Vegetation Patterns in Tropical Forests of the Rumpi Hills and Kimbi-Fungom National Park, Cameroon, West-Central Africa

Transcription

1 Vegetation Patterns in Tropical Forests of the Rumpi Hills and Kimbi-Fungom National Park, Cameroon, West-Central Africa Sainge Nsanyi Moses ( ) 9/30/2017 Thesis submitted in fulfillment of the requirements for the degree PhD in Environmental Health In the Faculty of Applied Sciences Department of Environmental and Occupational Studies At Cape Peninsula University of Technology Main Supervisor: Prof Felix Nchu Co-Supervisor 1: Prof Andrew Townsend Peterson Co-Supervisor 2: Dr. David Kenfack

2 DECLARATION I, Sainge Nsanyi Moses, student number , of Cape Peninsula University of Technology, hereby declare that this study for a Doctoral Degree (Plant Systematics and Ecology) in the Department of Environmental and Occupational Studies, Faculty of Applied Sciences, is wholly my research work. Supporting documents from other authors used in this work have been referenced and acknowledged. This piece of document has not been submitted to any academic institution for assessment. The Cape Peninsula University of Technology, Cape Town in the Republic of South Africa is the only institution with the mandate to assess this work for the award of PhD. Signature.. Date: 09/10/ i

3 ABSTRACT Western Cameroon is thought to hold rich biodiversity and diverse vegetation types, and contains two important forest reserves: Rumpi Hills Forest Reserve (RHFR), which is lowland to montane forest located in southwestern Cameroon and Kimbi Fungom National Park (KFNP), which is a semi-deciduous and savanna forest located in northwestern Cameroon. These forest blocks form part of the continental Cameroon Mountains. Thus far, few or limited studies have been undertaken at these two sites to characterise their floristic composition, vegetation patterns, biomass, and carbon stock. Hence, the vegetation of RHFR and KFNP were inventoried from February to November 2015 in detail with the view of describing and understanding the biodiversity and vegetation patterns vis-à-vis elevation gradient. This will enable us to answer the main research questions: How does elevation and vegetation patterns influence species composition, diversity, biomass and carbon in selected wet and dry tropical forests of the Congo Basin? Are plant species equitably distributed among life forms and elevations gradient? What are the extent of land cover changes in RHFR and the KFNP? The objectives of this study were: to characterise vegetation patterns, understand how elevation influences species distributions and diversity, and evaluate biomass and carbon stock per hectare. Furthermore, the study intended to assess the vegetation cover changes over the last few decades in RHFR and KFNP in western Cameroon. RHFR and KFNP were chosen as representative forests because limited ecological studies have been carried out on these forests, and each represented a tropical wet or dry forest, respectively. The floristic composition and vegetation patterns of the reserves were studied in 25 1-ha plots in the RHFR and ha plots in the KFNP spread along elevation gradient and different vegetation types. In each plot, the dbh of trees and lianas of diameter at breast height 10 cm were measured, and dbh of shrubs <10 cm were measured in nested plots of 10 m x 10 m. Remote sensing data (Landsat images) was downloaded from the Global Land Cover Facility (GLCF) and United States Geological Survey (USGS) websites to assess forest cover changes. Forest cover changes over time were compared for both sites. Satellite images from Rumpi Hills (2000 and 2015) and Kimbi Fungom forest (1979 and 2015) were used to compare past and present vegetation (forest cover changes over time). Phytosociological parameters such as basal area, relative density, relative dominance, and relative frequency were used to described forest structure and composition. The statistical program PAST version 2.17 was used to calculate species diversity and richness. Allometric equations were used to evaluate above ground biomass and carbon stock. ii

4 In all, 16,761 trees, shrubs and lianas with dbh 1 cm, representing 71 families, 279 genera, and 617 morphospecies were sampled in the RHFR. Floristic composition of trees (dbh 10 cm) ranged from species with a mean of 117.5±14.8 species ha -1 in lowland forest ( m); to species with a mean of 38.5±3.5 species ha -1 in montane cloud forest ( m) near the summit of Mount Rata were recorded. Two-way Indicator Species Analysis (TWINSPAN) classified the 25 plots into six vegetation types: lowland evergreen rainforest, lowland evergreen rainforest on basalt rocks, midelevation evergreen forest, submontane forest, transitional submontane forest, and montane cloud forest. Detrended correspondence analysis (DCA) of data demonstrated the importance of elevation in shaping vegetation patterns. A total above ground biomass of 9993 t with a mean of 400±100 t ha -1 and carbon 4997 t, mean of 200±50 t ha -1 were recorded in the RHFR. A negative relationship between carbon and elevation (r 2 = 6.18, P < 0.05) was detected, whereas a weak positive relationship between species diversity and carbon (r 2 = 39.13, P < 0.005) were recorded in the RHFR. Based on elevation gradient, we identified four forest types (lowland, mid-elevation, submontane and the montane cloud forest). The density of trees with dhb 10 cm decreased with higher elevation ( m asl). A strong significant negative relationship was noticed between species richness and elevation (r 2 = 75.6%, p < 0.05) across the 25 ha plot in the RHFR. Based on the satellite images, land cover and land use in the Rumpi Hills Forest Reserve for the years 2000 and 2015 showed no significant changes for the different vegetation types. In the Kimbi Fungom National Park, a total of 5551 trees, shrubs and lianas of dbh 1 cm belonging to 46 families, 121 genera, and 201 morphospecies were recorded. Multivariate analysis (TWINSPAN) revealed five vegetation types: primary semi-deciduous forest, secondary forest, mixed vegetation, gallery forest, and woody and grassland savanna. We found an average of tree stems ha -1 (range tree stems ha -1 ) and an average of 43.1 species ha -1 (range species ha -1 ) at dbh 10 cm. The five vegetation types had an average above-ground biomass of t ha -1 ranging from t ha -1, and carbon: 97.3 tc ha -1 ( tc ha -1 ) in the 17 ha for trees with dbh 10 cm. A regression analysis relating above-ground biomass to number of species per hectare across the five vegetation types yielded a significant positive relationship (r 2 = 0.712, P = 0.05). We also noted a weak positive association between elevation and above-ground biomass across vegetation types. In the Kimbi Fungom National Park, the extent of forest cover change was significant in parts of the forest when satellite data of 1979 and 2015 were compared. Extents of forest cover changes were remarkable in the west around Gayama and Munkep, bringing the size from km 2 in 1979 to km 2 in iii

5 These results follows the normal pattern that the wet and dry forest are composed of different vegetation types, and that these vegetation types show impressive variation in terms of structure, species composition, diversity, biomass, carbon, and spatial distribution. It also shows that the RHFR and KFNP are potential sites for the study of climate change and (Reducing Emissions from Deforestation and forest Degradation plus Conservation, sustainable forest management and enhancement of forest carbon stock) REDD+ mechanism. Furthermore, the fine-scale inventory data of species, biomass, and carbon obtained in this study could be useful in developing predictive models for efficient management of tropical rain and dry forests. iv

6 ACKNOWLEDGEMENTS I wish to express my sincere gratitude to my supervisor, Prof. Felix Nchu, co-supervisors, Prof. Andrew Townsend Peterson and Dr. David Kenfack for their intellectual criticism and guidance during the period of this study. Their interest and support on this subject has culminated in laying down a benchmark on the biodiversity, biomass and carbon stock assessment across the Cameroon Mountains, which ordinarily would have been very difficult. Special thanks to Prof. Felix Nchu for constructive criticism, technical aspects, and critical questions on this thesis. Prof. Felix, I am short of words. You are a real mentor. To my wife, Benedicta Jailughe Sainge, I say thank you for your words of encouragement, patience, hope and faith during this study period even when times were hard you still stood by me to propel me to higher heights. To all my lovely kids: Elizabeth, Thelma, Morgan, Carlson, and Vivian, I thank you all for your tolerance, hope and faith. To my lovely mother, Regina Nzofene Nsanyi, and father Samuel Nsanyi, I say thank you for being there when I needed your advice and words of encouragement. To Madame Catherine Metuge and Elizabeth Wotany for their encouragement, assistance, and advice. My brothers Nsanyi Joseph, and Nsanyi Paul for their words of encouragement. My sincere appreciation also goes to Pa Paul Tumenta, Elizabeth Tumenta and Rev. Simon Ngalle Njie for their endless prayers and encouragement towards the completion of this thesis. To my beloved friends, Ngoh Michael Lyonga and Ngoh Benis, Fon Julius and Fon Stella, Ekindi Moudingo and Moudingo Delphine, Peter Tifoubouh and Tifoubouh Confidence and Dr. Martin Sangoh and Sangoh Mercy. Dr. Dilonga Henry Meriki and Dilonga Mirabel, Prof. Chuyong George Bindeh and Bindeh Margaret, Dopbima Lewis Lavai and Dopbima Awah, Ali Amidou, Charles Obiyo, and Dr. Mafany George, thank you all for your inspiration. Special thanks to Dr. Mabel Nechia and Moses Minang for helping in times of need. Special thanks go to Ngoh Michael Lyonga and Sainge Benedicta Jailughe for reading and editing the final document. I will like to give special thanks to Mambo Peter Ekole, Motto Moses, Momene Manfred, Osang Kwelle, Ekpoh Innocent, Joseph Mulango and all who assisted me in the field for their endurance during field work. To all the chiefs and Fons of the villages of Dikome Balue (Chief Sakwe Cyril), Munyange (Chief Jonas Maliki Etongwe), Matamani (Chief Okanyene Christopher Itoe), Kpep (Chief Omah Jama Lucas), Fon of Esu Fondom (His Majesty Fon Kum-a-Ghuo II), and Chief council Tunka-Esu (Kum Olivier Ngah) for their logistics assistance, and the community members at large, I say thank you. v

7 I am grateful to the Government of Cameroon for granting us authorization through the Ministry of Scientific Research and Innovation (MINRESI) and the Ministry of Forestry and Wildlife (MINFOF) to carry out this study. To the Regional Delegate of Forestry and Wildlife for Northwest Cameroon (Mme Grace Mbah), Conservator of Kimbi-Fungom National Park, the Chief of post of Forestry and Wildlife for Fungom, and the Divisional Officers of Ekondo-Titi and Fungom sub divisions, I say thank you for providing logistical assistance during fieldwork. I am grateful to the JRS Biodiversity Foundation through Prof. A. Townsend Peterson for funding the training activities that led to this fruitful collaboration. I am indebted to Prof. Townsend of Kansas University, USA and Prof. Mathias Mwatterts of Gottengen University, Germany for inspiring and encouraging me to study the biodiversity of the Cameroon Mountains. My gratitude equally goes to the Rufford Small Grant Foundation (RSG) for supporting this study financially through the first booster grant (No B). Complementary grants were received from the Tropical Plant Exploration Group (TroPEG) Cameroon and the JRS Biodiversity Foundation through Prof. A. Townsend Peterson. I expressed my gratitude to colleagues at Tropical Plant Exploration Group (TroPEG), Cameroon. Above all, I thank the almighty God for giving me the strength, courage, knowledge, and showing me the way to carry out this work. Finally, I hope that this piece of work will add value to existing and future studies on the biodiversity, ecology, above-ground biomass, carbon stock and the payment for environmental services to Cameroon and the Congo Basin countries. Thanks, Merci, Miebucho, SA vi

8 DEDICATION This piece of work is dedicated to my wife Benedicta Jailughe Sainge, my kids: Elizabeth, Thelma, Morgan and Carlson for their patience, courage, love, hope and faith exhibited during my period of study. vii

9 TABLE OF CONTENTS DECLARATION...i ABSTRACT...ii ACKNOWLEDGEMENTS....v DEDICATION..... vii TABLE OF CONTENTS...viii ABBREVIATIONS AND ACRONYMS....xiv CHAPTER ONE. 1 1 Introduction Outline of the Thesis Literature Review Forest and forest types Definition of Forest Types of Forest Tropical forests Tropical rainforest Tropical dry forest Tropical montane forest Tropical forests and Ecosystem services Ecology of tropical forests Threats facing tropical forests Climate change and tropical forest The Cameroon Mountains: Rumpi Hills Forest Reserve (RHFR) and Kimbi Fungom National Park Climate and hydrology in the Rumpi Hills Forest Reserve (RHFR) and Kimbi Fungom National Park (KFNP) Elevation in the Rumpi Hills forest reserve (RHFR) and Kimbi Fungom National Park...13 viii

10 1.4 Significance of Study Rationale of Study Problem Statement Research Aim and Objectives Specific Research Objectives Field Methods Data Analysis..18 CHAPTER TWO Introduction Material and Methods Results Discussion. 33 CHAPTER THREE Introduction Materials and Methods Study area Field sampling Taxonomy and plant identification Data analysis Results Species accumulation curve and inventory completeness Composition and Floristic Structure Multivariate analysis Vegetation patterns Lowland evergreen rainforest rich in Oubanguia alata Lowland evergreen rainforest on basalt rocks rich in Crateranthus talbotii Mid-elevation evergreen forest rich in Strombosia grandifolia Submontane forest rich in Tabernaemontana ventricosa Transitional submontane forest rich in Trema orientalis Montane cloud forest rich in Carapa oreophila Discussion Vegetation Patterns and Floristic Composition..53 ix

11 3.4.2 Elevation Conservation Implications...54 CHAPTER FOUR Introduction Materials and Methods Study sites Field sampling Forest type classification Data analysis Results Plant density and basal area by forest type Elevational patterns Species richness within families Above-ground biomass and carbon estimation Discussion...91 CHAPTER FIVE Introduction Materials and Methods Study Site Field Sampling Data analysis Results Composition and Diversity Basal Area Forest Structure Classification and Vegetation Patterns Above-Ground Biomass and Carbon Discussion CHAPTER SIX Introduction Materials and Methods x

12 6.2.1 Study Area Image analysis Results Land cover classification Past and present vegetation: Forest cover changes in the wet Rumpi Hills Forest Reserve Past and present vegetation: Forest cover changes in the dry semi-deciduous and savanna Kimbi Fungom forest Discussion CHAPTER SEVEN General Discussions Recommendations REFERENCES LIST OF FIGURES Figure 1.1 Cameroon Volcanic Line.21 Figure 1.2 Climate representations over ten years ( )at Rumpi Hills Forest Reserve, Cameroon..22 Figure 1.3 Rainfall Representations for for Kimbi Fungom National Park, Cameroon..23 Figure 1.4 Rumpi Hills Forest Reserve Cameroon...24 Figure 1.5 Vegetation types based on present study in the Kimbi Fungom National Park, Cameroon...25 Figure 1.6 Vegetation types based on Letouzey (1985) in the Kimbi Fungom National Park, Cameroon Figure 1.7 Locations of Sample Plots in the Kimbi Fungom National Park and the Rumpi Hills Forest Reserve, Cameroon 27 Figure 1.8 A & B. Field designs. A. Transect 500 m long (orange) in the middle of two plots, with each plot measuring 500 x 20 m. B x 20 m at high elevation representing 1 ha. The red asterets at quadrat 1, 6.11,16 and 21 represents where nested plots of 10x10 m were established Figure 2.1 Diagrammatic representation of the Cameroon Mountain chain, extending from islands in the Gulf of Guinea to the coastal Mt Cameroon, and inland along the Cameroon-Nigeria border..36 Figure 2.2. Summary of highland areas in the continental portion of the Cameroon Mountains region. Areas above 1000 m elevation are shown in darker gray shading. Areas above 1600 m are shown colored red and outlined in black. The more important such highland areas are labeled for reference to the text..37 Figure 2.3. Map of the continental portion of the Cameroon Mountains region, showing points sampled in the xi

13 YA collections in blue. The different highland areas are indicated as well-sampled (dark red), some sampling (medium red), some information (light red) or no information (gray).38 Figure 2.4. Summary of gaps in coverage in terms of inventories of plants in the Cameroon Mountains region. White indicates well-known sites; darker shades of blue indicate greater distance (geographic space) or difference (environmental space) from well-surveyed sites. Left-hand column is the southwestern part of the chain; right-hand column is the northeastern part of the chain Figure 3.1 Vegetation stratification and permanent sample plot locations across the Rumpi Hills Forest Reserve, Cameroon 55 Figure 3.2. (A) Species area curve for observed species richness (S Obs), and estimated species richness (Chao 2 mean), (B) Inventory completeness for observed species richness (S Obs) and (C) a strong linear relationship between Estimated and observed species in the Rumpi Hills Forest Reserve, Cameroon.57 Figure 3.3 TWINSPAN dendrogram of 311 species of vascular plants with DBH 10 cm in 25 1-ha plots in the Rumpi Hills Forest Reserve, Cameroon..58 Figure 3.4 Floristic Detrended Correspondance Analysis (DCA) with 25 plots and 6 vegetation type from TWINSPAN analysis tree species data from the Rumpi Hills Forest Reserve, Cameroon 59 Figure 3.5 Association between elevation and numbers of species of trees of dbh >10 cm recorded in 25 1-ha plots in the Rumpi Hills Forest Reserve, Cameroon. The relationship has a slope of 78.3%, which is significantly different from a slope of zero (P < 0.005) 60 Figure 4,1. Sample points and four Plant Communities at the Rumpi Hills Forest Reserve, Cameroon..94 Figure 4.2. (A) Variation of tree density and Basal area, (B) Mean tree density/ha, Mean basal area m 2 /ha in four forest types in the Rumpi Hills Forest Reserve, Cameroon 95 Figure 4.3 Regression analyses showing a strong significant negative relationship between species richness and elevation (r 2 = 0.756, p < 0.05;) across 25 ha plot in the Rumpi Hills Forest Reserve, Cameroon..96 Figure 4.4 Correlation of number of lianas and number of large trees 60 cm (r 2 = , P <0.05) 97 Figure 4.5. Association between Carbon (tc/ha) and Elevation (m) of trees with dbh >10 cm recorded in ha plots in the Rumpi Hills Forest Reserve, Cameroon. The relationship has a slope of 6.1%, which is significantly different from a slope of zero (P < 0.05), y=-0.11x Figure 4.6. Association between Carbon (tc/ha) and number of species/ha with dbh >10 cm recorded in ha plots in the Rumpi Hills Forest Reserve, Cameroon...99 Figure 4.7 Correspondence analysis reflecting high tree density, number of species, carbon in lowland and Montane forest and low tree density, number of species, and carbon in mid-elevation and submontane forest of the Rumpi Hills forest Reserve in Cameroon..100 xii

14 Figure 5.1. Vegetation types and sample locations across the Kimbi-Fungom National Park Figure 5.2. TWINSPAN dendrogram of 120 species of vascular plants with dbh 10 cm in 17 plots of 1-ha each in the Kimbi-Fungom National Park, Cameroon Figure 5.3. Floristic Canonical Correspondence Analysis (CCA) with 17 1-ha sample plots and 5 vegetation types from TWINSPAN analysis in the Kimbi-Fungom National Park, Cameroon Figure 5.4. Association between above-ground biomass and numbers of species of trees of dbh >10 cm recorded in 17 1-ha plots at the Kimbi-Fungom National Park, Cameroon 143 Figure 5.5. Weak associations between carbon and elevation across different plots for tree species of dbh 10 cm recorded in 17 1-ha plots at the Kimbi-Fungom National Park, Cameroon..144 Figure 6.1A. 30 m Landsat ETM+ (Enhanced Thematic Mapper) of 2000, B. 30 m Landsat 8 of 2015 for Rumpi Hills Forest Reserve, Cameroon..167 Figure 6.2. Surface area and spatial extentof forest cover types in the Rumpi Hills Forest Reserve, Cameroon from 2000 and Figure 6.3. Forest cover classes from data collected in 2000 and in 2015 in the Rumpi Hills Forest Reserve, Cameroon..168 Figure 6.4a. 60 m Landsat MSS (Multispectral Scanner) of 1979, b. 30 m Landsat 8 of 2015 for Kimbi Fungom National Park, Cameroon Figure 6.5. Surface area and spatial extentof different cover typesd in the Kimbi Fungom National Park, Cameroon between 1979 (top) and 2015 (bottom) Figure 6.6. Forest covers change for thirty-seven years ( ) in the Kimbi Fungom National Park, Cameroon LIST OF TABLES Table 2.1. Highland areas in the continental portion of the Cameroon Mountains region with focus in Cameroon Table 3.1 Summary of number of hectares sampled, mean number of species, mean tree density, and mean basal area of trees 10 cm dbh recorded in each vegetation type in the Rumpi Hills Forest Reserve, Cameroon 61 Table 3.2 Summary of Detrended Correspondence Analysis (DCA or DECORANA) of 25 1-ha plots in the xiii

15 Rumpi Hills Forest Reserve, Cameroon Table 3.3 Six vegetation types generated from TWINSPAN with vegetation code, vegetation types and elevational range in the Rumpi Hills Forest Reserve, Cameroon. 62 Table 3.4 Monte Carlo test results showing eigenvalue, species elevational correlation, correlation, and biplot scores obtained following field sampling of the Rumpi Hills Forest Reserve, Cameroon...62 Table 3.5 Permanent sample plot locations in the Rumpi Hills Forest Reserve, Cameroon Table 4.1. Tree diversity, above ground biomass (AGB), and carbon in lowland, mid-elevation, submontane, and montane cloud forest. BA (basal area), N (tree density), SR (species richness), SW (Shannon-Weiner index) 101 Table 4.2. Summary of total number of individuals, basal area, number of species, and Shannon diversity for the different life forms recorded in four forest types at various size classes in the RHFR Cameroon 102 Table 4.3. Summary of Biomass, and Carbon in four forest types across Rumpi Hills Forest Reserve, Cameroon Table 4.4 Summary of 52 species with biomass, and carbon in the Rumpi Hills Forest Reserve, Cameroon..105 Table 5.1. Different vegetation cover types and corresponding numbers of species, stems, and mean Shannon diversity index in Kimbi-Fungom National Park forest, Cameroon.145 Table 5.2. Eleven plant families in terms of highest above ground biomass, and carbon stock in the Kimbi- Fungom National Park, Cameroon Table 5.3. Ten species with highest above-ground biomass, and carbon in 17 1-ha plots in Kimbi-Fungom National Park, Cameroon.146 Table 5.4. Mean above-ground biomass, carbon, basal area, and species richness, across five vegetation types in Kimbi-Fungom National Park, Cameroon Table 5.5. Vegetation types and corresponding mean tree height (in m; range in parentheses) in sample plots in Kimbi-Fungom National Park, Cameroon. 147 Table 6.1. Accuracy Assessment for Kimbi Fungom National Park, Cameroon..172 Table 6.2. Forest covers Change (2000 to 2015) Rumpi Hills Forest Reserve, Cameroon Table 6.3. Forest cover Changes (1979 to 2015) Kimbi Fungom National Park, Cameroon LIST OF APPENDICES Appendix 4.1. Species occurrence in the different forest types in the Rumpi Hills Forest Reserve, Cameroon Appendix 4.2. Summary table of distribution of biomass, and carbon Sequestered per hectare using Allometric xiv

16 models in four Tropical forest types in the Rumpi Hills Forest Reserve, Cameroon 128 Appendix 5.1. Sampling plot locations, number of species/ha and individual trees/ha in the Kimbi-Fungom National Park, Cameroon Appendix 5.2. Plant species recorded in observational efforts (i.e., outside of sampling plots) in the Kimbi- Fungom National Park, Cameroon Appendix 5.3. Summary of density, number of species, above-ground biomass, carbon, and Basal area in 17 ha plot across five vegetation types in Kimbi-Fungom National Park, Cameroon Appendix 5.4 Species list and abundance in the Kimbi-Fungom National Park, Cameroon 152 AGB ANOVA APG BA C CBD CBFP CCA CDC CDM CIFOR cm CO 2 COMIFAC COP CVL DAK dbh DCA DEM E Elev. (m) ENVI ETM FAO FD GESP GF ABBREVIATIONS AND ACRONYMS Above Ground Biomass Analysis of Variances Angiosperm Phylogeny Group Basal Area Carbon Convention of Biological Diversity Congo Basin Forest Partnership Canonical Correspondence Analysis Cameroon Development Cooperation Clean Development Mechanism Center for International Forestry Research Centimeter Carbon dioxide Central African Forest Commission Conference of Parties Cameroon Volcanic Line Digital Accessible Knowledge Diameter at Breast Height Detrended Correspondence Analysis Digital Elevation Model Pielous evenness Elevation Environment for Visualizing Images Enhanced Thematic Mapper Food and Agriculture Organization Flat Dry forest Global Environmental Sustainability Programme Gallery forest xv

17 GLCF Global Land Cover Facility GMTED Global Multi-resolution Terrain Elevation Data GOFC-GOLD Global Observation of Forest Cover and Land Dynamics GPFLR Global Partnership on Forest Landscape Restoration GtCyr -1 Gegatonnes of carbon per year G/WS Grassland/Woody savanna H Shannon-Weiner index of diversity ha Hectare ITTO International Timber Trade Organization IPCC Intergovernmental Panel on Climate Change IUCN International Union for Conservation of Nature KFNP Kimbi Fungom National Park km 2 Square kilometer L Lake Lidar Light Detention and Ranging laser ln Natural logarithm LULUCF Land-Use, Land-Use Change and Forestry m Meter MEA Millennium Ecosystem Assessment MINEF Ministry of the Environment and Forestry MINFOF Ministry of Forestry and Wildlife ms Microsoft MSS Multiple Spectral Scanner Mt Mountain MV Mixed Vegetation NBSAP National Biodiversity Strategy and Action Plan S(obs) Observed Number of Species PAST Paleontological Statistics Software package for education and data analysis PC-ORD DOS Program that performs Multivariate Analysis of Ecological Data Pg/yr Picogram per year PL Plateau PLC Public Limited Company PRSP Poverty Reduction Strategy Paper PSF Primary Semi-deciduous Forest PVC Polyvinyl Chloride QGIS Quantum Geographic Information System REDD Reducing Emissions from Deforestation and forest Degradation Reducing Emissions from Deforestation and forest Degradation plus Conservation, REDD+ sustainable forest management and enhancement of forest carbon stock Radar Radio Detention and Ranging RHFR Rumpi Hills Forest Reserve xvi

18 RSG Rufford Small Grant Foundation S Number of species SAR Synthetic Aperture Radar SD Stem Density SF Secondary Forest SL Slope SW Swamps t Tonnes TD Tree Density TroPEG Tropical Plant Exploration Group TWINSPAN Two-way Indicator Species Analysis UN United Nations UNEP United Nations Environmental Programme UNFCCC United Nations Framework Convention on Climate Change US United States USGS United States Geological Survey Vegtype Vegetation Types WD Wood Density WRI World Resource Institute YA National herbarium of Cameroon DECLARATION...i ABSTRACT.....ii ACKNOWLEDGEMENTS.....v DEDICATION....vii TABLE OF CONTENTS..viii ABBREVIATIONS AND ACRONYMS.... xiv CHAPTER ONE Introduction Outline of the Thesis Literature Review The Cameroon Mountains: Rumpi Hills Forest Reserve (RHFR) and Kimbi Fungom National Park..10 xvii

19 1.4 Significance of Study Rationale of Study Problem Statement Research Aim and Objectives Field Methods Data Analysis..18 CHAPTER TWO Introduction Material and Methods Results Discussion. 33 CHAPTER THREE Introduction Materials and Methods Results Discussion CHAPTER FOUR Introduction Materials and Methods Results Discussion...91 CHAPTER FIVE Introduction Materials and Methods Results Discussion CHAPTER SIX Introduction Materials and Methods Results Discussion CHAPTER SEVEN xviii

20 7.1 General Discussions Recommendations REFERENCES LIST OF FIGURES Figure 1.1 Cameroon Volcanic Line 21 Figure 1.2 Climate representations over ten years ( )at Rumpi Hills Forest Reserve, Cameroon..22 Figure 1.3 Rainfall Representations for for Kimbi Fungom National Park, Cameroon 23 Figure 1.4 Rumpi Hills Forest Reserve Cameroon.24 Figure 1.5 Vegetation types based on present study in the Kimbi Fungom National Park, Cameroon...25 Figure 1.6 Vegetation types based on Letouzey (1985) in the Kimbi Fungom National Park, Cameroon Figure 1.7 Locations of Sample Plots in the Kimbi Fungom National Park and the Rumpi Hills Forest Reserve, Cameroon.27 Figure 1.8 A & B. Field designs. A. Transect 500 m long (orange) in the middle of two plots, with each plot measuring 500 x 20 m. B x 20 m at high elevation representing 1 ha. The red asterets at quadrat 1, 6.11,16 and 21 represents where nested plots of 10x10 m were established Figure 2.1 Diagrammatic representation of the Cameroon Mountain chain, extending from islands in the Gulf of Guinea to the coastal Mt Cameroon, and inland along the Cameroon-Nigeria border Figure 2.2. Summary of highland areas in the continental portion of the Cameroon Mountains region. Areas above 1000 m elevation are shown in darker gray shading. Areas above 1600 m are shown colored red and outlined in black. The more important such highland areas are labeled for reference to the text 37 Figure 2.3. Map of the continental portion of the Cameroon Mountains region, showing points sampled in the YA collections in blue. The different highland areas are indicated as well-sampled (dark red), some sampling (medium red), some information (light red) or no information (gray).38 Figure 2.4. Summary of gaps in coverage in terms of inventories of plants in the Cameroon Mountains region. White indicates well-known sites; darker shades of blue indicate greater distance (geographic space) or difference (environmental space) from well-surveyed sites. Left-hand column is the southwestern part of the chain; right-hand column is the northeastern part of the chain Figure 3.1 Vegetation stratification and permanent sample plot locations across the Rumpi Hills Forest Reserve, Cameroon 55 Figure 3.2. (A) Species area curve for observed species richness (S Obs), and estimated species richness (Chao xix

21 2 mean), (B) Inventory completeness for observed species richness (S Obs) and (C) a strong linear relationship between Estimated and observed species in the Rumpi Hills Forest Reserve, Cameroon.57 Figure 3.3 TWINSPAN dendrogram of 311 species of vascular plants with DBH 10 cm in 25 1-ha plots in the Rumpi Hills Forest Reserve, Cameroon..58 Figure 3.4 Floristic Detrended Correspondance Analysis (DCA) with 25 plots and 6 vegetation type from TWINSPAN analysis tree species data from the Rumpi Hills Forest Reserve, Cameroon.. 59 Figure 3.5 Association between elevation and numbers of species of trees of dbh >10 cm recorded in 25 1-ha plots in the Rumpi Hills Forest Reserve, Cameroon. The relationship has a slope of 78.3%, which is significantly different from a slope of zero (P < 0.005) 60 Figure 4,1. Sample points and four Plant Communities at the Rumpi Hills Forest Reserve, Cameroon. 94 Figure 4.2. (A) Variation of tree density and Basal area, (B) Mean tree density/ha, Mean basal area m 2 /ha in four forest types in the Rumpi Hills Forest Reserve, Cameroon..95 Figure 4.3 Regression analyses showing a strong significant negative relationship between species richness and elevation (r 2 = 0.756, p < 0.05;) across 25 ha plot in the Rumpi Hills Forest Reserve, Cameroon...96 Figure 4.4 Correlation of number of lianas and number of large trees 60 cm (r 2 = , P <0.05) 97 Figure 4.5. Association between Carbon (tc/ha) and Elevation (m) of trees with dbh >10 cm recorded in ha plots in the Rumpi Hills Forest Reserve, Cameroon. The relationship has a slope of 6.1%, which is significantly different from a slope of zero (P < 0.05), y=-0.11x Figure 4.6. Association between Carbon (tc/ha) and number of species/ha with dbh >10 cm recorded in ha plots in the Rumpi Hills Forest Reserve, Cameroon Figure 4.7 Correspondence analysis reflecting high tree density, number of species, carbon in lowland and Montane forest and low tree density, number of species, and carbon in mid-elevation and submontane forest of the Rumpi Hills forest Reserve in Cameroon..100 Figure 5.1. Vegetation types and sample locations across the Kimbi-Fungom National Park Figure 5.2. TWINSPAN dendrogram of 120 species of vascular plants with dbh 10 cm in 17 plots of 1-ha each in the Kimbi-Fungom National Park, Cameroon Figure 5.3. Floristic Canonical Correspondence Analysis (CCA) with 17 1-ha sample plots and 5 vegetation types from TWINSPAN analysis in the Kimbi-Fungom National Park, Cameroon Figure 5.4. Association between above-ground biomass and numbers of species of trees of dbh >10 cm recorded in 17 1-ha plots at the Kimbi-Fungom National Park, Cameroon 142 Figure 5.5. Weak associations between carbon and elevation across different plots for tree species of dbh 10 cm recorded in 17 1-ha plots at the Kimbi-Fungom National Park, Cameroon..143 xx

22 Figure 6.1A. 30 m Landsat ETM+ (Enhanced Thematic Mapper) of 2000, B. 30 m Landsat 8 of 2015 for Rumpi Hills Forest Reserve, Cameroon..167 Figure 6.2. Surface area and spatial extentof forest cover types in the Rumpi Hills Forest Reserve, Cameroon from 2000 and Figure 6.3. Forest cover classes from data collected in 2000 and in 2015 in the Rumpi Hills Forest Reserve, Cameroon..168 Figure 6.4a. 60 m Landsat MSS (Multispectral Scanner) of 1979, b. 30 m Landsat 8 of 2015 for Kimbi Fungom National Park, Cameroon Figure 6.5. Surface area and spatial extentof different cover typesd in the Kimbi Fungom National Park, Cameroon between 1979 (top) and 2015 (bottom) Figure 6.6. Forest covers change for thirty-seven years ( ) in the Kimbi Fungom National Park, Cameroon LIST OF TABLES Table 2.1. Highland areas in the continental portion of the Cameroon Mountains region with focus in Cameroon Table 3.1 Summary of number of hectares sampled, mean number of species, mean tree density, and mean basal area of trees 10 cm dbh recorded in each vegetation type in the Rumpi Hills Forest Reserve, Cameroon 61 Table 3.2 Summary of Detrended Correspondence Analysis (DCA or DECORANA) of 25 1-ha plots in the Rumpi Hills Forest Reserve, Cameroon Table 3.3 Six vegetation types generated from TWINSPAN with vegetation code, vegetation types and elevational range in the Rumpi Hills Forest Reserve, Cameroon. 62 Table 3.4 Monte Carlo test results showing eigenvalue, species elevational correlation, correlation, and biplot scores obtained following field sampling of the Rumpi Hills Forest Reserve, Cameroon...62 Table 3.5 Permanent sample plot locations in the Rumpi Hills Forest Reserve, Cameroon Table 4.1. Tree diversity, above ground biomass (AGB), and carbon in lowland, mid-elevation, submontane, and montane cloud forest. BA (basal area), N (tree density), SR (species richness), SW (Shannon-Weiner index) 101 xxi

23 Table 4.2. Summary of total number of individuals, basal area, number of species, and Shannon diversity for the different life forms recorded in four forest types at various size classes in the RHFR Cameroon 102 Table 4.3. Summary of Biomass, and Carbon in four forest types across Rumpi Hills Forest Reserve, Cameroon Table 4.4 Summary of 52 species with biomass, and carbon in the Rumpi Hills Forest Reserve, Cameroon..105 Table 5.1. Different vegetation cover types and corresponding numbers of species, stems, and mean Shannon diversity index in Kimbi-Fungom National Park forest, Cameroon.144 Table 5.2. Eleven plant families in terms of highest above ground biomass, and carbon stock in the Kimbi- Fungom National Park, Cameroon Table 5.3. Ten species with highest above-ground biomass, and carbon in 17 1-ha plots in Kimbi-Fungom National Park, Cameroon.145 Table 5.4. Mean above-ground biomass, carbon, basal area, and species richness, across five vegetation types in Kimbi-Fungom National Park, Cameroon Table 5.5. Vegetation types and corresponding mean tree height (in m; range in parentheses) in sample plots in Kimbi-Fungom National Park, Cameroon. 146 Table 6.1. Accuracy Assessment for Kimbi Fungom National Park, Cameroon..172 Table 6.2. Forest covers Change (2000 to 2015) Rumpi Hills Forest Reserve, Cameroon Table 6.3. Forest cover Changes (1979 to 2015) Kimbi Fungom National Park, Cameroon LIST OF APPENDICES Appendix 4.1. Species occurrence in the different forest types in the Rumpi Hills Forest Reserve, Cameroon Appendix 4.2. Summary table of distribution of biomass, and carbon Sequestered per hectare using Allometric models in four Tropical forest types in the Rumpi Hills Forest Reserve, Cameroon 127 Appendix 5.1. Sampling plot locations, number of species/ha and individual trees/ha in the Kimbi-Fungom National Park, Cameroon Appendix 5.2. Plant species recorded in observational efforts (i.e., outside of sampling plots) in the Kimbi- Fungom National Park, Cameroon Appendix 5.3. Summary of density, number of species, above-ground biomass, carbon, and Basal area in 17 ha plot across five vegetation types in Kimbi-Fungom National Park, Cameroon Appendix 5.4 Species list and abundance in the Kimbi-Fungom National Park, Cameroon 152 xxii

24 AGB ANOVA APG BA C CBD CBFP CCA CDC CDM CIFOR ABBREVIATIONS AND ACRONYMS Above Ground Biomass Analysis of Variances Angiosperm Phylogeny Group Basal Area Carbon Convention of Biological Diversity Congo Basin Forest Partnership Canonical Correspondence Analysis Cameroon Development Cooperation Clean Development Mechanism Center for International Forestry Research cm Centimeter CO 2 Carbon dioxide COMIFAC Central African Forest Commission COP Conference of Parties CVL Cameroon Volcanic Line DAK Digital Accessible Knowledge dbh Diameter at Breast Height DCA Detrended Correspondence Analysis DEM Digital Elevation Model E Pielous evenness Elev. (m) Elevation ENVI Environment for Visualizing Images ETM Enhanced Thematic Mapper FAO Food and Agriculture Organization FD Flat Dry forest GESP Global Environmental Sustainability Programme GF Gallery forest GLCF Global Land Cover Facility GMTED Global Multi-resolution Terrain Elevation Data GOFC-GOLD Global Observation of Forest Cover and Land Dynamics GPFLR Global Partnership on Forest Landscape Restoration GtCyr -1 Gegatonnes of carbon per year G/WS Grassland/Woody savanna H Shannon-Weiner index of diversity ha Hectare ITTO International Timber Trade Organization IPCC Intergovernmental Panel on Climate Change xxiii

