the intensity of disturbance increased (Figure 11 and table 6). This decline may be due to a gradual and consistent increase in the extraction of fuel wood and construction wood for local consumption. This result were supported by the findings of Addo-Fordjour et al., (2009) and Kumar et al., (2016). The density seedlings and saplings of the forest is higher than similar forests in Ethiopia. But, the density of adult trees along disturbance is lower than similar forests like Yegof Forest (5574 stems/ha) (Mohammed and Abraha, 2013); Menagesha Amba Mariam Forest (4362 stems/ha) (Tilahun, 2015). The high density of seedlings and saplings may be attributed by the opening of regeneration gaps as a result of the ongoing selective logging of bigger mature trees and the high demand of mature trees for their local consumption, as species like, Juniperus procera, Olea europaea. Schefflera abyssinica, and Hagenia abyssinica are some the preferred timber species.

Table 6: Density of life stages (seedlings, saplings, and mature trees) along disturbance

Life stages Disturbance levels Less disturbed Moderately Highly disturbed Forest disturbed Forest Forest Seedlings 48565 37495 9312 Saplings 6666 5300 1885 Adult trees 3103 1878 528

Table 7: Density of life stages of all species along the gradients of disturbance

No Species name Density of Density of Sapling/ha Density of Tree/ha Seedling/ha LDF MDF HDF LDF MDF HDF LDF MDF HDF 1 Juniperus procera 3317 7025 368 1303 1219 265 1943 1083 305 2 Myrica salicifolia 1253 1170 312 670 691 210 655 298 143 3 Olea europaea 1675 975 496 333 291 0 13 9 8 4 Erica arborea 2069 1800 72 857 1197 25 285 316 10 5 Osyris quadripartita 1584 1005 184 460 394 20 77 78 5 6 Allophylus 832 85 80 153 19 0 33 9 8 abyssinicus 7 Dodonaea 1915 1060 2792 590 213 760 32 25 30 angustifolia 8 Myrsine africana 20699 18520 968 897 566 85 5 0 3 9 Clutia abyssinica 3173 3115 2096 467 481 255 0 0 0 10 Maytenus arbutifolia 7557 1770 96 723 122 130 3 8 0 11 Rhamnus staddo A. 1813 740 608 80 25 0 0 0 0 Rich. 12 Pittosporum 704 40 0 40 0 0 2 5 0 abyssinicum 13 Hypericum 0 0 40 10 38 20 15 27 3 revolutum 14 Buddleja 336 0 0 10 16 20 8 6 8 polystachya 15 Carissa spinarum 496 0 112 0 0 0 0 3 0 16 Rosa abyssincia 683 30 0 70 16 0 0 3 8 17 Rumex nervosus 0 160 1088 0 16 95 0 0 0 18 Calpurnia aurea 459 0 0 0 0 0 2 0 0 (Ait.) 19 Hagenia abyssinica 0 0 0 3 0 0 7 0 0 20 Schefflera abyssinica 0 0 0 0 0 0 5 0 0 21 Nuxia congeta 0 0 0 0 0 0 10 2 0 22 Dombeya torrida 0 0 0 0 0 0 3 2 0 23 Salix subserrata 0 0 0 0 0 0 5 3 0 Willd. 24 Acacia abyssinica 0 0 0 0 0 0 0 2 0 Hochst. Total 48565 37495 9312 6666 5300 1885 3103 1878 528

4.2.2.Frequency of woody species along disturbance

In all the three disturbance levels Juniperus procera was the most recurring species. The frequency of occurrence of woody species along disturbance (LDF, MDF, and HDF) was varied when the level of disturbance increases; for instance, in LDF the most frequent species were Juniperus procera, Myrica salicifolia, Erica arborea, Osyris quadripartita, and Allophylus abyssinica. In MDF Juniperus procera, Myrica salicifolia, Erica arborea, Osyris quadripartita, and Hypericum revolutum were the most frequently encountered species. Coming to HDF Juniperus procera, Dodonaea angustifolia, Myrica salicifolia, Olea europaea, and Erica arborea were the most frequent species which indicates that the area is seriously disturbed by many factors being such species are a common species of disturbed environments (Wassie, 2017) (Table 8).

Comparing these species with other forests, it was highly similar with Menagesha Amba Mariam Forest (Tilahun, 2015), Hugumburda Forest (Betemariam, 2011), Yegof Forest (Mohammed and Abraha, 2013), and Wof Washa Forest (Fisaha et al., 2013). This similarity may be attributed to the climatic condition of the study area and the study forest is resembled to the single dominant dry Afro montane forest (Friis, 1986). Except few variations, many of the most frequent species in the study forest along disturbance were similar one another.

4.2.3.Dominance of woody species along disturbance gradients

Dominance which is a function of basal area provides a better measure of the relative importance of the species than simple stem count (Tilahun, 2015). Thus, species with the largest contribution to basal area can be considered as the most ecologically important species in the forest. The basal area of the forest along disturbance (LDF, MDF, and HDF) were 8.96 m2/ha, 3.90 m2/ha, and 1.09 m2/ha respectively (Figure 12). In each of the three disturbance gradients Juniperus procera contributes the highest basal area and Dombeya torrida, and Hypericum revolutum were contributed to the lowest basal area/ha (Table 8).

10 9 8

/ha 7 2 6 5 4

3 Basal area M area Basal in 2 1 0 LDF MDF HDF Disturbance level

Figure 12: Basal area along disturbance levels

The cutting down of large trees in the MDF and HDF may contributed to the relatively lower basal area of woody species along disturbance. The total Basal area of the forest was very small and it is not comparable with typical dry Afro montane forests in Ethiopia like: Wof Washa Forest having 64.3 m2/ha (Fisaha et al., 2013); Menagesha Amba Mariam Forest having 84.17 m2/ha (Tilahun, 2015); Yegof Forest having 25.4 m2/ha (Mohammed and Abraha, 2013) and Ericaceous Forest of Semen Mountains (18.3m2//ha) (Teshome, 2007). The only forest which resembled with the study forests were Hugumburda Forest having a Basal area of nearly 9 m2/ha (Betemariyam, 2011). Moreover, the total basal area of woody species (ha-1) was generally in the lower range provided for tropical and subtropical dry and wet forests of the world (Murphy and Lugo, 1986). The basal area of the study forest decreased with increasing disturbance and which were in agreement with the findings of Addo-Fordjour et al., (2009) and Betemariyam (2011). The possible reason for having smaller Basal area may be due to selective removal of bigger woody species for fuel wood, construction wood and other local usages. Moreover, based on the existing situation of the forest shows that there was a major disturbance in the past times and this may be attributed to the indiscriminate destruction of woody species in the area.

Table 8: Relative Density, frequency, dominance and importance value of woody species across the three sites

species LDF MDF HDF RD RF RBA IVI RD RF RBA IVI RD RF RBA IVI Juniperus procera 62.6 15.8 71.1 149.5 57.7 16.3 62.9 136.8 57.8 25.7 65.7 149.2 Myrica salicifolia A. R 21.1 14.7 15.5 51.4 15.9 16.3 14.2 46.4 27.0 20.0 17.5 64.5 Olea europaea 0.4 4.2 2.7 7.4 0.5 5.1 3.2 8.8 1.4 8.6 8.4 18.4 Erica arborea L. 9.2 15.8 5.2 30.2 16.8 15.3 12.5 44.6 1.9 5.7 2.8 10.4 Osyris quadripartita 2.5 12.6 0.7 15.8 4.2 11.2 2.7 18.1 1.0 2.9 0.5 4.3 Allophylus abyssinicus 1.1 8.4 0.2 9.7 0.5 3.1 0.4 3.9 1.4 5.7 1.8 8.9 Dodonaea angustifolia 1.0 3.2 0.2 4.3 1.3 7.1 0.4 8.9 5.7 14.3 1.8 21.8 Myrsine africana L. ------0.5 2.9 0.0 3.4 Maytenus arbutifolia 0.3 3.2 0.1 3.5 0.4 3.1 0.2 3.6 Pittosporum abyssinin 0.1 1.1 0.0 1.1 0.3 3.1 0.1 3.4 Hypericum revolutum 0.5 3.2 0.1 3.8 1.4 8.2 0.9 10.4 0.5 2.9 0.1 3.5 Buddleja polystachya f 0.3 3.2 0.1 3.6 0.3 2.0 0.2 2.6 1.4 5.7 0.7 7.9 Carissa spinarum L. 0.2 2.0 0.1 2.3 Rosa abyssincia Lin. 0.2 2.0 0.1 2.3 1.4 5.7 0.7 7.9 Calpurnia aurea (Ait.) 0.1 1.1 0.0 1.1 Hagenia abyssinica 0.2 2.1 0.4 2.7 Schefflera abyssinica 0.2 2.1 0.9 3.2 Nuxia congeta R.Br. ex 0.3 4.2 2.0 6.6 0.1 1.0 0.5 1.6 Dombeya torrida (J. F. 0.1 2.1 0.0 2.2 0.1 1.0 0.0 1.1 Salix subserrata Willd. 0.2 3.2 0.8 4.1 0.2 2.0 1.2 3.4 Acacia abyssinica 0.1 1.0 0.6 1.7 100.0 100.0 100.0 300.0 100.0 100.0 100.0 300 100.0 100.0 100.0 300.0