25 IUCN KFNP km 2 L Lidar ln LULUCF m MEA MINEF MINFOF ms MSS Mt MV NBSAP S(obs) PAST PC-ORD Pg/yr PL PLC PRSP PSF PVC QGIS REDD REDD+ Radar RHFR RSG S SAR SD SF SL SW t TD TroPEG International Union for Conservation of Nature Kimbi Fungom National Park Square kilometer Lake Light Detention and Ranging laser Natural logarithm Land-Use, Land-Use Change and Forestry Meter Millennium Ecosystem Assessment Ministry of the Environment and Forestry Ministry of Forestry and Wildlife Microsoft Multiple Spectral Scanner Mountain Mixed Vegetation National Biodiversity Strategy and Action Plan Observed Number of Species Paleontological Statistics Software package for education and data analysis DOS Program that performs Multivariate Analysis of Ecological Data Picogram per year Plateau Public Limited Company Poverty Reduction Strategy Paper Primary Semi-deciduous Forest Polyvinyl Chloride Quantum Geographic Information System Reducing Emissions from Deforestation and forest Degradation Reducing Emissions from Deforestation and forest Degradation plus Conservation, sustainable forest management and enhancement of forest carbon stock Radio Detention and Ranging Rumpi Hills Forest Reserve Rufford Small Grant Foundation Number of species Synthetic Aperture Radar Stem Density Secondary Forest Slope Swamps Tonnes Tree Density Tropical Plant Exploration Group xxiv

26 TWINSPAN UN UNEP UNFCCC US USGS Vegtype WD WRI YA Two-way Indicator Species Analysis United Nations United Nations Environmental Programme United Nations Framework Convention on Climate Change United States United States Geological Survey Vegetation Types Wood Density World Resource Institute National herbarium of Cameroon xxv

27 CHAPTER ONE 1. Introduction The issues of biological diversity, climate change and agricultural expansion in tropical forest are concomitant and complex, and for decades have inspired many international conventions and agreements (COP03, 1997; COP06, 2002; COP09, 2008). Several conventions, such as the Convention of Biological Diversity (CBD) that tackles conservation and sustainable use of biological diversity, the United Nations Framework Convention on Climate Change (UNFCCC) that tackles the problem of global warming, and the International Plant Protection Convention that tackles genetic movement of plant species (Swanson, 1997) have provided insights into various issues associated with the management of the forest ecosystem. The implementation of CBD was adopted (CBD strategic plan-decision VI/26) in 2002 (COP06), and during the world summit in 2004 (COP07), goals and sub-targets were established according to focal areas (Decision VII/30). This was reenforced in the Bonn declaration in 2008 (COP09) in which firm decisions were taken to reduce the rapid loss of biological diversity at the global, regional and national levels to contribute to poverty alleviation and benefit life on earth before and beyond 2010 (Tim, 2010). Beside all these conventions, the biodiversity of tropical forest (lowland to Montane) has been of increasing priority due to its high level of forest depletion, and its important in species distribution, diversity and carbon storage. The challenges to meet the loss of biodiversity are still many, especially in the vast Congo basin forest, which has an estimated area of 198 million hectares. It is the second largest tropical forest after the Amazon rainforest in South America (Bikié et al., 2000), making up almost 91% of Africa s remaining moist forests (COMIFAC, 2009). This region is one of the most diversed landscapes in the sub-region (White, 1983; Barthlott et al., 2005; Cheek et al., 2004). This tropical forest served as a huge carbon sink that could help to mitigate global climate change (Brady and De Wasseige 2010). Unfortunately, in the last two decades over 1.1 million hectares of Congo basin forest have been cleared annually (Bikié et al., 2000) giving a deforestation rate of 16.5 million hectares in the Congo basin forest area (Bikié et al., 2000). The annual average deforestation rate for Cameroon between 1980 and 2000 was 0.6% (FAO, 2007). Most of these deforestation and degradation data mentioned above were predominantly in lowland forest, and excludes the montane ecosystem, which is the subject of this study. Similarly, forest carbon fluxes on ground sampling of carbon stock are mostly available for lowland forest in the region (Djomo, 2010; Djuikouo et al., 2010; Ekoungoulou, 2014; Memiaghe et al., 2016; Djomo et al., 2016; Tabue et al., 2016). Over the years, the phytogeographical studies conducted 1

28 in lowland and submontane forests, looked at species occurrence and richness (Cheek et al., 2000; Kenfack 2001; Cheek et al., 2004; Harvey et al., 2004). Only limited studies have focused on forest structure, distribution, composition and carbon stock for montane ecosystems in Cameroon (Tchouto, 1995; Tchouto et al., 1999; Sainge et al., 2014; Sainge, 2016). The present study was carried out in the wet tropical forest of the Rumpi Hills Forest Reserve (RHFR) ( m asl) and the dry semi-deciduous and savanna forest of the Kimbi Fungom National Park (KFNP) ( m asl), which falls within the continental Cameroon Mountains and spans from the coast of Mt Cameroon to the Mandara Mountains. Broadly, our study focused on the montane forest of western Cameroon. A detailed evaluation of the sampling gaps was carried out, and this subsequently led to the selection of two representative forest types the wet tropical forest (Rumpi Hills Forest Reserve) and the dry semi-deciduous and savanna forest (Kimbi Fungom National Park) for thorough study of species distribution and diversity, biomass and carbon storage. Firstly, this study will enable for data gathering, monitoring, and management of flora dynamics, biodiversity and ecology of the study area. It will complement the activities of the Central African Forest Commission (COMIFAC), Global Environmental Sustainability Programme (GESP), Congo Basin Forest Partnership (CBFP), and the Convention of Biological Diversity (CBD), whose collective vision is the sustainable management of forest and savanna ecosystems within its member states and the wellbeing of its inhabitants, preservation of biodiversity, and the protection of the global environment. Secondly, this study seeks to contribute and facilitate the achievement of some biodiversity targets stipulating that by 2020: about 80% of Cameroonians should be aware of the biodiversity of Cameroon, it is expected that the link and impact of human activities on the major ecosystems in Cameroon will be established, and that policy makers in Cameroon would be able to apply scientific information into biodiversity decision-making processes and management interventions (National Biodiversity Strategy and Action Plan, version II (NBSAP II), 2012). The approach adopted in this study centred on a series of measurements undertaken in tropical forests to understand the different vegetation patterns, forest structure, composition, diversity, biomass, and carbon stock in the present day scenario and evaluates their effects and threats on land cover changes resulting from human activities. The establishment of 42 permanent sampling plots of 1 ha each, i.e., 25 sampling plots at the Rumpi Hills Forest Reserve (RHFR) and 17 sampling plots at the Kimbi Fungom National Park (KFNP), provided the rationale to understand the different ecological parameters. Recording vegetation patterns, forest structure, composition, diversity, and the estimation of carbon stock on the Cameroon Mountain range through forest inventory, are ideal ways of understanding ecological systems of the studied areas. This will enable us to 2

29 answer with conviction the main research question: How does elevation and vegetation patterns influence species composition, diversity and biomass in selected wet and dry tropical forests of the Congo Basin? The following specific research questions guided data collection 1. What are the diversity, species composition, and forest structure at different elevational gradients? 2. What is the level of above-ground biomass, and carbon stock per hectare in tropical rainforest and tropical dry semi-deciduous and savanna forest in Africa? 3. What are the extent of land cover changes, the different land use cover types, and their effects on the species composition of the RHFR and the KFNP? 1.1 Outline of the Thesis My study sought to establish baseline data for long term ecological studies in tropical forest, focusing on the continental part of the Cameroon Mountains. This area is vast and covers approximately 37,879 km 2 in length with different mountain ranges, vegetation types and habitats. This thesis focused on the Rumpi Hills Forest Reserve and Kimbi Fungom National Park, and examined the different vegetation types and elevation gradient, forest structure, species composition, above-ground biomass and carbon. It consists of seven chapters. Chapter 1 and other five research chapters (chapters 2-6) and general discussion and recommendation in chapter 7. In chapter 2, we discussed the botanical sampling gaps across the continental part of the Cameroon Mountains using the concept of biodiversity informatics. This was the result of a reconnaissance field survey and review of existing literature (Letouzey, 1968, 1985; Sainge and Cooper, 2014). On the basis of the findings in Chapter 2, we choosed to focus on the Rumpi Hills Forest Reserve (RHFR) and the Kimbi Fungom National Park (KFNP), both of which have significant biodiversity sampling gaps, and therefore beckons for a detailed botanical survey. In chapter 3, we discussed the characterization, floristic composition and structure of the RHFR based on the six vegetation types (Montane, Submontane, transition submontane, mid-elevation forest, lowland basalt forest and the Atlantic Biafran forest with Caesalpiniaceae). Chapter 4 elaborates on plant community distribution, above-ground biomass (AGB) and carbon stock in four vegetation types of lowland ( m), Mid-elevation ( m), submontane ( m) and montane forest above 1600 m, along elevation gradient. Chapter 5 focused on floristic diversity, distribution, AGB and carbon stock across different vegetation types in the Kimbi Fungom National Park were analyze. While chapter 6 centered on the assessment of land cover changes 3

30 and land use covers in the RHFR and KFNP, and chapter 7 presented a general discussion and recommendations of the study. 1.2 Literature Review Forest and forest types Definition of Forest The definition of forest depends on the author or institution, landscape, and the different vegetation types present (Terborgh, 1992; Cameroon Forestry law, 1994; UNFCCC, 1997; FAO; 2008; GOFC-GOLD, 2008). FAO considers forest as any land of more than 0.5 hectare with trees >5 m tall with a canopy cover of >10%. However, by this definition, agricultural and urban lands are not considered as forests (FAO, 2010). Kyoto protocol on the other hand, describes forest land as forest with a minimum size of hectare, crown cover >10 30%, and a minimum mature tree height of 2 5 m tall (UNFCCC, 1997). Forest in the context of the Government of Cameroon is any land with vegetation cover mostly composed of trees, shrubs, grasses and species of non-agricultural products (Cameroon Forestry Law, 1994). According to Terborgh (1992), forest refers to all vegetation that supports a continuous tree canopy, while nonforest refers to everything else, including open woodlands, savanna, desert scrub, and alpine grassland. For the current study, forest is define following the concept of Food and Agricultural Organization (2008). Forest can be classified based on different vegetation types (lowland rainforest, mid-elevation, sub-montane, montane and afro-montane, wet and dry semi-deciduous, gallery, Sahel savanna, Miombo, Desert, Mediterranean woodland, and warm temperate humid), and succession (pioneer, early secondary, late secondary and climax) (Ngwa, 1981; Letouzey, 1985; Neba, 1999). Forest is such a high valued natural resource to humankind and nature because it acts as a source of medicinal plants, non-timber forest products, building materials, and food (Sainge et al., 2014). It also acts as carbon sinks, home for biodiversity, ameliorate the macroclimate of the environment, and provides other ecosystems services to millions of people (Reich, 2011). 4

31 Types of Forest Forests are grouped into different latitudinal zones. The polar zone where only lichens and mosses are recorded on rocks and in damp places. Hence, most studies have focused on cryptograms (Bednarek-Ochyra et al., 2000; Erdau, 2001; Newsham, 2010; Putzke et al., 2015). The temperate forest exist in two forms: the conifer forest occurring close to the polar zone, and the hardwood forest that shed their leaves in winter, occurring closed to the tropical zones. Temperate forests are mostly homogenous with few species, sparse undergrowth, absence of buttress trees and lianas. This forest is represented in Canada, Western and eastern USA, Europe and in northern Asia (Letouzey, 1986). The tropical forest occurs in Central-South America, West-Central and East Africa, Madagascar, the Arabia and Northern Central Asia, and Australia (Terborgh, 1992; Cole et al., 2014). This forest in the Africa continent occurs in the Sahara covering the Rio de Oro, Mauritania, part of Algeria, Mali, Chad, Libya, Egypt, part of Sudan and Eritrea and the Kalahari (Letouzey, 1986). The Sahel, which is close to the equatorial forest stretches to the Sudanian savanna with mostly small trees that lose their leaves in the dry season. Some authors (Letouzey, 1986; Murphy and Lugo, 1986; Janzen, 1988; Terborgh, 1992; Molle, 2008) called the abovedescribed forest as the dry tropical forest and the equatorial forest as the tropical rainforest. Generally, the equatorial forest occurs between the tropics of Capricorn in the south and the tropic of Cancer north of the equator. It is also called the tropical rainforest and is the most diverse, species-rich ecosystem on earth (Letouzey, 1986; Terborgh, 1992; Molle, 2008). The tropical rainforest still fascinate many botanists, ecologists and other researchers, who have been exploring challenging questions, hypotheses, and discovery of species (Carson and Schnitzer, 2008; Bryan and Shearman, 2015). This forest is mostly heterogeneous and they lost their leaves alternatively leaf after leaf or branch after branch. The undergrowth is dense, has trees with huge buttresses, and numerous lianas. It occurs in the Central and South America with the Amazonia forest being the largest followed by the Congo Basin Forest in Central Africa as the second largest forest block in the world. In West Africa, it stretches across Guinea, Liberia, Ivory Coast, Ghana, and Nigeria, while in Central Africa, it extends through Cameroon, Gabon, Equatorial Guinea, Congo, Democratic Republic of Congo and Central Africa Republic. It occurs in patches in East Africa: Kenya, Uganda, and Tanzania. It stretches southward to Angola and Madagascar, and in Asia it extends to Southern India, Japan, Indonesia, Papua-New Guinea, and northern Australia (Letouzey, 1986; Brandon 2014). 5

32 Tropical forests Globally, tropical forests are mostly located close to the equator and span between the tropics of cancer (23 28 N) and Capricorn (23 28 S), an area of ~50 million km 2 where temperatures and rainfall are high (Terborgh, 1992; Osborne, 2000; Brandon, 2014). In general, tropical and equatorial forest falls in the same latitude within the tropics. It can be sub-divided into tropical rainforest, which is most species-diversed and species-rich, and tropical dry forest, which is species-poor, and the tropical montane forest, which cuts across rainforest and dry forest. From time immemorial, scientists and researchers have studied tropical forests, yet they remain challenging and intriguing ecosystems to study due to their immensed richness of flora and fauna, their complex structures, and the high species distribution per surface area (Whittaker, 1956; 1960; 1965; Osborne, 2000; Carson and Schnitzer, 2008; Bryan and Shearman, 2015). Despite the numerous studies carried out in this ecosystem, there is still need for research on the systematics, ecological and biodiversity aspects of these ecosystems. Broadly, tropical forest used to cover about 12% of the earth s terrestrial surface, but this has now dropped to less than 5% (600 million hectares) due to extensive deforestation and land degradation (Corlett and Primack 2011; Hansen et al., 2013) Tropical rainforest The tropical rainforest also known as equatorial forest is home to the most fascinating species of plants and animals in the world. It is fondly described as a laboratory or open library for rainforest botanists, systematists, ecologist, tourists, and experts on ecosystem services (Carson and Schnitzer, 2008; Bryan and Shearman, 2015). Most rainforest occurs within 10 degrees latitude north and south of the equator (Letouzey, 1986; Molle, 2008). However, rainforest of Central America, Mexico, Southeastern Brazil, Eastern Madagascar, Southern India, and Northeastern Australia are located outside this equatorial zone (Molle, 2008). The largest tropical rainforest belt of the world is the Amazonian forest, while the Congo Basin forest of Central Africa is the second largest forest block (Terborgh, 1992; Molles, 2008; Brandon, 2014). Collectively, it is estimated to holds ~22 tons of biomass per hectare per year compared to only 13 tons of biomass per hectare per year in temperate forest (Montagnini and Jordan, 2005) Tropical dry forest Tropical dry forests are known for their most threatened and species-poor ecosystems on the planet (Murphy and Lugo, 1986; Janzen, 1988). Nevertheless, it remains the largest ecosystem on earth (42% surface area), accounting for almost half of the world s tropical and sub-tropical forest. It occurs between 10 and 25 latitude North and South of the equator (Murphy and Lugo, 1986; Molles, 2008; Blackie et al., 2014). They are 6

33 characterized by a tropical climate: annual rainfall of mm, 5-8 months of dry season, open forest canopies, and habitats that host some endangered species of plants, animals and insects (Janzen, 1988). They are located in the west coast of Central America, and North America, and west coast of Mexico. Most of the dry forests in South America are being converted to other land-use systems. In Africa, it is located both in the north and south of the west and central Africa rainforests, it extends to part of East Africa, the miombo woodland of southern Africa, and west of Madagascar (Molles, 2008; Blackie et al., 2014). In Asia, it occurs in India, Indochina peninsula, and in northern Australia (Molles, 2008; Blackie et al., 2014). The biome covers ~6 million Km 2, and over 50% of this land is used for animal grazing and agriculture (Ramankutty et al., 2008) Tropical montane forest Mountains formations are complex geological processes that results from earth crust movements and volcanism that are fold and elevated to the surface (Kapos, 2000; Körner et al., 2005; Richter, 2008). Globally, all the continents of the world are represented by mountain ecosystems that host tropical mountain forests (Kapos, 2000; Körner et al., 2005; Richter, 2008). Tropical montane forest are hotspots of biodiversity richness of flora and fauna, covering 32% of all protected area, with an area of 5, km 2 worldwide (Barthlott et al., 2005; Barthlott et al., 2007; Chape et al., 2008). In Africa, they spans from Mt Loma (1947 m), and Mt Nimba (1752 m) in West Africa. Cameroon mountain range (the peak of Bioko island [2850 m], Mt Cameroon [4095 m], Mt Rata [1778 m], Bakossi mountains [Bakossi National Park], Mt Nwoanengouba, Mt Kupe, Mt Bamboutos, Mt Nlonako and Mt Oku [3011 m]). In central Africa, a few massif east of Democratic Republic of Congo extend towards Uganda (Mt Ruwenzori 5119 m; Virunga 4531 m), the East African Mountains (Mt Kilimanjaro; 6010 m), the Ethiopia mountains (3000 m), Mt Kenya (5195 m), and the Mt Mulanje (3000 m) to Mt lovely and Chela (~2400 m) in Angola. The Atlas Mountains in northwest Africa and the South Africa mountains form a consortium of the African mountains. Most of the mountains in Africa have great flora and fauna affinity (Letouzey 1986; Dowsett- Lemaire, 1989; Stattersfield et al., 1998; Richter, 2008; Molles, 2008). In the Americas, tropical mountain forests are found on the west of Central America, extending north to Alaska and to South America. This biome has a vast range of vegetation from evergreen broadleaved rich in vascular tree species, cryptogram epiphytes to species poor dry woody savanna mountains. In Asia and Europe, the mountains ranges from east to west from the Eurasian mountain ranges, including the Pyrenees, Alps, Caucasus, and the highest peak which is the Himalayas. In Australia, the mountain ranges in the east of the continent. 7

34 1.2.2 Tropical forests and Ecosystem services Ecosystem services in tropical forest are direct or indirect benefits acquired by people living in the immediate vicinity of the tropical forest be it protected community or freelance forest (Daily, 1997; Millennium Ecosystem Assessment, 2005). These benefits can be related to ecosystem structure and functioning such as climate, topography, geology, supply of food and timber, resource availability (water and nutrients), disturbances (natural disasters such as floods and landslides), biotic communities (diverse species groups), carbon sequestration, pollination, pest control, and other human activities (Ricketts, 2004; Bunker et al., 2005). These services play vital roles in climate change strategies such as mitigation, adaptation and payment for environmental services (Turner et al., 2009; Locatelli, 2016). The world s terrestrial ecosystem can absorb approximately 3 billion tons of atmospheric carbon, 30% being anthropogenic CO 2 per year (Picogram/year, Pg/yr) (Canadell and Raupach 2008). This huge amount of carbon absorbed from the atmosphere qualifies tropical forest ecosystem as an important carbon sink in carbon sequestration. Unfortunately, the rate of deforestation in the tropics is alarming and is estimated at Pg/yr (Baccini et al., 2012; Harries et al., 2012) with 6-17% of global anthropogenic CO 2 emitted to the atmosphere (Van der Werf et al., 2009). In order to curb the effect of deforestation, which may lead to loss of biodiversity, afforestation and reforestation have been introduced, coupled with proper farm management, soil conservation, and increase in biodiversity in farms (Uprety et al., 2012). Furthermore, good governmental policies such as curbing the effect of climate change as reported by Locatelli (2016) will be of great importance in managing and protecting loss of biodiversity within the forest ecosystem. These changes in some cases are irreversible (Ellatifi, 2005) and will mostly affect the biodiversity, composition, forest structure and ecological processes of tropical forest ecosystems Ecology of tropical forests Ecological relationships influence the distribution, abundance and interaction of organism in their environment. The ecology of tropical forest has been studied for as far back as human history, and relying upon observed variation and predictions (Molles, 2008). Some theories or concepts tested in ecology over time are that of the Climax theory and the Holistic concept of Clement (1936), the individualistic concept of plant association of Gleason (1939), the concept of gradient analysis of vegetation invented by Whittaker (1956, 1960, 1965), the ecological principles of resource partitioning (MacArthur 1958; Connel, 1978; Grime, 1979; Davies, 1983, and Brown, 1984). These theories and hypotheses have shaped our understanding of tropical forest. 8

35 1.2.4 Threats facing tropical forests Tropical forests are important ecosystems for many species. Tropical forests also enhance climatic processes that contribute to controlling evaporation and temperature, maintenance of vegetation types, protect soil, landslide. They hold a considerable amount of biomass and carbon as sink and sustain biological diversity and cultural activities (Raven and Williams, 1995; Laurance, 1999; Bradshaw et al., 2007; FAO, 2012). However, anthropogenic activities are the main threats to these potentials of tropical forest. Deforestation and forest degradation are the major threats, and are driven by agro-industrial agriculture, ranches, plantations, selective logging, constructions of new roads, industrialization, exploitation of natural resources, and settlement expansion (Butler and Laurance 2008; Laurance and Balmford, 2013). If the current rate of industrialization and globalization continues in developing countries like China, Brazil, India and South Africa (MEA, 2005; Laurance, 2015) then biodiversity, biomass and carbon stock losses will continue to grow. In the Cameroon Mountains, the continuous exploitation of tropical forest for small and large scale agriculture are potential threats (Kenfack et al., 2014; Kupsh et al., 2014). Remedies such as mitigations, adaptations and payment for environmental services (Turner et al., 2009) will need to be implemented, which could including curbing the expansion of agricultural land, wood extraction and infrastructural development (Norris et al. 2010). A study conducted in West African Guinea forest shows that agricultural expansion is a major drive to deforestation and degradation in Africa (Norris et al., 2010). Demenou 1997 and Eba a Atyi et al., 2016 have demostrated that fuelwood and charcoal production for domestic and industrial uses are important causes of forest degradation Climate change and tropical forest The concept of linking forest loss and climate change in tropical forest was enforced in 2005 under the theme Reducing Emissions from Deforestation and forest degradation (REDD). At the climate change conference held in Poznań 2008 (COP14 in Poland), national and regional discussions on forest emissions reductions in developing countries led to a consensus of Reducing Emissions from Deforestation and forest Degradation plus. The role of conservation, sustainable forest management and enhancement of forest carbon stocks (REDD+) was initiated during COP13 in Bali, Indonesia (European Union Institute, 2014). This initiative has created a mechanism that encompasses both reduction of emissions caused by deforestation and forest degradation, the role of forest conservation, sustainable forest management and enhanced forest carbon stocks (Dkamela et al., 2009). Clean Development Mechanism (CDM), which emerged from the Kyoto Protocol, focuses on development (including agriculture systems) that will be as little damaging to the environment as possible. 9

36 Globally, the world s forest holds ~283 billion tons of carbon in living biomass (IPPC, 2007). Land-use changes in tropical Africa are estimated at 0.3 GtC yr -1 of CO 2, with deforestation as the primary contributor (IPCC, 2007). Clearly, reducing deforestation will be an important step towards conserving existing forest carbon stocks. These ideas underpin the role of the Land-Use, Land-Use Change and Forestry (LULUCF) sector as a credible part of the solution to addressing climate change (UNFCCC, 2008). Hence, sustainable agriculture in the form of agro-forestry and reforestation in farms are also vital to avoid deforestation and food insecurity (Foley et al., 2005). The conference of Parties 21 (COP21) held in Paris in December 2015 was another platform to discuss issues on mitigating climate change by keeping the world s temperature rise below 1.5 degree Celsius. Article 5.1 and 12 of this conference of parties 21 stress on the conservation and enhancements of sinks, reservoirs, ecosystems and forest in other to realize increase in biodiversity, biomass and carbon and how this information can be disseminated through education, training, and public hearing (UNFCCC, 2015). The enhancement of all these parameters will give the forest its actual value and improve other ecosystem services in terms of non-timber forest products, medicine, food security, generating income, biodiversity, biomass, protection of watershed, and carbon budget. The forest, including its vegetation and habitat is of great value to all living creatures. Incentives such as timber, fruits, vegetables, and medicial plants are important ecosystem services that are at the disposal of the local inhabitats, and the nation in income generation. The aesthetic value of tropical forest such as the beautifull waterfalls, cliffs, caves, animals and plants also attract tourist, and researchers. All these and a host of others are great sources of income, employment, and relaxation. 1.3 The Cameroon Mountains: Rumpi Hills Forest Reserve (RHFR) and Kimbi Fungom National Park The Cameroon Mountains, part of the Cameroon Volcanic Line (Ayonghe et al., 1999; Marzoli et al., 2000; Tsafack et al., 2009), have been called by different authors as the Cameroon Line (Nono et al., 2004), the Mountains of the Cameroons (Durrell, 1958), and Biafran forests and highlands (Cronin et al., 2014). This ecological zone is composed of a chain of volcanic and plutonic isolated mountain peaks that covers about 40,877 km 2 (Tsafack et al., 2009; Sainge, 2016; Sainge unpubl. data). It is the least well-studied mountain habitat in terms of consolidated large-scale biodiversity data (Figure 1.1). 10

37 This ecoregion consist of two parts: The oceanic portion of the mountain chain, which covers about 2998 km 2 (Frodin, 2001) that runs from Pagalu (17 km 2 ) through São Tomé (854 km 2 ), Principe (110 km 2 ), to Bioko (2017 km 2 ) Island (Ayonghe et al., 1999; Frodin, 2001). The continental portion of the mountain chain covers about 37,879 km 2, and extends from Mt Cameroon (summit at 4095 m above sea level) and Mt Etinde, (commonly called small Mt Cameroon, reaching 1474 m), on the coast, through a string of peaks including Mt Mwoanenguaba (2396 m), Mt Kupe (2050 m; Morgan et al., 2011; Neba 1999), Rumpi Hills (peak at Mt Rata at 1778 m; Sainge 2016), Mt Nlonako (1875 m; Tchouto and Ebwekoh, 1999, Kenfack, 2001), Mt Bamboutos (2740 m), Lebialem Highlands (2400 m), Malap Njibanchi Hills (1947 m), and the Nkounchankap Hills (1217 m). Other important highland areas include Mt Oku (3011 m; Thomas and Achoundong, 1994; Larison et al., 1996; Cheek et al., 2000), Rhum Rock (1603 m), Nkom Wum Forest (1197 m), Kagwene Wildlife Santuary (1800 m). Mt Mambila (1821 m), Tchabal-Mbabo (2456 m), Tchabal-Gandaba (1960 m) and the Adamawa Plateau, Tchabal Ngangha (1913 m), Hossere Vokré (2046 m), and the Mandara Mts (Gèze, 1943; Larison et al., 1996; Neba, 1999; Ayonghe et al., 1999; Nono et al., 2004). The Mandaras Mts are at 900 m, with two higher inselbergs at Mokolo (1442 m) and Mogode (1224 m). This mountain chain stretches into Nigeria at the Mambila Plateau, Obudu Plateau, Mbola Hills, and Mt Shebsi, on to the Biu Plateau (Hepper, 1965, 1966; Hall and Medler, 1975; Ayonghe et al., 1999) (Figure 1.1). Beyond this general view of the Cameroon Mountains, this study focused on the continental portion of the mountain region, specifically on the Rumpi Hills Forest Reserve (RHFR) in the South West Region and the Kimbi Fungom National Park (KFNP) in the North West Region of Cameroon. The RHFR is located in Ndian division, South west Region of Cameroon, at latitude N and longitude E with an elevation range of m at the summit of Mt Rata. It covers an area of km 2 (Forestry Ordinance 51, 1941), and was created in 1937 to protect its huge biodiversity. It is bordered to the north by Meta village in Mundemda sub-division and Iyombo village in Toko sub-division, in the east by Dikome Balue in Dikome Balue sub-division, in the south by Nalende, Kita and Munyange in the Ekondo titi sub-division, and to the west by Boa Yenge, Motindi, Meka, and Besingi in the Mundemba sub-division (Figure 1.4). The Mundemba settlements form a corridor between the Rumpi Hills Forest Reserve (RHFR) and the Korup National Park (KNP), with two agro-industrial plantations: Pamol Plantation (PLC) Limited in the south and Sithe Global Sustainable Oils Cameroon in the Northwest of the reserve (Kupsh et al., 2014). The Kimbi Fungom National Park is located in the Northwest Region of Cameroon at latitude N and longitude E with a size of km 2 (Sainge, 2016), created on 3 February 2015 under Prime 11

38 Ministerial decree number 2015/0024/PM, and is the only national park in the region. It cuts across three divisions: Boyo, Menchum, and Donga-Mantung, covering 4 Sub-divisions: Fonfuka, Fungom, Furu-Awa, and Misaje. In the north, it is bordered by Tumbo and Tosso in Nigeria, Baji, Nser, Kpep, Furubana, Supong, Akum, Edjong and river Katsina Ala in Furu Awa sub-division. In the east by Labo, Batari, and the Dumbo cattle ranch in the Misaje sub-division, and river Kimbi, Kimbi village, Su Bum in the Fonfuka sub division. In the south; Nkang, Esu, Kundzong, Iwo, and in the west by Munkep, and Gayama in the Fungom sub-division, and Nigeria. These two compartments connect to a corridor that stretches between Nkang and Nkannye on the Fungom end to the north west of Kimbi and southwest of Dumbo cattle ranch with river Kimbi being a natural boundary between the ranch and the National Park. This park extends (Donga Mentung extension) to the north of the Dumbo cattle ranch in the Donga Mantung division Climate and hydrology in the Rumpi Hills Forest Reserve (RHFR) and Kimbi Fungom National Park (KFNP) The climate across the continental part of the Cameroon Mountains is that of the Equatorial Cameroon type, hot and humid with two seasons. The dry and wet season varies in the amount of rainfall and temperature across the landscape from the wet rainforest in the south (Rumpi Hills) to the dry semi-deciduous and grassland/woody savanna in the north (Kimbi Fungom). The variation in temperature along different elevations and vegetation types in the Rumpi Hills Forest Reserve (RHFR) partly explains the high plant diversity in different forest communities in this reserve (Thomas, 1996; Sainge, 2016). Temperature within the RHFR varies with the lowest at the top of Mt Rata. Nembot and Tchanou (1998) reported a temperature of 22 C and an annual rainfall of 5000 mm for the reserve. The area exhibits a short period of dry season of approximately four months (December to March). In some years, only three months (January to March) of dry season are experienced (Sainge, pers. obser.). Rainfall is mostly high in the southwest corner of the reserve (Sainge, pers. obser.). Unfortunately, until date no proper weather station occurs in this area to obtain regular climatic data for this reserve (Sainge, 2016). We assumed that due to proximity (<50 km) with Korup National Park, Mundemba whose rainfall average ~ 6000 mm per year (Chuyong et al., 2000, Chuyong et al., 2004), this reserve has >5000 mm of rainfall. Climatic data from Mundemba, which is <50 km north of RHFR gives a mean rainfall record of mm, radiation of 9.3 and temperature: minimum (22.3 C) and maximum (30.2 C) (Figure 1.2) over a period of eleven years. 12

39 The RHFR is a watershed for different river (R.) sources and is Cameroon s principal watershed that continues through the mountains of Kupe, Manenguba, Bakossi Mountains, Bamboutos, the Bamenda highlands, Banyo, and Adamawa plateau to the Ubangi-Shari in Nigeria (Ngwa, 1981). The rivers of this reserve flow in five directions: North into Lake Chad, via river logone; North West into river Benue, South West into the Gulf of Guinea, via River Ndian (Moriba), Moko, Meme, Mungo, and Wouri; South east into Kadei, a tributary of the Congo River, and West into Nigeria, via River Munaya, and Mbo. Hence, Rumpi Hills is a watershed for five major basins of the world: Chad, Benue, Sanaga, Congo, and Manyu Basins (Ngwa, 1981). In the Kimbi-Fungom National Park, dry season runs from October to March, and the wet season from April to September (Sainge et al., 2014). No proper weather station is installed in the park. However, rainfall records from Nkambe (3,028 mm per year) and Ako, which has the same landscape like Kimbi Fungom and < 100 km away is mm per year with an average temperature record of C (Figure 1.3). This park is a watershed with numerous streams and rivers such as river Kimbi and Johga whose tributaries meet at Nkang village forming river Katsina Ala that flows through Kpep, Akum, Edjong, and Manga into Nigeria. The main rivers flowing through the park are Kimbi, Dumbo, and Johga in the Kimbi compartment. River Batum crosses Furu bana to Kpep where it flows into River Katsina Ala. Other rivers in the north are the Yamaha, Yakang, Sungsi, Ntuna, and Kenda, while in the south, River Yemene and Imea flow downward towards Nkang, and in the west, River Yaboa Elevation in the Rumpi Hills forest reserve (RHFR) and Kimbi Fungom National Park The Cameroon Mountains is an ecological zone that cuts diagonally across southwestern Cameroon ranging from sea level at the Gulf of Guinea to the top of Mt Cameroon at an elevation of 4095 m above sea level. Within this range, isolated mountain peaks exist such as Mt Etinde (commonly called Small Mt Cameroon, 1474 m) on the coast. Mt Mwoanenguaba (2396 m), Mt Kupe (2050 m; Morgan et al. 2011, Neba 1999), Bakossi National Park (Bakossi Moutains, 1896 m), Rumpi Hills (peak at Mt Rata, 1778 m; Sainge 2016), Mt Nlonako (1875 m; Tchouto and Ebwekoh 1999). Mt Bamboutos (2740 m), Lebialem Highlands (2400 m), Malap Njibanchi Hills (1947 m), the Nkounchankap Hills (1217 m) and Mt Oku (3011 m); (Thomas and Achoundong 1994; Larison et al. 199;, Cheek et al. 2000), which is the second highest mountain in the region. Rhum Rock (1603 m), Nkom Wum Forest (1197 m), Kagwene Wildlife Santuary (1800 m), Mt Mambila (1821 m), Tchabal-Mbabo (2456 m), Tchabal-Gandaba (1960 m), Adamawa Plateau, Tchabal Ngangha (1913 m), Hossere Vokré (2046 m), and the Mandara Mountains at 900 m, with higher inselbergs at Mokolo (1442 m) and Mogode (1224 m) (Gèze 1943, Larison et al. 1996, Neba 1999, Ayonghe et al. 1999, Nono et al. 2004). 13

40 In the RHFR, the elevation range from m. The southern part is flat, starting from the creeks of river Moko through Munyange to Nalende, and to the northwest from Matamani through river Moriba ( m a.s.l). West to east from Mbange, Lipenja- Mukete, Bossuga in the west to Dikome Balue, and Madie in the east, the topography is hilly with an elevation range of m to the top of Mt Rata. The KFNP ranges in elevation from m (Sainge, 2016). Most tree species occurring in semideciduous, gallery, grassland and woody savanna are resistance to extreme temperatures and fire regimes. 1.4 Significance of Study Surveys of these forest areas have been conducted in the past (Achoundong, 1995; Thomas and Thomas, 1996; Thomas, 1996, 1997; Cheek et al., 2000, 2004; Kenfack, 2001; Harvey et al., 2010; Kenfack et al., 2014); however, these were not carried out on permanent plots. Permanent plot establishment is paramount since they generate much data (Condit, 1998, Thomas et al., 2003) than temporal sample plots, used for the study of geographical distribution of species (Achoundong, 1995; Thomas and Thomas, 1996; Thomas, 1996, 1997; Cheek et al., 2000, 2004; Kenfack, 2001; Harvey et al., 2010; Kenfack et al., 2014). Studies on montane ecosystems are scant in Cameroon, which have resulted in data gaps. Furthermore, where the studies have been carried out, the findings are not readily available to the scientific community (Thomas and Cheek, 1992; Thomas and Achoundoung, 1994; Achoundoung, 1995; Cheek et al., 2000; Cheek et al., 2004; Forboseh et al., 2011). Thus, this study is setting a base for long term monitoring of the montane ecosystem that cuts across different vegetation types spanning from lowland to montane in the Rumpi Hills Forest Reserve and Kimbi Fungom National Park. Both are protected forests and the data is being hosted by Tropical Plant Exploration Group (TroPEG) Cameroon, an institution, which studies the plant biodiversity of the continental Cameroon Mountains in collaboration with students and other scientists. The present study will contribute greatly to the current database of plant diversity of Cameroon, by providing a comprehensive checklist of plants of the Rumpi Hills ranging from lowland to montane and in the Kimbi-Fungom National Park at different vegetation types. This study will also act as a booster for other scientists and researchers who are interested in studying plant taxa and other ecological aspects in the RHFR and the KFNP. The above factors will help the scientific community, as well assist the government of Cameroon to ratify its international agreements on the Convention of Biological Diversity (CBD), Climate change, REDD, and REDD+. 14

41 Cameroon lacks effective forestry laws for montane ecosystems, i.e. the forestry law of 1994, which is under review does not place emphasis on montane ecosystems (MINEF, 1994). This study lays the foundation and benchmark for policy development and implementation. Cameroon, commonly called, African in miniature is rich in biodiversity and carbon storage capacity. Thus far, few studies have tried to quantify the carbon stock in Cameroon forest from lowland to montane forest and in different vegetation types (Djomo 2010, Djuikouo et al., 2010, Djomo 2013, Djomo et al., 2016, Tabue et al., 2016). This present study also examined biomass and carbon in both tropical forests from lowland to Montane of the RHFR and tropical dry semi-deciduous to savanna forest of the KFNP in Cameroon. Cameroon has approximately 15,000 species of butterfly; species of mammals, 10 species endemic, 27 threatened species. 542 species of fish in 179 genus, 53 families with 96 species endemic and 294 species occurring in dense humid forest; 165 species of reptiles out of the 275 species in Africa; 200 species of amphibians with 63 endemic species; 3 species of crocodile; and 190 to 200 species of batrachian. 900 species of birds, 750 species of which are residual and 150 species migratory, 11 species are endemic and 17 species are threatened. Approximately 9000 species of plants (higher plants only) with 156 species endemic, 74 threatened. Within this 9000 species of plants, 630 species have potential commercial value (Nembot & Tchanou, 1998; Foahom, 2001; GEF, 2008). Thus, results of this study will increase our knowledge on the plant biodiversity of the study area and the country at large. It will also boost Cameroon s chances of meeting up with the goal on the Convention of Biological Diversity (CBD) in which she is a signatory. Capacity building, education and information management are other critical components for this study. During and after the field work, over fifteen Cameroonians (students, field assistants, and community members) were trained on forest mensuration, transect cutting, plot establishment, data and specimens collections, plant specimens description, drying, sorting and identification using floras, and monographs and herbarium specimens sheets. Reports of fieldwork were sent to funders, which have been published. Manuscripts have been submitted to reputable journals for review. After presentation of this thesis, a series of seminars and conferences will be organized, and decision makers and other stakeholders will be present at these meetings for better dissemination and appreciation of the findings of this study. 1.5 Rationale of Study Tropical montane forests are distributed in the tropics where mountains coincide with conditions sufficiently humid to permit forest growth. The vegetations of this biome are varied, and are modulated by elevation 15

42 gradient, rainfall patterns, temperature, soil type, and geology. This variety and combinations of conditions have led to high species diversity in tropical mountains (Lovett, 1996; Cheek et al., 2000; Tchiengué, 2004; Cheek et al., 2004; Richter, 2008), with some specific sites recognized as biodiversity hotspots (Mt Cameroon, Bakossi National Park, etc.). This study focuses on the continental part of the Cameroon Mountains, which falls within a mega biodiversity hotspot in western African with impressive species diversity of different taxa, and with high levels of endemism (Myers et al., 2000). Inspite of numerous biodiversity studies in Cameroon and particularly within this ecological zone (Cable and Cheek, 1998; Tchouto et al., 1999; Cheek et al., 2000; Thomas et al., 2003; Tchouto, 2004; Sainge, 2012; Sainge et al., 2014; Sainge, 2016), none of these studies has provided a broader view of the biodiversity of the continental part of the Cameroon Mountains. The Cameroon Mountains are thus relatively understudied and underappreciated. This is in contrast with other tropical montane regions, such as the Eastern Arc Mountains of East Africa and the Andes in South America (Lovett, 1996; Richter, 2008). A few studies, such as those by Cable and Cheek (1998), Tchouto et al. (1999) and Forboseh et al. (2011) focused on Mount Cameroon. Tchouto and Ebwekoh (1999) presented plants surveys of the Mwoanenguaba Mountains, Cheek et al., 2004 carried out a detailed study of plants of Mwoanenguaba, Bakossi Mountains and Mt. Kupe, and Tchiengué (2004) studied the plants of Mt Kupe. Thomas (1996) did a checklist from a reconnaissance survey of the Rumpi Hills, while Sainge (2016) studied the plants of Rumpi Hills and Kimbi Fungom National Park. Mount Oku was studied by Cheek and his team (Cheek et al., 2000), and only a preliminary botanical survey of the Lebialem Highlands has been published (Tchetcha, 2012, Fonge et al., 2013). More generally, no comprehensive overview on plants distribution and diversity exist across this broad and hyper-diverse biological landscape (Figure1.1). Until recently, the Cameroon Mountains have suffered considerable neglect from most researchers and funders who focus on tropical rain forest, and other lowland habitats (Thomas et al., 2003; Tchouto, 2004; Djomo, 2010). This has resulted to significant data gaps, such as data on species occurrences on the different mountains, species distributions across elevation gradients (lowlands to Mountain tops), and different vegetation types remain poorly understood. These gaps have influenced us to study the Cameroon Mountains with focus on the Rumpi Hills Forest Reserve (RHFR) and the Kimbi Fungom National Park (KFNP) in Cameroon. 16

43 1.6 Problem Statement A major challenge still exist in linking biological diversity, agricultural expansion, deforestation, forest degradation, biodiversity hotspots with the country land-use planning, land tenure and forestry reform strategies across the Cameroon Mountains. Estimations show that this landscape harbours species of vascular plants, based on biogeographical data of species presence and absence (Cable and Cheek 1998; Cheek et al., 2000; Cheek et al., 2004; Onana and Cheek 2011; Onana, 2013; Sainge, 2016). Hence, significant biodiversity data gaps relating to different habitats, vegetation types, and elevation gradient on permanent plots sampling across the Cameroon Mountains still exists. This gap of knowledge when bridged will be paramount to policy makers and biodiversity experts to understand the different diversity patterns, biodiversity hotspots and conservation priority sites. Establishment of permanent sampling sites as was done in this study will facilitate monitoring and tracking biodiversity changes over time. In addition, this will enable us to understand the diversity, distribution, amount of biomass and carbon stock that this landscape can store as carbon sinks. 1.7 Research Aim and Objectives The main objective of this work was to collect and analyse large-scale biodiversity data across different vegetation types and elevation gradients at the Rumpi Hills Forest Reserve and the Kimbi-Fungom National Park. This will give us the opportunity to identify, locate, explore the relationship between elevation and vegetation structure or species distribution and map the various plant species along the mountain range from lowland to montane Specific Research Objectives 1 To identify botanical sampling gaps across Cameroon Mountains, and select two gaps for a detailed study. 2 Investigate the forest structure, composition, and characterize the vegetation types across the Rumpi Hills Forest Reserve in South west Region of Cameroon. 3 Investigate the floristic diversity, distribution and carbon content across different vegetation and habitat types in the Rumpi Hills Forest Reserve. 4 Investigate the floristic diversity, distribution and carbon stock across different vegetation types in the Kimbi Fungom National Park. 5 Assess the land cover changes over time in tropical wet and dry forest of the Rumpi Hills and Kimbi Fungom forest. 17

44 1.8 Field Methods Smaller square and rectangular plots of different sizes can be used for rapid vegetation sampling and the study of different life forms (herbs, shrubs, lianas, and trees), and later projected to per hectare. Square plots of 100 x 100 m at could cover one or two habitats, while rectangular plots of 500 x 20 m may cover different habitats (flat land, slopes, plateau, swamps, streams, rivers), vegetation types (semi-deciduous forest, gallery, grassland/woody savanna), and forest types (primary, secondary, and disturbed forest). It is in this light that our approach to study the vegetation pattern of the RHFR and the KFNP was chosen to be rectangular, using line transects of 500 x 20 m, with 25 quadrats of 20 x 20 m representing one hectare. The above method can be appreciated when appropriate tools such as remote sensing and its application in forest inventory are used. Prior to detail field survey in the RHFR and KFNP, a thorough literature review and reconnaissance survey was carried out for both sites (Letouzey, 1985; Sainge and Cooper, 2014; Figure 1.6). All these were to identify sampling gaps, and the various vegetation types that may lead to a maximum and first hand gathering of biodiversity data (Figure 1.7). In general, our sampling design was in two parts: quantitative and qualitative sampling (Figure 1.8). 1.9 Data Analysis Identification of specimens were carried out at the National Herbarium of Cameroon (YA) by matching specimens collected with existing herbarium sheets, and by consulting floras, published documents, and keys for the plants of the region (Hutchinson and Dalziel 1954, 1958, 1963; Vivien and Faure, 1985; Keay1989; Thomas et al. 2003; Cheek et al. 2004; Harvey et al. 2004; Harvey et al. 2010; Onana 2011, 2013). The final checklist was consolidated following the Angiosperm Phylogeny Group (APG III) classification (Judd et al. 1999; Angiosperm Phylogeny Group (APG III, 2009). Basal Area, relative dominance, relative density, relative frequency was calculated using the formulas below. Fisher s alpha, Shannon-Wiener index (H ), and Simpson index were used as indices to compare species diversity among plant forms and elevations using the software package PAST version 2.17 (Hammer et al. 2001). This was based on 12 ha in lowland, 8 ha in mid-elevation, 3 ha in submontane and 2 ha in montane forest. Fisher s alpha was not calculated for lianas in montane forest because the values were too low. The distribution of variation in tree species diversity and carbon per hectare in different forest types was analysed using analysis of variance (ANOVA). The data were converted to binomials (0 and 1) and correspondent 18

45 analysis (CA) was performed to establish the relationship among elevation, species diversity and carbon. Regression analysis was conducted with the aid of PAST version A non-destructive method was used to estimate above-ground biomass, and carbon stock to diameter at breast height, and wood Specific density (WD). Above-ground biomass and carbon were estimated for trees (dbh 10 cm) across forest types using the allometric equation of Chave et al. (2015) and tree height were estimated following Djomo et al. (2016). Eq. (1) Basal Area (BA) = Area occupied by plant at breast height. (BA) = pi*(1/2dbh) 2 = pi*(dbh) 2 /4. Basal area is the area occupied by a species. Eq (2) Eq. (3) Eq. (4). Frequency is the number of quadrats in which a species is found in the entire sample Eq. (5) Importance Value Index (IVI) = Relative density + Relative dominance + Relative frequency Eq. (6) Shannon-Weiner index (H ) is the most convenient tool to measure diversity in 1-ha plots. This was achieved through the following formula: Where pi is the proportion of individual of a species (Number of individual of a species/total number of all species), ln is the natural logarithm. Thus, the natural logarithm of the number of species (lns) is the maximum value of H. Eq. (7) D 2 H) (Chave et al., 2015) Eq. (8) (Djomo et al., 2016) Where AGB is above-ground dry biomass; ρ is wood density; D is dbh; ln is the natural logarithm, and e indicates the exponential function. Wood specific density was assembled from published sources such as the Global Wood Density Database (Dryad identifier: (Zanne et al. 2009), and the African Wood Density Database 19

46 ( (Carsan et al. 2012). Species-specific wood densities were used for individuals identified to the species level. In cases in which species-specific wood densities were not available, mean values for the genus or family were used. For unidentified stems overall mean wood density for the data set was used (Baker et al. 2004). 20

47 Figure 1.1 Cameroon Volcanic Line (Cameroon Mountains) 21

48 Figure 1.2. Climate representations over 10 year s ( ) at Rumpi Hills Forest Reserve, Cameroon 22

49 Figure 1.3 Rainfall Representations for for Kimbi Fungom National Park, Cameroon (Jonathan Abe, personal communication). 23

50 Figure 1.4 Rumpi Hills Forest Reserve Cameroon 24

51 Figure 1.5 Vegetation types based on present study in the Kimbi Fungom National Park, Cameroon 25

52 Figure 1.6 Vegetation types based on Letouzey (1985) in the Kimbi Fungom National Park, Cameroon 26

53 Figure 1.7 Locations of Sample Plots in the Kimbi Fungom National Park and the Rumpi Hills Forest Reserve, Cameroon. 27

54 A Figure 1.8 A & B. Field designs. A. Transect 500 m long (orange) in the middle of two plots, with each plot measuring 500 x 20 m. B x 20 m at high elevation representing 1 ha. The red asterisks at quadrat 1, 6.11,16 and 21 represents locations of nested plots of 10x10 m. 28

55 CHAPTER TWO BOTANICAL SAMPLING GAPS ACROSS THE CAMEROON MOUNTAINS Moses Nsanyi Sainge 1, 2, Jean-Michel Onana 3,4, Felix Nchu 5, David Kenfack 6 and A. Townsend Peterson 7 1 Tropical Plant Exploration Group (TroPEG), P.O. Box 18, Mundemba, Ndian, South West Region, Cameroon 2 Department of Environmental and Occupational Studies, Faculty of Applied Science, Cape Peninsula University of Technology, Cape Town Campus, Keizersgracht, P.O. Box 652, Cape Town 8000, South Africa 3 The National Herbarium of Cameroon. P.O. Box 1601 Yaoundé, Centre Region, Cameroon 4 Department of Plant Biology, Faculty of Science, University of Yaoundé 1. P.O. Box 812 Yaoundé, 5 Department of Horticultural Sciences, Faculty of Applied Science, Cape Peninsula University of Technology, P.O Box 1906, Bellville 7535, South Africa 6 Center for Tropical Forest Science, Smithsonian Institution Global Earth Observatory, Washington, DC Biodiversity Institute, University of Kansas, 1345 Jayhawk Blvd., Lawrence, Kansas USA Corresponding Author: Moses N. Sainge, moses.sainge@gmail.com Tel: (+237) Preface This chapter addresses the first objective of the study, which was to determine the botanical sampling gaps across the continental part of the Cameroon Mountains. ABSTRACT With the emergence of a new field, biodiversity informatics, an important task has been to evaluate completeness of biodiversity information that is existing and available for various countries and regions. This paper offers a first and very basic assessment of sampling gaps and inventory completeness across the Cameroon Mountains. Because digital accessible knowledge is severely limited for the region, we relied on qualitative evaluations of inventory completeness, supplemented by large amounts of data from the National Herbarium of Cameroon (YA) database. Our findings shows that detailed botanical inventories have been developed for Mt Cameroon, the Kupe- Mwoanenguba Mountains, Mt Oku, and the Mambila Plateau, leaving substantial geographic and environmental coverage gaps corresponding to Rumpi Hills, Mt Nlonako, Kimbi Fungom National 29

56 Park, Bali and Bafut Ngemba, Mt Bamboutos, Kagwene, and Tchabal Mbabo. This gap resulted in the selection of Rumpi Hills and Kimbi Fungom National Park for a detailed survey. This paper provides a roadmap for a comprehensive botanical survey for this region. Completing this survey plan, the resulting data will allow researchers to track changes in biodiversity and identify priority areas for conservation on the various mountain ranges that make up this important biodiversity hotspot. We here conclude that the digital Accessible knowledge for Cameroon is limited. Keywords: biodiversity informatics, primary biodiversity data, sampling gaps, inventory completeness, mountains 2.1 INTRODUCTION The Cameroon Mountains, also known as the Cameroon Line (Nono et al., 2004) or Cameroon Volcanic Line (Ayonghe et al., 1999; Marzoli et al., 2000), is comprised of a chain of isolated volcanic and plutonic mountain peaks that covers ~40,877 km 2, stretching from Pagalu Island in the Gulf of Guinea to the Mandara Mountains in the interior. The oceanic portion of the mountain chain covers ~2998 km 2 in the form of four major islands: Pagalu, 17 km 2 ; São Tomé, 854 km 2 ; Principe, 110 km 2 ; Bioko, 2017 km 2 (Ayonghe et al., 1999; Frodin, 2001). The continental portion of the mountain chain is broader, covering ~37,879 km 2, extending from Mt Cameroon (4095 m) and Mt Etinde (commonly called Small Mt Cameroon, 1474 m) on the coast, through a string of peaks, extending to the north and east into the interior (Figure 1). The greatest part of the Cameroon Mountain chain lies within Cameroon, which is the focus of this study, but it extends into adjacent Nigeria at the Mambila Plateau, Obudu Plateau, Mbola Hills, Mt Shebsi, and the Biu Plateau (Hepper, 1965, 1966; Hall and Medler, 1975; Ayonghe et al., 1999, Figure 2.1). Numerous protected areas include portions of the Cameroon Mountain chain, including Mt Cameroon National Park, Bakossi National Park, Santchou Forest Reserve, Kimbi Fungom National Park, Mbam et Djerem National Park, and Vallée de la Mbéré National Park. However, although endemism is high in the region overall, particularly for plants, these protected areas were established based on minimal biodiversity information, and no quantitative conservation prioritization has been develop for the region. Botanists have invested considerable energy in basic surveys of the plants of this region. Mt Cameroon (Thomas and Cheek, 1992; Tchouto, 1995; Cable and Cheek, 1998; Tchouto et al., 1999; Cheek et al., 2006; Onana, 2010a, 2010b, 2011, 2013). Mt Kupe, Mt Mwoanenguba and Bakossi National Park (Thomas, 1993; 30

57 Tchouto and Ebwekoh, 1999; Cheek et al., 2004; Tchiengué, 2004). Mt Oku (Thomas, 1986a, b; Mbenkum and Fisey, 1992; Tame and Asonganyi, 1995; Asonganyi, 1995; Maisels and Forboseh, 1997; Cheek et al., 2000; Maisels et al., 2000; Tame and Thomas, unpubl. data) and the Mambila Plateau (Hepper, 1965 and 1966) have all been the subject of detailed, and reasonably complete botanical surveys. Sites for which at least some information exists include Rumpi Hills (Thomas, 1996; Sainge, 2016), Mt Nlonako (Kenfack, 2001), Lebialem highlands (Harvey et al., 2010; Tchetcha, 2012; Fonge et al., 2013), Bali and Bafut Ngemba (Harvey et al., 2004; Cheek et al., 2010), Tchabal Mbabo, and the Adamawa Plateau (Thomas and Thomas, 1996; Chapman et al., 2004). Areas with scanty botanical information are Mt Lefo, Malap Njibanchi Hills, Nkounchankap Hills, Kagwene, and Faro National Park. More generally, however, the Cameroon Mountains hold tropical montane forests known internationally for their rich flora and high levels of endemism, combined with high levels of threat (Myers et al., 2000; Onana and Cheek, 2011). This biome is part of a mega-hotspot of Western Africa cited as having impressive species diversity and hosting numerous endemic taxa (Mittermeier et al., 1999; Myers et al., 2000; Marchese, 2015). Still, only scanty primary biodiversity data are available for the region, and to our knowledge no quantitative analysis of conservation priority has been develop regarding any taxon for the region. As such, a detailed evaluation of gaps in knowledge regarding the plants of this region is presented here to lay a foundation for more detailed inventories and analyses in terms of structure and composition, species diversity, and richness. Because major sectors of the botanical knowledge of this region remain in non-digital or nonshared formats, our analyses are necessarily qualitative at this point, in hope of transition to digital and quantitative in coming years. The sampling gaps in this region were Rumpi Hills, Mt Nlonako, Kimbi Fungom National Park, Bali and Bafut Ngemba, Mt Bamboutos, Kagwene, and Tchabal Mbabo. These resulted to the selection of Rumpi Hills and Kimbi Fungom National Park for detailed studies. The main goal for this activity is to generate a plant database for this region in the future, understand its forest structure, composition, biomass and the carbon that its can holds per hectare in the different vegetation types. 2.2 MATERIAL AND METHODS We focused on montane areas with elevations >1600 m, and also considered information from those areas down to m. This was due to occurrence of endemics >1600 m and the high diversity and forest structure at m. However, many species found around 1000 m were also distributed over 1500 m up to m elevation. Data on previous botanical surveys across the region were derived from published documents, floras, monographs, reports (Hutchinson and Dalziel, 1954, 1958, 1963; Vivien and Faure, 1985; 31

58 Keay, 1989; Thomas et al., 2003; Cheek et al., 2004; Harvey et al., 2004; Harvey et al., 2010; Onana, 2011, 2013), and the databases of the National Herbarium of Cameroon (YA) (Holmgren et al., 1990). From the latter source, 23,667 records were assessed for accuracy, and records falling in highland areas extracted. This information was summarized in spreadsheets that included latitude, longitude, elevation, number of data records, and number of species documented, but numbers of records and species were unconvincingly low; the bulk of existing information for the plants of the region remains in non-digital formats. As a consequence, we resorted to a qualitative evaluation (our consensus opinion) of completeness of the inventory of each area, as 0 = never visited, 1 = incidental data, 2 = some records, and 3 = a published flora or otherwise comprehensive summary that appears to be reasonably complete. We produced a spatial dataset summarizing the distribution of montane areas within the Cameroon Mountains from the GMTED2010 digital elevation model (DEM; at a spatial resolution of 7.5 (~230 m at the Equator). We reclassified the raw DEM into elevational intervals of m (lowlands), m (foothills), and >1600 m (montane areas), and created vector-format shape-files of areas >1000 m and >1600 m to facilitate analysis. Finally, for analysis, we further subdivided two large areas into smaller subunits based on relatively low valleys that were nonetheless above 1600 m. We focused on areas with completeness at level 3, which we considered at least in a preliminary sense, as well sampled. Established which polygons were to be considered as sampled, the shape file were converted to raster (geotiff) format using custom scripts in R (R Foundation for Statistical Computing, 2004). This raster coverage was the basis for our identification of gaps, the proximity (raster distance) function in QGIS (version 2.4) was used to summarize geographic distances from all montane sites to those that are well sampled random points across the Cameroon Mountains were plotted to create a parallel view of environmental difference from well sampled areas, (i.e., in areas >1600 m), and used the point sampling tool in QGIS to link each point to the geographic distance raster, and to raster data layers at 30 (~1 km) spatial resolution summarizing annual mean temperature and annual precipitation from the WorldClim climate data archive (Hijmans et al., 2005). Rescaled values of each environmental variable to the overall range of the variable as (x i x min) / (x max - x min), where x i is the particular observed value in question, such that each environmental variable varied only 0-1. For all points not falling in well-sampled areas (geographic distance = 0), Euclidean distance were calculated in the twodimensional climate space to all points falling in well-sampled areas; minimum value of these Euclidean distances as the environmental distance to a well-known area were identified (Sousa-Baena et al., 2014). Finally, the environmental distances were imported into QGIS, and linked back to the random point shape-file. 32

59 2.3 RESULTS Among the 23,667 digital herbarium records from YA, 13,609 records correspond to highlands within the Cameroon Mountains region, representing 3995 species (Figure 2.3). Data generated from published papers, reports, floras, and monographs correspond to 3314 records representing 1175 species. The analysis shows significant sampling gaps (sites with few or no sampling history) along this mountain chain, ranging from the Rumpi Hills, Mt Nlonako, the Lebialem Highlands west of Mt Bamboutos, and the Bamenda Highlands (extending to Tchabal Mbabo and the Adamawa Plateau). Our explorations of the YA database (Figure 2.3) indicated that data is sparse for individual montane areas, preventing quantitative analysis (Colwell and Coddington 1994). As a result, qualitative analysis was used. A broad view of the Cameroon Mountains region with our qualitative inventory summary is presented in Figure 2.1. Four highland sites were identified whose inventory are considered well known: Mt Cameroon, Bakossi landscape (Mt Kupe, Mt Mwanenguba and the Bakossi National Park), Mt Oku (Kilum-Ijim ridge), and Mambila Plateau; Figure 2.4). Five sites for which some information exists (Rumpi Hills, Mt Nlonako, Bali Ngemba, Bafut Ngemba, and Tchabal Mbabo), and two sites that remain poorly known (Mt Bamboutos, and Adamawa Plateau). In five sites, there was no indication of any previous botanical work (Kagwene Forest, Nkom-Wum Forest Reserve, Faro Reserve, Kumbo Area, and Rhum Rock). Geographically, the sites most distant from well-known sites were the Adamawa Plateau, and Tchabal Mbabo (Figure 2.2). Interestingly, in terms of environmental distances, the sites most different from well-known sites were quite different from the list based on geographic distances. Specifically, the area most distinct in terms of environments was the Rumpi Hills, Mt Nlonako, and the Lebialem Highlands, which is west of Mt Bamboutos. These sites are centrally located in the Cameroon Mountains chain, and are geographically relatively close to well-known sites (Figure 2.4, Table 2.1). 2.4 DISCUSSION The continental part of the Cameroon Mountains with focus on Cameroon is one of the most diverse sites in Africa, and has been classified as a biodiversity hotspot (Cheek et al., 2000; Cheek et al., 2004; Barthlott et al., 2005). This region is the only part of Central Africa with an elevational range from sea level to over 4000 m, and holds a high diversity of plants > 6000 species out of the almost 9000 species in Cameroon (Cable and Cheek 1998; Cheek et al., 2000; Cheek et al., 2004; Onana and Cheek, 2011; Onana, 2011). At the continental level, this region has great affinity in species composition with other montane sites such as the mountains of 33

60 West Africa (e.g., Bersama abyssinica; Cheek et al., 2004) and East Africa (e.g., Polyscias fulva, Strombosia scheffleri, Schlefflera abyssinica, Alangium chinense, Maesa lanceolata; Dowsett-Lemaire, 1989; Thomas and Thomas, 1996; Cheek et al., 2000; Cheek et al., 2004; Sainge 2016). This region holds >200 species of plants that are considered as threatened, which is the highest in Cameroon and perhaps the highest in West and Central Africa (Cheek et al., 2004; Onana and Cheek, 2011), with >80 species endemic (Cheek et al., 2004; Franke, 2004; Sainge et al., 2005; Sainge et al., 2010; Sainge, 2012; Sainge, 2016). Examples of threatened and endemic plant species occurring in the region are Afrothismia saingei, A. fungiformis, Rhaptopetalium geophylax, Schefflera manni, Syzyzium staudtii, Ixora foliosa, Gambeya korupensis, Deinbollia angustifolia, and Begonia pseudoviola. The region hosts 3 of the 7 genera endemic to Cameroon (Hamilcoa, Medusandra, Platytinospora); finally, the only endemic family in Cameroon (Medusandraceae) is represented here, with its two species: Medusandra mpomiana and M. richardsiana (Cheek et al., 2004; Onana, 2013). However, this landscape has been visited and sampled by botanist mostly in terms of surveys of species occurrences, and most of the data are still in non-digitized formats. This information thus remains inaccessible to scientists with interest on African biodiversity, particularly those based in Africa. We attempted to develop two relatively independent views of botanical inventory completeness across the Cameroon Mountains region. Our attempt at a quantitative analysis (after Sousa-Baena et al., 2014; Idohou et al., 2015; Kouao et al., 2015) was stymied by the small numbers of primary occurrence data that are available for the region, and by the large proportion of such data that remains in non-digital formats in institutions in Europe. As a consequence, in our second effort, we used a literature review to detect and identify landmark studies that have documented Cameroon Mountain sites in good detail (Mt Cameroon, Mt Kupe, Mt Mwoanenguba, Bakossi National Park, Mt Oku, and Mambila Plateau). We found little or no access to the primary data that underlay those publications and that document the individual specimens collected. The 13,609 occurrence data points from the Cameroonian National Herbarium corroborated this view: good numbers of occurrence points concentrated in sites identified as well sampled in the qualitative analysis. In a bigger view, the online digital accessible knowledge (DAK; Sousa-Baena et al., 2014) for Cameroon remains entirely too sparse. With a conservative estimate of ~155,000 specimens (and likely many more) collected to date in the country (Onana, 2011), only ~65,000 (41.9%) are represented in the database of YA (J.M. Onana, unpubl. data.). This gap exists because large-scale collections made from the colonial era into the 34

61 1950s and subsequent decades are deposited in major herbaria in Europe and North America; although not without exceptions, most of these collections remain undigitized and largely inaccessible to the broader community interested in African biodiversity. Although some of the big data-holders have begun steps to make information available (Le Bras et al., 2017), the DAK impediment thus remains significant, with Kew Gardens being the single largest data holder from which no DAK are available. This blockage of information flow must be resolved if any quantitative analyses of biodiversity pattern, subregional endemism, and conservation priority are to be developed for the region. 35

62 Figure 2.1. Diagrammatic representation of the Cameroon Mountain chain, extending from islands in the Gulf of Guinea to the coastal Mt Cameroon, and inland along the Cameroon-Nigeria border. 36

63 Mwanenguba Mountains Figure 2.2. Summary of highland areas in the continental portion of the Cameroon Mountains region. Areas above 1000 m elevation are shown in darker gray shading. Areas above 1600 m are shown colored red and outlined in black. The more important such highland areas are labeled for reference to the text. 37

64 Figure 2.3. Map of the continental portion of the Cameroon Mountains region, showing points sampled in the YA collections in blue. The different highland areas are indicated as well-sampled (dark red), some sampling (medium red), some information (light red) or no information (gray). 38

65 Figure 2.4. Summary of gaps in coverage in terms of inventories of plants in the Cameroon Mountains region. White indicates well-known sites; darker shades of blue indicate greater distance (geographic space) or difference (environmental space) from well-surveyed sites. Left-hand column is the southwestern part of the chain; right-hand column is the northeastern part of the chain. 39

66 Table 2.1. Highland areas in the continental portion of the Cameroon Mountains region with focus in Cameroon Montane Unit Number of species sampled Latitude Longitude Elevation (m) Mandara Mountains Benue National Park Faro Reserve Tchabal Gandaba Tchabal Mbabo Ngel Nyaki Mountain forest Mambila Plateau Rhum Rocks Ijim and mt. Oku forest block Kumbo area Kagwene Bali & Bafut Ngemba Forest Reserve Mt. Bamboutos and Lebialem highlands Mt. Lefo Santchou forest reserve Bakossi National Park Mt Mwaneangumba Rumpi hills Mt. Nlonako Mt. Kupe Mt. Cameroon Degree of documentation 40

67 CHAPTER THREE CHARACTERISATION OF VEGETATION, FLORISTIC COMPOSITION, AND STRUCTURE IN TROPICAL MONTANE FOREST IN THE RUMPI HILLS FOREST RESERVE, CAMEROON Moses Nsanyi Sainge 1,2, Ngoh Michael Lyonga 1, Peguy Tchouto 3, David Kenfack 4, Felix Nchu 5*, and A. Townsend Peterson 6 1 Tropical Plant Exploration Group (TroPEG), P.O. Box 18, Mundemba, Ndian, South West Region, Cameroon. sainge2001@yahoo.com,nm.lyonga1@gmail.com, tropeg.cam@gmail.com. (+237) , (+237) Department of Environmental and Occupational Studies, Faculty of Applied Sciences, Cape Peninsula University of Technology, Cape Town Campus, Keizersgracht, P.O. Box 652, Cape Town 8000, South Africa. sainge2001@yahoo.com, moses.sainge@gmail.com, (+237) Programme for the Sustainable Management of Natural Resources (PSMNR-SWR) South West Region, Cameroon. peguy2000@yahoo.com, (+237) Center for Tropical Forest Science, Smithsonian Institution, Washington, DC , USA. kenfackd@si.edu, Department of Horticultural Sciences, Faculty of Applied Science, Cape Peninsula University of Technology, Bellville Campus, P.O Box 1906, Bellville 7535, South Africa. felixnchu@gmail.com Tel: (+27) Biodiversity Institute, University of Kansas, Lawrence, Kansas USA. town@ku.edu, * Corresponding Author: Felix Nchu, felixnchu@gmail.com Tel: (+27) Preface This chapter addresses in a broad view the vegetation patterns of the Rumpi Hills Forest Reseve. It analyses the vegetation, floristic composition and structure. 41

68 ABSTRACT The Rumpi Hills Forest Reserve (RHFR) is a Montane forest area located in southwestern Cameroon. Although it is thought to hold rich biodiversity and diversed vegetation types, few studies have been undertaken to characterise its floristic composition and vegetation patterns. The study characterised vegetation patterns in the reserve with a view to understanding how elevation influences species distributions and diversity. Also the study intended to provide the first detailed plant species inventory in this important tropical rainforest in Southwestern Cameroon. The floristic composition and vegetation patterns of the reserve were studied in 25 1-ha plots spread along an elevational gradient. In each plot trees and lianas of diameter at breast height (dbh) 10 cm were measured, and shrubs <10 cm were measured in nested plots of 10 x 10 m. Lianas with dbh 1 cm were measured with the help of a caliper and diameter tape. They were measured above the last roots that get into the canopy. In all, 16,761 trees, shrubs and lianas with dbh 1 cm were census, representing 71 families, 279 genera, and 617 morphospecies. Floristic composition ranged between species with a mean of species ha -1 in lowland forest ( m), to species with a mean of 38.5 species ha -1 in montane cloud forest ( m) near the summit of Mount Rata. Two-way Indicator Species Analysis (TWINSPAN) classified the 25 plots into six vegetation types: lowland evergreen rainforest, lowland evergreen rainforest on basalt rocks, mid-elevation evergreen forest, submontane forest, transitional submontane forest, and montane cloud forest. Detrended correspondence analysis (DCA) highlighted the importance of elevation in shaping vegetation patterns, with a strong positive relationship to species turnover. Results of this study revealed that the different vegetation types in the RHFR show impressive variation in terms of structure, species composition, diversity, and spatial distribution. Furthermore, the fine-scale inventory data of species obtained in this study could be useful in developing predictive models for efficient management of tropical rainforests. Key words: Floristic composition, Vegetation patterns, Montane forest, Rumpi Hills, Cameroon Mountains 3.1 INTRODUCTION Floristic composition, forest structure and vegetation characterisation have been treated widely in tropical forest ecosystems (Campbell et al., 2006; Lutz et al., 2012; Noumi, 2013; Neelo et al., 2015; Zhong et al., 2015), with such importance that they need to be obtained for tropical forest zones that are less studied. One such area is the tropical montane forest along the continental part of the Cameroon Volcanic Line (Ayonghe et al., 1999; Marzoli et al., 2000, Sainge et al., 2017),where relatively few studies have assessed forest structure and composition along elevational gradients using permanent sampling plots (Sunderland et al., 2003; 42

69 Tchouto, 2004; Gonmadje at el., 2011; Sainge, 2016), as opposed to the more common general plant collection assessments (e.g., Thomas, 1996; Thomas, 1997; Cheek et al., 2000; Kenfack, 2001; Cheek et al., 2004; Harvey et al., 2004; Harvey et al., 2010). Some studies have examined changes in species composition and diversity across environmental and geographic gradients (Gentry, 1988; Imani et al., 2017) but vegetation structure and composition are also influenced strongly by elevation (Richter, 2008; Imani et al., 2017). Vegetation systems at different elevations on different substrates in montane ecosystems differ in biomass production, carbon storage, and biodiversity conservation value (Richter, 2008). Globally, the concepts of elevational gradients along tropical mountain ranges have been treated (Vallèrié, 1971; Thomas, 1984; Hemp, 2002; Desalegn and Beierkuhnlein 2003; Lovett et al., 2006; Fischer et al., 2011; Rutten et al., 2015). Still, this subject remains little-study in the Cameroon Mountains. The study of elevational gradients in different mountain systems in the tropics began as far back as 1800, with the work of Alexander von Humboldt and Charles Darwin. This early work led many scientists to center research on these questions (Fischer et al., 2011), aiming to understand distributional changes of species along different mountain gradients (Hemp, 2002; Lovett et al., 2006, Rutten et al., 2015). In east Africa, the work of Lovett et al., (2006), Rutten et al., (2015), and Desalegn and Beierkuhnlein, (2003) outlined basic patterns, and classified and delimited the montane forest as starting at m elevation. The Rumpi Hills Forest Reserve (RHFR) covers much of an important patch of wet montane forest in the continental part of the Cameroon Mountains that is considered a critical site for biodiversity conservation (Oates et al., 2004; Birdlife International, 2016). White, (1983) classified the RHFR as part of the Guineo- Congolian region of endemism, and Barthlott et al., (2005) placed it under the Mount Cameroon sub-center of endemism. Despite its high diversity potential, RHFR has received little attention in terms of detailed botanical and ecological surveys, when compared to surrounding mountain masses, such as Mt Nta Ali (Achoundong, 1995; Thomas, 1996), Mt Kupe (Cheek et al., 2004; Tchiengué, 2004), Bakossi National Park and Mt Mwanenguba (Cheek et al., 2004; Tchiengué, 2004; Tchouto and Ebwekoh, 1999), Mt Nlonako (Kenfack, 2001), Mount Etinde (Thomas and Cheek, 1992; Tchouto, 1995), Mt Oku (Tame and Asonganyi, 1995; Maisels and Forboseh, 1997; Maisels et al., 2000; Cheek et al., 2000), Mt Cameroon (Cable and Cheek, 1998; Tchouto et al., 1999), and Tchabal Mbabo (Thomas and Thomas, 1996; Chapman et al., 2004). Understanding the Rumpi Hills ecosystem in detail will thus allow detailed comparisons across the region in the future, as well as identification of crucial conservation elements represented there. 43

70 A few brief botanical collections were made in the Rumpi Hills area between 1976 and 1984, with a total of 57 botanical specimens collected by D. Dang, R. Letouzey, S. Polhill, B. Satabie, and D. Thomas (National Herbarium of Cameroon database). It was not until 1996 that Thomas (1996) did a more intensive, rapid botanical survey of the area. Later, during , botanical reconnaissance trips were made by D. Thomas, D. Kenfack, and M. Sainge, during which 68 specimens were collected and deposited at the Missouri Botanical Garden Herbarium and the National Herbarium of Cameroon. The RHFR and surrounding areas are threatened by encroachment from both small farm estates and large-scale agro-industrial companies (Kupsch et al., 2014). In this study, we aimed to characterise vegetation patterns across RHFR, and to present a first detailed assessment of its vegetation structure and composition. We sought to assemble an understanding of the flora of the region. The specific objectives of this study were thus to understand composition, structure, classification and patterns in the RHFR along an elevational gradient. 3.2 MATERIALS AND METHODS Study area The RHFR lies near the southwestern extreme of the Cameroon Mountain range. It is located in Ndian Division, South West Region, Cameroon, and stretches across latitudes N and longitudes E, with an elevational range of m. It covers an area of km 2 (Forestry Ordinance ). Data were collected in clusters of 1-ha sampling plots: southern plots (numbers 1-8) were located on level terrain at elevations of m; northern plots (numbers 18-25) were located on fairly level terrain at elevations of m; four plots (numbers 14-17) were on basaltic rocks at elevations of m; and eastern plots (numbers 9-13) were on undulating terrain at elevations of m. Plots were sampled based on accessibility and availability of funds (Figure 3.1). The climate of the RHFR is typical of that of equatorial Cameroon, being hot and humid, with two distinct seasons: dry (December to March) and wet (April to November). An annual average rainfall of 5000 mm has been reported for the reserve (Nembot and Tchanou, 1998). Temperature fluctuates with elevation, with coldest temperatures at the top of Mt Rata; although no climate station is located in this area, Nembot and Tchanou, (1998) reported a mean temperature of 22 C for the reserve. This reserve forms a topographic platform for different river sources that supply the Chad, Benue, Sanaga, Congo, and Manyu Rivers (Ngwa, 1978). Rivers originating in this reserve flow in five directions: north into Lake Chad via the Logone River; 44