RD=relative density RF=relative frequency RBA=relative basal area IVI=importance value index

4.2.4.Importance Value Index (IVI) of woody species in each study sites

Based on the IVI value as shown in (Table 8 and Figure 13); the most dominant species in LDF and MDF in decreasing order were Juniperus procera, Myrica salicifolia, Erica arborea, Osyris quadripartita, and Allophylus abyssinica. Whereas the most dominant species in HDF in order of descending were Juniperus procera, Myrica salicifolia, Dodonaea angustifolia, Olea europaea, and Erica arborea. These shows that much of IVI was attributed by a few woody species. These tree individuals were tolerant species that resist high pressure of disturbance, natural and environmental factors, and the effect of local communities Benti (2011).

As most research findings suggested that species having the highest IVI value are ecologically dominant and less demanded by the local communities and due to this less priority should be given for conservation. Whereas those rare species having lesser IVI values are needed urgent conservation priority as they are on the verge to be endangered (Gurmessa et al., 2012). But, on the case of the study forest those woody species having the higher IVI value are the most valuable and preferable woody species which are adoptable to the existing situation of the area. So, priority for conservation of these dominant species should be given (the first priority for species with the highest IVI value and the last priority of conservation for species with least IVI values) to these species (Benti, 2011; Zegeye et al., 2011; Aynekulu et al., 2016) (Table 8). Almost 50% of the IVI value was contributed by Juniperus procera which is one of the highly valuable woody species. Hence, this species is the most ecologically important species of the study area (Figure 12).

LDF MDF HDF 160

140

120

100

IVI IVI value 60

Dominant species

Figure 13: The proportion of the dominant woody species along disturbance gradient 4.3. Population Structure of the Common Woody Species along the Gradients of Disturbance

4.3.1.Diameter and height class distribution of the common woody species along disturbance levels

The DBH class distribution of woody species density in the study area showed an Inverted-J shaped structure (Figure 14). All DBH classes were represented by at least some individuals even though the highest density was concentrated in DBH classes below 15.5 cm. when the three disturbance levels compared with one another LDF and MDF fully satisfies the inverted ‘J’ shape structure. It is an indicator of healthy forest; thus, intermediate disturbance is best for good recruitment of woody species especially for light demanders. If there is good space or gap for regeneration the population structure became inverted ‘J’ shaped, that is the healthy and normal functioning of the forest ecosystem.

2000 1800 1600 1400 1200 1000 LDF 800 MDF

600 Number Number stems of 400 HDF 200 0 1 2 3 4 5 6 DBH-Class [CM]

1=2-8.5cm, 2=8.6-15.1, 3=15.2-21.7, 4=21.8-28.3, 5=28.4-34.9, 6=35-41.5 Figure 14: Horizontal structure of the forests

Looking in to the population structure of the representative dominant woody species, the DBH class profile along disturbance can be grouped in three exhibited patterns. The first pattern is the inverted ‘J’ shaped pattern, i.e., species having many individuals at the lower diameter classes and decreasing number of individuals at successively higher diameter classes, which is an indication for good biological functions of species. In LDF; species like, Juniperus procera, Myrica salicifolia, Osyris quadripartita, Erica arborea, and Allophylus abyssinicus exhibited this pattern (Figure 15). Similarly, In MDF Juniperus procera, Myrica salicifolia, Erica arborea, and Olea europaea exhibited the same trend; although the density of each species varied following the site characteristics (Figure 16). But in HDF only Juniperus procera, and Myrica salicifolia followed inverted ‘J’ shape pattern (Figure 17). The second pattern is bell shaped distribution pattern having less number of individuals in lower and higher DBH classes; but, high number of individuals in middle classes that indicate a poor reproduction and recruitment potential. Species like Dodonaea angustifolia in HDF and Osyris quadripartita in MDF exhibited this pattern (Figure 16 and 17). The third pattern is interrupted bell-shaped distribution pattern having low number of individuals in the lower DBH class and high number of individuals in the middle DBH class. But there are very limited and missing individuals in the higher DBH class. Species like Erica arborea and Allophylus abyssinicus confirmed this pattern in HDF (Figure 17)

Juniperus procera ) Myrica Salcifolia Erica arborea 1000 120 900 350 100 800 300 700 250 80 600 200 60

500 150 Density/ha

400 100 40 Density/ha Density/ha 300 50 20 200 0 100 0 0

DBH Class in cm DBH class in cm DBH Class in cm.

Osyris abyssinica Alloyphayllus abysinicus 25 12 20 10 8 15 6

10 Density/ha 4 Density/ha 5 2

0 0

DBH Class in cm DBH Class in cm

Figure 15: population structure (DBH class distribution) of representative species in LDF site

Juniperus procera Myrica Salicifolia Erica arborea 600 140 140 500 120 120 400 100 100 80 80

300 Density/ha

Density/ha 60

Density/ha 60 200 40 40 100 20 20 0 0 0

DBH Class in cm DBH Class in cm DBH class in cm

Olea europaea Osyris quadripartita 5 16 14 4 12 3 10 8 2

Density/ha 6 Density/ha 1 4 2 0 0

DBH Class in cm DBH Class in cm

Figure 16: population structure (DBH class distribution) of representative species in MDF site

Juniperus procera Myrica salicifolia Dodonaea angustifolia 60 30 5 50 25 4 40 20 3 30 15

10 2 Density/ha

20 Density/ha Deensity/ha 5 10 1 0 0 0

DBH Class in cm DBH Class in cm DBH Class

Erica arborea Alloyphayllus abyssinicus 2.5 3.5 2 3 2.5 1.5 2

1 1.5 Density/ha 0.5 Density/ha 1 0.5 0 0

DBH class incm DBH in cm

Figure 17: population structure (DBH class distribution) of representative species in HDF site

The height class distribution of LDF and MDF showed similar pattern to DBH classes. However, in HDF the last two height classes were missing at all. This indicate in highly disturbed forest those big trees were removed from the forest for local uses. The majority of individuals were accumulated at the lower height classes (<7 m) (Figure 18). As it was observed the upper canopy of the forest were dominantly occupied by Juniperus procera, Myrica salicifolia and Erica arborea species; whereas, the middle canopy was occupied by Olea europaea, Allophylus abyssinicus, Hypericum revolutum, Osyris quadripartita, and Buddleja polystachya and others and the lower canopy were dominated by Clutia abyssinica, Calpurnia aurea, Dodonaea angustifolia, Rumex nervosus, Maytenus arbutifolia, and Rhamnus staddo.

1800

1600

1400

1200 LDF 1000 MDF 800 HDF Number Number stems of 600

400

200

0 1 2 3 4 5 6 Height class

1=2-4.08m, 2=4.09-6.17, 3=6.18-8.26, 4=8.27-10.35, 5=10.36-12.44, 6=12.45-14.53 Figure 18: Vertical structure of the forests

The overall population structure of the three disturbance levels indicates the dominance of small-sized trees and shrubs because of selective logging of bigger trees. This result agrees with the findings of Mekuria (2007) in Douga Tembein Forest and Betemariyam (2011) in Desa forest. By observing the inverted ‘J’ shape pattern, it can be said that the forests are in a good recruitment capacity. But, it is difficult to generalize based on the shape of the population pattern of the whole forest alone as indicator of forest health; because a highly skewed Inverted-

‘J’ shaped population pattern may to some extent indicate abnormal distribution of population densities in the forests undergoing secondary development after forest clearance or disturbance such as selective logging of individuals having higher DBH and height classes (Friss, 1986 cited in Wassie, 2017). The situation on the studied forest reflects this argument. Therefore, if such condition is continued the fate of the forest will be in danger.

4.4. Regeneration Status of Woody Species along Gradients of Disturbance

The total density of seedling and saplings along variability of disturbance were 48,565 ha-1 and 6666 ha-1 in LDF; 37495 ha -1 and 5300 ha-1 in MDF and 9312 ha-1 and 1885 ha-1 in HDF respectively (Figure 19). So, this result indicates that the distribution of seedling population is greater than that of sapling and sapling population is greater than mature individuals. Hence according to the standard used by Fisaha et al., (2013) the regeneration status of the forests became in a good condition. The distribution of seedlings, saplings and mature trees shows an inverted ‘J’ shaped distribution patterns; and hence the forest is healthy in those study sites with variable degree of disturbance intensities.