71 northwest via the Benue River, the Kimbi River, and the Katsina Ala River; southwest into the Gulf of Guinea via the Ndian (Moriba), Moko, Meme, Mungo, and Wouri rivers; southeast via the Kadei River, a tributary of the Congo River; and west into Nigeria via the Munaya and Mbo rivers Field sampling A reconnaissance survey based on topographic and vegetation maps of the reserve (Letouzey, 1985) was carried out to identify homogeneous areas of putatively different vegetation types (Sainge and Cooper, 2014). Data collection was done from February to June 2015, using line transects. Each transect (plot) measured 500 m long x 20 m wide, and was subdivided into 25 quadrats of 20 x 20 m to ease data collection and to reduce error margins. All 625 quadrat in the entired 25 hectare permanent plots were sampled (Condit 1998, Thomas et al., 2003, Sunderland et al., 2003). For each plot, GPS coordinates were recorded for the four corners, including start and end points, via careful, repeated measures to assure accuracy. In each plot, all trees, and lianas with diameter at breast height (dbh, 1.3 m above ground) 10 cm were identified, measured with a diameter tape, tagged, recorded, and mapped using their GPS coordinates. Small trees, shrubs, and lianas with dbh <10 cm were sampled and measured with calipers in 10 x 10 m quadrats located in every fifth 20 x 20 m quadrat in each 1 ha plot. Plots were all permanent with permanent poles and GPS coordinates of the four corners recorded (Condit, 1998, Thomas et al., 2003). The forest was divided into 4 vertical strata: trees <10 cm dbh as understory, cm dbh as mid-canopy, cm dbh as canopy and 60 cm dbh as emergent species. This was to have an idea on the degree of regeneration of the forest. This was more accurate than the height because dbh were measured using standard dbh tapes and calipers whereas height was based on estimate. Plots were set based on accessibility to sample different vegetation types and elevations (Figure 3.1). To assess the sampling effort of our study, a sampling intensity of 0.006% was carried out (Branthomme, 2004). Finally, we recorded non-plot-based, observational data (general plant collection) to detect and include species not recorded in the standardised plots Taxonomy and plant identification In the field, plant identification was done using five-letter codes, including the first three letters of the genus and the first two letters of the species. In cases where the genus and species were not known or only the genus or family was known, arbitrary codes were generated to represent morphospecies. For unknown species (those that could not be identified, partly identified, or with doubtful identification), herbarium specimens were collected. These specimens were labeled, pressed, dried, sorted, and classified for proper identification at the National Herbarium of Cameroon, in Yaoundé (YA). Flowers and fruits were collected when the species was 45

72 possibly new to science, endangered, or endemic to the area. In the field two Botanists (Sainge Moses and Mambo Peter) were responsible for all plant identification, and in the herbarium, Sainge Moses and Mezili Paul carried out detailed identification of all specimens. Identification in the herbarium was accomplished by comparing and matching specimens with existing collections and available floras and monographs (Hutchinson and Dalziel, 1954, 1958, 1963; Vivien and Faure, 1985; Keay, 1989; Thomas et al., 2003; Cheek et al., 2004; Harvey et al., 2004; Harvey et al., 2010; Onana, 2011, 2013). Plant classification followed species lists in the Angiosperm Phylogeny Group (APG III, 2009), with the Papilionaceae, Caesalpiniaceae and Mimosaceae merged into Fabaceae, and Sterculaceae, Tiliaceae, and Malvaceae merged into Malvaceae (Judd et al., 1999; APG III, 2009) Data analysis Regression analysis was used to assess relationships between numbers of species and elevation, in PAST, version 2.17 (Hammer et al., 2001). Individuals not identified to the species level (1271 individuals with dbh 10 cm, 10.6%) and singletons (57 individuals, 0.5%) were excluded from the inventory completeness calculations. Thus, 10,709 fully identified individual trees (dbh 10 cm, 88.9%) of 311 species were used in the analysis. Classification of the vegetation was achieved using two-way indicator species analysis (TWINSPAN; Hill, 1979). Detrended Correspondence Analysis (DCA or DECORANA; Hill and Gauch 1980) was used to examine relationships between vegetation types and elevation, via PC-ORD for Windows, version 5.10 (McCune and Mefford, 2006). Forest structure and composition were described using basal area, relative density, relative basal area, relative frequency, and importance value index (Dallmeier, 1992). Inventory completeness was assessed using EstimateS version (Colwell, 2013: via the Chao2 estimator of expected species richness (S exp), which is calculated from the number of species actually known (S obs), and frequency of detection of rare species; completeness was calculated as: 46

73 3.3 RESULTS Species accumulation curve and inventory completeness Inventory completeness varied considerably among plots, from a low of 0.36 in lowland plot 2 to 1.0 in lowland plot 1, with an overall mean of Most of the plots had completeness values in the range of , resulting in a strong positive relationship between estimated and observed species across the Rumpi Hills Forest Reserve, with an inventory completeness of about 80%.(Figure 2A, B, and C) Composition and Floristic Structure In lowland evergreen rainforest, 4086 individual trees were recorded, while we recorded 3600 in mid-elevation evergreen forest, 1831 in lowland evergreen rainforest on basalt rocks, 1191 in submontane forest, 1066 in montane cloud forest, and 263 in transitional forest. For all vascular plants 10 cm, mean number of trees per hectare were not the same across vegetation types: 596 in submontane forest (range ), 533 in montane cloud forest (range ), 511 in lowland evergreen rainforest (range ), 458 in lowland evergreen rainforest on basalt rocks (range ), 450 in mid-elevation evergreen forest (range ), and 263 in transitional forest (263) (Table 3.1). The mean number of shrubs/0.05 ha varied as follows: 180 shrubs/0.05 ha in lowland evergreen rainforest (range ), 146 in lowland evergreen forest on basalt rocks ( ), to 71 in montane cloud forest (59-83). The mean number of lianas ranged from 6 lianas per hectare (range 0-12) in lowland evergreen rainforest, 5 in mid-elevation evergreen forest (0-11), to no lianas in montane or transitional forest at dbh 10 cm. Mean basal area ranged from 37.5 m 2 per hectare in lowland evergreen rainforest (range m 2 ha -1 ), to 34.4 m 2 ha -1 in montane cloud forest ( ) (Table 1), for trees 10 cm dbh. For shrubs <10 cm dbh, mean basal areas were low, ranging from 0.27 in lowland evergreen rainforest (range m 2 /0.05 ha), to 0.09 m 2 ha -1 in montane cloud forest ( m 2 ha -1 ). Lianas 10 cm dbh ranged from basal areas of 0.11 m 2 ha -1 in lowland evergreen rainforest to 0.01 m 2 ha -1 in submontane forest. Species composition by family for trees 10 cm dbh was Fabaceae (954 individuals of 58 species), Rubiaceae (310 individuals of 35 species), Annonaceae (422 individuals of 26 species), and Sapotaceae (385 individuals) and Malvaceae (381 individuals) with 19 species each. Importance in terms of family (importance value index) for the most common 15 families was highest in Lecythidaceae (39.8), Fabaceae (20.9), Phyllanthaceae (19.9), and Malvaceae (13.7), and lowest in Sapotaceae (6.3), Ebenaceae (5.3), Sapindaceae (4.7), and Melastomataceae (1.8). The most broadly distributed family was Fabaceae, occurring in 390 of 625 quadrats 47

74 sampled, whereas the rarest families in terms of number of species and abundance were Dilleniaceae, Malpighiaceae, Erythroxylaceae, and Lepidobotryaceae, each represented by a single individual of a single species. Lower-elevation vegetation types held many more species than highest-elevation vegetation types. The linear multivariate model reveals a strong significant negative relationship between species richness and elevation (r 2 = P = 0.005; Figure 5), with numbers declining from species ( species) in lowland rainforest to 38.5 species (36-41 species) at highest-elevation sites (Table 1). In all, 16,761 individuals of trees, shrubs, and woody lianas were recorded in 25 1-ha plots across the RHFR; casual observational data provided records of an additional 254 individual plants in 62 families, 129 genera, and 210 species. Indeed, 132 species were recorded only outside of the standardized sampling plots and quadrats. A total of 109 individuals in 21 morphospecies (97 trees/shrubs, and 12 lianas) could not be identified to species level, and 21 individuals could not be identified even to family. The most common 15 families accounted for 62% of total individual abundance across all plots. Species composition changed drastically between vegetation types, as only a few species occurred in all vegetation types: Bridelia grandis (Phyllanthaceae), Cola verticillata (Malvaceae), Sapium ellipticum (Euphorbiaceae), and Symphonia globulifera (Clusiaceae). At the other end of the spectrum, many species were recorded only in single vegetation types, such as Afrostyrax kamerunensis, Afzelia bipindensis, A. pachyloba, Alexis cf. cauliflora, Allanblackia gabonensis, and Allophylus megaphyllus in lowland forest; Beilschmiedia gabonensis, Leptonychia lasiogyne, Maesa kamerunensis, M. lanceolata, Syzygium staudtii, and Tricalysia amplexicaulis in submontane forest; Alangium chinense, Alchornea floribunda, and Elaeophorbia drupifera in transitional submontane forest; and Acridocarpus macrocalyx and Allophylus grandifolius in midelevation evergreen forest Multivariate analysis Classification The 25 1-ha plots were classified in the TWINSPAN analysis into six groups at 50% similarity (Figure 3.3). Plots 1-8 corresponded to lowland evergreen forest (sensu Letouzey 1968, 1985), characterized by an abundance of Oubanguia alata (556 individual trees). Plots grouped together, corresponding to lowland evergreen forest on basalt rocks that was abundant in Crateranthus talbotii (103 individual trees). Plots clustered together, and can be termed mid-elevation forest, abundantin Strombosia grandifolia (324 individual 48

75 trees) and Leonadoxa africana (192 individual trees). Plots 9-10 were in submontane forest abundant in Tabernaemontana ventricosa (102 individual trees), Cola verticilata (75 individual trees), and Dasylepis thomasii (68 individual trees). Plots were in montane cloud forest abundant in Strombosia sp. (146 individual trees), Carapa oreophila (125 individuals), and Xylopia africana (121 individual trees). Lastly, plot 13 was in transitional submontane forest rich in Macaranga sp. (56 individual trees), Trema orientalis (46 individual trees), Bridelia grandis (33 individual trees), and Pauridiantha viridiflora (31 individual trees). Ordination The floristic dataset of 25 plots was also subjected to DCA analysis and plotted along axes 1 and 2. Variation was expressed along axis 1 with an eigenvalue of and a gradient length of 4.183, which reflects high variation among vegetation types and species composition. Vegetation types 4, 5, and 6 (submontane, montane, and transitional forest, respectively) separated toward the positive side of DCA axis 1, whereas vegetation types 1, 2, and 3 (lowland, basalt, and mid-elevation respectively) separated toward the negative end (Figure 3.4). DCA axis 2 showed a weaker eigenvalue of 0.478, with a gradient length of (Table 3.2). Figure 3.4 shows patterns suggesting that vegetation types 1-3 are more closely related than vegetation 4-6. A high species-environment correlation for axis 1 indicates a strong association between vegetation types and elevation (Table 3.3), which can be verified from the biplot record (Table 3.4). A Monte Carlo permutation test (998 runs) with an eigenvalue of axis 1 and significant at P < confirmed the strong relationship between species composition and elevation (Table 3.4) Vegetation patterns Lowland evergreen rainforest rich in Oubanguia alata This vegetation type occurs along the southern edge of RHFR, mainly at elevations below 250 m. The vegetation is intact, evergreen, and continuous from the ground layer of herbaceous plants in dense undergrowth to emergent tree species. The canopy is more or less continuous, with only a few emergents. In all, 4086 trees were recorded, belonging to 51 families, 174 genera, and 291 species. Seven species were not fully identified, and 2 individuals were not identified either to family, genus or species. Shrubs with dbh <10 cm totaled 1065 individual trees in 44 families, 126 genera, and 208 species. Only one individual of this group of plants was not identified to genus or species. Lianas of <10 cm dbh included 7 individuals in 3 species, 2 genera, and 2 families. 49

76 Dominant tree families in lowland evergreen rainforest were Fabaceae (61 species), Annonaceae (26 species), and Euphorbiaceae, Malvaceae, and Phyllanthaceae, with 19 species each. Frequent genera included Cola (14 species), Diospyros (13 species), and Drypetes and Trichoscypha (10 species each). Our TWINSPAN analysis presented Oubanguia alata (556 individuals or 13.6% of total individuals in this vegetation type), Protomegabaria stapfiana (263 individuals, 6.4%), and Korupodendron songweanum (164 individuals, 4%), as dominant species in this vegetation type. Some emergent and upper canopy species were m tall, with huge buttresses that spanned 3-10 m on the forest floor, and trunk diameters of up to 2.5 m. Such tree species included Microberlinia bisulcata, Irvingia gabonensis, Desbordesia glaucescens, Pycnanthus angolensis, Saccoglottis gabonensis, Omphalocarpum cf elatum, and Engomegoma gordonii. The canopy, m tall, was composed of tree species such as Tetraberlinia bifoliolata, Korupodendron songweanum, Santiria balsamifera, Eriocoelum macrocarpum, and Pellegriniodendron diphyllum. The midstory was made up of small trees, mostly immature emergent and canopy trees, of approximately 10 m tall, composed of Uvariopsis bakeriana, Craterispermum aristatum, Napoleonaea talbotii, Deinbollia angustifolia, Phyllobotryon spathulathum, and Gaertnera bieleri. The ground layer, composed of herbaceous species, seedlings of the above layers, lianas, and creepers, was dominated by Deinbollia angustifolia Impatiens niamniamensis, Costus afer, Anubias barteri, Begonia quadrialata, and Palisota hirsuta. Seven large woody liana species were recorded in this forest type, of which Connarus staudtii, Dichapetalum affine, Rhaphiostylis beninensis, Strychnos tricalysioides, and Santalinoides afzelii were most common Lowland evergreen rainforest on basalt rocks rich in Crateranthus talbotii This vegetation type occurs along the western edge of the RHFR, at elevations of m, and dominated by Crateranthus talbotii (103 individuals, 5.6%), Trichilia prieureana (65 individuals, 3.5%), and Anthonotha macrophylla (64 individuals, 3.4%). This vegetation type is characterised by trees, shrubs, and lianas with dbh <22 cm, and is made up of patches of huge basalt rocks. In all, 1831 trees with dbh 10 cm were recorded in this vegetation type, in 39 families, 122 genera, and 175 species. Two tree species could not be identified to genus. Shrubs with dbh <10 cm amounted to 40 families, 85 genera, and 119 species. This forest type was characterised by a rich diversity of Fabaceae (234 individuals of 21 species), Rubiaceae (29 individuals of 11 species), Phyllanthaceae (98 individuals of 10 species), and Anacardiaceae (72 individuals of 10 species). 50

77 Mid-elevation evergreen forest rich in Strombosia grandifolia This vegetation type occurs in the northern parts of RHFR, at elevations of m, and is dominated by Strombosia grandifolia (324 individuals) and Leonadoxa africana (192 individuals). We sampled 3600 trees in 42 families, 119 genera, and 176 species, and 1153 shrubs in 41 families, 98 genera, and 150 species. Most frequent shrub genera were Strombosia, Leonodoxa, Trichilia, and Tabernaemontana; the most common species were Strombosia grandifolia, Oubanguia alata, Trichilia prieureana, Tabernaemontana brachyantha, and Mammea africana Submontane forest rich in Tabernaemontana ventricosa This forest type was found along the eastern edge of RHFR at m elevation. It occurs close to the villages of Dikome Balue, Madie, and Iyombo, and along the western edge of the reserve around Mbange, Bossunga, Boa Yenge, and Motindi village. RHFR submontane forest occurs on steep cliffs and inaccessible slopes. We sampled 1191 trees of 38 families, 75 genera, and 98 species; shrubs totaled 240 individuals in 32 families, 49 genera, and 60 species. This forest type was rich in Rubiaceae (9 species), Euphorbiaceae (7 species), and Meliaceae (6 species) for trees; common species included Tabernaemontana ventricosa, Dasylepis thomasii, Cola verticillata, Diospyros cinnabarina., and Garcinia conrauana. The submontane forest forms part of a mosaic of forest and grassland savanna above m. The canopy here is relatively short (20-25 m tall), with only a few tall trees, such as Cylicomorpha solmsii, Eriocoelum macrocarpum, Guarea cedrata, and Sapium ellipticum, reaching m tall. Two species of large woody lianas were recorded: Dichapetalum heudelotii, and Salacia pyriformis Transitional submontane forest rich in Trema orientalis This vegetation type is a mosaic of forest and grassland occurring along the eastern edge of RHFR at m, close to Dikome Balue and is characterized by low diversity. We sampled 263 individual trees in 20 families, 31 genera, and 32 species with dbh 10 cm, and 61 individual trees in 16 families, 22 genera, and 24 species with dbh <10 cm. Dominant tree families with dbh 10 cm were Euphorbiaceae and Rubiaceae (4 species each); all other families were represented by only 1-2 tree species. Dominant genera were Macaranga, Trema, Pauridiantha, Bridelia, and Margaritaria; most common species included Macaranga sp., Trema orientalis, Pauridiantha viridiflora. Rare species recorded here included Thunbergia affinis, Alangium chinense, Trichoscypha preussii, Monodora myristica, and Uvariastrum pynaertii. 51

78 Montane cloud forest rich in Carapa oreophila Montane forest is found on the eastern edge of RHFR, at elevations above 1600 m, up to the summit of Mount Rata (~1778 m). We sampled 1066 individual trees with dbh 10 cm, in 25 families, 38 genera, and 48 species, and 146 individual trees with dbh <10 cm in 19 families, 24 genera, and 31 species. This forest was characterised by species of Rubiaceae (6 species), Clusiaceae (5 species), and Salicaceae (4 species). Most abundant species included Strombosia sp. (146 individuals, 13.7%), Carapa oreophila (125 individuals, 11.7%), Xylopia africana (121 individuals, 11.4%), and Dasylepis thomasii (91 individuals, 8.5%). The upper canopy layer was m tall, with most species covered with numerous species of epiphytes (orchids, mosses, lichens, liverworts, bryophytes, and Piperaceae). Understory vegetation was sparse. Woody lianas were scarce, with only two species recorded (Dichapetalum rudatisii, and an unknown species). 3.4 DISCUSSION The diverse and heterogeneous vegetation structure and composition in RHFR are likely related to the complex physical features, elevational differences (from 50 m to the top of Mount Rata at 1778 m), and climatic factors, such as the southwest monsoonal (warm wet) winds that are weak in the dry season and strong in the wet season (Ngwa, 1978; Neba, 1999). Rainfall is highest in the southwestern corner of the reserve; however, to date, no weather station has been installed to provide detailed climatic data for the reserve. We assume that, given its proximity to Korup National Park, with an average yearly rainfall of ~5 m (Newbery and Gartlan 1996; Chuyong et al., 2000; Thomas et al., 2003; Chuyong et al., 2004), RHFR also has high rainfall. Temperature within the reserve is variable, with temperatures lowest at the top of Mount Rata. The reserve per se is free from human settlements, as no villages are located within its core area. However, 12 villages are within 1-3 km of the reserve margins: Matamani in the northwest; Mata in the north; Madie and Dikome Balue in the east; Munyange and Nalende in the south; Mbange, Bossunga, Motindi, Lipenja Mukete in the west; and Meka and Besingi in the northwest. The above parameters (i.e., the enclaved nature of RHFR, its importance as a watershed, and lack of human settlements) are indicators of conservation importance. This importance is emphasized by our findings of high tree, liana, and shrub diversity (Table 3.1), and the large number of species of high conservation priority and the fact that it occurs in a recognized biodiversity hotspot (White, 1983; Myers et al., 2000; Barthlott et al., 2005; Onana, 2013; Marchese, 2015). 52

79 The sampling intensity of 0.006% is actually low compared to the standard for management inventory procedure for Cameroon % with a sampling unit of one hectare (Branthomme, 2004) Vegetation Patterns and Floristic Composition Multivariate analyses (TWINSPAN and DCA) classified RHFR vegetation into six types. Letouzey, (1985) recognized seven vegetation types in RHFR: the Atlantic Biafran forest; Atlantic littoral forest, (which in the current study is classified as lowland forest); piedmont forest; degraded submontane forest; submontane forest; highly degraded evergreen forest, (which was not sampled in the present study); and submontane grassland (also not sampled), as well as various combinations of fallow, grazed, and human-inhabited areas (Figure 3.1). The RHFR is part of the chain of mountains of Cameroon and Nigeria that includes the Cameroon Mountains and associated highland biomes (Burgess et al., 2007; Cronin et al., 2014). It forms part of the Lower Guinea Forest, with high levels of species richness and endemism (Barthlott et al., 2005; Plumptre et al., 2007; Burgess et al., 2007). The occurrence of a mosaic of forest and grassland on the upper slopes of Mount Rata, at elevations of m, is not surprising, as grassland savanna begins at 1500 m in the Takamanda Forest Reserve, in the South West Region of Cameroon (Sunderland et al., 2003), and above 2000 m on Mount Cameroon (Richards, 1963). Administratively, Mount Rata falls outside of the RHFR; and represents the highest peak in the Rumpi Hills range. Our results differ from patterns in lowland forest at other sites, like Takamanda Forest Reserve (South West Region, Cameroon) where lowland forest is dominated by Huaceae (Afrostyrax kamerunensis) and Irvingiaceae (Irvingia, Klainedoxa, Desbordosea) (Sunderland et al., 2003). However, the RHFR lowland forest showed the same dominance trends as the lowland forest in nearby Korup National Park (Thomas et al., 2003). Forest structure changed from lowland evergreen forest ( m), with huge, emergent trees m tall, to montane cloud forest (1778 m), with a lower and more even canopy m tall, comprising tree with branches covered by Piperaceae, Orchidaceae, ferns, liverworts, lichens, etc. Our results agree with Letouzey, (1985) that RHFR is composed of different vegetation types, and show that these vegetation types demonstrate impressive variation in structure, species composition, and distribution. Furthermore, RHFR contains a distinct montane vegetation type, as defined by the TWINSPAN analysis, at elevations above 1600 m. This result 53

80 concurs with Vallèrié, (1971) and Thomas, (1984), who both classified upper montane forest as starting from 1600 m in the Cameroon Mountains region Elevation The effect of elevation on the vegetation of the RHFR was pronounced, as it influences vegetation pattern, vegetation structure, species diversity and species composition of the area (Figure 3.3) across an elevational range of m. We documented marked changes in species composition with elevation: lowland evergreen rainforest on basalt and lowland evergreen rainforest rich in O. alata were relatively richer in species than the other vegetation types. The inverse relationship between species richness and elevation recorded in RHFR was consistent with results obtained in many studies (Hamilton, 1975; Henrik et al., 2006; Chuyong et al., 2011). Dauby et al., (2013) investigated tree diversity patterns in communities of evergreen forest trees in five landscapes of western Gabon, and concluded that mean alpha and gamma diversities were much higher in the hilly region, with differences in elevation explained a significant part of species turnover. The decrease of alpha diversity with elevation within the hilly region could be associated with mass effects, which are expected to enrich valley and slopes (Dauby et al., 2013). Overall, in RHFR lowland forest is characterized by large trees with huge buttresses and lianas (Tchouto, 2004, Ewango, 2010). At high elevations, shrubs and tree branches are covered with bryophytes and vascular epiphytes, and tree boles and leaves are covered by moss and liverworts, with a minimal liana population. Our detailed sampling across vegetation types and elevations within and near the RHFR makes our data useful both for ecological understanding and for guiding management decisions. Given that our plots are permanent, with GPS-based outlines of each sampling plot ( the opportunity arises to repeat these censuses in the future (e.g., every 5-10 years) to understand the dynamics of the forest (Condit, 1998). Such detailed monitoring would allow a far more nuanced understanding of the status and condition of these forests, as well as of the effects of global change on their composition and structure Conservation Implications Conservation prioritisation usually takes in consideration a number of factors for designing conservation areas, including species composition, structure, vegetation patterns, and socioeconomic and cultural importance of the sites. Our study confirmed that the RHFR is a rich site in terms of vegetation; analysis of species composition segregated the community into six vegetation types, some of which are found outside of the administrative boundaries of the reserve (e.g., Mt Rata). Our records of rare species, such as Deinbollia 54

81 angustifolia, Korupodendron songweanum, Gambeya korupensis, and Oubanguia alata, in lowland forest; Crateranthus talbotii in lowland basalt forest; Rhaptopetalum geophylax and Cylicomorphia solmsii in submontane forest; and Carapa oreophila, Oncoba lophocarpa, and Xylopia africana in montane forest may upgrade the conservation interest in the Rumpi Hills forest area. Particular species occured only in specific vegetation types, or in a few adjacent types. Hence protection of the different vegetation types including Mt Rata is paramount for conservation prioritisation. Figure 3.1 Permanent sampling plot locations in different vegetation types across Rumpi Hills Forest Reserve, Cameroon. 55

82 56

83 Figure 3.2 (A) Species area curve for observed species richness (S obs), and estimated species richness (S exp), (B) inventory completeness for observed species richness (S obs) and (C) a strong linear relationship between estimated and observed species in the Rumpi Hills Forest Reserve, Cameroon 57

84 Figure 3.3 TWINSPAN dendrogram of 311 species of vascular plants with dbh 10 cm in 25 1-ha plots in the Rumpi Hills Forest Reserve, Cameroon. 58

85 Figure 3.4 Floristic Detrended Correspondance Analysis (DCA) with 25 plots and six vegetation type from TWINSPAN analysis tree species data following field sampling of the Rumpi Hills Forest Reserve, Cameroon 59

86 Number of species Elevation (m) Figure 3.5 Association between elevation and numbers of species of trees of dbh >10 cm recorded in 25 1-ha plots in the Rumpi Hills Forest Reserve, Cameroon. The relationship has a slope of 78.3%, which is significantly different from a slope of zero (P < 0.05). 60

87 Table 3.1 Summary of number of hectares sampled, mean number of species, mean tree density, and mean basal area of trees 10 cm dbh recorded in each vegetation type in the Rumpi Hills Forest Reserve, Cameroon. Vegetation type No of ha Mean # species Mean # tree Mean BA/ha density/ha Lowland evergreen (94-132) ( ) ( ) rainforest Lowland evergreen ( ) ( ) 30.8 ( ) rainforest on basalt rocks Mid-elevation rainforest (73-88) 450 ( ) ( ) Submontane forest 2 75 (68-82) ( ) ( ) Transitional submontane 1 32 (32) 263 (263) (15.93) Forest Montane cloud forest (36-41) 533 ( ) ( ) Mean # species = Mean number of species, Mean # tree density/ha = Mean number of trees density/ha, Mean BA/ha = Mean Basal Area/ha Table 3.2 Summary of Detrended Correspondence Analysis (DCA) of 25, 1-ha plots in the Rumpi Hills Forest Reserve, Cameroon Axis 1 Axis 2 Axis 3 Total inertia DCA Eigenvalue Length of gradient Inter-set correlation Correlation Variance 1. Elevation

88 Table 3.3 Six vegetation types generated from TWINSPAN with vegetation code, vegetation types and elevational range in the Rumpi Hills Forest Reserve, Cameroon. Vegetation code Vegetation type Elevation (m) Lowland Lowland evergreen rainforest abundant in Oubanguia alata Basalt Lowland evergreen rainforest on basalt rocks abundant in Crateranthus talbotii Mid-Elevation Mid-elevation rainforest abundant in Strombosia grandifolia Submontane Submontane forest rich in Tabernaemontana ventricosa Montane Montane cloud forest abundant in Carapa oreophila Transitional Transitional submontane Forest abundant in Trema orientalis Table 3.4 Monte Carlo test results showing eigenvalue, species elevational correlation, correlation and biplot scores obtained following field sampling of the Rumpi Hills Forest Reserve, Cameroon Randomized Data Real data Monte Carlo test Axis Eigenvalue Mean Minimum Maximum P Monte Carlo Test. species environmental correlation Real data Randomized data Axis spp-envt corr. mean minimum maximum P Correlation and biplot scores Correlation Biplot scores Variable Axis 1 Axis 2 Axis 3 Axis 1 Axis 2 1 Elevation

89 Table 3.5 Permanent sample plot locations in the Rumpi Hills Forest Reserve, Cameroon Plot Vegetation type Location Latitude (N) Longitude (E) Elevation (m) 1 Lowland Munyange Lowland Munyange Lowland Munyange Lowland Munyange Lowland Munyange Lowland Munyange Lowland Munyange Lowland Munyange Submontane Madie River Submontane Madie River Montane Mt Rata Montane Mt Rata Transitional Submontane Mt Rata Basalt Matamani Basalt Matamani Basalt Matamani Basalt Matamani Mid-Elevation Matamani Mid-Elevation Matamani Mid-Elevation Matamani Mid-Elevation Matamani Mid-Elevation Matamani Mid-Elevation Matamani Mid-Elevation Matamani Mid-Elevation Matamani

90 Appendix 3.1. Species occurrence in the different forest types in the Rumpi Hills Forest Reserve, Cameroon Family Species Acanthaceae Thunbergia affinis Achariaceae Dasylepis thomasii Obama & Breteler Achariaceae Scottellia klaineana Pierre Alangiaceae Alangium chinense (Lour.) Harms Anacardiaceae Antrocaryon micraster A.chev. & Guillaum. Anacardiaceae Antrocaryon sp. Anacardiaceae Lannea welwitschii (Hiern) Engl. Anacardiaceae Pseudospondias microcarpa (A.Rich.) Engl. Anacardiaceae Sorindeia grandifolia Engl. Anacardiaceae Sorindeia juglandifolia (A.Rich.) Planch.ex Oliv. Anacardiaceae Trichoscypha acuminata Engl. Anacardiaceae Trichoscypha oliveri Engl. Anacardiaceae Trichoscypha klainei Anacardiaceae Trichoscypha patens (Oliv.) Engl. Anacardiaceae Trichoscypha preussii Engl. Anacardiaceae Trichoscypha sp.10 Anacardiaceae Trichoscypha sp.12 Anacardiaceae Trichoscypha sp.22 Anacardiaceae Trichoscypha sp.31 Anacardiaceae Trichoscypha sp.4 Anacardiaceae Trichoscypha sp.6 Anacardiaceae Trichoscypha sp.72 Anacardiaceae Trichoscypha sp.9 Anacardiaceae Trichoscypha sp.i Anisophylleaceae Anisophyllea meniaudii Aubr. & Pelleg. Anisophylleaceae Anisophyllea polyneura Floret Anisophylleaceae Anisophyllea purpurascens Hutch. & Dalziel Anisophylleaceae Anisophyllea sororia Pierre Anisophylleaceae Poga oleosa Pierre Annonaceae Annickia chlorantha (Oliv.) Setten & P.J. Maas 64

91 Annonaceae Cleistopholis patens (Benth.) Engl. & Diels Annonaceae Cleistopholis staudtii Engl. & Diels Annonaceae Hexalobus sp. Annonaceae Isolona campanulata Engl. & Diels Annonaceae Isolona thonneri (De Wild & Th. Dur.) Engl. & Diels Annonaceae Monodora brevipes Benth. Annonaceae Monodora myristica (Gaertn.) Dunal Annonaceae Pachypodanthium staudtii Engl. & Diels Annonaceae Piptostigma oyemense Pellegrin Annonaceae Piptostigma pilosum Oliv. Annonaceae Piptostigma sp. Annonaceae Polyanthia suaveolens (Engl. & Diels) Verdc Annonaceae Polyceratocarpus parviflorus (Baker f.) Ghesq. Annonaceae Uvariastrum pynaertii De Wild. Annonaceae Uvariodendron connivens (Benth.) R.E. Fr. Annonaceae Uvariodendron giganteum (Engl.) R.E. Fr. Annonaceae Uvariopsis bakeriana (Hutch. & Diels) Robyns & Ghesq. Annonaceae Uvariopsis korupensis Gereau & Kenfack Annonaceae Uvariopsis submontana Kenfack, Gosline & Gereau Annonaceae Xylopia acutiflora (Dunal) A.Rich. Annonaceae Xylopia aethiopica (Dunal) A.Rich. Annonaceae Xylopia africana (Benth.) Oliv. Annonaceae Xylopia hypolampra Mildbr. Annonaceae Xylopia sp.3 Annonaceae Xylopia sp.i Annonaceae Xylopia staudtii Engl. & Diels Annonaceae Xylopia villosa Chipp Apocynaceae Alstonia boonei De Wild Apocynaceae Funtumia elastica (Preuss) Stapf Apocynaceae Hunteria umbellata (K. Schum) Hallier f. Apocynaceae Landolphia congolensis (Stapf.) Pichon Apocynaceae Landolphia dulcis (Sabine) Pichon 65

92 Apocynaceae Apocynaceae Apocynaceae Apocynaceae Apocynaceae Apocynaceae Apocynaceae Apocynaceae Apocynaceae Apocynaceae Apocynaceae Apocynaceae Apocynaceae Araliaceae Araliaceae Araliaceae Asteraceae Asteraceae Bignoniaceae Bignoniaceae Bombacaceae Bombacaceae Boraginaceae Burseraceae Burseraceae Burseraceae Burseraceae Caricaceae Cecropiaceae Cecropiaceae Cecropiaceae Celastraceae Landolphia landolphioides (Hallier f.) A.Chev. Pleiocarpa bicarpellata Stapf. Pleiocarpa rostrata Benth. Pleiocarpa sp. Rauvolfia caffra Sond. Rauvolfia mannii Stapf. Rauvolfia vomitoria Afzel. Tabernaemontana brachyantha Stapf. Tabernaemontana crassa Benth. Tabernaemontana ventricosa Hochst. ex A.DC. Tabernanthe iboga Baill. Voacanga africana Stapf. Voacanga psilocalyx Pierre ex Stapf. Polycias fulva (Hiern) Harms Schefflera abyssinica (Hochst.ex. A.Rich.) Harms Schefflera barteri (Seem) Harms Vernonia conferta Benth. Vernonia frondosa Oliv. & Hiern Allophylus grandifolius (Baker) Radlk. Kigelia africana (Lam.) Benth. Bombax buenopozense P. Beauv. Ceiba pentandra (L.) Gaertn. Cordia sp. Canarium schweinfurthii Engl. Dacryodes edulis (G. Don) H.J.Lam Dacryodes klaineana (Pierre) H.J.Lam Santiria balsamifera Oliv. Cylicomorpha solmsii (Urb.) Urb. Musanga cecropioides R.Br. ex Tedlie Myrianthus arboreus P. Beauv. Myrianthus preussii Engl. Salacia loloensis Loes 66

93 Celastraceae Salacia pyriformis (Sabine) Steud. Chrysobalanaceae Chrysobalanus icaco L. Chrysobalanaceae Dactyladenia pallescens (Baill.) G.T. Prance & F. White Chrysobalanaceae Dactyladenia staudtii (Engl.) G.T.Prance & F.White Chrysobalanaceae Magnistipula aff. cuneatifolia Hauman Chrysobalanaceae Magnistipula glaberrima Engl. Chrysobalanaceae Maranthes kerstingii (Engl.) G.T.Prance Chrysobalanaceae Parinari chrysophylla Oliv. Chrysobalanaceae Parinari excelsa Sabine Clusiaceae Allanblackia gabonensis (Pellegr.) Bamps Clusiaceae Endodesmia calophylloides Benth. Clusiaceae Garcinia cf polyantha Clusiaceae Garcinia conrauana Engl. Clusiaceae Garcinia gnetoides Hutch. & Dalziel Clusiaceae Garcinia kola Heckel Clusiaceae Garcinia mannii Oliv. Clusiaceae Garcinia polyantha Clusiaceae Garcinia sp. Clusiaceae Garcinia sp.1 Clusiaceae Garcinia staudtii Engl. Clusiaceae Mammea africana Sabine Clusiaceae Pentadesma butryacea Sabine Clusiaceae Pentadesma grandifolia Baker f. Clusiaceae Symphonia globulifera L.f. Combretaceae Combretum sp. Combretaceae Strephonema pseudocola A.Chev. Combretaceae Terminalia ivorensis A.Chev. Combretaceae Terminalia superba Engl. & Diels Connaraceae Connarus sp. Connaraceae Connarus sp.1 Connaraceae Connarus staudtii Gilg. Connaraceae Jollydora duparquetiana (Baill.) Pierre 67

94 Connaraceae Santalinoides afzelii R.Br.ex Planch. Convolvulaceae Neuropeltis acuminata (P.Beauv.) Benth. Convolvulaceae Neuropeltis pseudovelutina Lejoly & Lisowski Convolvulaceae Neuropeltis velutina Hallier f. Dichapetalaceae Dichapetalum affine (Planch.ex Benth.) Breteler Dichapetalaceae Dichapetalum dewevrei De Wild. & Th. Dur. Dichapetalaceae Dichapetalum heudelotii (Planch. ex Oliv.) Baill. Dichapetalaceae Dichapetalum phyiotrica Dichapetalaceae Dichapetalum rudatisii Engl. Dichapetalaceae Dichapetalum tomentosum Engl. Dichapetalaceae Tapura africana Oliv. Dilleniaceae Tetracera potatoria Afzel. Ex G.Don Ebenaceae Diospyros bipindensis Gürke Ebenaceae Diospyros abyssinica (Hiern) F. White Ebenaceae Diospyros canaliculata De Wild Ebenaceae Diospyros cinnabarina (Gürke) F. White Ebenaceae Diospyros conocarpa Gürke & K. Schum. Ebenaceae Diospyros gabunensis Gürke Ebenaceae Diospyros gracilescens Gürke Ebenaceae Diospyros hoyleana White Ebenaceae Diospyros iturensis (Gürke) Letouzey & F. White Ebenaceae Diospyros kamerunensis Gürke Ebenaceae Diospyros korupensis Gosline Ebenaceae Diospyros obliquifolia (Hiern ex Gürke) F. White Ebenaceae Diospyros preussii Gürke Ebenaceae Diospyros sp. Ebenaceae Diospyros sp.3 Ebenaceae Diospyros sp.4 Ebenaceae Diospyros sp.5 Ebenaceae Diospyros suaveolens Gürke Ebenaceae Diospyros zenkeri (Gürke) F. White Erythroxylaceae Erythroxylum mannii Oliv. 68

95 Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Fabaceae Fabaceae Fabaceae Fabaceae Fabaceae Fabaceae Fabaceae Fabaceae Fabaceae Fabaceae Alchornea floribunda Croton sylvaticus Hochst. Croton sylvaticus Hochst. Crotonogyne preussii Pax Dichostemma glaucescens Pierre Discoclaoxylon hexandrum (Mull.Arg.) Pax & K. Hoffm. Discoglypremna caloneura (Pax) Prain Elaeophorbia drupifera (Thonn.) Stapf. Euphorbia sp.2 Jatropha curcas L. (Cult.) Klaineanthus gaboniae Pierre ex Prain Macaranga monandra Müll.Arg. Macaranga sp. Macaranga spinosa Müll.Arg. Mallotus oppositifolius (Geiseler) Müll.Arg.var.oppositifolius Mareya micrantha (Benth.) Müll.Arg. Mareyopsis longifolia (Pax) Pax & K. Hoffm. Neoboutonia glabrescens Prain Pycnocoma macrophylla Benth. Ricinodendron heudelotii (Baill.) Pierre ex Baill. Sapium ellipticum (Krauss) Pax Tetrorchidium didymostemon (Baill.) Pax & K.Hoffm. Acacia pennata Willd. Afzelia bella Harms Afzelia bipindensis Harms Afzelia pachyloba Harms Albizia adianthifolia (K. Schum.) W.F. Wright Albizia ferruginea Benth. Albizia sp. Albizia sp.4 Albizia zygia (DC.) J.F. Macbr. Amphimas ferrugineus Pellegr. 69

96 Fabaceae Amphimas pterocarpoides Fabaceae Angylocalyx oligophyllus (Baker) Baker f. Fabaceae Anthonotha cladantha (Harms)J.Léonard Fabaceae Anthonotha fragrans (Baker f.) Exell & Hillc. Fabaceae Anthonotha lamprophylla (Harms)J.Léonard Fabaceae Anthonotha macrophylla P. Beauv. Fabaceae Aphanocalyx microphyllus Harms Fabaceae Baikiaea insignis Benth. Fabaceae Baphia buettneri Harms Fabaceae Baphia capparidifolia Baker f. Fabaceae Baphia laurifolia Baill. Fabaceae Baphia sp. Fabaceae Berlinia auriculata Benth. Fabaceae Berlinia bracteosa Benth. Fabaceae Berlinia hollandii Benth. Fabaceae Calpocalyx dinklagei Harms Fabaceae Calpocalyx heitzii Pellegr. Fabaceae Copaifera mildbraedii Harms Fabaceae Crudia sp. Fabaceae Cryptosepalum congolanum Fabaceae Dialium bipindense Harms Fabaceae Dialium dinklagei Harms Fabaceae Dialium pachyphyllum Harms Fabaceae Didelotia africana Baill. Fabaceae Didelotia morelii Aubr. Fabaceae Distemonanthus benthamianus Baill. Fabaceae Erythrina milbraedii Harms Fabaceae Erythrophleum ivorensis A. Chev. Fabaceae Eurypetalum unijugum Harms Fabaceae Gilbertiodendron demonstrans (Baill.) J.Léonard Fabaceae Gilbertiodendron dewevrei (De Wild.) J. Léonard Fabaceae Gilbertiodendron sp.2 70

97 Fabaceae Hylodendron gabunense Taub. Fabaceae Hymenostegia korupensis Wer. Fabaceae Leonardoxa africana (Baill.) Aubrév. Subsp. Rumpiensis Fabaceae Leptoderris ledermannii Harms Fabaceae Microberlinia bisulcata Aubr. Fabaceae Milletia sp. Fabaceae Millettia hypolampra Harms Fabaceae Newtonia griffoniana (Baill.) Baker f. Fabaceae Parkia bicolor A. Chev. Fabaceae Parkia sp. Fabaceae Pellegriniodendron diphyllum (Harms) J. Léonard Fabaceae Pentaclethra macrophylla Benth. Fabaceae Piptadeniastrum africanum (Hook.f.) Brenan Fabaceae Plagiosiphon longitubus (Harms) J.Léonard Fabaceae Pterocarpus soyauxii Taub. Fabaceae Talbotiella korupensis Mackinder & Wieringa Fabaceae Tetraberlinia bifoliolata (Harms Hauman Fabaceae Tetraberlinia korupensis Wieringa Fabaceae Tetraberlinia polyphylla Fabaceae Tetrapleura tetraptera (Schum. & Thonn.) Taub. Fabaceae Zenkerella citrina Taub. Gentianaceae Anthocleista schweinfurthii Gilg. Gentianaceae Mostuea brunonis Didr. Hoplestigmataceae Hoplestigma klaineanum Pierre Huaceae Afrostyrax kamerunensis Perkins & Gilg. Huaceae Afrostyrax lepidophyllus Mildbr. Humiriaceae Saccoglottis gabonensis (Baill.) Urban Hypericaceae Harungana madagascariensis Poir. Icacinaceae Iodes africana Welw. ex Oliv. Icacinaceae Lasianthera africana P. Beauv. Icacinaceae Rhaphiostylis beninensis (Hook.f.ex Planch.) Planch ex Bench. Icacinaceae Rhaphiostylis viridiflora 71

98 Irvingiaceae Desbordesia glaucescens Van Tiegh. Irvingiaceae Irvingia gabonensis (Aubrey-Leconte ex O.Rorke) Baill. Irvingiaceae Irvingia grandifolia (Engl.) Engl. Irvingiaceae Klainedoxa gabonensis Pierre ex Engl. Irvingiaceae Klainedoxa trillesii Pierre ex van Tiegh. Lamiaceae Vitex grandifolia Gürke Lamiaceae Vitex lehmbachii Gürke Lamiaceae Vitex oxycuspis Baker f. Lamiaceae Vitex sp.2 Lamiaceae Vitex sp.3 Lamiaceae Vitex sp.4 Lamiaceae Vitex sp.5 Lamiaceae Vitex sp.i Lauraceae Beilschmiedia acuta Kostermans Lauraceae Beilschmiedia gabonensis (Meissn.) Benth. & hook.f. Lauraceae Beilschmiedia mannii (Meisn.) Benth. & Hook.f. Lauraceae Beilschmiedia sp. Lauraceae Beilschmiedia sp.2 Lauraceae Beilschmiedia sp.22 Lauraceae Beilschmiedia sp.6 Lauraceae Beilschmiedia talbotiae (S. Moore) Robyns & Wilczeck Lauraceae Hypodaphnis zenkeri Stapf. Lauraceae Persea americana Mill. (Cult.) Lecythidaceae Crateranthus talbotii Baker f. Lecythidaceae Napoleonaea cf heudelotii A. Juss Lecythidaceae Napoleonaea ergortonii Baker f. Lecythidaceae Napoleonaea talbotii Baker f. Lecythidaceae Oubanguia alata Baker f. Lecythidaceae Oubanguia laurifolia (Pierre ex De Wild.) Tiegh. Lecythidaceae Rhaptopetalum coriaceum Oliv. Lecythidaceae Rhaptopetalum depressum Letouzey Lecythidaceae Rhaptopetalum geophylax Cheek & Gosline 72

99 Lecythidaceae Lecythidaceae Lecythidaceae Leeaceae Lepidobotryaceae Leptaulaceae Leptaulaceae Loganiaceae Loganiaceae Loganiaceae Loganiaceae Loganiaceae Loganiaceae Malpighiaceae Malvaceae Malvaceae Malvaceae Malvaceae Malvaceae Malvaceae Malvaceae Malvaceae Malvaceae Malvaceae Malvaceae Malvaceae Malvaceae Malvaceae Malvaceae Malvaceae Malvaceae Malvaceae Rhaptopetalum sp. Rhaptopetalum sp.nov. Scytopetalium klaineanum Pierre ex Engl. Leea guineensis G. Don Lepidobotrys staudtii Engl. Leptaulus daphnoides Benth. Leptaulus grandifolius Engl. Strychnos angolensis Gilg. Strychnos camptoneura Gilg. & Busse Strychnos congolana Gilg. Strychnos johnsonii Hutch. & M.B. Moss Strychnos memecyloides S.Moore Strychnos tricalysioides Hutch. & M.B. Moss Acridocarpus macrocalyx Engl. Cola altissima Engl. Cola sp. nov.4 Cola cauliflora Mast. Cola chlamydantha K. Schum. Cola digitata Cola flaviflora Engl. & K. Krause Cola lateritia K. Schum. Cola lepidota K. Schum. Cola megalophylla Brenan & Keay Cola nitida (Vent.) Schott & Endl. Cola pachycarpa K. Schum. Cola rostrata K. Schum. Cola semecarpophylla K. Schum. Cola sp. Cola sp.nov. Cola sp.nov.2 Cola sp.nov.3 Cola suboppositifolia Cheek & Kenfack 73

100 Malvaceae Cola verticillata (Thonn.) Stapf ex A. Chev. Malvaceae Duboscia viridifolia Malvaceae Glyphaea brevis (Spreng) Monachino Malvaceae Leptonychia echinocarpa K. Schum Malvaceae Leptonychia lasiogyne K. Schum. Malvaceae Leptonychia macrantha (C.H. Wright) Burret Malvaceae Microcos coriacea (Mast.) Burret Malvaceae Octolobus spectabilis Welw. Malvaceae Scaphopetalum blackii Mast. Malvaceae Sterculia tragacantha Lindl. Medusandraceae Soyauxia gabunensis Baker f. Medusandraceae Soyauxia sp.1 Melastomataceae Memecylon afzelii G.Don Melastomataceae Memecylon calophyllum Gilg. Melastomataceae Memecylon candidum Gilg. Melastomataceae Memecylon englerianum Cogn. Melastomataceae Memecylon lateriflorum (G. Don) Brem. Melastomataceae Memecylon laurentii De Wild Melastomataceae Memecylon sp.1 Melastomataceae Memecylon sp.2 Melastomataceae Memecylon zenkeri Gilg. Melastomataceae Warneckea cinnamomoides (G. Don) Jacq.-Fél. Melastomataceae Warneckea jasminoides (Gilg.) Jacq-Fél. Melastomataceae Warneckea membranifolia (Hook.f.) Jacq-Fél Melastomataceae Warneckea pulcherrima (Gilg.) Jacq-Fél. Melastomataceae Warneckea sp.i Meliaceae Carapa angustifolia Harms Meliaceae Carapa dinklagei Harms Meliaceae Carapa oreophila Kenfack Meliaceae Carapa parviflora Harms Meliaceae Carapa zemagoana Kenfack Meliaceae Entandrophragma cylindricum (Sprague) Sprague 74

101 Meliaceae Meliaceae Meliaceae Meliaceae Meliaceae Meliaceae Meliaceae Meliaceae Meliaceae Meliaceae Menispermaceae Menispermaceae Monimiaceae Moraceae Moraceae Moraceae Moraceae Moraceae Moraceae Moraceae Myristicaceae Myristicaceae Myristicaceae Myristicaceae Myristicaceae Myristicaceae Myrsinaceae Myrsinaceae Myrsinaceae Myrtaceae Myrtaceae Myrtaceae Guarea cedrata (A. Chev.) Pellegr. Guarea glomerulata Harms Guarea sp. Guarea thompsonii Sprague & Hutch. Heckeldora staudtii (Harms) Staner Khaya ivorensis A. Chev. Trichilia aff.gilgiana Harms Trichilia prieureana A. Juss Turraeanthus africanus (Welw.ex A.DC.) Pellegr. Turraeanthus mannii Baill. Penianthus camerounensis A. Dekker Syrrhonema fasciculatum Miers Glossocalyx brevipes Benth. Ficus mucuso Welw. ex Ficalho Ficus sp. Ficus sp.3 Ficus sur Forssk. Milicia excelsa (Welw.) C.C. Berg. Treculia africana (Engl.) J. Léonard Treculia obovoidea N.E.Br. Coelocaryon preussii Warb. Pycnanthus angolensis (Welw.) Exell Scyphocephalium mannii Warb. Staudtia gabunensis Warb. Staudtia kamerunensis Warb. Staudtia sp. Ardisia cymosa Mez Maesa kamerunensis Mez Maesa lanceolata Forssk. Eugenia callophyloides DC Eugenia sp.2 Eugenia talbotii Keay 75

102 Myrtaceae Myrtaceae Ochnaceae Ochnaceae Ochnaceae Ochnaceae Ochnaceae Ochnaceae Octoknemaceae Octoknemaceae Octoknemaceae Olacaceae Olacaceae Olacaceae Olacaceae Olacaceae Olacaceae Olacaceae Olacaceae Olacaceae Olacaceae Olacaceae Pandaceae Pandaceae Passifloraceae Phyllanthaceae Phyllanthaceae Phyllanthaceae Phyllanthaceae Phyllanthaceae Phyllanthaceae Phyllanthaceae Syzygium rowlandii Sprague Syzygium staudtii (Engl.) Mildbr. Campylospermum calanthum (Gilg.) Farron Campylospermum elongathum Campylospermum flavum Campylospermum mannii Lophira alata Banks ex Gaertn.f. Rhabdophyllum affine (Hook.f.) Tiegh. Octoknema affinis Pierre Octoknema anuwiniencis Octoknema bakosiensis Gosline & Malécot Coula edulis Baill. Diogoa zenkeri (Engl.) Exell & Mendonca Engomegoma gordonii Breteler Heisteria parvifolia Sm. Olax triplinervia Oliv. Strombosia grandifolia Hook.f. Strombosia pustulata Oliv. Strombosia scheffleri Engl. Strombosia sp. Strombosia sp.1 Strombosiopsis tetrandra Engl. Microdesmis puberula Hook.f. ex Planch. Panda oleosa Pierre Barteria fistulosa Mast Antidesma chevaleri Antidesma laciniatum Müll.Arg. Antidesma sp. Antidesma vogelianum Müll. Arg. Bridelia grandis Pierre ex Hutch. Bridelia micrantha (Hochst.) Baill. Cleistanthus letouzeyi J. Leonard 76

103 Phyllanthaceae Phyllanthaceae Phyllanthaceae Phyllanthaceae Phyllanthaceae Phyllanthaceae Phyllanthaceae Phyllanthaceae Phyllanthaceae Phyllanthaceae Phyllanthaceae Phyllanthaceae Polygalaceae Putranjivaceae Putranjivaceae Putranjivaceae Putranjivaceae Putranjivaceae Putranjivaceae Putranjivaceae Putranjivaceae Putranjivaceae Putranjivaceae Putranjivaceae Putranjivaceae Rhamnaceae Rhizophoraceae Rhizophoraceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Keayodendron bridelioides (Mildbr..ex Hutch. & Dalz.) Léandri Maesobotrya barteri (Baill.) Hutch. Maesobotrya dusenii (Pax) Hutch. Maesobotrya sp. Maesobotrya staudtii (Pax) Hutch. Margaritaria discoidea (Baill.) Webster Protomegabaria stapfiana (Beille) Hutch. Thecacoris batesii Hutch. Thecacoris leptobotrya (Müll.Arg.) Brenan Uapaca acuminata (Hutch.) Pax & K. Hoffm. Uapaca guineensis Müll.Arg. Uapaca staudtii Pax Carpolobia lutea G. Don Drypetes aframensis Hutch. Drypetes afzelii (Pax) Hutch. Drypetes gossweileri S. Moore Drypetes ivorensis Hutch. & Dalz. Drypetes laciniata (Pax) Hutch. Drypetes molunduana Pax & K.Hoffm. Drypetes paxii Hutch. Drypetes sp.3 Drypetes sp.4 Drypetes sp.6 Drypetes sp.7 Sibangea similis (Hutch.) Radcl.-Sm. Maesopsis eminii Engl. Cassipourea gummiflua Tul. Cassipourea schizocalyx C.H. Wright Aoranthe cladantha (K. Schum) Somers Aulacocalyx caudata Aulacocalyx jasminiflora Hook.f. Aulacocalyx talbotii (Wernham) Keay 77

104 Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Belonophora coriacea Belonophora wernhamii Hutch. & Dalziel Bertiera laxa Benth. Bertiera racemosa (G. Don) K. Schum. Calycosiphonia spathicalyx (K. Schum.) Robbr. Canthium sp. Coffea sp. Craterispermum aristatum Wernham Cuviera subuliflora Benth. Cuviera trilocularis Hiern Euclinia longiflora Salisb. Gaertnera bieleri (De Wild) E.M.A. Petit Gaertnera sp. Hallea ledermannii (K.Krause) Verdc. Heinsia crinita (Afzel.) G. Taylor Ixora bauchiensis Hutch. & Dalziel Ixora guineensis Benth. Ixora hippoperifera Bremek Ixora nematopoda K. Schum. Massularia acuminata (G.Don) Bullock ex Hoyle Morelia senegalensis A. Rich. Morinda lucida Benth. Nauclea diderrichii (De Wild.& T.Durand) Merrill Nauclea sp. Oxyanthus laxiflorus K.Schum.ex Hutch. & Dalziel Pauridiantha afzelii (Hiern) Bremek. Pauridiantha floribunda (K. Schum. & K.Krause) Bremek. Pauridiantha sp. Pauridiantha viridiflora (Schweinf. ex Hiern) Hepper Pausinystalia macroceras (K. Schum.) Pierre ex Beille Pavetta baconiella Bremek Pavetta staudtii Hutch. & Dalziel 78

105 Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Rubiaceae Ruscaceae Ruscaceae Ruscaceae Rutaceae Rutaceae Petitiocodon parviflorum (Keay) Robbr. Petitiocodon sp. Petitiocodon sp.2 Psilanthus mannii Hook.f. Psychotria bimbiensis Bridson & Cheek Psychotria brachyantha Hiern Psychotria darwiniana Cheek Psychotria dorotheae Wernham Psychotria geophylax Cheek & Sonké Psychotria peduncularis (Salisb.) Steyerm Psychotria sp.2 Psychotria sp.3 Psychotria sp.5 Psychotria sp.7 Psychotria sp.9 Psydrax sp. Rothmannia hispida (K. Schum.) Fagerlind Rothmannia lujae (De Wild.) Keay Rothmannia sp. Schumanniophyton magnificum (K. Schum.) Harms Stipularia africana P. Beauv. Tarenna baconoides Wernham var. baconoides Tarenna glandiflora (Benth.) Hiern Tricalysia biafrana Hiern Tricalysia coriacea (Benth.) Hiern subs. coriacea Tricalysia amplexicaulis Robbr. Vangueriella chlorantha (K. Schum.) Verdc. Dracaena arborea (Wild.) Link Dracaena deisteliana Engl. Dracaena talbotii Rendle Vepris adamaoua Vepris lecomteana (Pierre) Cheek & T. Heller 79

106 Rutaceae Rutaceae Rutaceae Rutaceae Rutaceae Rutaceae Rutaceae Salicaceae Salicaceae Salicaceae Salicaceae Salicaceae Salicaceae Salicaceae Salicaceae Salicaceae Salicaceae Salicaceae Salicaceae Sapindaceae Sapindaceae Sapindaceae Sapindaceae Sapindaceae Sapindaceae Sapindaceae Sapindaceae Sapindaceae Sapindaceae Sapindaceae Sapindaceae Sapindaceae Vepris soyauxii (Engl.) Mziray Vepris sp. Zanthoxylon gilletii (De Wild.) P.G. Waterman Zanthoxylon heitzii Aubrév. Pellegr. Zanthoxylon lemairei (De Wild.) P.G. Waterman Zanthoxylon sp.2 Zanthoxylon sp.i Casearia barteri Mast. Homalium africanum (Hook.f.) Benth Homalium letestui Pellegr. Homalium longistylum Mast. Homalium macroptenum Homalium sp.1 Oncoba glauca (P.Beauv.) Planch. Oncoba lophocarpa Oliv. Oncoba mannii Oliv. Oncoba ovalis (Oliv.) Chipp Oncoba sp. Phyllobotryon spathulathum Müll.Arg. Allophylus africanus P. Beauv. Allophylus grandifolius (Baker) Radlk. Allophylus megaphyllus Hutch. & Dalz. Allophylus sp. Allophylus sp.2 Allophylus sp.3 Blighia sapida Koenig Blighia welwitschii (Hiern) Radlk. Chytranthus angustifolius Exell Chytranthus edulis Chytranthus sp. Chytranthus talbotii (Baker f.) Keay Deinbollia pycnophylla Gilg ex Radlk. 80

107 Sapindaceae Deinbollia unijuga D. W. Thomas Sapindaceae Eriocoelum kerstingii Gilg. ex Engl. Sapindaceae Eriocoelum macrocarpum Gilg. ex Radlk. Sapindaceae Laccodiscus ferrugineus (Baker) Radlk. Sapindaceae Laccodiscus pseudostipularis Radlk. Sapindaceae Placodiscus cf. caudatus Pierre ex Radlk Sapindaceae Radlkofera calodendron Gilg. Sapotaceae Pradosia spinosa Ewango & Breteler Sapotaceae Englerophytum kennedyi Sapotaceae Englerophytum sp.nov. Sapotaceae Gambeya africanum A.DC Sapotaceae Gambeya boukokoensis Aubbrév. & Pellegr. Sapotaceae Gambeya delevoyi De Wild. Sapotaceae Gambeya korupensis Ewango & Kenfack Sapotaceae Gambeya lacourtianum De Wild. Sapotaceae Gambeya subnudum Baker f. Sapotaceae Lecomtedoxa klaineana (Pierre & Engl.) Dubard Sapotaceae Manilkara argentea Sapotaceae Manilkara lososiana Kenfack & Ewango Sapotaceae Manilkara pellegriniana Tisserant & Silans Sapotaceae Omphalocarpum cf. elatum Miers Sapotaceae Omphalocarpum elatum Miers Sapotaceae Pouteria sp. Sapotaceae Synsepalum dulcificum (Schum. & Thonn.) Daniell Sapotaceae Synsepalum letouzeyi Aubrév. Sapotaceae Synsepalum longecuneatum De Wild. Simaroubaceae Odyendya gabonensis (Pierre) Engl. Simaroubaceae Pierreodendron africanum (Hook.f.) Little Simaroubaceae Quassia sanguinea Cheek & Jongkind Simaroubaceae Quassia silvestris Cheek & Jongkind Thymelaeaceae Dicranolepis pulcherrima Gilg. Ulmaceae Trema orientalis (L.) Blume 81

108 Violaceae Violaceae Violaceae Violaceae Violaceae Violaceae Violaceae Violaceae Violaceae Violaceae Violaceae Vochysiaceae Vochysiaceae Alexis cf cauliflora (Oliv.) Pierre Rinorea dentata (P.Beauv.) O. Kuntze Rinorea kamerunensis Engl. Rinorea leophylla Brandt. Rinorea oblongifolia (C.H. Wright) Marqua Rinorea sp.3 Rinorea sp.33 Rinorea sp.4 Rinorea sp.5 Rinorea subintegrifolia (P.Beauv.) O.Kuntze Rinorea woermanniana (Butin) Engl. Erismadelphus exsul Mildbr. Korupodendron songweanum Litt & Cheek 82

109 CHAPTER FOUR PATTERNS OF TREE DIVERSITY, ABOVE GROUND BIOMASS AND CARBON ALONG AN ELEVATIONAL GRADIENT IN TROPICAL FOREST: A CASE STUDY OF THE RUMPI FOREST RESERVE IN CAMEROON, WESTERN CENTRAL AFRICA Moses Nsanyi Sainge 1,2, Felix Nchu 3*, and A. Townsend Peterson 4 1 Tropical Plant Exploration Group (TroPEG), P.O. Box 18, Mundemba, Ndian, South West Region, Cameroon. sainge2001@yahoo.com, moses.sainge@gmail.com, tropeg.cam@gmail.com 2 Department of Environmental and Occupational Studies, Faculty of Applied Sciences, Cape Peninsula University of Technology, Cape Town Campus, Keizersgracht, P.O. Box 652, Cape Town 8000, South Africa. sainge2001@yahoo.com, moses.sainge@gmail.com, tropeg.cam@gmail.com 3 Department of Horticultural Sciences, Faculty of Applied Sciences Cape Peninsula University of Technology, P.O Box 1906, Bellville 7535, South Africa. felixnchu@gmail.com 4 Biodiversity Institute, University of Kansas, Lawrence, Kansas USA. town@ku.edu *Author of correspondence Felix Nchu felixnchu@gmail.com Tel: (+27) Preface An overview of tree diversity, above-ground biomass, and carbon stock along elevation gradients were evaluated and analyze in this chapter. ABSTRACT Tropical forest remains the most diverse ecosystem on earth, and stores considerable amounts of biomass and carbon. Despite its importance, sizable knowledge gaps surround the relationships among species diversity, plant biomass and carbon in different forest ecosystems. These knowledge gaps are particularly wide in tropical systems, and even more so in the African tropics. This study seeks to provide baseline data on species composition and vegetation structure, and to evaluate variation along elevational gradients for tree species diversity, above-ground biomass and carbon in 25 1-ha plots in the Rumpi Hills Forest Reserve, in Cameroon. Plots were sampled in 500 x 20 m transects, measuring trees and lianas with diameter at breast height 10 cm 83

110 in four types of forest in the reserve. Results revealed high diversity, particularly in lowland forest. Overall, the study examined 12,037 individuals (Trees 10 cm dbh) in 441 species; mean numbers of species per plot were 112 species ha -1 in lowland, 81 species ha -1 in mid-elevation, 60 species ha -1 in submontane and 38 species ha -1 in montane forest. Above-ground biomass averaged ±100 t ha -1, and carbon averaged ±50 t ha -1. We found negative relationship between carbon and elevation (r 2 = , P < 0.05), and a weak positive relationship between species diversity and carbon (r 2 = 0.34, P < 0.05), following regression analysis. Correspondence analysis suggested that lowland elevation, high species diversity and high carbon are associated while low carbon and low species diversity are associated. Our results indicate that high species diversity and occurrence of larger tree species are more important in carbon storage than elevation in RHFR. These findings are useful for management and land use planning of the unique forests in the Rumpi Hills Forest Reserve. Keywords: Carbon, Lowland, Montane ecosystem, Tropical forest, Rumpi Hills Forest Reserve 4.1 INTRODUCTION Terrestrial forest ecosystems remain major carbon sinks, and hold large stock of biomass. They mitigate climate change through sequestration of carbon in vegetation biomass. The vegetation gains carbon from productivity investment in aerial and roots of live trees, and loses carbon owing to aging, mortality, harvest, etc. (Myneni et al., 2001). Although tropical forests have high carbon storage capacity (Pan et al., 2011, Reich, 2011), they are threatened by anthropogenic activities such as deforestation, farming and urbanization. Currently, gaps in tropical forest dynamics and ecology exist; these gaps may hinder reliable predictions of forest responses to global change and development of efficient management strategies to minimize anthropogenic threats and optimize carbon storage capacity of forests (Zuidema et al., 2013). Empirical findings pertaining to forest ecology have revealed that biodiversity correlates broadly with aboveground biomass and carbon stocks in forests, and that species diversity declines along elevational gradients (Asase et al., 2012; Day et al., 2013; Imani et al., 2017). Relationships between elevational gradients and carbon should be straightforward, but recent studies have revealed that it is not: different elevational patterns documented include hump-shaped, reversed hump-shaped, increasing multimodal relationships and no relationship at all (Lee and Chun 2016). In a recent study in tropical montane forests in southern Ecuador, biomass of lianas and the relative contribution of lianas to total above-ground biomass decrease with elevation 84

111 and with a decrease of host trees (Fadrique and Homeier, 2016). Some studies have found positive correlations between tree diversity and above-ground biomass (Day et al., 2013), and others have established positive relationships between tree diversity and carbon stock in tropical forest (Asase et al., 2012). To fill these knowledge gaps and to provide additional points of information from the African tropics, we carried out fine-scale and accurate measurements of biomass and carbon, tree species, tree sizes and tree densities along an elevational gradient in the Rumpi Hills Forest Reserve. The Rumpi Hills Forest Reserve is a protected tropical rainforest that is rich in endemic tree species; it is also relatively intact and understudied, and thus is an ideal choice for our investigation. This study seeks to establish baseline data on species composition and vegetation structure, and explore relationships among species biodiversity, elevation and carbon in the diverse tropical forests of the Rumpi Hills Forest Reserve. 4.2 MATERIALS AND METHODS Study sites The study area was in the Rumpi Hills Forest Reserve (RHFR) in Ndian Division, South West Region of Cameroon, stretching across latitudes N and longitudes E, with an elevational range of m. It covers an area of 453 km 2 (Sainge, 2016). The annual rainfall at the nearby village of Dikome Balue is 4933 mm, with August being the wettest month and December the driest (Thomas, 1996). The study area was stratified into four vegetation communities (Letouzey, 1985), namely lowland evergreen rainforest, mid-elevation evergreen forest, submontane forest and montane cloud forest. The lowland forest covers the southern and part of the northeastern sections of the reserve, with an approximate extent of 185 km 2 at elevations of m and 12 one-hectare plots were established. The mid-elevation evergreen forest covers the northern part of the reserve (ca. 133 km 2 ), at elevations of m; eight one-hectare plots were established. Submontane forest occurs in the central, northeastern and eastern sectors of the reserve, with an extent of ca. 132 km 2 at elevations of ; three one-hectare plots were established. Finally, the montane cloud forest was in the eastern part of the reserve near Dikome Balue village with an extent of ca. 3 km 2 at elevations of m; two one-hectare plots were established (Figure 4.1). Administratively, the montane cloud forest ( m, at Mt Rata) falls outside of the Reserve, but it is a unique element in this montane system and should be considered with the other forest types. 85

112 4.2.2 Field sampling In all, twenty-five 1-ha plots were sampled between February and June 2015 in the four forest types. Four types of vegetation cover (life forms) were considered in our survey: lianas, and trees 10 cm, 30 cm and 60 cm. Precise GPS coordinates were recorded at the beginning of each plot to assure repeatable plot locations; these coordinates are available in tabular form at Plots comprised transects that were 500 x 20 m, and sub-divided into 25 quadrats of 20 x 20 m. In mountainous areas, plots were located strategically so that 25 quadrats of 20 x 20 m representing one-hectare fit within a particular sampling area. Twenty-five plots were sampled for all trees and lianas 10 cm dbh. Information on vegetation, diameter at breast height, habitat, and species identification to morphospecies was recorded in the field. All live trees and lianas with dbh 10 cm were measured using a diameter tape, tagged using tree tag numbers and nails, and identified by two botanists. Voucher specimens were collected for all morphospecies. Plot data were used to estimate above-ground biomass, and carbon using the allometric equation of Chave et al., (2015). The height of each individual tree was calculated according to Djomo et al., (2016). Wood Specific Density (WSD) was assembled from the global wood density database (Zanne et al., 2009) and the African wood density database (Carsan et al., 2012). Biomass and carbon estimations were achieved for every individual tree, including for multiple stems. These values were summarized by plot and by forest type (Losi et al. 2003). Observational data were collected within our broader survey area, representing individuals with flowers or fruits, and particularly species that were not recorded in the general plot census. Plots were not established at m due to time constraint, difficulty to access this elevation and limited funding Forest type classification A previous study recognized 267 forest types in Cameroon (Letouzey, 1985), of which seven are represented in our study area, broadly divided into four forest types: lowland, mid-elevation, submontane, and montane. Detailed descriptions of the different forest types are available in Letouzey, Data analysis Identification of specimens were carried out at the National Herbarium of Cameroon (YA) by matching specimens collected with existing herbarium sheets, and by consulting floras, published documents, and keys for the plants of the region (Hutchinson and Dalziel, 1954, 1958, 1963; Vivien and Faure, 1985; Keay, 1989; Thomas et al., 2003; Cheek et al., 2004; Harvey et al., 2004; Harvey et al., 2010; Onana, 2011, 2013). The 86

113 final checklist was consolidated following the Angiosperm Phylogeny Group (APG III) classification (Judd et al., 1999; Angiosperm Phylogeny Group (APG III, 2009)). Basal Area, Relative dominance, relative density, relative frequency was calculated using the formulas below. Fisher s alpha, Shannon-Wiener index (H ), and Simpson index were used as indices to compare species diversity among plant forms and elevations using the software package PAST version 2.17 (Hammer et al., 2001). This was based on 12 ha in lowland, 8 ha in mid-elevation, 3 ha in submontane and 2 ha in montane forest. Fisher s alpha was not calculated for lianas in montane forest because the values were too low. The distribution of variation in tree species diversity and carbon per hectare in different forest types was analysed using analysis of variance (ANOVA). The data were converted to binomials (0 and 1) and correspondent analysis (CA) was performed to establish the relationship among elevation, species diversity and carbon. Regression analysis was conducted with the aid of PAST version A non-destructive method was used to estimate above-ground biomass, and carbon stock to diameter at breast height, and wood Specific density (WD). Above-ground biomass and carbon were estimated for trees (dbh 10 cm) across forest types using the allometric equation of Chave et al., (2015) and tree height were estimated following Djomo et al., (2016). Eq. (1) Basal Area (BA) = Area occupied by plant at breast height. (BA) = pi*(1/2dbh) 2 = pi*(dbh) 2 /4. Basal area is the area occupied by a species. Eq (2) Eq. (3) Frequency is the number of quadrats in which a species is found in the entire sample Eq. (4). Eq. (5) Importance Value Index (IVI) = Relative density + Relative dominance + Relative frequency 87

114 Eq. (6) Shannon-Weiner index (H ) is the most convenient tool to measure diversity in 1-ha plots. This was achieved through the following formula: Where pi is the proportion of individual of a species (Number of individual of a species/total number of all species), ln is the natural logarithm. Thus, the natural logarithm of the number of species (lns) is the maximum value of H. Eq. (7) D 2 H) (Chave et al., 2015) Eq. (8) (Djomo et al., 2016) Where AGB is above-ground dry biomass; ρ is wood density; D is dbh; ln is the natural logarithm, and e indicates the exponential function. Wood specific density was assembled from published sources such as the Global Wood Density Database (Dryad identifier: (Zanne et al. 2009), and the African Wood Density Database ( (Carsan et al., 2012). Species-specific wood densities were used for individuals identified to the species level. In cases in which species-specific wood densities were not available, mean values for the genus or family were used. For unidentified stems overall mean wood density for the data set was used (Baker et al., 2004). We calculated carbon content as Carbon = Total above-ground biomass/2 (Brown, 1997; Chave et al., 2005; Lewis et al., 2009). 4.3 Results Plant density and basal area by forest type Density of trees with dbh 10 cm decreased with elevation (Figure 2). Mean density varied across forest types and elevation: 493 trees ha -1 ( trees) in lowland, 450 trees ha -1 ( trees) in mid-elevation, 485 trees ha -1 ( trees) in sub-montane and 533 trees ha -1 ( trees) in montane (Table 1). Mean basal area was associated with elevation: 35.3 m 2 ha -1 ( m 2 ) in lowland, 26.9 m 2 ha -1 ( m 2 ) in midelevation, 27.5 m 2 ha -1 ( m 2 ) in sub-montane and 34.4 m 2 ha -1 ( m 2 ) in montane forest. Overall tree density (multiple stems excluded) decreased with elevation: lowland (5916 trees), mid-elevation (3600 trees), sub-montane (1454 trees) and montane (1066 trees) (Table 4.1, Figure 4.2). 88

115 4.3.2 Elevational patterns All trees with dbh 10 cm totaled 12,037 individuals in 441 species, 229 genera, and 63 families across the four forest types. Ninety-two individuals were not identified to species and genus, and nine individuals were not identified to family. The 92 individuals corresponded to 17 morphospecies, so the overall species richness of all trees 10 cm dbh was 458 morphospecies. The observational data provided records of an additional 254 individual trees in 62 families, 129 genera and 210 species, of which 132 species were not recorded in the sampling plots. The Reserve overall holds a mean of 92 species ha -1 (ranging species) for trees with dbh 10 cm. Lowland forest holds a mean of 115 species ha -1 (ranging species), 84 species ha -1 (75-91 species) in mid-elevation forest, 59 species ha -1 (32-80 species) in submontane and 38 species ha -1 (35-40 species) in montane forest. Lianas with dbh 10 cm represent a mean of 4 species ha -1 (0-8 species) in lowland forest, 3 species ha -1 (0-7 species) in mid-elevation, and 1 species ha -1 (1-1 species) in submontane forest, no liana species were recorded in montane forest in this diameter class. Regression analysis showed a strong significant negative relationship between species richness and elevation (r = 0.756, p < 0.05;) across 25 ha plot in the Rumpi Hills Forest Reserve, Cameroon (Figure 3). Overall we found 24 species of lianas with dbh 10 cm in lowland forest, 17 species in mid-elevation, two species in submontane and no species in montane forest. The number of lianas correlated strongly with the number of trees with dbh 60 cm along the elevational gradient (Figure 4.4). In all, 88.3% of trees and lianas were identified to species level, 89% to genus level, 99.3% to family level, and <1% entirely unidentified. A detailed summary of occurrence of species across different forest types is presented in Appendix Species richness within families Lowland forest had the most families (53), followed by mid-elevation and submontane with 42 families each and montane with 25 families. In lowland forest, Fabaceae had the highest number of species, with 55 species in 566 individual trees, Annonaceae hold 21 species with 136 trees, Rubiaceae 21 species with 164 trees, Phyllanthaceae 17 species in 455 trees and Malvaceae 15 species in 190 trees. Rare families with one species each were Bignoniaceae (62 trees), Boraginaceae (2 trees), Cecropiaceae (55 trees), Medusandraceae (121 trees) and Erythroxylaceae (1 tree). In mid-elevation, Fabaceae was still the dominant family with 21 species in 370 trees, followed by Annonaceae 12 species in 139 trees, Malvaceae 10 species in 61 trees and 89

116 Euphorbiaceae (130 trees), Phyllanthaceae (182 trees), and Rubiaceae (28 trees) in nine species each. The rarest family in this vegetation type was Vochysiaceae with one species. In submontane vegetation, Rubiaceae was the dominant family with 14 species in 100 trees, Euphorbiaceae nine species (97 trees), Meliaceae eight species (125 trees), Annonaceae (24 trees) and Apocynaceae (160) with 6 species each. The rarest families in this vegetation type with one species each were Achariaceae (68 trees), Alangiaceae (three trees), Medusandraceae (11 trees), Ochnaceae (one tree), Octoknemaceae (eight trees), and Simaroubaceae (eight trees). In montane forest, Rubiaceae was the dominant family with six species in 18 trees, Clusiaceae (169 trees) in five species, Salicaceae (13 trees) in four species and Malvaceae (47 trees) in three species. The rarest families with one species each were Achariaceae 91 trees; Anacardiaceae 54 trees; Putranjivaceae 37 trees; Asteraceae 14 trees; Phyllanthaceae 12 trees; Araliaceae six trees; Fabaceae and Melastomataceae with three trees each; and Cecropiaceae, Leptaulaceae, Sapindaceae, and Thymelaeaceae with one tree each Above-ground biomass and carbon estimation The mean wood density of species in this study was 0.63 g cm -3, ranging from 0.21 g cm -3 to 0.96 g cm -3. Within the entire data set, 78.7% of species had WSD ranging from >0.5 to 0.8 g cm -3, 12.3% had densities g cm -3 and 9% >0.8 g cm -3. Overall, above-ground biomass in Rumpi Hills was tons in 25 ha. There was great variation among the twenty-five hectares at different forest types (Table 1): mean AGB per hectare was t ha -1 ranging from t ha -1 in submonane to t ha -1 in lowland forest. In lowland forest, the mean was 386.1±95.5 t ha -1 ranging from t ha -1 to t ha -1, at mid-elevation, the mean was ±48.58 t ha -1 ( t ha -1 ), the mean in submontane was ± t ha -1 ( t ha -1 ) and montane forest records a mean of ±5.96 t ha -1 ( t ha -1 ). These values varied significantly among forest types (ANOVA, F 8.414=10.14, P <0.01): biomass of mid-elevation and submontane forest was significantly lower than the biomass of the other forest types (Welch F test of unequal variances, df=15.4, P <0.02). The overall total carbon estimate in the 25 1-ha plot was t, ranging from t in submontane forest to t in lowland forest; overall mean was ±50.42 t per hectare. A weak significant negative association between carbon and elevation (r 2 = , p <0.05; Figure 5) was observed, with the amount of carbon declining from (292.6) tc ha -1 in lowland forest to (64.55) tc ha -1 in submontane forest (Table 4.1). Thus, the four forest types presented a mean carbon density of t ha -1 (ranging t ha -1 ) in 90