The density and composition of seedlings and saplings indicate the status of regeneration of the common species along the gradients of disturbance. From the total identified 24 woody species about 14 in LDF, and 14 in both MDF and HDF had seedlings of variable densities. On the other hand, nearly 10 woody species did not have seedlings; some of the woody species that lack seedlings include Hypericum revolutum, Hagenia abyssinica, Schefflera abyssinica, Nuxia congeta, Dombeya torrida, and Salix subserrata. Moreover, about 16 species in LDF, 15 in MDF, and 13 in HDF did not have saplings; some of those saplings lacking woody species are Carissa spinarum, Calpurnia aurea, Schefflera abyssinica, Nuxia congeta, Dombeya torrida, and Salix subserrata. About 18 species in LDF, 17 in MDF, and 11 in HDF didn’t have adult trees at all. Among the species that lacks tree/shrub stage are Rumex nervosus, Rhamnus staddo, and Clutia abyssinica (Table 7 in section 4.2.1). The seedling density for each species ranged between 2 and 20,699 individuals’ ha-1. The highest seedling density was exhibited by Myrsine africana, Maytenus arbutifolia, followed by Clutia abyssinica, Juniperus procera, Rhamnus staddo and Erica arborea.

The possible reason for the absence of seedling and sapling for the aforementioned species might be due to habitat preference and microsite conditions; moreover, the ever-increasing human induced disturbance also has an impact on the seedling mortality rate of such species. In addition, species that lacks seedling and sapling like Hypericum revolutum, Dombeya torrida, Salix subserrata and Carissa edulis may be due to the grazing pressure of the area as these species are palatable by herbivores and lowers production capacity of viable seeds.

The good regeneration status may be attributed to opening of canopy due to disturbance and availability of seeds. The high density of seedling and sapling in LDF and MDF forest may be due to the low disturbance intensities. This result were in agreement with Lalfakawma et al. (2009) observed in semi evergreen forest of India and Wassie et al., (2009) in church forests of South Gondar. As it was closely observed species like Juniperus procera and Erica arborea regenerated well in areas having mosses and ferns and abundant sun light. This may be due to the moisture holding and soil improvement capacity of mosses (Teshome, 2007).

In less disturbed forest, the density of matured trees exceeds the density of seedlings and saplings. Which may be due to the closeness of the canopy that hampers regeneration whereas in MDF and HDF the density of seedlings and saplings decreased with increasing intensity of disturbance. This may be due to several reasons; the first is the high restriction effort of grazing in LDF and MDF by the district Office of Agriculture that favors survival of seedling and saplings. The second reason is the opening of regeneration gap by the ongoing selective logging of big trees creates conducive environment for the emergence of seedlings (Wassie et al., 2009). The third reason may be the less demand of seedling and sapling by local communities for their local usages. And another reason may be availability of viable seed in the soil seedbank and presence of seed dispersal agents like herbivores animals. This argument were supported by the findings Senbeta and Teketay (2001).

60 LDF MDF HDF

30 Density/ha 20

0 Seedlings Saplings Trees Life form

Figure 19: The proportion of seedlings, saplings, and adult trees along the gradient of disturbance 4.5. Investigates of the effect of anthropogenic disturbance

During the study, the number of dead and coppiced stumps were recorded so as to see the human pressure on the forest species. Based on the counted data the overall density of Dead stumps /ha were 32 in LDF; 310 in MDF and 468 in HDF; and the density of Coppiced stumps/ha in those sites were 147 in LDF; 305 in MDF, and 81 in HDF sites. As shown in the result (Figure 12) the highest removal of woody species was exhibited in highly disturbed forest site as a result of high intensity of disturbance. The intensity of selective removal varies from species to species. Among those species the most cut out species in the three disturbance levels were Juniperus procera, Myrica salicifolia, Olea europaea, and Erica arbore. From these Juniperus procera were the most logged out species than the other woody species (Figure 20 and 21) as Juniperus procera is a preferred timber species. This result were supported by similar findings of Wassie et al., (2007), in church forests reveal that the most exploited tree species in the higher altitudes were the most dominant species (Juniperus and Olea), which are known for their valuable woods harvested for church building construction and maintenance.

250 LDF MDF HDF

200

150

100 No of dead dead stumps/ha of No 50

Figure 20: Density of dead stumps/ha in along disturbance

120 LDF MDF HDF

100

40 Number of coppiced stump/haNumber coppiced of 20

Figure 21: Density of coppiced stumps/ha along disturbance

Even if Juniperus procera is the most hunted species, it resisted the human pressure by acquiring an ability of stem and root sprouting. So, an important finding of the study is that, unlike other conifer species Juniperus procera has high coppicing capacity. The possible reason for high number of dead and coppiced stump on these forests might be due to the suitability and economic importance of such species for construction and other local uses. As clearly depicted on (Figure 21), high density of coppice was exhibited in moderately disturbed forest; which could be resulted from availability of gap which facilities coppicing ability of those woody species. In fact, there is higher gap in HDF than MDF but, in HDF there is higher grazing and encroachment pressure that affects coppicing.

4.6. The relationship between Species Diversity, Richness, Evenness, and Density along environmental variables

4.6.1.Effect of elevation on species parameters

In the study forest a significance difference of species richness and diversity were exhibited along elevation gradient at (p=0.047) and (p=0.016) respectively. But, there was no significant variation on woody species density and species evenness index. Again, there was no interaction effect of species parameters between life stages with change in elevation gradient (p>0.05). The species diversity and evenness in between life stages significantly differ at (p=0.018) and (p=0.006) following change in elevation. According to the test result and close observation the mean density of species decreases with increasing altitude. These may be due to the decrease in temperature as altitude increases; and the low temperature has a negative effect on the growth and development of woody species (Kebede et al., 2013). There was also high diversity of saplings in each disturbance levels. This may be due to low anthropogenic pressure on saplings. There was also high species richness in mid altitude of the study sites. Similar results were found in Moist Afromontane forest of Wondo Genet revealed that density and diversity of species was significantly varied with variability of elevation (Kebede, and Lemenih et al., 2013). Another finding coined by Wassie et al. (2007), in church forests of South Gondar revealed that, in Ethiopia, at least within the altitudinal range of (1800-3200 m a.s.l.) the trend of linearly decreasing species richness and diversity with altitude is clear. This result agrees with the general concept that as altitude decreases plant species parameters increases (Ghildiyal et al., 2009).

Table 9: Mean table to show the effect of elevation on woody species diversity, evenness, richness and density Distur Low altitude Mid altitude High altitude

bance Densi Dive Species Species Densi Dive Species Species Densi Dive Species Species level life stages ty /ha rsity Evenness Richness ty /ha rsity Evenness Richness ty /ha rsity Evenness Richness LDF Seedling 4164 1.63 0.69 10.64 4036 1.79 0.75 11.00 4036 1.78 0.73 10.00 Sapling 727 1.95 0.86 9.36 470 2.02 0.88 10.00 450 1.90 0.85 10.00 Tree 580 0.99 0.58 6.07 353 0.82 0.37 9.00 350 0.80 0.34 9.00 MDF Seedling 3550 1.50 0.63 10.00 4662 1.63 0.69 10.50 3126 1.62 0.70 10.10 Sapling 591 1.48 0.69 9.00 695 1.54 0.71 9.00 563 1.66 0.80 8.30 Tree 350 0.99 0.68 6.00 490 1.17 0.72 6.00 301 1.18 0.69 6.30 HDF Seedling 2272 1.56 0.81 7.11 1180 1.64 0.79 8.00 1175 1.63 0.78 8.00 Sapling 481 1.25 0.76 5.00 186 1.69 0.87 7.00 160 1.60 0.80 7.00 Tree 240 0.83 0.54 3.33 150 1.21 0.75 5.00 120 0.90 0.65 5.00 Mean 1439 1.35 0.69 7.39 1358 1.50 0.73 8.39 1142 1.45 0.70 8.19 CV 22 7.75 10.21 7.95 23 6.99 9.75 7.00 28 7.22 10.04 7.17 LSD 549 0.18 0.12 1.02 549 0.18 0.12 1.02 549 0.18 0.12 1.02 Significan ce [0.05] * ** * ** * ** * ** * ** * * *

Low altitude=3060-3121m. a.s.l Mid altitude=3123-3185 m a.s.l High altitude (area specific classification)

4.6.2.Effect of Slope on species parameters

Based on the test result in the study forests the change in slope class did not have significant main effect as well as interaction effect on Species diversity, Evenness, species Richness and density (p>0.05) (Table 10). But, quantitatively there was high species diversity and Evenness in slope class one (10-28%) and there was also high density of species in slope class two (29- 47%) followed by slope class one (10-28%) (Table 10). Whereas there was significant main effect on species richness between life stages along slope gradient (p<0.04). As described by Hosseinzadeh et al. (2016), an increase in the slope (especially slopes more than 45 %) yielded a slight reduction in the mean of the species diversity, Richness, evenness and Density; perhaps from adverse growing conditions such as reduced soil depth and fertility.