117 lowland forest, mean carbon density of t ha -1 (ranging t ha -1 ) in mid-elevation, mean carbon density of t ha -1 (ranging t ha -1 ) in submontane, and a mean carbon density of t ha -1 (ranging t ha -1 ) in montane forest (Table 4.3). A weak positive relationship was manifested between carbon and number of species per hectare (r 2 = 0.34, P < 0.05; Figure 4.6). Correspondence analysis indicated axes 1 and 2 accounted for 60% of the total variance of the data, and it revealed two opposite associations: the correlations among lowland forest, high species diversity and high carbon in axis 1, and on the other hand mid-elevation, low species diversity and low carbon correlated (Figure 4.7). Large trees (dbh 10 cm) in our study plots ranged cm, with most trees representing dbh <50 cm (93.3%); only 6.7% of trees had dbh 50 cm. This difference reveals a high rate of regeneration of trees in the Reserve, which brings us to an assumption that the level of disturbance in the RHFR is low. However, of the 441 species with dbh 10 cm, 52 species had 5070 individuals out of the stems (42.1%), representing 66.6% of above-ground biomass; and 66.6% of carbon in the entire 25 ha plots (Table 4.4). 4.4 Discussion Many authors have documented different tree densities across the different landscapes in Cameroon (Newberry and Gartlan, 1996; Thomas et al., 2003; Kenfack et al., 2007; Gonmadje et al., 2011; Djuikouo et al., 2014). Newberry and Gartlan, (1996) worked in Korup National Park ( m asl) and Douala-Edea reserve ( m), documenting tree densities of trees ha -1 and 295 trees ha -1 respectively. Thomas et al., (2003) and Kenfack et al., (2007), in their studies of a 50-ha plot in Korup National Park, registered a mean tree density of 487 trees ha -1. Generally, these results are in agreement with the overall mean density of stands (481 trees ha -1 ) in the Rumpi Hills (lowland to Montane forest). This study revealed that the RHFR is rich in species endemic to Cameroon. In total, 17 species were endemic to Cameroon with species such as Deinbollia angustifolia, and Gambeya korupensis that are endemic to the Korup and Rumpi Hills area. A further 43 species were endemic to the Lower Guinea Forest Block (Appendix 4.1). This high level of species diversity and endemism is corroborated by Mittermeier et al., (1999), Myers et al., (2000), Bergl et al., (2007) and Marchese, (2015), who classified the lowland and highland forests of west and central Africa as a global biodiversity hotspot and concentration of endemic species. We found a decrease in overall tree density and tree species diversity with elevation (Figure 4.2). Although species richness patterns generally decrease monotonically with elevation, some previous studies have found mid-elevational peaks 91

118 along elevational gradients (McCain and Grytnes, 2010). It is worth noting that methodological factors including scale and sampling, and geographic factors can strongly influence elevational species richness. Elevational gradients are modulated by cascading and interlinked effects of biotic and abiotic factors such as rainfall, temperature and humidity, which vary among ecosystems and with spatial and temporal scale (Ali and Yan 2017). The present study demonstrated a strong correlation between lianas and large tree diversity (Figure 4.6), suggesting a liana host tree interaction, which agrees with studies that explain how large trees are important for liana abundance in tropical forest (Ewango, 2010; Ewango et al., 2015; Fadrique and Homeier, 2016). Lianas as structural parasites generally rely on trees for support (Parren, 2003; Ewango, 2010). These ideas align with the niche complementarity and mass ratio hypotheses of Ali and Yan, (2017), which explain the effect of functional diversity on carbon stock. Most previous studies on biomass and carbon have been carried out in lowland forest (Brown, 1997; Djomo, 2010; Djomo et al., 2010; Djuikouo et al., 2010; Day et al., 2013; Chave et al., 2015; Djomo et al., 2016; Memiahge et al., 2016). The present study covers different forest types at different elevations, from the lowlands at 50 m to montane forest at 1778 m. Although biomass and carbon decreases with elevation, our results were not straight forward as lowland and montane forest reveals high biomass and carbon than midelevation and submontane forest (Figure 4.7). Based on the results obtained in this study, it plausible to suggest that lowland elevation, high species diversity and high carbon are associated while low carbon and low species diversity are associated. Generally, the overall totals for biomass and carbon were high in RHFR per hectare compared to other sites in Africa (Sonwa et al., 2011). Our lowland forest with a mean biomass of t ha -1 was slightly lower compared to other lowland tropical forests, with 402 t ha -1 in (Djuikouo et al., 2010), 404 t ha -1 in (Lewis et al., 2009), and 429 t ha -1 in the Congo Basin (Lewis et al., 2013). This is not surprising as the previous studies used the earlier formula by Chave et al., (2005) which differs from the more recently revised version of Chave et al., (2015) that was used in this study. Discrepancies are bound when the same data set is analyzed with different allometric equations (Brown et al., 1989; Brown, 1997; Chave et al., 2005; Djomo, 2010; Chave et al., 2015; Djomo et al., 2016). 92

119 Although mid-elevation and submontane forest showed lower values compared to lowland and montane forest, they still fell within the high end of the range among other studies in Africa (Brown, 1997; Makana, 2010; Djuikoko et al., 2010; Lewis et al., 2009; Lewis et al., 2013; Kupsch et al., 2014; Ekoungoulou et al., 2014; Kenfack et al., 2014; Memiaghe et al., 2016). The Rumpi Hills, on a per hectare basis, had its maximum biomass in a lowland plot of t ha -1, which accorded well with a hectare plot in the Monts de Cristal National Park in Gabon (619 t ha -1 ), the highest value in the entire summary (Day et al., 2013). In a similar study in the Albertine Rift (Imani et al., 2017), biomass ranged from 168 t ha -1 in upper montane to 290 t ha -1 in middle montane forest, lower than the values we recorded. Thus, the intact lowland to montane forest continuum in the Rumpi Hills is a potential carbon sink and site for the implementation of REDD+ mechanisms, climate change and forest dynamics. 93

120 Figure 4.1 Sample points and four plant communities at the Rumpi Hills Forest Reserve, Cameroon 94

121 Figure 4.2 (A) Variation of tree density and basal area, (B) Mean tree density/ha, (C) Mean basal area m 2 /ha in four forest types in the Rumpi Hills Forest Reserve, Cameroon 95

122 Shannon diversity (Trees => 10 cm) Elevation (m) Figure 4.3 Regression analyses showing a strong significant negative relationship between species richness and elevation (r 2 = 0.756, p < 0.05;) across 25 ha plot in the Rumpi Hills Forest Reserve, Cameroon. 96

123 Figure 4.4 Correlation of number of lianas and number of large trees 60 cm (r 2 = , P <0.05). 97

124 Carbon (t/ha) Elevation (m) Figure 4.5 Association between carbon (tc/ha) and elevation (m) of trees with dbh >10 cm recorded in 25 1-ha plots in the Rumpi Hills Forest Reserve, Cameroon. The relationship has a slope of 0.061, which is significantly different from a slope of zero (P < 0.05), y=-0.11x

125 y = x Carbon (t/ha) Number of species/ha Figure 4.6 Association between carbon (tc/ha) and number of species/ha with dbh >10 cm recorded in 25 1-ha plots in the Rumpi Hills Forest Reserve, Cameroon; r 2 = 0.34; y = x

126 Figure 4.7 Correspondence analysis reflecting high tree density, number of species, carbon in lowland and Montane forest and low tree density, number of species, and carbon in mid-elevation and submontane forest of the Rumpi Hills forest Reserve in Cameroon. 100

127 Table 4.1 Tree diversity, above ground biomass (AGB), and carbon in lowland, mid-elevation, submontane, and montane cloud forest. BA (basal area), N (tree density), SR (species richness), SW (Shannon-Weiner index). Plot Forest type Tree/ha (N) BA (m 2 /ha) SR SW AGB (t/ha) Carbon (t/ha) 1 Lowland Lowland Lowland Lowland Lowland Lowland Lowland Lowland Lowland Lowland Lowland Lowland Submontane Submontane Submontane Montane Montane Mid-elevation Mid-elevation Mid-elevation Mid-elevation Mid-elevation Mid-elevation Mid-elevation Mid-elevation Total Mean Standard deviation

128 Table 4.2 Summary of total number of individuals, basal area, number of species, and Shannon diversity for the different life forms recorded in four forest types at various size classes in the Rumpi Hills Forest Reserve, Cameroon Forest type Life form Size class (cm dbh) N BA S Fisher s alpha H Lowland Lianas Lowland Trees Lowland Trees Lowland Trees Mid-elevation Lianas Mid-elevation Trees Mid-elevation Trees Mid-elevation Trees Submontane Lianas Submontane Trees Submontane Trees Submontane Trees Montane Lianas Montane Trees Montane Trees Montane Trees Fisher s alpha was based on 12 ha in lowland, 8 ha in mid-elevation, 3 ha in submontane and 2 ha in montane forest. Fisher s alpha was not calculated for lianas in montane forest (0) because the values were too low. N=Number of individuals stands, BA=Basal Area, S= Number of Species, Fisher s= Fisher s alpha, H = Shannon-wiener diversity index. Basal area includes all multiple stems for each individual stem. Multiple stems excluded in all other parameters. 102

129 Table 4.3 Summary of biomass, and carbon in four forest types across Rumpi Hills Forest Reserve, Cameroon Plot Forest Types AGB (t ha -1 ) Carbon (t ha -1 ) 1 Lowland Lowland Lowland Lowland Lowland Lowland Lowland Lowland Lowland Lowland Lowland Lowland Total 4,633 2,317 Mean Standard deviation Mid-Elevation Mid-Elevation Mid-Elevation Mid-Elevation Mid-Elevation Mid-Elevation Mid-Elevation Mid-Elevation Total 2, ,011 Mean Standard deviation Submontane Submontane Submontane Total

130 Mean Standard deviation Montane Montane Total Mean Standard deviation

131 Table 4.4 Summary of 52 species with biomass, and carbon in the Rumpi Hills Forest Reserve, Cameroon Species Abundance AGB (t ha -1 ) Carbon (t ha -1 ) Scyphocephalium mannii Pycnanthus angolensis Cola verticilata Oubanguia alata Korupiodendron songweanum Protomegabaria stapfiana Piptadeniastrum africanum Vepris soyauxii Berlinia brateosa Santiria balsamifera Syzygium rowlandii Strombosia sp Irvingia gabonensis Pseudospondias microcarpa Erythrophleum ivorensis Strombosiopsis tetrandra Strombosia grandifolia Eriocoelum macrocarpum Staudtia kamerunensis Carapa oreophilia

132 Pellegriniodendron diphyllum Coula edulis Homalium longistylum Xylopia africana Vitex grandifolia Margaritaria discoidea Lecomtedoxa klaineana Trichoscypha cf oliveri Anthonotha macrophylla Coelocaryon preussi Albizia adianthifolia Hypodaphnis zenkeri Sapium ellipticum Trichoscypha sp Desbordesia glaucescens Gambeya africanum Musanga cecropioides Gambeya subnudum Poga oleosa Terminalia ivorensis Dacryodes edulis Mammea africana Vitex sp

133 Strombosia pustulata Alstonia boonei Talbotiella korupensis Zanthoxylon gilletii Afrostyrax lepidophyllus Staudtia gabunensis Barteria fistulosa Omphalocarpum elatum Symphonia globulifera Total General Total in survey Percentage

134 Appendix 4.1. Species occurrence in the different forest types in the Rumpi Hills Forest Reserve, Cameroon Family Species Lowland Cameroon Midelevation Status Cameroon Line/Guinea IUCN Endemic to Volcanic Montane Submontane Forest Achariaceae Dasylepis thomasii VU - Endemic Achariaceae Scottellia klaineana LC - - Alangiaceae Alangium chinense LC - - Anacardiaceae Antrocaryon micraster LC - - Anacardiaceae Antrocaryon sp Anacardiaceae Pseudospondias microcarpa NE - Endemic Anacardiaceae Sorindeia grandifolia LC - - Anacardiaceae Sorindeia juglandifolia LC - - Anacardiaceae Trichoscypha acuminata LC - - Anacardiaceae Trichoscypha cf oliveri LC - - Anacardiaceae Trichoscypha patens LC - - Anacardiaceae Trichoscypha preussii Anacardiaceae Trichoscypha sp Anacardiaceae Trichoscypha sp Anacardiaceae Trichoscypha sp Anacardiaceae Trichoscypha sp

135 Anacardiaceae Trichoscypha sp Anisophylleaceae Anisophyllea meniaudii DD - - Anisophylleaceae Anisophyllea polyneura LC - - Anisophylleaceae Anisophyllea purpurascens DD - - Anisophylleaceae Anisophyllea sororia DD - - Anisophylleaceae Poga oleosa DD - - Annonaceae Annickia chlorantha LC - - Annonaceae Cleistopholis patens LC - - Annonaceae Cleistopholis staudtii NT - - Annonaceae Hexalobus sp Annonaceae Isolona campanulata LC - - Annonaceae Monodora brevipes LC - - Annonaceae Monodora myristica LC - - Annonaceae Pachypodanthium staudtii LC - - Annonaceae Piptostigma oyemense Annonaceae Piptostigma pilosum LC - - Annonaceae Piptostigma sp Annonaceae Polyanthia suaveolens LC - - Annonaceae Polyceratocarpus parviflorus LC - - Annonaceae Uvariastrum pynaertii LC - - Annonaceae Uvariodendron connivens NT - Endemic Annonaceae Uvariodendron giganteum NT - - Annonaceae Uvariopsis korupensis EN Endemic - 109

136 Annonaceae Uvariopsis submontana EN Endemic - Annonaceae Xylopia acutiflora LC - - Annonaceae Xylopia aethiopica LC - - Annonaceae Xylopia africana VU - Endemic Annonaceae Xylopia hypolampra LC - - Annonaceae Xylopia sp Annonaceae Xylopia sp.i Annonaceae Xylopia staudtii LC - - Annonaceae Xylopia villosa LC - - Apocynaceae Alstonia boonei LC - - Apocynaceae Funtumia elastica LC - - Apocynaceae Hunteria umbellata LC - - Apocynaceae Pleiocarpa bicarpellata LC - - Apocynaceae Pleiocarpa sp Apocynaceae Rauvolfia caffra LC - - Apocynaceae Rauvolfia vomitoria LC - - Apocynaceae Tabernaemontana brachyantha LC - - Apocynaceae Tabernaemontana crassa LC - - Apocynaceae Tabernaemontana ventricosa LC - - Apocynaceae Voacanga africana LC - - Araliaceae Polyscias fulva NT - - Araliaceae Schefflera abyssinica LC - - Asteraceae Vernonia conferta LC

137 Asteraceae Vernonia frondosa LC - - Bignoniaceae Kigelia africana LC - - Bombacaceae Bombax buenopozense LC - - Bombacaceae Ceiba pentandra LC - - Boraginaceae Cordia sp Burseraceae Canarium schweinfurthii LC - - Burseraceae Canthium sp Burseraceae Dacryodes edulis LC - - Burseraceae Dacryodes klaineana LC - - Burseraceae Santiria balsamifera LC - - Caricaceae Cylicomorpha solmsii VU Endemic - Cecropiaceae Musanga cecropioides LC - - Cecropiaceae Myrianthus arboreus LC - - Cecropiaceae Myrianthus preussii NT - Endemic Chrysobalanaceae Chrysobalanus icaco LC - - Chrysobalanaceae Dactyladenia pallescens LC - - Chrysobalanaceae Dactyladenia staudtii LC - - Chrysobalanaceae Magnistipula aff. cuneatifolia EN - - Chrysobalanaceae Magnistipula glaberrima NT - - Chrysobalanaceae Maranthes kerstingii LC - - Chrysobalanaceae Parinari chrysophylla LC - - Chrysobalanaceae Parinari excelsa LC - - Clusiaceae Allanblackia gabonensis VU - Endemic 111

138 Clusiaceae Endodesmia calophylloides LC - - Clusiaceae Garcinia cf polyantha Clusiaceae Garcinia conrauana NT - - Clusiaceae Garcinia gnetoides LC - - Clusiaceae Garcinia kola VU - - Clusiaceae Garcinia mannii LC - - Clusiaceae Garcinia polyantha Clusiaceae Garcinia sp Clusiaceae Garcinia staudtii NT - Endemic Clusiaceae Mammea africana LC - - Clusiaceae Pentadesma butryacea LC - - Clusiaceae Pentadesma grandifolia Clusiaceae Symphonia globulifera LC - - Combretaceae Strephonema pseudocola LC - - Combretaceae Terminalia ivorensis LC - - Combretaceae Terminalia superba LC - - Dichapetalaceae Tapura africana LC - - Ebenaceae Diospyros bipindensis LC - - Ebenaceae Diospyros cinnabarina LC - - Ebenaceae Diospyros gabunensis LC - - Ebenaceae Diospyros gracilescens LC - - Ebenaceae Diospyros hoyleana LC - - Ebenaceae Diospyros iturensis LC

139 Ebenaceae Diospyros kamerunensis LC - - Ebenaceae Diospyros korupensis VU Endemic - Ebenaceae Diospyros sp Ebenaceae Diospyros sp Ebenaceae Diospyros sp Ebenaceae Diospyros suaveolens LC - - Ebenaceae Diospyros zenkeri LC - - Erythroxylaceae Erythroxylum mannii LC - - Euphorbiaceae Cleistanthus letouzeyi VU - Endemic Euphorbiaceae Croton sylvaticus LC - - Euphorbiaceae Dichostemma glaucescens LC - - Euphorbiaceae Discoclaoxylon hexandrum LC - - Euphorbiaceae Discoglypremna caloneura LC - - Euphorbiaceae Elaeophorbia drupifera LC - - Euphorbiaceae Euphorbia sp Euphorbiaceae Jatropha curcas LC - - Euphorbiaceae Klaineanthus gaboniae LC - - Euphorbiaceae Macaranga monandra LC - - Euphorbiaceae Macaranga sp Euphorbiaceae Macaranga spinosa LC - - Euphorbiaceae Mareya micrantha LC - - Euphorbiaceae Mareyopsis longifolia LC - - Euphorbiaceae Neoboutonia glabrescens

140 Euphorbiaceae Ricinodendron heudelotii LC - - Euphorbiaceae Sapium ellipticum Euphorbiaceae Tetrorchidium didymostemon LC - - Fabaceae Afzelia bella LC - - Fabaceae Afzelia bipindensis VU - - Fabaceae Afzelia pachyloba VU - - Fabaceae Albizia adianthifolia LC - - Fabaceae Albizia ferruginea LC - - Fabaceae Albizia sp Fabaceae Albizia sp Fabaceae Albizia zygia LC - - Fabaceae Amphimas ferrugineus LC - - Fabaceae Angylocalyx oligophyllus LC - - Fabaceae Anthonotha cladantha LC - - Fabaceae Anthonotha fragrans LC - - Fabaceae Anthonotha lamprophylla LC - - Fabaceae Anthonotha macrophylla LC - - Fabaceae Aphanocalyx microphyllus LC - - Fabaceae Baikiaea insignis LC - - Fabaceae Baphia buettneri NT - Endemic Fabaceae Baphia capparidifolia LC - - Fabaceae Baphia laurifolia LC - - Fabaceae Baphia sp

141 Fabaceae Berlinia auriculata LC - - Fabaceae Berlinia bracteosa LC - - Fabaceae Berlinia hollandii LC - - Fabaceae Calpocalyx dinklagei LC - - Fabaceae Calpocalyx heitzii NT - Endemic Fabaceae Copaifera mildbraedii LC - - Fabaceae Crudia sp Fabaceae Cryptosepalum cougolanum Fabaceae Dialium bipindense LC - - Fabaceae Dialium dinklagei LC - - Fabaceae Dialium pachyphyllum LC - - Fabaceae Didelotia africana LC - - Fabaceae Didelotia morelii LC - - Fabaceae Distemonanthus benthamianus LC - - Fabaceae Erythrina milbraedii LC - - Fabaceae Erythrophleum ivorensis LC - - Fabaceae Gilbertiodendron demonstrans NT - Endemic Fabaceae Gilbertiodendron dewevrei LC - - Fabaceae Gilbertiodendron sp Fabaceae Hylodendron gabunense LC - - Fabaceae Hymenostegia korupensis LC - - Fabaceae Leonardoxa africana LC - Endemic Fabaceae Microberlinia bisulcata VU - Endemic 115

142 Fabaceae Newtonia griffoniana LC - - Fabaceae Parkia bicolor LC - - Fabaceae Parkia sp Fabaceae Pellegriniodendron diphyllum LC - - Fabaceae Pentaclethra macrophylla LC - - Fabaceae Piptadeniastrum africanum LC - - Fabaceae Plagiosiphon longitubus LC - Endemic Fabaceae Pterocarpus soyauxii LC - - Fabaceae Talbotiella korupensis EN Endemic - Fabaceae Tetraberlinia bifoliolata LC - - Fabaceae Tetraberlinia korupensis EN Endemic - Fabaceae Tetraberlinia polyphylla Fabaceae Tetrapleura tetraptera LC - - Fabaceae Zenkerella citrina NT - Endemic Gentianaceae Anthocleista schweinfurthii LC - - Hoplestigmataceae Hoplestigma klaineanum NT - Endemic Huaceae Afrostyrax kamerunensis LC - - Huaceae Afrostyrax lepidophyllus LC - - Humiriaceae Saccoglottis gabonensis LC - - Hypericaceae Harungana madagascariensis LC - - Irvingaceae Desbordesia glaucescens LC - - Irvingaceae Irvingia gabonensis LC - - Irvingaceae Irvingia grandifolia LC

143 Irvingaceae Klainedoxa gabonensis LC - - Irvingaceae Klainedoxa trillesii LC - - Lamiaceae Vitex grandifolia LC - - Lamiaceae Vitex sp Lamiaceae Vitex sp Lamiaceae Vitex sp Lauraceae Beilschmiedia acuta DD - Endemic Lauraceae Beilschmiedia gabonensis DD - Endemic Lauraceae Beilschmiedia mannii LC - - Lauraceae Beilschmiedia sp Lauraceae Beilschmiedia sp Lauraceae Beilschmiedia sp Lauraceae Beilschmiedia sp Lauraceae Beilschmiedia talbotiae LC - - Lauraceae Hypodaphnis zenkeri LC - - Lauraceae Persea americana LC - - Lecythidaceae Crateranthus talbotii VU - Endemic Lecythidaceae Napoleonaea cf heudelotii Lecythidaceae Napoleonaea ergortonii VU - Endemic Lecythidaceae Napoleonaea talbotii NT - Endemic Lecythidaceae Oubanguia alata LC - - Lecythidaceae Oubanguia laurifolia LC - - Lecythidaceae Rhaptopetalum coriaceum NT - Endemic 117

144 Lecythidaceae Rhaptopetalum depressum CR Endemic - Lecythidaceae Rhaptopetalum geophylax EN Endemic - Lecythidaceae Rhaptopetalum sp Lecythidaceae Rhaptopetalum sp.nov Lecythidaceae Scytopetalium klaineanum LC - - Lepidobotryaceae Lepidobotrys staudtii LC - - Leptaulaceae Leptaulus daphnoides LC - - Leptaulaceae Leptaulus grandifolius VU - Endemic Loganiaceae Strychnos congolana LC - - Malvaceae Cola altissima LC - - Malvaceae Cola cauliflora LC - Endemic Malvaceae Cola chlamydantha LC - - Malvaceae Cola flaviflora NT - Endemic Malvaceae Cola lateritia LC - - Malvaceae Cola lepidota LC - - Malvaceae Cola megalophylla EN - Endemic Malvaceae Cola nitida LC - - Malvaceae Cola pachycarpa Malvaceae Cola rostrata LC - Endemic Malvaceae Cola semecarpophylla LC - Endemic Malvaceae Cola sp Malvaceae Cola sp.nov Malvaceae Cola verticillata LC

145 Malvaceae Leptonychia lasiogyne LC - - Malvaceae Leptonychia macrantha LC - - Malvaceae Microcos coriacea LC - - Malvaceae Octolobus spectabilis LC - - Malvaceae Sterculia tragacantha LC - - Melastomataceae Memecylon afzelii LC - - Melastomataceae Memecylon candidum LC - - Melastomataceae Memecylon lateriflorum Melastomataceae Memecylon laurentii LC - - Melastomataceae Memecylon sp Melastomataceae Warneckea cinnamomoides LC - - Melastomataceae Warneckea jasminoides LC - - Melastomataceae Warneckea membranifolia LC - - Melastomataceae Warneckea pulcherrima LC - - Melastomataceae Warneckea sp.i Meliaceae Carapa angustifolia VU - Endemic Meliaceae Carapa cf dinklagei NE - - Meliaceae Carapa oreophila NT - Endemic Meliaceae Carapa parviflora NE - - Meliaceae Carapa zemagoana VU Endemic - Meliaceae Entandrophragma cylindricum VU - - Meliaceae Guarea cedrata VU - - Meliaceae Guarea sp

146 Meliaceae Guarea thompsonii LC - - Meliaceae Khaya ivorensis VU - - Meliaceae Trichilia aff.gilgiana LC - - Meliaceae Trichilia prieureana LC - - Meliaceae Turraeanthus africanus LC - - Melianthaceae Bersama abyssinica LC - - Monimiaceae Glossocalyx brevipes var letouzeyi NT Endemic - Moraceae Ficus mucuso LC - - Moraceae Ficus sp Moraceae Ficus sp Moraceae Ficus sur LC - - Moraceae Milicia excelsa LC - - Moraceae Treculia africana LC - - Moraceae Treculia obovoidea LC - - Myristicaceae Coelocaryon preussii LC - - Myristicaceae Pycnanthus angolensis LC - - Myristicaceae Scyphocephalium mannii LC - - Myristicaceae Staudtia gabunensis LC - - Myristicaceae Staudtia kamerunensis LC - - Myristicaceae Staudtia sp Myrsinaceae Maesa kamerunensis LC - - Myrsinaceae Maesa lanceolata LC

147 Myrtaceae Eugenia callophyloides Myrtaceae Eugenia sp Myrtaceae Eugenia talbotii Myrtaceae Syzygium rowlandii LC - - Myrtaceae Syzygium staudtii LC - - Ochnaceae Campylospermum calanthum CR Endemic - Ochnaceae Lophira alata VU - - Ochnaceae Rhabdophyllum affine LC - - Octoknemataceae Octoknema affinis LC - - Octoknemataceae Octoknema anuwinlencis Octoknemataceae Octoknema bakosiensis EN Endemic - Olacaceae Coula edulis LC - - Olacaceae Diogoa sp Olacaceae Diogoa zenkeri LC - - Olacaceae Engomegoma gordonii LC - - Olacaceae Strombosia grandifolia LC - - Olacaceae Strombosia pustulata LC - - Olacaceae Strombosia scheffleri LC - - Olacaceae Strombosia sp Olacaceae Strombosia sp Olacaceae Strombosiopsis tetrandra LC - - Pandaceae Panda oleosa LC - - Passifloraceae Barteria fistulosa LC

148 Peridiscaceae Soyauxia gabunensis VU - Endemic Peridiscaceae Soyauxia talbotii VU - Endemic Phyllanthaceae Antidesma chevaleri LC - - Phyllanthaceae Antidesma laciniatum LC - - Phyllanthaceae Antidesma sp Phyllanthaceae Antidesma vogelianum LC - - Phyllanthaceae Bridelia grandis LC - - Phyllanthaceae Bridelia micrantha LC - - Phyllanthaceae Keayodendron bridelioides LC - - Phyllanthaceae Maesobotrya barteri LC - - Phyllanthaceae Maesobotrya dusenii Phyllanthaceae Maesobotrya sp Phyllanthaceae Maesobotrya staudtii LC - - Phyllanthaceae Margaritaria discoidea LC - - Phyllanthaceae Protomegabaria stapfiana LC - - Phyllanthaceae Thecacoris leptobotrya LC - - Phyllanthaceae Uapaca acuminata LC - - Phyllanthaceae Uapaca guineensis LC - - Phyllanthaceae Uapaca staudtii LC - - Putranjivaceae Drypetes aframensis Putranjivaceae Drypetes afzelii Putranjivaceae Drypetes gossweileri Putranjivaceae Drypetes ivorensis

149 Putranjivaceae Drypetes laciniata LC - - Putranjivaceae Drypetes molunduana NT - - Putranjivaceae Drypetes paxii LC - - Putranjivaceae Drypetes sp Putranjivaceae Drypetes sp Putranjivaceae Drypetes sp Putranjivaceae Sibangea similis LC - - Rhamnaceae Maesopsis eminii LC - - Rubiaceae Aoranthe cladantha LC - - Rubiaceae Aulacocalyx jasminiflora LC - - Rubiaceae Aulacocalyx talbotii LC - - Rubiaceae Bertiera racemosa LC - - Rubiaceae Calycosiphonia spathicalyx LC - - Rubiaceae Coffea sp Rubiaceae Craterispermum aristatum VU - Endemic Rubiaceae Cuviera subuliflora LC - - Rubiaceae Hallea ledermannii Rubiaceae Heinsia crinita LC - - Rubiaceae Ixora guineensis LC - - Rubiaceae Morelia senegalensis LC - - Rubiaceae Morinda lucida LC - - Rubiaceae Nauclea diderrichii VU - - Rubiaceae Nauclea sp

150 Rubiaceae Pauridiantha floribunda LC - - Rubiaceae Pauridiantha sp Rubiaceae Pauridiantha viridiflora LC - - Rubiaceae Pausinystalia macroceras LC - - Rubiaceae Pavetta staudtii LC - Endemic Rubiaceae Petitiocodon parviflorum LC - Endemic Rubiaceae Psychotria bimbiensis CR Endemic - Rubiaceae Psychotria brachyantha LC - - Rubiaceae Psychotria peduncularis LC - - Rubiaceae Psychotria sp Rubiaceae Psydrax sp Rubiaceae Rothmannia hispida LC - - Rubiaceae Rothmannia sp Rubiaceae Schumanniophyton magnificum LC - - Rubiaceae Tarenna baconoides EN - Endemic Rubiaceae Tarenna grandiflora LC - - Rubiaceae Tricalysia amplexicaulis Rubiaceae Tricalysia coriacea LC - - Ruscaceae Dracaena arborea LC - - Ruscaceae Dracaena deisteliana NT - Endemic Rutaceae Vepris adamaoua Rutaceae Vepris lecomteana NT - Endemic Rutaceae Vepris soyauxii VU - Endemic 124

151 Rutaceae Vepris sp Rutaceae Zanthoxylon gilletii LC - - Rutaceae Zanthoxylon heitzii LC - - Rutaceae Zanthoxylon lemairei LC - - Rutaceae Zanthoxylon sp Rutaceae Zanthoxylon sp.i Salicaceae Casearia barteri LC - - Salicaceae Homalium africanum LC - - Salicaceae Homalium letestui LC - - Salicaceae Homalium longistylum LC - - Salicaceae Homalium macroptenum Salicaceae Homalium sp Salicaceae Oncoba glauca LC - - Salicaceae Oncoba lophocarpa VU Endemic - Salicaceae Oncoba mannii LC - - Salicaceae Oncoba sp Sapindaceae Allophylus africanus LC - - Sapindaceae Allophylus grandifolius LC - - Sapindaceae Allophylus sp Sapindaceae Allophylus sp Sapindaceae Blighia sapida LC - - Sapindaceae Blighia welwitschii LC - - Sapindaceae Chytranthus angustifolius LC

152 Sapindaceae Chytranthus talbotii LC - - Sapindaceae Deinbollia pycnophylla EN - Endemic Sapindaceae Eriocoelum kerstingii LC - - Sapindaceae Eriocoelum macrocarpum LC - - Sapindaceae Laccodiscus pseudostipularis LC - - Sapindaceae Placodiscus cf. caudatus EN - Endemic Sapotaceae cf Pradosia spinosa NE - - Sapotaceae Englerophytum kennedyi Sapotaceae Englerophytum sp.nov Sapotaceae Gambeya africanum LC - - Sapotaceae Gambeya boukokoensis LC - - Sapotaceae Gambeya delevoyi Sapotaceae Gambeya korupensis VU Endemic - Sapotaceae Gambeya lacourtianum LC - - Sapotaceae Gambeya subnudum LC - - Sapotaceae Lecomtedoxa klaineana VU - Endemic Sapotaceae Manilkara argentea Sapotaceae Manilkara lososiana cr Endemic - Sapotaceae Manilkara pellegriniana DD - - Sapotaceae Omphalocarpum cf elatum LC - - Sapotaceae Omphalocarpum elatum LC - - Sapotaceae Pouteria sp Sapotaceae Synsepalum letouzeyi EN Endemic - 126

153 Sapotaceae Synsepalum longecuneatum Simaroubaceae Odyendya gabonensis LC - - Simaroubaceae Pierreodendron africanum LC - - Simaroubaceae Quassia silvestris LC - - Thymelaeaceae Dicranolepis pulcherrima LC - - Ulmaceae Trema orientalis LC - - Violaceae Alexis cf cauliflora LC - - Violaceae Rinorea dentata LC - - Violaceae Rinorea oblongifolia LC - - Vochysiaceae Erismadelphus exsul LC - - Vochysiaceae Korupiodendron songweanum EN - Endemic 127

154 Appendix 4.2. Summary table of distribution of biomass, and carbon Sequestered per hectare using Allometric models in four Tropical forest types in the Rumpi Hills Forest Reserve, Cameroon Plot Forest Types AGB (t/ha) Carbon (t/ha) 1 Lowland Lowland Lowland Lowland Lowland Lowland Lowland Lowland Lowland Lowland Lowland Lowland Mid-Elevation Mid-Elevation Mid-Elevation Mid-Elevation Mid-Elevation Mid-Elevation Mid-Elevation Mid-Elevation Submontane Submontane Submontane Montane Montane Total 8, ,072.1 Mean Standard deviation

155 CHAPTER FIVE DIVERSITY, ABOVE-GROUND BIOMASS, AND VEGETATION PATTERNS IN THE TROPICAL DRY FOREST OF KIMBI- FUNGOM NATIONAL PARK, CAMEROON Moses N. Sainge 1,2, Felix Nchu 3*, and A. Townsend Peterson 4 1 Tropical Plant Exploration Group (TroPEG), South West Region, Cameroon 2 Department of Environmental and Occupational Studies, Faculty of Applied Sciences, Cape Peninsula University of Technology, Cape Town 8000, South Africa 3 Department of Horticultural Sciences, Faculty of Applied Sciences, Cape Peninsula University of Technology, Bellville 7535, South Africa 4 Biodiversity Institute, University of Kansas, Lawrence, Kansas USA *Corresponding author, felixnchu@gmail.com Preface This chapter addresses the fourth objective of this study, which was to tackle the diversity, above ground biomass, and vegetation patterns in dry forest. ABSTRACT Considerable consensus exists regarding the importance of forests in carbon storage; however, the relationship between forest tree species composition and species richness and carbon Stock has not been fully established in the study area. The present study investigated the associations between above-ground biomass and species composition. The study further investigated how different vegetation types vary in terms of species composition, diversity, and carbon storage, in a dry forest in Kimbi-Fungom National Park, Cameroon. Vegetation was inventoried in 17 permanent 1-ha plots; using transects of 500 m long x 20 m wide. All trees and lianas with dbh 10 cm were measured with the aid of a diameter tape. In nested plots of 10x10 m, dbh 10 cm of trees and lianas were recorded with the help of a caliper. The multivariate analysis (TWINSPAN) revealed five vegetation types: semi-deciduous forest, secondary forest, mixed vegetation, gallery forest, and woody and grassland savanna. We found a mean of tree 129

156 stems ha -1 (range tree stems ha -1 ) and a mean of 43.1 species ha -1 (range species ha -1 ). The Shannon-Weiner diversity index was 2.5, with an average of 3.1, ranging A total of 148 specimen vouchers were collected, including 144 species collected outside plot sampling; 116 species recorded during observational data did not occur in the different sample plots. Allometric equations were used to calculate above-ground biomass and carbon. The five vegetation types had an average aboveground biomass of t ha -1 ranging t ha -1, and Carbon: 74.6 tc ha -1 ( tc ha -1 ) in the 17 ha analyzed. Correspondence analysis of data revealed two somewhat contrasting associations on the Axis 1: high elevation ( m), low tree density, low above-ground biomass, low carbon and low species diversity were associated, whereas low elevation ( ), high tree density, high species diversity, high carbon and high above-ground biomass showed weak correlations. All in all, the forests of the Kimbi-Fungom National Park were simultaneously poor in plant diversity and in both biomass and carbon, especially in the secondary, mixed vegetation, gallery, and grassland/woody savanna forest. Key words: Dry forest, Bamenda Highlands, Kimbi-Fungom National Park, carbon, semi-deciduous, tree composition, diversity 5.1 INTRODUCTION Globally, forest inventory and monitoring are key tools in understanding the structure, composition, diversity, above-ground biomass, and carbon storage of different vegetation types and habitats, and also serve as instruments by which to achieve targets for international agreements (UNEP/CBD, 2012). Dry forests are known to rank among the most threatened ecosystems globally, creating a need for detailed assessments of biodiversity hotspots, carbon stocks, and extent and preservation of these forests (Murphy and Lugo 1986; Janzen, 1988). Anthropogenic factors, such as land degradation, pastoral nomadism, and population expansion, are depleting dry forest extents, and natural factors like drought and fire also affect this biome. What is more, these ecosystems appear highly vulnerable even to small increases in temperature (UNFCCC, 2015). Cameroon, situated at the juncture of West and Central Africa, holds important extents of Lower Guinean forest (White, 1983; Poorter et al., 2003), holding rich biodiversity totaling almost 9000 species, 1800 genera, and 230 families of vascular plants (Olivry, 1986; Onana, 2011; Onana, 2015). Cameroon holds three main biomes: dry savanna, moist savanna, and tropical rain forest. Dry savanna covers the northern parts of the country, and moist savanna and tropical rainforest form a mosaic across much of the rest of 130