4.6.3.The effect of Aspect on plant parameters

Based on the test result in the study forests the change in Aspect have significant main effect on Species Richness and density (p=0.038) and (p< 0.018) respectively (Table 11). The recorded results showed that there was high seedling and sapling density on the north east and south East aspect of the study forests (Table 11). This may be due to the light demanding nature of those dominant woody species like Juniperus procera, Clutia abyssinica, Erica arborea, Dodonaea angustifolia and others. Similar findings presented by Hosseinzadeh et al. (2016) in Himalayan forest of Nepal shows that woody species density significantly varied between northeast- and southeast-facing aspect as a result of wider exposure to sun radiation. It has been suggested that different aspect has its own effect in species richness, density, and evenness of tree species. Similar result found in Moist Afromontane forest of Wondo Genet revealed that elevation and aspect was significantly varied with density and species richness (Kebede, and Lemenih et al., 2013).

Table 10: Mean table to show the effect of slope on plant parameters

Disturbance Life stages 10-28% 29-47% 48-65% level D H E R D H E R D H E R LDF Seedling 3735 1.63 0.69 11 4201 1.65 0.70 10.57 6782 1.67 0.75 9 Sapling 746 2.05 0.87 10 685 1.90 0.86 9.14 621 1.69 0.87 7 Tree 488 1.10 0.60 7 591 0.95 0.58 5.71 915 0.26 0.15 5 MDF Seedling 3215 1.61 0.69 10 4108 1.68 0.73 10.14 4753 1.34 0.54 12 Sapling 535 1.59 0.79 8 667 1.63 0.72 9.71 850 1.73 0.83 8 Tree 449 1.09 0.73 6 95 0.23 0.10 1.39 350 1.19 0.66 6 HDF Seedling 1440 1.58 0.88 6 8587 1.29 0.59 9.00 1453 1.61 0.81 7.5 Sapling 489 0.94 0.65 3 940 0.98 0.61 5.00 351 1.52 0.85 6 Tree 467 1.4 0.67 3 144 1.06 0.76 4.00 128 1.28 0.80 5 Mean 1285 1.44 0.73 7.06 2224 1.26 0.63 7.19 1800 1.37 0.70 7 CV 47.7208 27.88 37.48 18.32 48 31.83 43.60 18.02 48 31.83 43.60 18 LSD 2564.59 0.70 0.47 2.24 2565 0.70 0.47 2.24 2565 0.70 0.47 2 Significance [0.05] * * ns * * * ns * * * * *

D=Density/ha, H= Species diversity index, E=Species Evenness index, R=Species richness index

Table 11:The effect of aspect on species parameters

Disturbance NE SE NW SW Overhead (perpendicular) level Life stages D H E R D H E R D H E R D H E R D H E R LDF Seedling 3972 1.69 0.72 10.6 6109 1.5 0.64 11 4007 1.69 0.72 10.8 3742 1.5 0.7 9 4327 1.5 0.63 12 Sapling 730 1.84 0.85 8.9 641 2.2 0.92 11 580 2.02 0.88 10.0 413 2.0 0.8 8 1080 2.1 0.90 10 Tree 773 0.86 0.59 4.6 245 1.3 0.54 11 423 0.98 0.46 8.0 271 1.2 0.6 7 422 1.1 0.64 6 MDF Seedling 4299 1.66 0.71 10.5 4753 1.3 0.54 12 3124 1.76 0.76 10.0 2774 1.5 0.7 9 2749 1.7 0.72 11 Sapling 675 1.74 0.79 9.1 850 1.7 0.83 8 478 1.70 0.86 7.5 491 1.5 0.8 7 438 1.7 0.82 8 Tree 403 1.24 0.70 6.6 350 1.2 0.66 6 390 1.03 0.61 5.5 348 1.1 0.7 6 208 1.3 0.72 6 HDF Seedling 1027 1.39 0.77 6.0 8587 1.3 0.59 9 840 1.51 0.84 6.0 1597 1.6 0.8 8 1587 1.6 0.90 6 Sapling 650 0.86 0.53 5.0 940 1.0 0.61 5 700 1.09 0.99 3.0 334 1.4 0.8 6 217 1.7 0.96 6 Tree 138 1.08 0.78 4.0 144 1.1 0.76 4 120 0.90 0.60 3.0 310 1.1 0.7 4 175 0.9 0.80 1 Mean 1407 1.37 0.71 7.3 2513 1.4 0.68 8.6 1185 1.41 0.75 7.1 1142 1.4 0.7 7 1245 1.5 0.79 7 CV [%] 61 10.80 9.89 14.5 34 10.6 10.45 12.3 72 10.54 9.47 14.9 75 10.4 9.7 15 69 9.7 8.98 14 LSD 1433 0.25 0.12 1.8 1433 0.2 0.12 1.8 1433 0.25 0.12 1.8 1433 0.2 0.1 2 1433 0.2 0.12 2 Significance [0.05] * * * * * * * * * * * * * * * * * * * * NE=North East, SE=South East, NW=North West, SW=South West, Overhead /perpendicular D=Density/ha, H= Species diversity index, E=Species Evenness index, R=Species richness index

5. CONCLUSION AND RECOMMENDATION

5.1. Conclusion

The result of the present study revealed that the species composition of the study forest was largely predominated by those species which are common to single dominated Afromontane forests. Woody species like, Juniperus procera, Myrica salicifolia, Erica arborea, Olea europaea, Maytenus arbutifolia, Osyris quadripartita, Myrsine africana and others were recorded in the area. But, as a result of the ongoing anthropogenic disturbance the growth habit of most of the species especially the moderately and highly disturbed sites became shrubby and stunted. So that, disturbance negatively affects the woody species of the study forest and finally the fate of the existing single dominated afromontane forest will be dominated by those disturbance resistant species and elimination of the most valuable canopy woody species like Juniperus and Olea species.

Species diversity, richness and evenness are important indicators for knowing the status of the forest for future management plan. The finding of the present study revealed that, there was no significant variation of overall species diversity along the gradients of disturbance. As disturbance continues species are developing a coping up mechanism by evenly distributing on the area throughout the whole disturbance levels. This was much pronounced in the highly disturbed site where species diversity is quantitatively larger than the other two sites having a proportionate abundance of disturbance resistant species like Dodonaea angustifolia, Rumex nervosus, and Clutia abyssinica. Coming to the species diversity in between life stages along disturbance levels significant variation were exhibited. As disturbance increases species diversity between life stages became decreased. Based on the result species richness and density of individuals significantly varied and follows a decreasing trend as the intensity of disturbance increases. Therefore, disturbance has a negative effect on species diversity, richness and density of woody species in the study forest.

Basal area provides a better measure of the relative importance of tree species in a given forest. In the present study; Juniperus procera, Myrica salicifolia, Erica arborea, Osyris quadripartita, and Allophylus abyssinica were among the dominant woody species and consecutively ecologically the most important woody species in the area. The Basal area along

64 the gradient of disturbance follows a decreasing trend as intensity of disturbance increases and there was a sharp decrease in basal area from less disturbed to highly disturbed forest. This shows how much anthropogenic disturbance is affecting the productivity of the forest.

Although there were irregularities in between representative woody species; the population structure across disturbance followed the normal inverted ‘J’ shape curve; and these is an indication of healthiness of the forest. But, observing the low density of higher diameter and height classes the forest will face extinction of mother trees in the future. Considering the regeneration status of the forest the density of seedling and sapling exceeds the density of trees and hence the forest has high regenerating capacity due to the opening of regeneration gaps as a result of disturbances. Therefore, one can conclude that to have a good regeneration in a certain forest mild disturbance is important and we can say that mild disturbance has a positive effect on regeneration.

Human induced disturbance like selective logging has a negative impact on growth and productivity of the forest. Observing the actual human pressure by counting dead and coppiced stumps species like Juniperus procera, Myrica salicifolia, Olea europaea, Erica arborea, Allophylus abyssinicus are the most hunted species where the local community most preferred. Therefore, the ongoing anthropogenic disturbance is seriously affecting the most valuable afromontane woody species. Compared to other logged out woody species Juniperus procera were highly affected by selective logging. But, the species resist such impact by having coppicing ability unlike other coniferous species.

Based on the result environmental variables like elevation, topography(slope), and aspect has its own effect on species diversity, Evenness, Richness and Density. Therefore, elevation has a significant main effect on species diversity and evenness. But, has no interaction effect on woody species parameters. High species richness was exhibited on mid altitude of the study areas. In relation to aspect, there was a significant main effect on species richness and plant density. Whereas, slope has nothing to do with change in woody species parameters. To generalize the less disturbed closed forest is unquestionably found in best situation and this shows that disturbance in these forests have a negative effect on the ecological stability as well as productivity of the forests.

5.2. Recommendations

 Although the study forest is being affected by anthropogenic disturbance, Juniperus procera is still the dominant woody species in the study area. So, the regional government as well as the local government should give special attention to include this forest under National forest priority area list for a better management and protection of the forest.  Regarding species diversity, species richness, density and other species parameters decreased as intensity of disturbance increases. This indicates that disturbance has a strong negative effect on species composition, diversity, richness, density and other species parameters. Therefore, there must be an enforcement of forest protection law to limit the human pressure on the forest and to conserve such valuable forest resources.  Because of the ongoing selective logging the basal area of the forest decreases as intensity of disturbance increases. Therefore, there must be serious attention to conserve those species having bigger basal area as these species are ecologically more important than the other species. To do this there must be urgent mobilization of the local communities to build a sense of ownership and formulate forest management plan.  As the population structure shows, there is high density of seedling and sapling; but low number of big mother trees. Unless it is controlled it has a consequence for the future seed availability as those big mother trees are being eliminated. So, there must be strict regulation to limit anthropogenic disturbance and there should be a regulation that which type of species and what size should be logged.  If the forest is conserved and managed it may be used as an alternative provenance, so the regional forest research center and forest enterprise must do in collaboration with the local authority for better conservation and management of the forest.  Finally, I recommend that further study should be conducted in relation to provenance trial in comparison with other Juniperus provenances in terms of germination, early survival and establishment and growth performance and socioeconomic importance of the forest and attitude of the local community towards the forest.

6. REFERENCES

Addo-Fordjour, P., Obeng, S., Anning, A. K., & Addo, M. G. (2009). Floristic composition, structure and natural regeneration in a moist semi-deciduous forest following anthropogenic disturbances and plant invasion. International Journal of Biodiversity and Conservation, 21–37. Retrieved from http://www.academicjournals.org

Adugna, A. (2017). Demography and Health Surveys, 2014(25/08), 1–20. Retrieved from http://www.dhsprogram.com

Afua, E. K. (2011). Patterns of Woody Plant Species Richness, Diversity and Structure Along a Disturbance Gradient in the Atiwa Range Forest Reserve, Eastern Region, Ghana. Thesis.

Alelign, A., Teketay, D., Yemshaw, Y., & Edwards, S. (2007). Diversity and status of regeneration of woody plants on the peninsula of Zegie, northwestern Ethiopia. Tropical Ecology, 37–49.

Amare, D., Mekuria, W., T/wold, T., Belay, B., Teshome, A., Yitaferu, B., … Tegegn, B. (2016). Perception of local community and the willingness to pay to restore church forests: the case of Dera district, northwestern Ethiopia. Forests Trees and Livelihoods, 25(3), 173–186. https://doi.org/10.1080/14728028.2015.1133330

Anne E.Magurran. (1988). Ecological Diversity and its measurement. Princeton, New Jersey: Princeton University press.

Anne E.Magurran. (2004). Measuring biological Diversity. Blackwell Publishing.

Ayalew, A., Bekele, T., & Demissew, S. (2006). The undifferentiated afromontane forest of Denkoro in the central highland of Ethiopia: a floristic and structural analysis. SINET: Ethiopian Journal of Science, 29(1), 45–56. https://doi.org/10.4314/sinet.v29i1.18258

Aynekulu, E., Aerts, R., Denich, M., Negussie, A., Friis, I., Demissew, S., & Boehmer, H. J. (2016). Plant diversity and regeneration in a disturbed isolated dry Afromontane forest in northern Ethiopia. Folia Geobotanica, 115–127.

Aynekulu, E., Denich, M., & Tsegaye, D. (2009). Regeneration Response of Juniperus procera and Olea europaea subsp cuspidata to Exclosure in a Dry Afromontane Forest in Northern Ethiopia. Mountain Research and Development, 143–152. https://doi.org/10.1659/mrd.1076

Azamal Huse, V.K. Mishra, Kemal Semwal, D. K. (2012). Biodiversity Status in Ethiopia and Challenges. In A. C. Pawan Kumar (Ed.), Environmental pollution and biodiversity (first). New Delhi, India: Discovery Publishing House PVT.LTD.

BDoA. (2017). Beyeda District annual report. Dilyibza,Beyeda.

Bekele, T. (1993). Vegetation Ecology of Remnant Afromontane Forests on the Central Plateau of Shewa, Ethiopia. (Erik Sjogren, Ed.). TK Uppsala AB, Uppsala: ACTA

phytogeographica suecica.

Bekele, T. (2000). Plant population ynamics of Dodonaea angustifolia and Olea europaea ssp . cuspidata in ry afromontane forests of Ethiopia.

Benti, D. B. (2011). Floristic Composition, Diversity and Structure of Woody Plant species in Menagesha Suba state Forest, Central Ethiopia. Thesis,Addis Ababa University School of Graduate Studies Plant Biology and Biodiveristy Management Program Unit.

Berhanu, A., Demissew, S., Woldu, Z., & Didita, M. (2017). Woody species composition and structure of Kuandisha afromontane forest fragment in northwestern Ethiopia. Journal of Forestry Research, 343–355. https://doi.org/10.1007/s11676-016-0329-8

Betemariam, E. A. (2011). Forest diversity in fragmented landscapes of northern Ethiopia and implications for conservation . Most, 1–32.

Bishaw, B. (2001). Deforestation and land degredation in the Ethiopian highlands: A Strategy for Physical Recovery. Northeast African Studies,7–25. https://doi.org/10.1353/nas.2005.0014

BoFED. (2016). Amhara National Regional State: Bureaue of Finance and Economic Development GIS_based Regional Map. Bahir dar,Ethiopia.

BoFED. (2017). Anual projected population. Bahir Dar,Ethiopia.

Chandrakar, A. K. (2012). Biodiversity Conservation in India. https://doi.org/10.13140/RG.2.1.1490.3208

Demissew, S. (2014). Overview of the flora of ethiopia and eritrea: the long road to a completion. The Biological Society of Ethiopia, 1–27.

EBI. (2005). Government of the Federal Democratic Republic of Ethiopia Institute of Biodiversity Conservation National Biodiversity. Current, (December), 115.

EBI. (2009). Convention on Biological Diversity (CBD) Ethiopia’s 4th Country Report. Addis Ababa,Ethiopia.

El-juhany, L. I., Aref, I. M., & Al-ghamdi, M. A. (2009). Effects of Different Pretreatments on Seed Germination and Early Establishment of the Seedlings of Juniperus procera Trees, 616–624.

Ellenberg, D. M. and H. (1974). Aims Methods of Vegetation Ecolology.Jhon Wiley and sons, New York, London, Sydney,Toronto.

Etefa, L. (2011). Floristic Composition and Diversity of Herbaceous Flowering Plants in Menagesha Suba State Forest, Oromia Region, Ethiopia. Aau.

EWCA. (2015). Simien Mountains National park: Grazing Pressure Reduction Strategy.

Fisaha, G., Hundera, K., & Dalle, G. (2013). Woody plants’ diversity, structural analysis and regeneration status of Wof Washa natural forest, North-east Ethiopia. African Journal of Ecology, 599–608. https://doi.org/10.1111/aje.12071

Friis, I., & Demissew, S. (2001). Vegetation maps of Ethiopia and Eritrea. A review of existing maps and the need for a new map for the flora of Ethiopia and Eritrea. Biologiske Skrifter, 54(February), 399–439. Retrieved from http://books.google.

Friss, I. (1986). The forest vegetation of Ethiopia,.Symb.Bot. Ups XXVI:2

Gebrehiwot, K. (2003). Ecology and Management of Boswellia papyrifera ( Del .) Hochst . Dry Forests in Tigray , Northern Doctoral Dissertation submitted for the degree of Doctor of Forest Sciences of the Faculty of Forest Sciences and Forest Ecology Georg-August- University.

Gebremicael Fisaha, K. H. and G. D. (2013). Woody plants diversity , structural analysis and regeneration status of Wof Washa Natural Woody plants ’ diversity , structural analysis and regeneration status of Wof Washa natural forest. https://doi.org/10.1111/aje.12071

Gebretsadik, T. (2016). Causes for Biodiversity Loss in Ethiopia : A Review from Conservation Perspective, 32–40.