157 the country, except for montane areas. The dry savanna and moist semi-deciduous forest of Kimbi- Fungom National Park (KFNP) in the northwestern part of the country are assumed to be relatively species-poor, albeit with relatively few comparable studies (Sainge et al., 2014; Sainge, 2016). Indeed, fewer than 70 herbarium sheets, representing few than 60 plant species/100 km 2, have been collected from the KFNP area (Onana and Cheek, 2011; Onana, 2011; Onana, 2015). Although the vegetation of the Bamenda Highlands has been studied extensively (Cheek et al., 2000; Harvey et al., 2004; Cheek et al., 2010; T.F. Mbenkum and C.F Fisiy, pers. comm.), much remains to be understood regarding forest structure, species composition, species diversity, distribution, and carbon stocks across different vegetation types (Sainge et al., 2014; Sainge, 2016). Furthermore, few or no studies on carbon storage have been undertaken in dry forests in this region. Finally, the vegetation of the newly-established KFNP remains poorly understood; and since this park represents an important conservation effort in the region, a detailed understanding of its vegetation is paramount. Hence, this study aimed to understand patterns of plant species distribution and carbon storage across vegetation types, and their interrelationships in KFNP. Specific objectives were (i) to characterize forest structure; (ii) to characterize species composition, abundance, and distributions across vegetation types; and (iii) to estimate existing biomass and potential carbon storage. 5.2 Materials and Methods Study Site The study site lies within the Bamenda Highlands, in the North West Region of Cameroon, at latitude N and longitude E (Figure 1), covering km 2. This forest is a mixture of humid semievergreen forest, woody savanna, grassland savanna, and gallery forest of the Sudano-Zambazien forest (Letouzey, 1985), with different habitats such as swampy Pandanus forest, Raphia forest, and inselberg. KFNP is the only national park in the Bamenda Highlands that is surrounded by other protected areas: Mt Oku, (peak at 3011 m), Mbembe Forest Reserve, Mt Tabenken, Nkom-Wum Forest Reserve, Mbi Crater, Kagwene Wildlife Sanctuary, Bali Ngemba Forest Reserve, and Bafut Ngemba Forest Reserve. Plots were selected randomly and sampled, with 3-ha sampled in a dry semi-deciduous forest and the remaining 14-ha distributed in the dry forest of woody savanna, grassland savanna, gallery forest, swamp and secondary forest (Appendix 5A). 131

158 The climatic pattern in KFNP has two seasons within the Equatorial Cameroon climate type (Sainge, 2016), with dry season running November-mid March with < 100 mm for the season; December to February are the driest months. The rainy season occurs in April-October each year, with August and September the wettest months. No proper climatic data are available for this area; however, it is not expected to deviate much from the nearby Mbembe Forest, <100 km away, which has the following climatic conditions: rainfall ( mm), temperature (21-24 C). Soils of KFNP forest are ferruginous, brown to gray in color (Yerima and Ranst 2005) and soil acidity of around ph 5.6. Plots were established at the central part, west of the national park at elevations of m. During the study period, the woody and grassland savanna areas of the reserve were under intense cattle grazing, and the semi-deciduous forest was exposed to subsistence crop farming Field Sampling We used line transects of 500 x 20 m as plots, established at irregular intervals in the various vegetation types. Each plot was divided into 25 quadrats of 20 x 20 m. We established 17 of these plots (425 quadrats), on which all trees and lianas 10 cm diameter at breast height (1.3 m, dbh) were sampled. Five 10 x 10 m quadrats within each 1 ha were selected at every 5 th quadrat (i.e., quadrats 1, 6, 11, 16, and 21) where all trees and lianas of dbh <10 cm were measured, representing 0.05 ha sampled per hectare. Trees, shrubs, and lianas with dbh <10 cm were measured with calipers, whereas trees and lianas 10 cm were measured with a diameter tape. Lianas were measured above the last rooting points at a height of 1.3 m from the ground (Ewango, 2010; Ewango et al., 2015; Thomas et al., 2015). Tree height was determined as the average of visual estimates by 3 field staff base on an estimated guess. All sampled species were measured, and identified to morphospecies using field codes: voucher specimens were collected for all morphospecies. Dominant species were defined as species with the highest abundance of stems; rare species were those with only single stems for each vegetation type. Habitat type (swamp SW, flat dry forest FD, slope SL, plateau PL) was recorded for each plot. Finally, outside of plots sampled observational data were accumulated as we traversed the area to enhance the general species list for the area Data analysis TWINSPAN multivariate analysis was used to classify vegetation types using the PC-ORD package (McCune and Mefford 2006). Species diversity estimates, correspondence analysis were achieved using PAST version 2.17 (Hammer et al., 2001). Correspondent analysis was carried out by converting data into 132

159 binomials (0 and 1) and establishing relationship among elevation, species diversity, and carbon stock estimation using PAST version Data for each vegetation type were separated into different life forms: trees 10 cm, shrubs 10 cm, and lianas 1 cm. Forest structure was classified into three strata (life forms): <10 m, m (10-29 m), and 30 m height. Above-ground biomass (AGB) was estimated for all trees with dbh 10 cm following the allometric equation of Chave et al., 2015 (equation 1). Tree height was estimated following Djomo et al., 2016 (equation 2): D 2 H) (Chave et al., 2015 equation 5).Eq. 1 (Djomo et al., 2016).Eq. 2 AGB = Above-ground biomass (Tones), ρ = Wood specific density (g/cm 3 ) at 0% humidity D (dbh) = diameter at breast height (1.30 m), e indicates the exponential function, H = Height in meters. Carbon was estimated for trees 10 cm as C = Total Biomass/2...Eq. 3 The forest structure and composition was described using phytosociological parameters such as basal area, relative density, relative dominance, relative frequency, and the important value index (Dallmeier, 1992). Basal Area (BA) = Area occupied by plant (species) at breast height. Basal area (BA) = pi*(1/2dbh) 2 = pi*(dbh) 2 /4..Eq. 4 Shannon-Weiner index (H ) is the most convenient tool to measure diversity in 1-ha plots. This can be achieved through the following formula:..eq. 5. Where pi is the proportion of individual of a species (Number of individual of a species/total number of all species) and ln is the natural logarithm. Thus, the natural logarithm of the number of species (lns) is the maximum value of H. 133

160 5.3 RESULTS Composition and Diversity In total, 5551 stems of trees, shrubs, and lianas were recorded in 17 1-ha plots, dbh 1 cm stems of trees and lianas, dbh 1 cm, belonging to 201 morphospecies with an average density of 293 stems ha -1 was recorded. However, 564 trees and lianas had multiple stems. In summary, 4607 tree, were recorded with dbh 10 cm, representing 178 species, 110 genera, and 42 families; 350 trees with dbh <10 cm representing 84 species, 72 genera, and 33 families, and 30 stems of lianas 1 cm (27 stems with dbh 10 cm and 3 stems with dbh <10 cm) representing 15 species, 15 genera, and 11 families (Table 5.1). In all, some 147 species (76.9%) were identified to the species level, as were 118 genera (83.6%). The mean number of trees ha -1 with dbh 10 cm in KFNP was 270±74 trees ha -1 ( trees ha -1 ). Shrubs with dbh <10 cm had an average of 135 trees ha -1, with a range of trees ha -1. Lianas with dbh 1 cm had a mean of 2.8 stem ha -1, with a range of 1-6 stems ha -1. Species richness and diversity varied among plots and life forms, with a mean of 43±13 species ha -1, ranging from species ha -1. The Shannon-Weiner diversity index was 2.5, with an average of 3.1, ranging A total of 148 specimen vouchers were collected, including 144 species collected outside plot sampling; 116 species recorded during observational data did not occur in the different sample plots (Appendix 5B) Basal Area The 17 ha plots gave a total basal area (dbh 10 cm) of m 2, with a mean of 15.1 m 2 ha -1 (range m 2 ha -1 ). The dominant family was Fabaceae (87.0 m 2, 33.2%; Table 2), followed by Chrysobalanaceae (27.3 m 2 ), >Phyllanthaceae (21.6 m 2 ), >Anacardiaceae (19.0 m 2 ), >Combretaceae (11.0 m 2 ). Dominant genera were Brachystegia (31.4 m 2 ), Maranthes (26.3 m 2 ), Uapaca (17.7 m 2 ), Daniellia (17.1 m 2 ), Pseudospondias (11.3 m 2 ) and Terminalia (10.3 m 2 ) (Table 5.3). Dominant species were Brachystegia eurycoma, Maranthes glabra, Daniellia oliveri, Uapaca togoensis, Pseudospondias microcarpa and Terminalia glaucescens. The total basal area for trees <10 cm dbh was 1.7 m 2 ha -1 whereas lianas gave 3.6 m 2 ha -1. Semi-deciduous forest had the largest basal area (Table 5.4) Forest Structure Tree height within the five vegetation types ranged 2 45 m. The 17 ha plots resulted in 4607 trees of 178 species, 110 genera, and 42 families in morphospecies for trees 10 cm dbh. Trees <10 m tall formed the 134

161 bulk of abundance, representing 66.6% (3068 tree stems), >10 m (10-29 m) was 29.0% (1336 tree stems), and 30 m gave 4.1% (190 stems). At the vegetation level, for trees <10 m, gallery forest represented 6.6% (201 stems), woody and grassland savanna 66.2% (2029 stems), mixed vegetation 2.1% (63 stems), semi-deciduous forest 19.8% (606 stems), and secondary forest 5.5% (169 stems; Tables 5.5) Classification and Vegetation Patterns Multivariate analyses using TWINSPAN revealed five vegetation types, dry semi-deciduous forest, here termed primary forest and four dry forest types (secondary forest, gallery forest, mixed vegetation, grassland/woody savanna) in Cameroon (Figure 5.2), with 4607 stems in 178 morphospecies, 110 genera, and 42 families. Twenty-one and twelve individuals were not identified to genus and family, respectively. Main and secondary forest matrices were based on abundances of tree species 10 cm that were all identified to species, measured for dbh, and with data on elevation (Figures 5.2 & 5.3). The floristic canonical correspondence analysis (CCA) reveals great variation within the different vegetation type with an eigenvalue of 0.8 expressed toward axis 2 and 0.2 expressed toward axis 1 along elevational gradient (Figure 5.3). In all, 7 plots had elements of semi-deciduous forest representing 4.52 ha with a total of 1559 stems in 130 species, 89 genera, and 39 families. Dominant species were Maranthes glabra, with 227 stems, Sorindeia grandifolia (95 stems), Spondianthus preussii (93 stems), Pseudospondias microcarpa (85 stems), Chrysophyllum ubanguiense (75 stems), and Brachystegia eurycoma (70 stems). In this vegetation type, 37 species were rare, with one individual each, such as Beilschmiedia gabonensis, Bridelia atroviridis, Daniellia oliveri, Englerophytum stelechanthum, and Shirakiopsis elliptica. Secondary forest (three plots) representing 0.96 ha, had 259 stems pertaining to 55 species, 44 genera, and 26 families. One morphospecies was identified only to genus and one only to family. Dominant species were Hallea stipulosa (36 stems), Ricinodendron heudelotii (18 stems), Albizia zygia (17 stems), Trema orientalis (12 stems), and Anthocleista djalonensis, and Sterculia tragacantha with 10 stems each. Eighteen species were rare in this vegetation type, with one individual each, such as Alstonia boonei, Daniellia oliveri, Erythrophleum suaveolens, Irvingia wombulu, and Quassia sylvestris. Gallery forest found in 8 plots representing 0.8 ha, had a total of 276 stems belonging to 53 species, 44 genera, and 24 families. Dominant species were Uapaca togoensis (96 stems), Daniellia oliveri (20 135

162 stems), Vitex doniana (13 stems), and Hymenocardia acida (11 stems); 19 rare species included Afzelia africana., Albizia adianthifolia, Cassia arereh, Cola cordifolia and Pterocarpus erinaceus. Mixed vegetation (4 plots) representing 0.4 ha, had 129 stems belonging to 37 species, 31 genera, and 21 families, with 2 morphospecies not identified to genus or family. Dominant species were Uapaca togoensis (20 stems), Maranthes glabra (15 stems), Vitex doniana (9 stems), and Nauclea latifolia (8 stems); 16 rare species included Annona senegalensis., Beilschmiedia anacardioides, Brachystegia eurycoma, Elaeis guineensis., and Vitex rivularis. Grassland/woody savanna (14 plots) representing ha, had a total of 2383 stems belonging to 77 species, 55 genera, and 29 families (3 morphospecies were not identified to genus, 28 morphospecies were not identified to family). Dominant species in this vegetation type were Hymenocardia acida (237 stems), Terminalia glaucescens (231), Crossopteryx febrifuga, Nauclea latifolia (225 stems), Lophira lanceolata (186 stems), Daniellia oliveri (147 stems), Entada abyssinica (116 stems), Piliostigma thonningii (111 stems), Cussonia arborea (105 stems), and Uapaca togoensis (86 stems). Rare species totaled 21, including Albizia adianthifolia, Antidesma chevalieri, Erythrina senegalensis Maesopsis eminii., Magnistipula butayei, Milicia excelsa, Morelia senegalensis, Pterocarpus erinaceus, and Uapaca paludosa (Table 4). Five quadrats were empty in the entire study representing 0.2 ha. In all, 16.8 ha were fully sampled. At an elevation of , high values of number of species, tree density, and aboveground biomass was experiences with an eigenvalue of 65% at axis 1 while at , low number of species, tree density, and above-ground biomass was recorded with at eigenvalue of 17% at axis 2 (Figure 5.4 a & b) Above-Ground Biomass and Carbon The 17 ha of sampled plots in KFNP yielded a total above-ground biomass of t, and carbon of t (Appendix 5C). Among the 11 families with highest above-ground biomass (AGB), Fabaceae had the highest AGB (914.9 t ha -1 ), corresponding to t ha -1 of carbon (Table 5.2). Brachystegia eurycoma recorded the highest AGB of any species (439.3 t ha -1 ) equivalent to t ha -1 of carbon (Table 5.3). Mean AGB by vegetation type was t ha -1 in mixed vegetation forest, 72 t ha -1 in grassland/woody savanna, 141 t ha -1 in gallery forest, t ha -1 in secondary forest, and t ha -1 in semi-deciduous forest (Table 5.4). The correspondence analysis shows that the first two axes accounted for 82% (axis 1 = 65% and axis 2 = 17%) of total variation in this study. Correspondence analysis of data revealed two somewhat contrasting associations on the Axis 1: high elevation ( m), low tree 136

163 density, low above-ground biomass and low species diversity were associated, whereas low elevation ( ), high tree density, high species diversity and high above-ground biomass showed weakly correlations. An overall species list and abundance in the 17 1-ha permanent plots were represented in appendix 5D. 5.4 DISCUSSION Tree diversity, density, and diameter are important indicators in assessing forest carbon and other ecological processes in tropical forests. However, these indicators vary across regions, vegetation types, and habitats. Generally, average tree density in the dry forest of KFNP was lower compared to similar tropical dry forests in other regions: for example mean tree densities of 994 stems ha -1 (dbh > 10 cm) and 3486 stems ha -1 (dbh > 1 cm) were documented in the tropical dry forest of Bannerghatta National Park of the Eastern Ghats in southern India and Hawaiian lowland dry forest, respectively (Ostertag et al., 2014; Gopalakrishna et al., 2015). A similar study in nearby Mbembe Forest Reserve at different vegetation types, gave an average of 741 stems ha -1 in woody savanna, 236 stems ha -1 in grassland savanna and 4963 stems ha -1 in semi-deciduous forest at dbh 1 cm (Sainge et al., 2014). Trees with dbh 10 cm in the Mbembe forest gave an average of 311 stems ha -1 in woody savanna, 124 stems ha -1 in grassland savanna and 408 stems ha -1 in semi-deciduous forest (Sainge et al., 2014). The low tree density in KFNP and indeed in the greater Bamenda Highlands could be attributed to unsustainable practices such as gathering of fuel wood, timber exploitation, pastoral nomadism, and subsistence agriculture. The tree species that were most affected were Afzelia africana, Daniella oliveri, Erythrophleum suaveolens, Hymenocardia acida, Terminalia glaucescens (Sainge et al., 2014; Sainge, 2016). Nevertheless, globally, tropical dry forests are highly threatened: recent reports are that tropical dry forests in Latin America and the Caribbean have been reduced to <10% of their original extent (Banda et al., 2016). The current results highlight the poor state of KFNP and the need for appropriate interventions. It worth-mentioning that this is one of the few studies in Cameroon and the Congo Basin sub-region that looks at diversity, aboveground biomass and carbon in a dry forest (Sainge et al., 2012; Sainge et al., 2014; Ekoungoulou et al., 2014; Sainge, 2016). Mean tree species richness (for trees with dbh 10 cm) of 43.1±13.3 species ha -1 (27-65 species ha -1 ) in KFNP were comparable to those obtained in the dry forest of the Western Ghats, in Milodai & Courtallam Forest Reserve, India, which ranged from species ha -1 (Chandrashekara and Ramakrishna 1994; Parthasarathy and Karthikeyan 1997). Studies of mature tropical forests revealed a minimum value for 137

164 species richness for trees with dbh 10 cm of 56 species (Philips and Gentry 1994). However, mean tree species richness of 43.1 species ha -1 (27-65 species ha -1 ) in KFNP, in comparison to rainforests of Rumpi Hills (lowland forest species ha -1, and submontane 75 species ha -1 ), and Korup National Park (88.5 species ha -1 ), is lowland (Newbery and Gartlan 1996; Thomas et al., 2003; Chuyong et al., 2004; Kenfack et al., 2007; Gonmadje et al., 2011). Thus, KFNP with a mean of 43.1 species ha -1 and Mbembe forest species ha -1 can be considered as quite species-poor. Species richness and diversity in KFNP shows results similar to nearby Mbembe Forest Reserve with a Shannon index of 3 (Sainge et al., 2014). In this study, the number of species and basal area was higher in dry semi-deciduous forest than in the other vegetation types (Table 5.4). This is evident, since semi-deciduous forest is more closed to lowland or mid-elevation rainforest with large trees, dense with fewer gaps than the open grassland/woody savanna that is prone to fire annually. The most abundant families were Fabaceae with 887 individual trees, Rubiaceae (783 trees) and Phyllanthaceae (503 trees) (Table 5.2), while the most abundance species were Terminalia glaucescens (271 trees), Maranthes glabra (Oliv.) (263 trees), and Uapaca togoensis (256 trees, Table 5.3). Ten species in our study yielded t of above-ground biomass (63.8%) of the total AGB ( t) with an overall abundance of 1489 tree stems (Table 5.3), suggesting that biomass is relatively dependent to tree diameter. In the 17 ha sampled, we calculated a mean AGB of t ha -1 ( t ha -1 ), and carbon 74.6 tc ha - 1 ( tc ha -1 ). These values are far lower compared to the minimum values of AGB (429 t ha -1 ) and carbon stock estimate of 249 tc ha -1 documented previously for Central African forests (Lewis et al., 2013). Although the present study revealed KFNP is poor in mean AGB and carbon, exceptions were observed for some specific plots: for example, plots 1-3 had high mean AGB of t ha -1 ( t ha -1 ), a mean carbon of tc ha -1 ( tc/ha), which is only slightly lower than values documented by Lewis et al (Appendix C). In Congo Brazzaville (Iboukikro and Ngambali Forest), a study conducted by Ekoungoulou et al. (2014) in 6 1-ha plots in a gallery forest revealed a mean of tc ha -1 ( tc ha -1 ), which is still higher than that of the gallery forest in KFNP (70.5 tc ha -1 ), in which the ten most abundant species showed high density overall and per vegetation type when compared to the general and respective totals (Table 5.6). We noted a close association among variables: low tree density, low above-ground biomass, low species diversity and high altitude in KFNP; conversely, high tree density, high species diversity and high above- 138

165 ground biomass were linked. Low carbon was closely associated with low tree density, low species diversity, and high altitude as opposed to high carbon, high tree density, high species diversity at low altitude. The semi-deciduous forest recorded the highest AGB and number of species, and mixed vegetation the lowest AGB (Table 5.4). Although previous studies of the relationship between species richness and carbon storage have yielded mixed results, most studies have found a positive association between species richness and above-ground productivity (Thompson et al., 2009; Strassburg et al., 2010). Whereas the low carbon content in the grassland/woody savanna, gallery, and mixed forest may be attributed to the open vegetation, and anthropogenic activities, other factors such as rainfall, duration of wet season, and topography can also influence net primary productivity of tropical dry forest (Murphy and Lugo1986). In conclusion, this study demonstrated a positive association between above-ground biomass and number of species per hectare. The current study revealed that the forest of the Kimbi-Fungom National Park is simultaneously poor in plant diversity, biomass, and carbon. 139

166 Figure 5.1. Vegetation types and sample locations across the Kimbi-Fungom National Park, Cameroon. 140

167 Figure 5.2. TWINSPAN dendrogram of 120 species of vascular plants with dbh 10 cm in 17 plots of 1- ha each in the Kimbi-Fungom National Park, Cameroon. 141

168 Figure 5.3. Floristic Canonical Correspondence Analysis (CCA) with 17 1-ha sample plots and 5 vegetation types from TWINSPAN analysis in the Kimbi-Fungom National Park, Cameroon. 142

169 Figure 5.4. Association between above-ground biomass and numbers of species of trees of dbh >10 cm recorded in 17 1-ha plots at the Kimbi-Fungom National Park, Cameroon. 143

170 Figure 5.5 Weak association (R 2 = , P<0.05) between carbon and elevation across different plots for tree species of dbh 10 cm recorded in 17 1-ha plots at the Kimbi-Fungom National Park, Cameroon. 144

171 Table 5.1. Different vegetation cover types and corresponding numbers of species, stems, and mean Shannon diversity index in Kimbi-Fungom National Park forest, Cameroon Vegetation cover No. of species No. of stems dbh Mean Shannon diversity index shrubs cm 1.8 (0-3.4) lianas cm 0.26 ( ) trees cm 3.12 ( ) Table 5.2. Eleven plant families in terms of highest above ground biomass, and carbon stock in the Kimbi-Fungom National Park, Cameroon Family Biomass (t/ha) Carbon (t/ha) BA (m 2 /ha) Abundance Fabaceae Chrysobalanaceae Anacardiaceae Phyllanthaceae Euphorbiaceae Combretaceae Myristicaceae Moraceae Ochnaceae Arecaceae Rubiaceae

172 Table 5.3. Ten species with highest above-ground biomass, and carbon in 17 1-ha plots in Kimbi-Fungom National Park, Cameroon. Biomass Carbon Basal Area Abundance Species (t/ha) (t/ha) (m 2 /ha) Brachystegia eurycoma Harms Maranthes glabra (Oliv.) G.T. Prance Daniellia oliveri (Rolfe) Hutch. & Dalziel Pseudospondias microcarpa (A.Rich.) Engl Erythrophleum suaveolens (Guill. & Perr.) Brenan Uapaca togoensis Pax Terminalia glaucescens Planch.ex Benth Pycnanthus angolensis (Welw.) Exell Ricinodendron heudelotii (Baill.) Pierre ex Baill. Lophira lanceolata Tiegh. Ex Keay

173 Table 5.4. Mean above-ground biomass, carbon, basal area, and species richness, across five vegetation types in Kimbi-Fungom National Park, Cameroon. Vegetation Types Biomass (t/ha) Carbon (t/ha) Basal Area (m 2 ) Number of Species Semi-deciduous forest Secondary forest Grassland/woody savanna Mixed vegetation Gallery forest Table 5.5. Vegetation types and corresponding mean tree height (in m; range in parentheses) in sample plots in Kimbi-Fungom National Park, Cameroon. Vegetation types Mean tree height (range) Semi-deciduous forest 21.6 (2-45) Secondary forest 15 (4-45) Grassland/woody savanna 12.2 (2-25) Mixed vegetation 15.1 (4-34) 147

174 Appendix 5.1. Sampling plot locations, number of species/ha and individual trees/ha in the Kimbi- Fungom National Park, Cameroon Plot Vegetation type Site Location Latitude (N) Longitude (S) Elevation (m) 1 PSF KFNP Kpep PSF KFNP Kpep PSF KFNP Kpep GS_MV_PSF KFNP Kpep PSF_GS KFNP Kpep GS_GAL_SF KFNP Kpep GS_GAL_SF KFNP Kpep WS_MV_GAL KFNP Kpep WS KFNP Kpep WS_PSF KFNP Kpep SF_GAL_WS KFNP Kpep GS_GAL KFNP Tunka-Esu GS_GAL KFNP Tunka-Esu GS KFNP Tunka-Esu GS_GAL KFNP Tunka-Esu GS_MV KFNP Tunka-Esu GS_PSF KFNP Tunka-Esu GF=Gallery forest, G/WS=Grassland/Woody savanna, MV=Mixed vegetation, PSF=Primary Semideciduous Forest, SF=Secondary Forest 148

175 Appendix 5.2 Plant species recorded in observational efforts (i.e., outside of sampling plots) in the Kimbi- Fungom National Park, Cameroon. Family Species Acanthaceae Acanthus montanus (Nees) T.Anderson Acanthaceae Asystasia decipiens Heine Amaranthaceae Amaranthus sp. Anacardiaceae Lannea kerstingii Engl. & K. Krauce Annonaceae Artabotrys aurantiacus Engl. & Diels. Annonaceae Uvaria sp. Annonaceae Xylopia sp. Apocynaceae Baissea axillaris (Benth.) Hua Apocynaceae Landolphia sp. Asclepiadaceae Marsdenia sp. Asparagaceae Asparagus flagellaris Asparagaceae Chlorophytum macrophyllum (A.Rich.) Aschers. Asteraceae Chromolaena odorata (L.) R.M.King & H.Robinson (nat.) Asteraceae Vernonia kotschyana Sch. Bip. Bignoniaceae Crescentia cujute Billb. & Beurl. (exo.) Chrysobalanaceae Dactyladenia barteri (Hook.f.ex Oliv.) G.T. Prance & F. White Chrysobalanaceae Magnistipula cuneatifolia Hauman Clusiaceae Garcinia cf barteri Colchicaceae Gloriosa simplex Combretaceae Agelaea pseudobliqua Commeliniaceae Palisota ambigua (P.Beauv.) C.B. Clarke Connaraceae Connarus griffonianus Baill. Connaraceae Jaundea pubescens Costaceae Costus spectabilis (Fenzl) K. Schum. Dichapetalaceae Dichapetalum sp. Dilleniaceae Tetracera masuiana De Wild. & T. Durand Dilleniaceae Tetracera sp. Dioscoreaceae Dioscorea alata L. Ebenaceae Diospyros monbuttensis Gürke Euphorbiaceae Shirakiopsis elliptica (Hochst.) Esser Fabaceae Albizia adianthifolia (Shum.) W.F. Wright Fabaceae Anthonotha macrophylla P. Beauv. Fabaceae Crotalaria macrocalyx Benth. Fabaceae Dalbergia sp. Fabaceae Dalbergiella welwitschii Baker Fabaceae Desmodium hirtum Guill. & Perr. 149

176 Fabaceae Fabaceae Fabaceae Fabaceae Fabaceae Fabaceae Fabaceae Gentianaceae Hypericaceae Hypericaceae Hypericaceae Lamiaceae Lamiaceae Lamiaceae Lamiaceae Leeaceae Loganiaceae Loganiaceae Malvaceae Malvaceae Malvaceae Malvaceae Marantaceae Melastomataceae Moraceae Musaceae Myristicaceae Myrtaceae Ochnaceae Ochnaceae Ochnaceae Ochnaceae Olacaceae Orchidaceae Orchidaceae Orchidaceae Orchidaceae Orchidaceae Orchidaceae Orchidaceae Orchidaceae Desmodium velutinum (Wild.) DC. Dialium zenkeri Harms Oddoniodendron micranthum (Harms) Baker Pseudarthria hookeri Wright & Arn. Sesbania sp. Tamarindus indica Linn. Tephrosia barbigera Welw.ex Bak. Anthocleista liebrechtsiana De Wild Psorospermum glaucum Psorospermum sp. Psorospermum sp.3 Lippia africana Solamestemum latifolius Vitex myrmecophila Vitex thyrsifolia Baker Leea guineensis G.Don Strychnos spinosa Lam. Strychnos tricalysioides Hutch. & M.B.Moss Cola millenii K. Schum. Microcos mollis Juss. Sida corymbosa Sterculia setigera Delile Megaphrynium macrostachyum (Benth.) Milne-Redh. Dissotis brazzae Cogn. Ficus craterostoma Mildbr. & Burret Ensete livingstonianum (J.Kirk) Cheesman Coelocaryon botryoides Verm. Eugenia obanensis Baker.f. Campylospermum calanthum (Gilg.) Farron Campylospermum excavatum (Van Tiegh.) Farron Campylospermum flavum (Schum. & Thonn.) Farron Rhabdophyllum affine (Hook.f.) Van Tiegh. Strombosia grandifolia Hook.f. Ancistrochilus rothschildianus O'Brien Ancistrorhynchus serratus Summerh Bulbophyllum colubrinum (Rchb.f.) Rchb.f. Bulbophyllum vuleanicum Eulophia euglossa (Rchb.f.) Rolfe Habenaria longinostris Habenaria longirostris Summerhayes Habenaria malacophylla Rchb.f. 150

177 Orchidaceae Liparis caillei Finet Orchidaceae Liparis guineensis Orchidaceae Nervilia sp. Orchidaceae Polystachya odorata Lindl. Orchidaceae Vanilla imperialis Kraenzl. Passifloraceae Adenia cissampeloides (Planch.ex Hook.) Harms Passifloraceae Adenia sp.1 Petiveriaceae Hilleria latifolia H. Walter Phyllanthaceae Bridelia micrantha (Hochst.) Baill. Phyllanthaceae Macaranga assas Amougou Phyllanthaceae Phyllanthus muellerianus (Kuntze) Exell Pittosporaceae Pittosporum viridiflorum Sims subsp. Dalzielii (Hutch.) Cuf. Proteaceae Protea madiensis Oliv. Rhizophoraceae Cassipourea zenkeri (Engl.) Alston Rosaceae Prunus africana (Hook.f.) Kalkman Rubiaceae Canthium cf henriquerianum Rubiaceae Euclinia longiflora Salisb. Rubiaceae Gardenia lutea Rubiaceae Gardenia vogelii Hook.f.ex Planch Rubiaceae Ixora anemodesma K. Schum Rubiaceae Ixora bauchiensis Hutch. & Dalziel Rubiaceae Leptactina sp. Rubiaceae Polysphaeria arbuscula K. Schum. Rubiaceae Psychotria cf ebensis K. Schum Rubiaceae Psychotria peduncularis (Salisb.) Steyerm. Rubiaceae Psychotria sp. Rubiaceae Psychotria vogeliana Benth. Rubiaceae Rothmannia ebamutensis Sonké Rubiaceae Rothmannia malaensis Ruscaceae Dracaena aubryana Brongn.ex E. Morren Ruscaceae Dracaena surculosa Lindl. Rutaceae Clausena anisata (Wild.) Hook.f.ex Benth. Sapindaceae Paullinia pinnata L. Sapotaceae Porteria pierrei (A.Chev.) Baehni Smilacaceae Smilax kraussiana Meisn. Thymelaeaceae Dicranolepis disticha Planch. Violaceae Rinorea dentata (P.Beauv.) O.Kuntze Zingerberaceae Aframomum daniellii (Hook.f.) K.Schum. Zingiberaceae Renealmia sp. 151

178 Appendix 5.3. Summary of density, number of species, above-ground biomass, carbon, and Basal area in 17 ha plot across five vegetation types in Kimbi-Fungom National Park, Cameroon. Plot Vegetation type Elevation (m) Stem Density Tree Density Number of Species AGB (t/ha) Carbon (t/ha) BA (m2/ha) 1 PSF PSF PSF GS_MV_PSF PF_GS GS_GAL_SF GS_GAL_SF WS_MV_GAL WS WS_PSF SF_GAL_WS GS_GAL GS_GAL GS GS_GAL GS_MV GS_PSF Total Mean Standard dev GF=Gallery forest, G/WS=Grassland/Woody savanna, MV=Mixed vegetation, PSF=Primary semideciduous Forest, SF=Secondary Forest Vegtype=Vegetation types, Elev. (m) =Elevation, SD=Stem Density, TD=Tree Density, #spp=number of species/ha, AGB (t/ha) = above ground biomass, CO 2 (t/ha) =Carbon dioxide, BA (m 2 /ha) =Basal area. 151

179 Appendix 5.4 Species list and abundance in the Kimbi-Fungom National Park, Cameroon Family Species GF G/WS MV PSF SF Total Anacardiaceae Lannea microcarpa Engl. & K. Krauce Anacardiaceae Lannea schimperi (Hochst.ex. A. Rich.) Engl Anacardiaceae Lannea sp Anacardiaceae Lannea sp Anacardiaceae Pseudospondias microcarpa (A.Rich.) Engl Anacardiaceae Sorindeia grandifolia Engl Annonaceae Annona senegalensis Pers Annonaceae Cleistopholis patens (Benth.) Engl. & Diels Annonaceae Cleistopholis sp Annonaceae Cleistopholis staudtii Engl. & Diels Annonaceae Monodora sp Annonaceae Monodora tenuifolia Benth Apocynaceae Alstonia boonei De Wild Apocynaceae Funtumia elastica (Preuss) Stapf Apocynaceae Holarrhena floribunda (G.Don) Dur & Schinz Apocynaceae Rauvolfia caffra Sond Apocynaceae Rauvolfia sp Apocynaceae Rauvolfia vomitoria Afzel Apocynaceae Voacanga africana Stapf Araliaceae Cussonia arborea Hochst.ex.A.Rich Araliaceae Polycias fulva (Hiern) Harms Arecaceae Elaeis guineensis Jacq Bignoniaceae Markhamia tomemtosa (Benth.) K. Schum Bignoniaceae Newbouldia laevis (P. Beauv.) Seeman ex Bureau Bignoniaceae Spathodea campanulata P. Beauv Bignoniaceae Stereospermum kunthianum Cham Bombacaceae Bombax buenopozense P. Beauv Burseraceae Canarium schweinfurthii Engl Burseraceae Dacryodes sp Cecropiaceae Musanga cecropioides R.Br.ex Tedlie Cecropiaceae Myrianthus arboreus P. Beauv Chrysobalanaceae Magnistipula butayei De Wild Chrysobalanaceae Magnistipula butayei subsp. balingembeaensis De Wild Chrysobalanaceae Maranthes glabra (Oliv.) G.T. Prance Chrysobalanaceae Parinari curatellifolia Planch.ex Benth Chrysobalanaceae Parinari sp Clusiaceae Garcinia cf mannii Oliv

180 Clusiaceae Garcinia epunctata Stapf Clusiaceae Mammea africana Sabine Clusiaceae Symphonia globulifera L.f Combretaceae Combretum sp Combretaceae Terminalia glaucescens Planch. ex Benth Ebenaceae Diospyros sp Euphorbiaceae Alchornea cordifolia (Schum. & Thonn.) Müll.Arg Euphorbiaceae Macaranga spinosa Müll.Arg Euphorbiaceae Neoboutonia velutina Prain Euphorbiaceae Ricinodendron heudelotii (Baill.) Pierre ex Baill Fabaceae Afzelia africana Sm Fabaceae Afzelia bipindensis Harms Fabaceae Albizia adianthifolia (Schum.) W.F. Wright Fabaceae Albizia sp Fabaceae Albizia zygia (DC.) J.F. Macbr Fabaceae Angylocalyx pynaertii De Wild Fabaceae Anthonotha macrophylla P.Beauv Fabaceae Baphia buettneri Harms subsp. hylophila (Harms) Soladoye Fabaceae Baphia sp Fabaceae Brachystegia eurycoma Harms Fabaceae Cassia arereh Delile Fabaceae Daniellia oliveri (Rolfe) Hutch. & Dalziel Fabaceae Dialium cf pachyphyllum Harms Fabaceae Dialium sp Fabaceae Entada abyssinica Steud. ex A. Rich Fabaceae Erythrina senegalensis A.DC Fabaceae Erythrophleum ivorense A. Chev Fabaceae Erythrophleum suaveolens (Guill. & Perr.) Brenan Fabaceae Parkia africana R.Br Fabaceae Parkia filicoidea Welw. ex Oliv Fabaceae Penthaclethra macrophylla Benth Fabaceae Pericopsis laxiflora (Benth.) Van Meeuwen Fabaceae Piliostigma thonningii (Schum.) Milne-Redh Fabaceae Pterocarpus erinaceus Poir Fabaceae Pterocarpus osun Craib Fabaceae Pterocarpus soyauxii Taub Gentianaceae Anthocleista djalonensis A. Chev Hymenocardiaceae Hymenocardia acida

181 Hyperaceae Harungana madagascariensis Poir Hypericaceae Psorospermum febrifugum Spach Hypericaceae Psorospermum sp Icacinaceae Rhaphiostylis sp Irvingiaceae Irvingia grandifolia (Engl.) Engl Irvingiaceae Irvingia wombulu Vermoesen Irvingiaceae Klainedoxa gabonensis Pierre Irvingiaceae Klainedoxa sp Lamiaceae Vitex cf simplicifolia Lamiaceae Vitex doniana Sweet Lamiaceae Vitex grandifolia Gürke Lamiaceae Vitex rivularis Gürke Lauraceae Beilschmiedia anacardioides (Engl. & Krause) Robyns & Wilczeck Lauraceae Beilschmiedia gaboonensis (meissn.) Benth. & Hook.f Lecythidaceae Napoleonaea imperialis P.Beauv Loganiaceae Strychnos sp Loganiaceae Strychnos sp Malvaceae Cola caricaefolia K. Schum Malvaceae Cola cordifolia (Cav.) R.Br Malvaceae Cola sp Malvaceae Cola sp Malvaceae Microcos flavescens Juss Malvaceae Sterculia tragacantha Lindl Meliaceae Entandrophragma angolense (Welw.) C. DC Meliaceae Entandrophragma candollei Harms Meliaceae Trichilia rubescens Oliv Meliaceae Trichilia sp Moraceae Antiaris toxicaria Lesch Moraceae Ficus abutilifolia (Miq.) Miq Moraceae Ficus adolfi-friderici Mildbr Moraceae Ficus bubu Warb Moraceae Ficus cf sur Forssk Moraceae Ficus exasperata Vahl Moraceae Ficus glumosa Delile Moraceae Ficus mucuso Welw.ex Ficalho Moraceae Ficus natalensis Hochst Moraceae Ficus sp Moraceae Ficus sp Moraceae Ficus sp