Gurmessa, F., Soromessa, T., & Kelbessa, E. (2012). Structure and regeneration status of Komto Afromontane moist forest, East Wollega Zone, west Ethiopia. Journal of Forestry Research, 205–216. https://doi.org/10.1007/s11676-012-0242-8

Hill, J. L., & Hill, R. A. (2001). Why are tropical rain forests so species rich? Classifying, reviewing and evaluating theories, 326–354. https://doi.org/10.1177/030913330102500302

Hosseinzadeh, R., Soosani, J., Alijani, V., Khosravi, S., & Karimikia, H. (2016). Diversity of woody plant species and their relationship to physiographic factors in central Zagros forests ( Case study : Journal of Forestry Research, 27(5), 1137–1141. https://doi.org/10.1007/s11676-016-0243-0

IB Friis. (1986). The forest vegetation of Ethiopia. Symbolae Botanicae Upsalienses, 31–47.

Ib Friis, S. D. and P. van B. (2010). Atlas of the Potential Vegetation of Ethiopia. Atlas of the Potential Vegetation of Ethiopia, 65(2), 321–322. https://doi.org/10.1093/aob/mcq242

Inga Hedberg, S. E. (1989). Flora of ethiopia, Pittosporaceae to Araliaceae (Vol. 3). Addis Ababa and Asmara, Ethiopia Uppsala, Sweden: The National Herbarium, BiologyDepartment, Science Faculty, Addis Ababa University, Ethiopia, and The Department of Systematic Botany, Uppsala University, Sweden.

Inga Hedberg, S. E. & S. N. (2003). Flora of Ethiopia and Eritrea, Apiaceae to Dipsacacea (Vol. 4 part 1). Addis Ababa,Ethiopia; Uppsala, Sweden: The National Herbarium, Biology Department, Science Faculty, Addis Ababa University, Ethiopia and The Department of

Systematic Botany Uppsala University, Sweden.

Kebede, M., Yirdaw, E., Luukkanen, O., & Lemenih, M. (2013). Plant community analysis and effect of environmental factors on the diversity of woody species in the moist Afromontane forest of Wondo Genet , South Central Ethiopia, (12), 63–80. https://doi.org/10.2478/biorc-2013-0003

Kent, M. (2012). Vegetation Description and Data Analysis. The Atrium, Southern Gate, Chichester,West Sussex, PO19 8SQ, UK: JohnWiley & Sons, Ltd.

Krebs, C. J. (1989). Ecological Methodology. The Benjamin/Cummings Series in the Life Sciences, 650. https://doi.org/10.1007/s007690000247

Krebs, C. J. (1999). Ecological Methodology. The Benjamin/Cummings Series in the Life Sciences, 650. https://doi.org/10.1007/s007690000247

Kumar, M., Kumar, R., & Kumar, A. (2016). Patterns of diversity and regeneration in unmanaged moist deciduous forests in response to disturbance in Shiwalik Himalayas, India. Journal of Asia-Pacific Biodiversity, 9(2), 144–151. https://doi.org/10.1016/j.japb.2016.01.004

Lalfakawma, Sahoo, U. K., Roy, S., Vanlalhriatpuia, K., & Vanalalhluna, P. C. (2009). Community Composition and Tree Population Semi-Evergreen Forest Stands of North- East India. Applied Ecology and Environmental Research, 7(4), 303–318.

Magurran, A. E. (2004). Measuring biological biversity, 256.

Mamo, N., Mihretu, M., Fekadu, M., Tigabu, M., & Teketay, D. (2006). Variation in seed and germination characteristics among Juniperus procera populations in Ethiopia Variation in seed and germination characteristics among Juniperus procera populations in Ethiopia, (April). https://doi.org/10.1016/j.foreco.2006.01.026

Mamo, Y., Mengesha, G., Shimelis, A., Authority, W. C., Box, P. O., Ababa, A., … Ababa, A. (2015). Woody plant diversity along disturbance gradients in the northern afromontane forests of the Bale Mountains, Ethiopia, 5(3), 3745–3754. International Journal of Development Research Vol.5,Issue, 03,pp.3745-3754

Mckee, J. (2007). Ethiopia country environmental profile ", Addis Ababa.

Mekuria, W. (2007). Vegetation Restoration in Area Closures: The Case of Douga Tembein, Central Tigray, Ethiopia. Energy, (ii), 50–53.

Melese, S. M., & Ayele, B. (2017). Woody plant diversity, structure and regeneration in the Ambo State Forest, South Gondar Zone, Northwest Ethiopia. Journal of Forestry Research, 28(1), 133–144. https://doi.org/10.1007/s11676-016-0280-8

Mohammed, S., & Abraha, B. (2013). Floristic Composition and Structure of Yegof Mountain Forest , South Wollo ,. Ethiopian Journal of Science and Technology, 6(1), 33–45.

Morris, R. J. (2010). Anthropogenic impacts on tropical forest biodiversity : a network structure and ecosystem functioning perspective, 3709–3718. https://doi.org/10.1098/rstb.2010.0273

Mueller-Dombois, D. (2016). Aims Methods of Vegetation Ecolology ch5. ResearchGate,.

Murphy, P. G., & Lugo, A. E. (1986). Ecology of Tropical Dry Forest. Annual Review of Ecology and Systematics, 17(1), 67–88. https://doi.org/10.1146/annurev.es.17.110186.000435

Neelo, J., Teketay, D., Masamba, W., & Kashe, K. (2013). Diversity , Population Structure and Regeneration Status of Woody Species in Dry Woodlands Adjacent to Molapo Farms in Northern Botswana, 3(4), 138–151.

Ordonez, J. C., Luedeling, E., Kindt, R., Tata, H. L., Harja, D., Jamnadass, R., & Noordwijk, M. Van. (2014). Constraints and opportunities for tree diversity management along the forest transition curve to achieve multifunctional agriculture. Environmental Sustainability, 54–60.

Peters, C. M. (1996). The Ecology and Management of Wetlands (May No. 322) (Vol. 1). Washington, D.C.

Pitchairamu, M. and S. (2008). Floristic Inventory and Quantitative Vegetation Analysis of Tropical Dry Deciduous Forest in Piranmalai Forest, Eastern Ghats, Tamil Nadu, India Pitchairamu. Ethnobotanical Leaflets, 2(12), 204–216.

Pohjonen, V., & Pukkala, T. (1992). Juniperus procera Hocht. ex. Endl. in Ethiopian forestry. Forest Ecology and Management, 49(1–2), 75–85. https://doi.org/10.1016/0378- 1127(92)90161.

Senbeta, F. (2006). Biodiversity and ecology of Afromontane rainforests with wild Coffea arabica L. populations in Ethiopia. Ecologyand Development, 49(38), 0–9.

Senbeta, F., & Teketay, D. (2001). Regeneration of indigenous woody species under the canopies of tree plantations in Central Ethiopia. Tropical Ecology, 42(2), 175–185.

Sertse, D., Gailing, O., Eliades, N. G., & Finkeldey, R. (2011). Anthropogenic and natural causes influencing population genetic structure of Juniperus procera Hochst. ex Endl. in the Ethiopian highlands. Genetic Resources and Crop Evolution, 58(6), 849–859. https://doi.org/10.1007/s10722-010-9623-z

Sharew, H. (1994). Regeneration of Juniperus procera and afrocarpus gracilior. in the afromontane forests of Ethiopia.Dissertations.

Sharew, H., Legg, C. J., & Grace, J. (1997). Effects of ground preparation and microenvironment on germination and natural regeneration of Juniperus procera and Afrocarpus gracilior in Ethiopia. Forest Ecology and Management, 93(3), 215–225. https://doi.org/10.1016/S0378-1127(96)03962-X

Sharma, C. M., Suyal, S., Gairola, S., & Ghildiyal, S. K. (2009). Species richness and diversity along an altitudinal gradient in moist temperate forest of Garhwal Himalaya . Journal of American Science, 5(5), 119–128.

Shrestha, K. B., Måren, I. E., Arneberg, E., Sah, J. P., & Ole, R. (2013). Effect of anthropogenic disturbance on plant species diversity in oak forests in Nepal , Central Himalaya, 3732. https://doi.org/10.1080/21513732.2012.749303

Sterck, F., Couralet, C., Nangendo, G., Wassie, A., Sahle, Y., Sass-Klaassen, U., … Bongers, F. (2010). Juniperus procera (Cupressaceae) in Afromontane forests in Ethiopia: From tree growth and population dynamics to sustainable forest use. Degraded Forests in Eastern Africa: Management and Restoration, 291–303. https://doi.org/10.4324/9781849776400

Teketay, D. (2001). Deforestation, Wood Famine, and Environmental Degradation in Ethiopia’s Highland Ecosystems: Urgent Need for Action. Northeast African Studies, 8(1), 53–76. https://doi.org/10.1353/nas.2005.0020

Teketay, D. (2005). Seed and regeneration ecology in dry Afromontane forests of Ethiopia : I . Seed production - population structures. Tropical Ecology, 46(1), 29–44.

Teketay, D. (2005). Seed ecology and regeneration in dry Afromontane forests of Ethiopia. Acta Universitatis Agriculturae Sueciae. Silvestria (Sweden), 46(1), 29–44. https://doi.org/10.1016/j.foreco.2012.09.025

Tesfaye, G., Teketay, D., Fetene, M., & Beck, E. (2009). Regeneration of seven indigenous tree species in a dry Afromontane forest, southern Ethiopia. Tropical Ecology, 52(3), 229–241. https://doi.org/10.1016/j.flora.2008.12.006

Teshome, E. and G. G. (2007). Regeneration Ecology of Erica arborea L.in Simen Mountains National Park and Surroundings, North Western Ethiopia,.

Tilahun, A. (2015). Structure and Regeneration Status of Menagesha Amba Mariam Forest in Central Highlands of Shewa, Ethiopia. Agriculture, Forestry and Fisheries, 4(4), 184. https://doi.org/10.11648/j.aff.20150404.16

Tilman, D. (2000). Causes, consequences and ethics of biodiversity. Nature, 405(May), 208+.

Uniyal, P., Pokhriyal, P., Dasgupta, S., Bhatt, D., & Todaria, N. P. (2010). Plant diversity in two forest types along the disturbance gradient in Dewalgarh Watershed, Garhwal Himalaya. Current Science, 98(7), 938–943.

Vishal, P. G. (2011). A study on the effect of protection on regeneration status, species diversity, structure and biomass pattern of three sites of tropical deciduous forest. College of agriculture Indira Gandhi krishi vishwavidyalaya raipur (C.G.) 492006.

Wagner, S., Fischer, H., & Huth, F. (2010). Canopy effects on vegetation caused by harvesting and regeneration treatments, 17–40. https://doi.org/10.1007/s10342-010-0378-z

Wassie, A. B. (2017). Vegetation Ecology and Conservation Status of Evergreen Afromontane Forest Patches in Awi Zone of Amhara Region , Northwestern Ethiopia. Addis Ababa University.

Wassie, A., Sterck, F. J., & Bongers, F. (2010). Species and structural diversity of church forests in a fragmented Ethiopian Highland landscape. Journal of Vegetation Science, 21(5), 938– 948. https://doi.org/10.1111/j.1654-1103.2010.01202.x

Wassie, A., Sterck, F. J., Teketay, D., & Bongers, F. (2009a). Effects of livestock exclusion on tree regeneration in church forests of Ethiopia. Forest Ecology and Management, 257(3), 765–772. https://doi.org/10.1016/j.foreco.2008.07.032

Wassie, A., Sterck, F. J., Teketay, D., & Bongers, F. (2009b). Tree Regeneration in Church Forests of Ethiopia : Effects of Microsites and Management, 41(1), 110–119.

Wassie, A., Sterck, F. J., Teketay, D., & Bongers, F. (2009c). Tree Regeneration in Church Forests of Ethiopia: Effects of Microsites and Management. Forest Ecology and Management, 257(3), 765–772. https://doi.org/10.1016/j.foreco.2008.07.032

Wassie, A., & Teketay, D. (2006). Soil seed banks in church forests of northern Ethiopia: Implications for the conservation of woody plants. Flora: Morphology, Distribution, Functional Ecology of Plants, 201(1), 32–43. https://doi.org/10.1016/j.flora.2005.04.002

Wassie, A., (2007). Ethiopian Church Forests Opportunities and Challenges for Restauration. https://doi.org/10.1017/S0014479702003046

Woldu, Z., & Mohammed Saleem, M. A. (1999). Grazing induced biodiversity in the highland ecozone of East Africa. Agriculture, Ecosystems and Environment, 79(1), 43–52. https://doi.org/10.1016/S0167-8809(99)00141-3

Worku, G. A. (2006). Population Status and Socio-economic Importance of Gum and Resin Bearing Species in Borana Lowlands, Southern Ethiopia. Addis Ababa. Thesis

Yineger, H., Kelbessa, E., Bekele, T., & Lulekal, E. (2008). Floristic Composition and Structure of the Dry Afromontane Forest At Bale Mountains National Park, Ethiopia. J. Sci, 31(2), 103–120. https://doi.org/10.4314/sinet.v31i2.66551

Zegeye, H., Teketay, D., & Kelbessa, E. (2006). Diversity, regeneration status and socioeconomic importance of the vegetation in the islands of Lake Ziway, south-central Ethiopia. Flora: Morphology, Distribution, Functional Ecology of Plants, 201(6), 483– 498. https://doi.org/10.1016/j.flora.2005.10.006

Zegeye, H., Teketay, D., & Kelbessa, E. (2011). Diversity and regeneration status of woody species in Tara Gedam and Abebaye forests, northwestern Ethiopia. Journal of Forestry Research, 22(3), 315–328. https://doi.org/10.1007/s11676-011-0176-6

Zhang, J. Ã. (2005). Succession analysis of plant communities in abandoned croplands in the Eastern Loess Plateau of China, 458–474. https://doi.org/10.1016/j.jaridenv.2005.03.027

7. APPENDICES

Appendix 1: Photo shows data collection team during field work

Appendix 2: GPS points of the study forest and sample plots

Main forest area delineation LDF_site plots No X_Reading Y_Reading Elevation Plot-No X_Reading Y_Reading Elevation 1 440564 1447801 3016 1 437284 1448265 3053 2 440615 1447100 3065 2 437194 1448214 3069 3 440649 1446672 3046 3 437085 1448151 3093 4 440258 1446368 3058 4 436953 1448105 3104 5 439148 1445601 3055 5 436829 1448063 3112 6 439060 1445565 3058 6 436710 1448243 3121 7 438737 1445705 3087 7 436788 1448340 3093 8 438632 1445761 3104 8 436886 1448377 3072 9 438520 1445801 3072 9 436776 1448529 3078 10 438331 1446197 3067 10 436756 1448438 3092 11 437444 1448243 3061 11 436724 1448340 3061 12 437141 1447412 3071 12 436648 1448266 3125 13 437063 1446794 3039 13 436434 1448331 3122 14 435779 1445550 3117 14 436469 1448426 3082 15 435365 1446050 3152 15 436553 1448497 3060

Main forest area delineation LDF_site plots No X_Reading Y_Reading Elevation Plot-No X_Reading Y_Reading Elevation 16 434833 1446422 3130 MDF_site 17 434736 1447185 3181 1 435976 1446675 3150 18 434732 1447183 3181 2 435869 1446621 3179 19 434788 1447423 3217 3 435735 1446619 3186 20 435002 1447714 3271 4 435552 1446749 3196 21 435111 1448104 3246 5 435640 1446824 3169 22 435355 1448373 3217 6 435709 1446906 3149 23 435728 1448713 3199 7 435635 1447092 3152 24 436011 1449044 3147 8 435524 1447091 3166 25 436607 1449862 3116 9 435426 1447065 3192 26 436727 1449551 3168 10 435326 1447047 3216 27 437094 1449073 3105 11 435271 1447245 3215 28 437474 1449060 3134 12 435163 1447243 3232 29 437692 1448565 3140 13 435060 1447270 3239 30 437835 1448269 3101 14 435082 1447492 3246 31 438284 1448263 3106 15 435187 1447555 3230 32 438420 1448038 3154 16 435234 1447658 3221 33 438660 1448009 3161 HDF_site 34 438771 1447998 3177 1 438438 1446204 3087 35 438792 1447710 3173 2 438557 1446261 3107 36 438795 1447494 3176 3 438660 1446326 3121 37 438602 1447393 3137 4 438778 1446360 3120 38 438489 1447195 3127 5 438938 1446442 3108 39 438629 1446912 3133 6 438930 1446529 3082 40 438747 1446938 3142 7 439090 1446577 3125 41 439092 1446579 3122 8 439018 1446641 3102 42 439533 1446607 3101 43 439802 1446715 3107 44 440282 1446545 3107 45 440360 1447591 3088

Appendix 3: Data collection sheet for diversity

Data collection sheet for trees

District______Kebele______Got______Cluster/site______Plot No______

Agro ecology_____ Land use______Location _X_____Y______Elevation_____ Slope______

soil type _____ Stoniness______Aspect______Land form ______Special feature______No Species Name Local Name DBH at Height Remarks 1.3 m [m] [CM]

Name of Data collector______signature______Date______

Appendix 4: Regeneration data collection sheet

Data collection sheet for Regeneration District______Kebele______Got______Cluster/site______Plot No______Agro ecology______Land use______Location _X______Y______Elevation_____ Slope______soil type ______Stoniness______Aspect______Land form ______Special feature______

Life form Sapling in 2 quadrates Seedling in 5 quadrates of 5m x 5m area of 10m x 10m

NO Species name Local name 1 2 3 4 5 1 2

Name of Data collector’s ______signature______Date______

Appendix 5: ANOVA table for species parameter difference in life form

Sum of Squares df Mean Square F p-value H Between Groups 1.447 2 .723 4.961 .054 Within Groups .875 6 .146 Total 2.322 8 E Between Groups .159 2 .079 39.000 .000 Within Groups .012 6 .002 Total .171 8 D Between Groups 7390288.222 2 3695144.111 6.707 .030 Within Groups 3305772.667 6 550962.111 Total 10696060.889 8 R Between Groups 2.667 2 1.333 .176 .842 Within Groups 45.333 6 7.556 Total 48.000 8

Appendix 6: ANOVA Table for species parameter difference across Disturbance level

Sum of Squares df Mean Square F P_value Species diversity Between .587 2 .294 1.015 .417 index Group Within Groups 1.735 6 .289 Total 2.322 8 Species eveness Between .003 2 .001 .046 .955 index Groups Within Groups .168 6 .028 Total .171 8 Density in hectar Between 382732500.667 2 191366250.333 .548 .605 Groups Within Groups 2095324615.333 6 349220769.222 Total 2478057116.000 8 species richness Between 34.667 2 17.333 7.800 .021 Groups Within Groups 13.333 6 2.222 Total 48.000 8

Appendix 7: MANOVA Table for difference in species parameters across elevation

Dependent Type III Sum Source Variable of Squares df Mean Square F P value DIS Mean density /ha 5753064.093 2 2876532.047 28.603 .000 Species Diversity .108 2 .054 4.870 .041 Species Evenness .028 2 .014 2.737 .124 Species Richness 50.586 2 25.293 73.320 .000 Elv Mean density /ha 423900.769 2 211950.384 2.108 .184 Species diversity .102 2 .051 4.599 .047 Species Evenness .005 2 .002 .474 .639 species richness 5.028 2 2.514 7.288 .016 LF Mean density /ha 44835055.338 2 22417527.669 222.911 .000 Species diversity 2.712 2 1.356 122.261 .000 Species Evenness .207 2 .104 20.383 .001 Species Richness 50.107 2 25.053 72.626 .000 DIS * Mean density /ha 782945.807 4 195736.452 1.946 .196 Elv Species diversity .081 4 .020 1.817 .219 Species Evenness .024 4 .006 1.181 .388 Species Richness 2.483 4 .621 1.799 .222 DIS * LF Mean density /ha 6083446.625 4 1520861.656 15.123 .001 Species diversity .376 4 .094 8.465 .006 Species Evenness .140 4 .035 6.893 .010 Species Richness 2.000 4 .500 1.449 .303 Elv * LF Mean density /ha 239223.436 4 59805.859 .595 .677 Species diversity .018 4 .004 .396 .807 Species Evenness .006 4 .002 .313 .862 Species Richness 1.820 4 .455 1.319 .342 Error Mean density /ha 804537.647 8 100567.206 Species diversity .089 8 .011 Species Evenness .041 8 .005 Species Richness 2.760 8 .345 Corrected Mean density /ha 58922173.715 26 Total Species diversity 3.485 26 Species Evenness .451 26 species Richness 114.782 26 a. R Squared = .986 (Adjusted R Squared = .956) b. R Squared = .975 (Adjusted R Squared = .917) c. R Squared = .910 (Adjusted R Squared = .707) d. R Squared = .976 (Adjusted R Squared = .922)

Appendix 8: MANOVA Table for difference in species parameters across Slope

Dependent Type III Sum of Source Variable Squares df Mean Square F Sig. DIST Species richness 50.963 2 25.481 15.204 .002 Mean Density/ha 1398308.074 2 699154.037 .318 .736 Species diversity .406 2 .203 1.258 .335 Species evenness .004 2 .002 .029 .971 SLOPE Species richness .963 2 .481 .287 .758 Mean Density/ha 3983135.630 2 1991567.815 .906 .442 Species diversity .051 2 .026 .159 .856 Species evenness .021 2 .010 .139 .873 LIF Species richness 112.963 2 56.481 33.702 .000 Mean Density/ha 83504472.519 2 41752236.259 19.003 .001 Species diversity 3.511 2 1.756 10.868 .005 Species evenness .429 2 .214 2.847 .116 DIST * SLOPE Species richness 28.148 4 7.037 4.199 .040 Mean Density/ha 11269957.037 4 2817489.259 1.282 .353 Species diversity .913 4 .228 1.412 .313 Species evenness .233 4 .058 .774 .572 DIST * LIF Species richness 8.148 4 2.037 1.215 .376 Mean Density/ha 910252.148 4 227563.037 .104 .978 Species diversity .484 4 .121 .749 .586 Species evenness .062 4 .016 .206 .928 SLOPE * LIF Species richness 6.815 4 1.704 1.017 .454 Mean Density/ha 8228948.593 4 2057237.148 .936 .490 Species diversity .052 4 .013 .081 .986 Species evenness .025 4 .006 .082 .986 Error Species richness 13.407 8 1.676 Mean Density/ha 17576654.074 8 2197081.759 Species diversity 1.292 8 .162 Corrected Total Species richness 221.407 26 Mean Density/ha 126871728.074 26 Species diversity 6.710 26 Species evenness 1.377 26 a. R Squared = .939 (Adjusted R Squared = .803) b. R Squared = .861 (Adjusted R Squared = .550) c. R Squared = .807 (Adjusted R Squared = .374) d. R Squared = .562 (Adjusted R Squared = -.422)

Appendix 9: MANOVA Table for difference in species parameters across Aspect

Dependent Type III Sum of Source Variable Squares df Mean Square F Sig. DIS Mean Density/ha 3586032.933 2 1793016.467 2.440 .119 Species Diversity .891 2 .446 20.722 .000 Species Evenness .022 2 .011 2.045 .162 Species richness 138.756 2 69.378 62.567 .000 Aspect Mean Density/ha 11949772.311 4 2987443.078 4.065 .018 Species Diversity .115 4 .029 1.336 .299 Species Evenness .059 4 .015 2.727 .066 Species richness 14.576 4 3.644 3.286 .038 LIF Mean Density/ha 96879529.200 2 48439764.600 65.917 .000 Species Diversity 2.630 2 1.315 61.149 .000 Species Evenness .202 2 .101 18.499 .000 Species richness 109.415 2 54.708 49.337 .000 DIS * Mean Density/ha 7072839.289 8 884104.911 1.203 .357 Aspect Species Diversity .311 8 .039 1.805 .150 Species Evenness .057 8 .007 1.314 .305 Species richness 18.440 8 2.305 2.079 .101 DIS * LIF Mean Density/ha 3879064.267 4 969766.067 1.320 .305 Species Diversity .891 4 .223 10.362 .000 Species Evenness .106 4 .026 4.846 .009 Species richness 4.265 4 1.066 .962 .455 Aspect * Mean Density/ha 20641617.689 8 2580202.211 3.511 .016 LIF Species Diversity .291 8 .036 1.690 .177 Species Evenness .124 8 .016 2.846 .036 Species richness 9.403 8 1.175 1.060 .436 Error Mean Density/ha 11757703.511 16 734856.469 Species Diversity .344 16 .022 Species Evenness .087 16 .005 Species richness 17.742 16 1.109 Corrected Mean Density/ha 155766559.200 44 Total Species Diversity 5.473 44 Species Evenness .658 44 Species richness 312.597 44 a. R Squared = .925 (Adjusted R Squared = .792) b. R Squared = .937 (Adjusted R Squared = .827) c. R Squared = .867 (Adjusted R Squared = .635) d. R Squared = .943 (Adjusted R Squared = .844)

UNIVERSITY OF GONDAR POSTGRADUATE DIRECTORATE

APPROVAL SHEET “Woody Species Diversity, Structure and Regeneration Status of Juniperus dominated Dry

Afromontane Forest in Beyeda District, North Gondar Administrative Zone, Amhara National

Regional State: North West Highlands of Ethiopia”.

Submitted by:

______

Name of Student Signature Date

Approved by:

1. ______

Name of Major Advisor Signature Date

2. ______

Name of Co-Advisor Signature Date

3. ______

Name of Chairman, DGC Signature Date

4. ______

Name of Coordinator, CART PGC Signature Date