182 Moraceae Ficus sp Moraceae Ficus sp Moraceae Ficus sur Forssk Moraceae Ficus vallis-choudae Delile Moraceae Ficus vogeliana (Miq.) Miq Moraceae Milicia excelsa (Welw.) C.C. berg Moraceae Trilepisium madagascariense DC Myristicaceae Pycnanthus angolensis (Welw.) Exell Myrtaceae Syzygium guineense (Wild.) DC Ochnaceae Lophira lanceolata Tiegh. Ex Keay Ochnaceae Ochna afzelii R.Br. ex Oliv Ochnaceae Ochna sp Olacaceae Olax subscorpioidea Oliv Pandanaceae Pandanus candelabrum P. Beauv Passifloraceae Adenia sp Passifloraceae Barteria fistulosa Mast Phyllanthaceae Antidesma chevalieri Phyllanthaceae Bridelia atroviridis Müll. Arg Phyllanthaceae Bridelia ferruginea Benth Phyllanthaceae Bridelia grandis Pierre ex Hutch Phyllanthaceae Bridelia scleroneura Müll.Arg Phyllanthaceae Bridelia sp Phyllanthaceae Macaranga monandra Müll.Arg Phyllanthaceae Margaritaria discoidea (Baill.) Webster Phyllanthaceae Spondianthus preussii Engl Phyllanthaceae Uapaca guineensis var. guineensis Müll. Arg Phyllanthaceae Uapaca paludosa Aubrév. & Léandri Phyllanthaceae Uapaca togoensis Pax Rhamnaceae Maesopsis eminii Engl Rhizophoraceae Cassipourea zenkeri (Engl.) Alston Rubiaceae Aidia genipiflora (DC.) Dandy Rubiaceae Aidia sp Rubiaceae Craterispermum laurinum (Poir) Benth Rubiaceae Crossopteryx febrifuga (Afzel. ex G.Don) Benth Rubiaceae Cuviera sp Rubiaceae Hallea stipulosa (DC) Leroy Rubiaceae Ixora euosmia K. Schum Rubiaceae Ixora sp Rubiaceae Macrosphyra longistyla (DC.) Hiern Rubiaceae Morelia senegalensis A. Rich.ex DC

183 Rubiaceae Nauclea latifolia SM Rubiaceae Pavetta baconiella Bremek Rubiaceae Pavetta calothyrsa Bremek Rubiaceae Rothmannia sp Salicaceae Homalium africanum (Hook.f.) Benth Sapindaceae Allophyllus bullatus Radlk Sapotaceae Chrysophyllum ubanguiense (De Wild.) Govaerts Sapotaceae Englerophytum stelechanthum Krause Sapotaceae Pouteria alnifolia (Baker) Roberty Sapotaceae Synsepalum stipulatum (Radlk.) Engl Simaroubaceae Quassia sanguinea Cheek & Jongkind Simaroubaceae Quassia sylvestris Cheek & Jongkind Ulmaceae Celtis philippensis Blanco Ulmaceae Trema orientalis (L.) Blume Shirakiopsis elliptica (Hochst. Esser Grand Total GF=Gallery forest, G/WS=Grassland/Woody savanna, MV=Mixed vegetation, PSF=Primary Semideciduous Forest, SF=Secondary Forest 156

184 CHAPTER SIX TEMPORAL LAND-COVER CHANGES IN TROPICAL FORESTS OF THE CONGO BASIN: CASE STUDIES OF A WET FOREST OF RUMPI HILLS FOREST RESERVE AND A DRY FOREST OF KIMBI FUNGOM NATIONAL PARK IN CAMEROON Moses Nsanyi Sainge 1,2, Mabel Nechia Wantim 3, Felix Nchu 4*, and A. Townsend Peterson 5 1 Tropical Plant Exploration Group (TroPEG), P.O. Box 18, Mundemba, Ndian, South West Region, Cameroon. sainge2001@yahoo.com, moses.sainge@gmail.com, tropeg.cam@gmail.com 2 Department of Environmental and Occupational Studies, Faculty of Applied Sciences, Cape Peninsula University of Technology, Cape Town Campus, Keizersgracht, P.O. Box 652, Cape Town 8000, South Africa. sainge2001@yahoo.com, moses.sainge@gmail.com 3 Department of Environmental Science, Faculty of Science, University of Buea,P.O. Box 63 Buea, Cameroon. mabnechia@yahoo.com 4 Department of Horticultural Sciences, Faculty of Applied Science Cape Peninsula University of Technology, P.O Box 1906, Bellville 7535, South Africa. felixnchu@gmail.com 5 Biodiversity Institute, University of Kansas, Lawrence, Kansas USA. town@ku.edu *Author of correspondence Felix Nchu felixnchu@gmail.com Tel: (+27) Preface Chapter six deals with the temporal evaluation of land cover changes in the wet Rumpi Hills Forest Reserve and the dry Kimbi Fungom National Park. ABSTRACT The concept of linking deforestation in Tropical forest to anthropogenic activities has been a major driver to biodiversity loss, biomass and forest carbon depletion in the tropics. This problem is becoming 157

185 alarming in the Congo Basin. Therefore, continuous assessment of spatial and temporal vegetation changes is warranted for understanding the interaction between humans and the forests, as well as the effectiveness of long-term forest management practices. In this chapter, temporal and spatial land-cover changes in two tropical forests: wet forest of Rumpi Hills Forest Reserve and a dry forest of Kimbi Fungom National Park in Cameroon were assessed, compared and discussed. Satellite images from Rumpi Hills (2000 and 2015) and Kimbi Fungom forests (1979 and 2015) were used to compare past and present vegetation (forest cover changes over time). Remote sensing data (Landsat images) were downloaded from the Global Land Cover Facility (GLCF) and United States Geological Survey (USGS) websites and forest cover changes compared over time at both sites in relation to land-use changes that resulted from anthropogenic activities. Result from Rumpi Hills Forest Reserve over time ( ) shows no representative difference on forest types and other parameters, though there were slight changes in the sizes bringing the entire size (area) from 453 km 2 in 1936 to km 2 in This finding suggests that Rumpi Hills Forest Reserve is generally intact and is encouraging. In contrast, in the dry forest of Kimbi Fungom National Park there was evidence of deforestation all over the park with locations in the west around Gayama and Munkep being the most affected. Interestingly, forest regeneration was observed south of the Park. An increase in deforestation over this time ( ) in the Fungom compartment and in the Kimbi compartment from coincided with the presence of access roads, and population increase. In the Kimbi Fungom National Park, forest cover changes from km 2 in 1979 to km 2 in 2015 with a degree of accuracy of 82.41% in 1979 and 71.52% in The lack of representativechange in the vegetation of Rumpi Hills Forest Reserve can be ascribed to its inaccessibility. This has led to its high diversity, and endemism. The numerous linking roads and villages that occur within and around the Kimbi Fungom National Park are the main causes for the high anthropogenic activities within the park. Thus, effective land-use planning and management practices are needed to enhance the park s conservation and biodiversity status. Key words: Land use, Forest cover changes, management practices, Landsat images, Rumpi Hills, Kimbi Fungom, Gayama, and deforestation. 6.1 INTRODUCTION Land-use and land-cover changes have been identified as major threats to biodiversity, biomass, and forest carbon on a national, regional, and global scale, with forest ecosystems being the most affected 158

186 (Lugo, 1988; Meyer et al., 1996; Foley et al., 2005, Aukema et al., 2017). The vegetation of the wet forest is generally of closed forest that ranges from herbaceous to canopy trees of up to 50 m tall, while the vegetation of dry forest range from semi-deciduous forest with sparse undergrowth, savanna and woody savanna forest with open undergrowth mostly with tall grasses; trees which hardly exceed 20 m tall. Although generally, both dry and wet tropical forests are disappearing rapidly (Turner et al., 1994; WRI and IUCN 1998; Bikié et al., 2000; FAO, 2007; De Wasseige et al., 2009; Haggar et al., 2013), comparatively, dry tropical forests have received heavier impact through forest fire, and overgrazing than wet tropical forests (Risser, 1988). Many reasons have been attributed to these diverse impacts in the tropical wet and dry forest. Dry forests are exposed to higher human activities and tend not to benefit from the same protection status as wet forests. Hence, the wet and dry forest of the Rumpi Hills Forest Reserve (RHFR) and the Kimbi Fungom National Park (KFNP), respectively, were selected for this study. Temporal and spatial land-cover changes in the two tropical forests were assessed, compared and discussed. Analysis of land-cover, land-use, and land management practices using long-term remote sensing analysis to identify high risk zones were used. This method is appropriate because images are obtained from the optical multispectral satellite. This instrument has more than one sensor and the impact to biodiversity is that the sensor takes images of the biodiversity of a place at the time of image capture and changes can not be made in a later time. For instance clould covers on images can not be removed after image capture. The study sought to answer the following research question: To what extent has landcover changes occurred in the RHFR and the KFNP over time? 6.2 MATERIALS AND METHODS Study Area The wet forest of the Rumpi Hills Forest Reserve (RHFR) is located at N and E in a tropical lowland to montane forest at km 2 with an elevation range of m above sea level (a.s.l). East of the reserve, is an uplifted cone that forms a plateau at 1778 m a.s.l called Mt Rata. South of this plateau, is a gentle trail, which ascends to the summit. Halfway to the summit is a carbonate-rich vertical cliff and a mosaic of savanna at about m. West of the summit is covered with huge vertical cliffs, inaccessible that descend into a piedmont forest at mid-elevation ( m). This forest continues to the west to a lowland rain forest of 50 m a.s.l. The mean annual temperature is 22 C, but decreases at the summit of Mt Rata. The mean annual rainfall ranges from mm (Thomas, 1996; Nembot and Tchanou 1998), occurring between April to November. Weather records from 159

187 Mundemba, which is < 50 km west of RHFR ( ) gave a mean temperature range of C yr -1, mean radiation of 9.3 yr -1, and mean rainfall of mm yr -1. The topography and relief of the reserve, and the strong southwest monsoon wind that blows into the reserve from the Ndian creeks have great influence on the different vegetation types and species diversity. This reserve was classified into eight vegetation types (Letouzey, 1985), and is characterized by plant species such as Oubanguia alata (Lecythidaceae), Crateranthus talbotii (Lecythidaceae), Strombosia grandifolia (Olacaceae), Leonadoxa africana (Fabaceae), Tabernaemontana ventricosa (Apocynaceae), Dasylepis thomasii (Achariaceae), Strombosia sp. (Olacaceae), Carapa oreophila (Meliaceae), Xylopia africana (Annonaceae), Macaranga sp. (Euphorbiaceae), Trema orientalis (Ulmaceae), Bridelia grandis (Phyllanthaceae) and Pauridiantha viridiflora (Rubiaceae). The dry semi-deciduous and savanna forest of the Kimbi Fungom National Park (KFNP) is located at N, and E at km 2 with an elevation of m a.s.l. The size of Kimbi Fungom National Park is km 2. Weather record from Ako, which is < 100 km east of KFNP has a mean annual temperature of C with mean annual rainfall record for at mm yr -1. The dry season here is severe and runs from November to February, with < 100 mm of rainfall during these months. December to February is the driest period, with a peak in January and eight months of rainfall (March to October) each year with a peak in July to September. Despite the eight months of rainfall, monthly readings hardly exceed 350 mm month -1 with January recording no rainfall. The vegetation of the dry semi-deciduous and savanna forest of the KFNP is characterized by Brachystegia eurycoma (Fabaceae), Maranthes glabra (Chrysobalanaceae), Daniellia oliveri (Fabaceae), Hymenocardia acida, Terminalia glaucescens (Combretaceae), Crossopteryx febrifuga (Rubiaceae), Nauclea latifolia (Rubiaceae), and Lophira lanceolata (Ochnaceae) Image analysis To understand forest cover changes in the two protected areas over time, optical multispectral satellite images for Rumpi Hills Forest Reserve (RHFR) and Kimbi Fungom National Park (KFNP) were downloaded from the Global Land Cover Facility (GLCF) and United States Geological Survey (USGS) websites (www. Glcf.umiacs.und.edu) and analyzed using the ENVI 4.4 software for land cover changes. The images used comprised of the 60 m Landsat MSS of 1979 (acquired on 14 of March 1979) and a 30 m Landsat 8 image of 2015 (acquired on 6 March 2015) for KFNP to assess land cover changes for a 160

188 period of 36 years. Landsat ETM+ (30 m) of 2000 (acquired on 10 December 2000) and a 30 m Landsat 8 image of 2015 (acquired on 10 January 2015) for RHFR were also used to assess land cover changes for a period of 15 years. These images were all process and all limitations such as shadows removed before uploading to the website (Teillet et al., 2001). In both studies, only the limits of the study area were taken into consideration. Thus, our results represent all land cover in the core study area excluding the surroundings. The 2015 image used in the analysis for KFNP did not have data for the southern part that represent Kimbi and Donga Mantung extension which is represented here as rivers when compared with the 2000 image. This therefore led to a misrepresentation of water bodies during the analysis as this section was automatically classified as a water body due to the similarity of its reflectance with that of water bodies. The Normalized Differential Vegetation Index (NDVI) that ranges from -1 to 1 was used for vegetation classification. It s usually calculated using bands of 3 and 4 of the optical multispectral satellite images. The NDVI tells us that any index with a negative value indicates no vegetation while indicates vegetation cover. The usual range for green vegetation is , hence, the higher the index, the thicker the vegetation canopy (Cui et al., 2013). The classification method used was supervised classification using the Minimum Distance Classification type. The land-cover analysis carried out allowed verification of pixel values that differ on the different maps over time, 36 years for KFNP and 15 years for RHFR. In the classification scheme, six classes were identified: Springs/streams/rivers, evergreen forest, Atlantic biafran forest, piedmont forest, submontane forest and cloud cover for RHFR. For KFNP, five classes that comprised of bare soil/burnt area, springs/streams/rivers; savanna type 1, savanna type 2, and evergreen forest were identified. In this study, we define bare soil/burnt area as land that has been cleared for agricultural practices and settlements; savanna type 1 represent grassland savanna and fallow farmland;, savanna type 2 represent woody savanna and fallow farmland, and evergreen forest represent dry semi-deciduous forest. Land-use and land-cover changes in the Kimbi Fungom area are prominent and could be ascribed to human activities and natural factors. Post classification comprised of an accuracy assessment to calculate the amount of error that exists in the information that is being produced for KFNP and RHFR. This was realized using the confusion matrix tool in ENVI 4.4. Areas of the various classes were calculated and summarized in the results. 6.3 RESULTS Land cover classification To understand the level of precision in this study, an accuracy assessment test was carried out for images at both sites to classify land-use and forest cover changes. This was classified as percentage of producer 161

189 accuracy which is the accuracy of the original image and the user accuracy which is the accuracy of the outcome image. To get the overall accuracy, the sum of pixels classified correctly was divided by the total number of pixels. The Kappa coefficient (k) is also an accuracy parameter. The accuracy assessment for the Kimbi Fungom forest gave an overall accuracy of 82.4 % for 1979 image and 71.5 % for 2015 image with a Kappa coefficient of 0.74 and 0.65 respectively (Table 6.1). Although, this is below the threshold of 85% (USGS guideline), the records presented here is still appreciated, and a Kappa value of above 0.8 is strongly agreed. The NDVI for KFNP 1979 ranges from indicating very spare or shrubby vegetation as the value for green vegetation starts from The 2015 NDVI for KFNP reveals more negative NDVI values that relate the high rate of deforestation mostly in the west of figure 6.4. The index ranges from indicating more spare or shrubby vegetation than in the year In Rumpi Hills Forest Reserve, the NDVI for 2000 ranges from , and due to cloud cover and haze found on the 2015 image, most of the NDVI values were negative. However, the vegetation was underneath the clouds thereby preventing actual vegetation values Past and present vegetation: Forest cover changes in the wet Rumpi Hills Forest Reserve In 2015 the RHFR was realized to contain more than 98.6% of its forest cover (446.7 km 2 ) of the total land area of 453 km -1 in This was synthesized into six forest cover classes using the minimum distance classification: rivers/streams/springs, clouds, Atlantic Biafran forest, evergreen forest, piedmont forest, and submontane forest. Forest cover classes/forest cover changes are presented in Table 6.2 (class changes over time) and Figure 1 (a & b; the supervised classified images). No representative changes were observed for the different classes analyzed in this reserve between 2000 and 2015 (Table 6.2). This can be attributed to the inaccessibility of this reserve since its creation in 1937 which has limited anthropogenic activities. Graphical representations using pie charts (Figure 6.2) and bar graphs (Figure 6.3) were also made to show the surface area occupied by the different forest cover classes and the changes that have taken place over time from 2000 to The Atlantic biafran forest dominated the forest cover with km 2 total surface area for 2000 and The total area occupied by springs/streams/rivers (70.65 km 2 ) is slightly lower than the size of submontane forest (79 km 2 ). This actually indicates the quantity of water that RHFR is holding for such a long period since its creation in The impact of cloud covers on images (figure 6.1 A and B) obtained in tropical areas are common (Bastero and Lagmay 2006). This is due to the fact that we are in the tropics and secondly, images were 162

190 obtained from the optical multispectral satellite that uses more than one sensor (Bastero and Lagmay 2006) Past and present vegetation: Forest cover changes in the dry semi-deciduous and savanna Kimbi Fungom forest Broadly, the forest cover of the Kimbi Fungom National Park changed from km 2 in 1979 to km 2 in 2015 (Table 6.3) giving a reduction of 5.7 km 2. The class with the greatest observed and calculated change was bare soil/burnt area with an initial value of 19.6 km 2 in 1979 and a value of km 2 in This gave an increase of km 2. The western part of the park around Gayama and Munkep was greatly deforested (degraded) with slight changes in the center and southeast of the park. This is strongly attributed to increase population, commercial agriculture, cattle grazing, timber logging, and unsustainable land use management practices. Savanna type 1 changes from km 2 in 1979 to 46.4 km 2 in 2015 with a reduction of km 2. Savanna type 2 changes from km 2 in 1979 to 44.5 km 2 in 2015 with a reduction of 88.3 km 2. This indicates a high rate of deforestation in these forest cover classes. This can be attributed to the fact that these areas are favorable for grazing, settlement, agricultural practices and gathering of fuel wood (Table 6.3 and Figure 6.4A and B). Another factor for this reduction could be the fact that in the 2015 image the Kimbi and Donga Mantung extension here represented as rivers were mostly savanna. The evergreen forest changes from km 2 in 1979 to km 2 in 2015 with an increment of 96.6 km 2. This implies a high rate of reforestation or it could be as a consequence of the resolution of the images used. The 1979 image had a resolution of 60 m which makes it less accurate than the 30 m 2015 image. The change in springs/streams/rivers from km 2 in 1979 to km 2 in 2015 with an increment of km 2 is not realistic as the area classified as water bodies in 2015 Kimbi and Donga Matung extension had no data (Figures 6.4B, 6.5 and 6.6). 6.4 DISCUSSION Forest cover changes over a time period of 8 to 10 years and at accurate scale using imageries presented at different time periods remains the most suitable tool for detecting changes in vegetation cover (Mahmood et al., 2010). To date, limited or no study on this subject has been undertaken in the Rumpi Hills and Kimbi Fungom forest. The results obtained in the current study revealed that the Rumpi Hills did not show any representative change in vegetation cover (Figures 6.1 to 6.3). The low net deforestation rate in Rumpi Hills over 79 years ( ) corroborates the study of Rogers (2011), which showed a low deforestation rate of 0.05% in Central Africa. This study further insinuates that the lack of permanent 163

191 access roads due to the difficult terrain has led to high biodiversity, and low development in protected areas in rural zones (Rogers, 2011). Due to the lack of motorable access roads, villages such as Muyange, Kita and Nalende in the south, Lipenja Mukete, Mbange and Meka in the west, Matamani and Iyombo in the north and Madie 2, Bwembe, and Dikome-Balue in the east depend on subsistence agriculture, which has only little effect on the reserve. Human population growth over time has been a major problem for tropical forest even in areas that are set aside as protected (Rogers, 2011). The 79 years from the creation of Rumpi Hills Forest Reserve ( ) assessment period used in this study sufficiently accommodates for a reliable determination of the impact of population growth on vegetation cover in a particular landscape. However, it s possible that reduction in population in settlements such as Motindi, Toko, Boa Yenge, Bweme and Bossunga due to migration to larger towns like Meka, Mundemba, Ekondo-Titi, and Kumba might have reduced the pressure on the forest. Thus, the forest cover change for the wet Rumpi Hills Forest Reserve for fifteen years ( ) reveals no significant difference (Figures 6.1 to 6.2). These images bring us to the conclusion that Rumpi Hills is a principle watershed with springs/streams/rivers occupying a surface area of 70.7 km 2 and supplying five basins: Chad, Benue, Sanaga, Congo and Manyu rivers (Ngwa, 1978). This water body flows from the east down to the south supplying rivers Meme and Mungo that empties into the Douala estuary, and to the north and west, rivers Moko, Moriba, and Mana that empties into the Ndian creeks. To the east, it flows down to river Madie that form rivers Mbo and Munaya and continues to Nigeria. The enclave nature of this reserve may have saved it from anthropogenic activities that could potentially lead to climate change and loss of biodiversity. In the Fungom Native Administration Forest Reserve ( ), and the Kimbi Wildlife Sanctuary ( ) that made-up the Kimbi Fungom National Park in 2015 (Forestry Ordinance, 42 of 1936; Rogers, 2011; Sainge, 2016), deforestation was rather massive, albeit evidence of forest regeneration south of the Fungom compartment. This park is made-up of different vegetation types such as the dry semi-deciduous forest with huge trees that ranges in height from 2-45 m tall, and cm in diameter at breast height, gallery forest, secondary forest, and forest savanna where the understory is mostly composed by tall grasses in the wet season and bare soil with short grasses in the dry season after burning. This vegetation type is exposed to direct sunlight and big trees hardly exceed 15 m in height. Land-use and land-cover changes in the Kimbi Fungom area are pronounced and could be attributed to human activities and natural factors. A change in vegetation from savanna type 1 and 2 (Figure 6.4) to 164

192 bare soil/burnt area is a clear indication of human activities that resulted from agriculture, deforestation (timber extraction), cattle grazing, over flooding of the numerous streams and rivers. There was increase in bare soil in the west, which can be attributed to population growth and the need for livelihood improvement. However, the analysis indicated that forest regeneration occurs in some areas of Kimbi Fungom National Park. An increment of 96.6 km 2 was recorded in the forest which indicates reforestation (Dami et al., 2014). This is partly due to the abandonment of this reserve since its creation in 1936 as the Fungom forest reserve. Generally, the land-cover map for the Kimbi Fungom National Park for thirty-seven years ( ) represence differences in land-use patterns (settlements, agriculture, cattle grazing, fuel wood and charcoal gathering, timber extraction, fishing, gathering of non-timber forest products, medicinal plants, etc.). Although there was reforestation in the south, the rate of deforestation in the entire park was more than reforestation rate (Table 6.3, Figure 6.4). The increase of human population, anthropogenic activities cause by agriculture, deforestation (timber exploitation), fuel wood and charcoal was pronounced, especially in the west around Gayama and Munkep villages in 2015 than These activities will lead to loss of plant and animal species, disrupted forest structure, and diversity; and in a long run may affect the climate of KFNP. 165

193 166

194 Figure 6.1A. 30 m Landsat ETM+ (Enhanced Thematic Mapper) of 2000, B. 30 m Landsat 8 of 2015 for Rumpi Hills Forest Reserve, Cameroon. 167

195 Figure 6.2. Surface area and spatial extentof forest cover types in the Rumpi Hills Forest Reserve, Cameroon from 2000 and 2015 Figure 6.3. Forest cover classes from data collected in 2000 and in 2015 in the Rumpi Hills Forest Reserve, Cameroon. 168

196 169

197 Figure 6.4A. 60 m Landsat MSS (Multispectral Scanner) of 1979, B. 30 m Landsat 8 of 2015 for Kimbi Fungom National Park, Cameroon 170

198 Figure 6.5. Surface area and spatial extentof different cover typesd in the Kimbi Fungom National Park, Cameroon between 1979 (top) and 2015 (bottom) 171

199 Figure 6.6. Forest covers change for thirty-seven years ( ) in the Kimbi Fungom National Park, Cameroon. Table 6.1. Accuracy Assessment for Kimbi Fungom National Park, Cameroon Land cover classes Producer User Accuracy Producer User Accuracy Accuracy (%) (%) Accuracy (%) (%) Forest Springs / Streams / Rivers Bare soil / burnt area Savanna Type I Savanna Type II Over all Accuracy % 71.52% Kappa Co-efficient

200 Table 6.2. Forest covers change (2000 to 2015) in Rumpi Hills Forest Reserve, Cameroon Land Cover Classes km 2 Evergreen Atlantic Piedmont Submontane Rivers Cloud forest Biafran forest forest forest cover Initial state image Final state image Image difference Table 6.3. Forest cover Changes (1979 to 2015) Kimbi Fungom National Park, Cameroon Land Cover Classes km 2 Savanna Savanna Forest Bare soil / Rivers Type 1 Type 2 burnt area Initial state image Final state image Image difference

201 7.1 GENERAL DISCUSSION CHAPTER SEVEN The continental part of the Cameroon Mountains is among the most species-rich and biodiverse region in Africa (Cheek et al., 2000; Myers et al., 2000; Cheek et al., 2004; Barthlott et al., 2005). This region is the only part of Central Africa with an elevational range from sea level to over 4000 m, and holds a high diversity of plants >6000 species out of the almost 9000 species in Cameroon (Cable and Cheek, 1998; Cheek et al., 2000; Cheek et al., 2004; Onana and Cheek, 2011; Onana, 2011). At the continental level, the continental Cameroon Mountains have species that are represented in mountains of west Africa (e.g., Bersama abyssinica; Cheek et al., 2004) and East Africa (e.g., Polyscias fulva, Strombosia scheffleri, Schlefflera abyssinica, Alangium chinense, Maesa lanceolata; Dowsett-Lemaire, 1989; Thomas and Thomas 1996; Cheek et al., 2000; Cheek et al., 2004; Sainge, 2016). This region holds >200 species of plants that are considered as threatened (Cheek et al., 2004), which is the highest in Cameroon and perhaps the highest in west and Central Africa (Cheek et al., 2004; Onana and Cheek, 2011), with >80 species endemic (Cheek et al., 2004; Franke, 2004; Sainge et al., 2005; Sainge et al., 2010; Sainge, 2012; Sainge, 2016). Examples of threatened and endemic plant species occurring in the region are Afrothismia saingei, A. fungiformis, Rhaptopetalium geophylax, Schefflera manni, Syzyzium staudtii, Ixora foliosa, Gambeya korupensis, Deinbollia angustifolia, and Begonia pseudoviola. The region hosts three of the seven genera endemic to Cameroon (Hamilcoa, Medusandra, Platytinospora). Finally, the only endemic family in Cameroon (Medusandraceae) is represented here, with its two species: Medusandra mpomiana and M. richardsiana (Cheek et al., 2004; Onana, 2013). This landscape has been visited and sampled by botanists; nevertheless, their works were mostly centered on surveys of species occurrences, and most of the data are still in non-digitized formats. Data generated from expeditions by these earlier researchers, remains inaccessible to scientists with interest in Africa s biodiversity, particularly those based in Africa. Consequently, attempts to quantitatively analyzed (after Sousa-Baena et al., 2014; Idohou et al., 2015; Kouao et al. 2015) botanical inventory data are stymied by the small numbers of primary occurrence data that are available for the region, and by the large proportion of such data that remain in non-digital formats in institutions in Europe, and America. This study identified botanical inventory gaps, collect detailed biodiversity data and explored the ecological 174

202 relationships among species diversity, forest structure, carbon storage and altitudinal gradient. Furthermore, the study builds on the pioneering works of other authors (Newberry and Gartlan 1996; Thomas et al., 2003; Kenfack et al., 2006; Gonmadje et al., 2011; Djuikouo et al., 2014) who have worked, documented and found different tree densities across the different landscapes in Cameroon. The present study revealed that the RHFR is rich in species endemism. In total, seventeen species are endemic to Cameroon with species such as Deinbollia angustifolia, and Gambeya korupensis endemic in the Korup and Rumpi area, while 43 species were endemic to the lower Guinea forest. This high levels of species diversity and endemism is corroborated by studies of Mittermeiers et al. (1999), Myers et al., (2000), Bergl et al. (2007) and Marchese (2015) who classified the lowland and highland forest of west and central Africa as biodiversity hotspot and concentration sites of endemic species. The present study also covers different forest types at different elevations from lowland at 50 m to montane at 1778 m. Above Ground Biomass (AGB) decreases with elevation: lowland (57.3%), Midelevation (25.3%), Submontane (9.7%) and Montane (7.7%), this represent high AGB, carbon, and carbon dioxide pool in RHFR per hectare compared to other sites in Africa (Sonwa et al., 2011). In lowland forest, a mean of 468 t ha -1 AGB was higher when compared to some lowland tropical forest with 402 t ha -1 (Djuikouo et al., 2010), 404 t ha -1 (Lewis et al., 2009), and the Congo Basin forest 429 t ha -1 (Lewis et al., 2013). Based on the allometric equation of Chave et al. (2005), which is widely employed to estimate AGB for the past decade (Djuikouo et al., 2010; Memiahge et al., 2016), our 25 ha gave 9993 t with a mean of 400 t ha -1 AGB, and 4997 t of carbon with a mean of 200 t ha -1. These values did not change much when compared with a 25 ha plot in lowland forest of Rabi, Gabon, which yielded an AGB of t and a mean of t ha -1 (Memiaghe et al., 2016). This revealed that the Rumpi Hills forest stores high AGB and carbon on the same surface area as the Rabi plot in Gabon. In the 17 ha sampled in KFNP, a mean AGB of t ha -1 ( t ha -1 ), and carbon 97.3 tc ha -1 ( tc ha -1 ) were calculated. These values are far lower compared to the minimum values of AGB (429 t ha -1 ) and carbon stock estimate of 249 tc ha -1 documented previously for Central African forests (Lewis et al., 2013). Although the present study revealed KFNP is poor in mean AGB and carbon, exceptions were observed for some specific plots. For example, plots 1-3 had high mean AGB of t ha -1 ( t ha -1 ), a mean carbon of tc ha -1 ( tc/ha), which is only slightly lower 175

203 than values documented by Lewis et al. (2013). In Congo Brazzaville (Iboukikro and Ngambali Forest), a study conducted by Ekoungoulou et al. (2014), in 6 1-ha plot in a gallery forest revealed a mean of tc ha -1 ( tc ha -1 ), which is higher than that of KFNP. The present study have greatly contributed to understand the biodiversity of Cameroon; a comprehensive checklist of plants of the Rumpi Hills ranging from lowland to montane and in the Kimbi Fungom National Park at different vegetation types were established for the first time. Biodiversity indicators such as forest structure and composition, characterization of the different vegetation types, species diversity, Above Ground Biomass (AGB), carbon and the carbon dioxide sequestered were achieved in the RHFR and KFNP. The above factors will help the government of Cameroon to ratified its international agreements on the Convention of Biological Diversity (CBD), Climate change, REDD, and REDD+. In Cameroon, studies of montane ecosystems are somewhat scant and even if they exist, the findings of these studies are not easily available to the scientific community (Thomas and Cheek 1992; Thomas and Achoundoung 1994; Achoundoung, 1995; Cheek et al., 2000; Cheek et al., 2004; Forboseh et al., 2011). Thus, this study has set a base for long-term monitoring of the montane ecosystem in Cameroon. Results of this study will help conservation and biodiversity stakeholders in the region to set-up better policies for the proper management of RHFR and the KFNP. This can be achieved by, drawing of management plans, lay down of proper land-use management plan and policies, and set-up different models for ecological monitoring, sustainable agriculture, conservation, education and training. This study has created awareness on biodiversity and forest inventory in the communities, and during this period over fifteen Cameroonians were trained (students, field assistants, and community members) on forest mensuration, transect cutting, plot establishment, data and specimens collections, plant specimens description, drying, sorting and identification of plot specimens using floras, monographs and herbarium specimens sheets. The creation of a biodiversity database is underway, and this is in addition to scheduled monitoring cycles every five years. The current study is a booster for other scientists and researchers to develop interest in studying other aspects such as soils, climate, forest regeneration and taxa in the RHFR and the KFNP, such as understanding the ecological interaction of plants and other taxa, the effects of climate change, land-use patterns, and other edaphic factors. While this project was a success, there were setbacks that are worth reporting: difficult terrain and topography in some areas hindered the 176

204 establishment of many plots at high elevation as planned; nevertheless, twenty-five quadrats of 20 x 20 m were established systematically per plot. This study was based on forest inventory only, and lacked data on climate and edaphic factors due to funding constrains. 7.2 RECOMMENDATIONS Conserve and protect the natural sites and habitat of the RHFR as much as possible. Update the status of RHFR to a sanctuary or a National Park because of its high biodiversity potentials and carbon storage capacity. Complete restoration of habitat lost at the KFNP that will enhance climate change. Special focus on the Kimbi Fungom Forest to enhance its level of biodiversity, since it s the only national park in the region. Implementation of forest, wildlife, andland-use policies; and legislations should be re-enforced at both the RHFR and the KFNP. All stakeholders involved in the management of these protected areas should develop a collaborative approach for effective management. Scientists and students should be given the opportunity to continue research at the RHFR and the KFNP as multi data layers of different taxa and aspects are necessary. Long-term study on the climate, soils, and other future research needs couple with appropriate equipment to study these aspects should be set-up in the RHFR and the KFNP. 177

205 REFERENCES Achoundong, G Les formations submontagnardes du Nta-Ali au Cameroon. Bois et Forêts des Tropiques. 243: Ali, A., & Yan, E-R Relationship between biodiversity and carbon stocks in forest ecosystems: a systematic literature review. Tropical Ecology. 58: Angiosperm Phylogeny Group III (APG III) An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants. Botanical Journal of the Linnean Society. 161: Asase, A., Asitoakor, B.K., & Ekpe, P.K Linkages between tree diversity and carbon stocks in unlogged and logged West African tropical forests. International Journal of Biodiversity science, Ecosystem services and Management. 8: Asonganyi, J. N A Report on the Vegetation Survey of Ijim Mountain Forest carried out by Mr. Simon Tame and J. N. Asonganyi. Herbier National, Yaounde. Unpublished report. Aukema, J.E., Pricope, N.G., Husak, G.J., & Lopez-Carr, D Biodiversity areas under threat: overlap of climate change and population pressures on the world s biodiversity priorities. PLoS ONE. 12: Ayonghe, S.N., Mafany, G.T., Ntasin, E. & Samalang, P Seismically activated swarm of landslides, tension cracks, and a rockfall after heavy rainfall in Bafaka, Cameroon. Natural Hazards. 19: Baccini, A., Goetz, S.J., Walker, W.S., Laporte, N.T., Sun, M., Sulla-Menashe, D., Hackler, J., Beck, P.S.A., Dubayah, R., Friedl, M.A., Samanta, S. & Houghton, R.A Estimated carbon dioxide emissions from tropical deforestation improved by carbon-density maps. Nature climate change. 2: Baker, T.R., Philips, O.L., Malhi, Y., Almeida, S., Arroyo, L., Di Fiore, A., Erwin, T., Higuchi, N., Killeen, T.J., Laurance, S.G., Laurance, W.F., Lewis, S.L., Monteagudo, A., Neill, D.A., Vargas, P.N., Pitman, N.C.A., Silva, J.N.M., & Martinez, R.V Variation in wood density determines spatial patterns in Amazonian forest biomass. Global Change Biology. 14: Banda, K.R., Delgado-Salinas, A., Dexter, K.G., Linares-Palomino, R., Oliveira-Filho, A., Prado, D., et al Plant diversity patterns in neotropical dry forests and their conservation implications. Science. 353: Barthlott, W., Mutke, J., Rafiqpoor, D., Kier, G., & Kreft, H Global centers of vascular plant diversity. Nova Acta Leopoldina. 92:

206 Barthlott, W., Hostert, A., Kier, G., Kuper, W., Kreft, H., Mutke, J., Rafiqpoor, M.D.,& Sommer, H Geographic patterns of vascular plant diversity at continental to global scales. Erdkunde. 61: Bastero, C.F. & Lagmay, A.M.F.A Estimating silica content of lava deposits in the humid tropics using remotely sensed imagery. Journal of volcanology and Geothermal Research. 151: Bednarek-Ochyra, H., Váňa, J., Ochyra, J.V.R. & Smith, R.I.L The liverwort flora of Antarctica. Polish Academy of Sciences, Institute of Botany, Cracow. P Bergl, R.A., Oates, J.F., & Fotso, R Distribution and protected area coverage of endemic taxa in West Africa's Biafran forests and highlands. Biological Conservation. 134: Bikié, H., Collomb, J-G., Djomo, L., Minnemeyer, S., Ngoufo, R., & Nguiffo S An overview of logging in Cameroon. Global forest watch, World Resourcces Institutte. p.69. BirdLife International CM024: Mount Rata and Rumpi Hills Forest Reserve. on 01/05/2016. Blackie, R., Baldauf, C., Gautier, D., Gumbo, D., Kassa, H., Parthasarathy, N., Paumgarten, F., Sola, P., Pulla, S., Waeber, P., & Sunderland, T Tropical dry forests: The state of global knowledge and recommendations for future research. Discussion Paper. Center for International Forestry Research (CIFOR), Bogor, Indonesia. p. 30. Brady, M., & De Wasseige, C., Monitoring forest carbon stocks and fluxes in the Congo Basin. Conference Report. Central Africa Forest Commission (COMIFAC). Brazzaville, Republlic of Congo. Bradshaw, C.J.A., Sodhi, N.S., Peh, K., & Brook, B.W Global evidence that deforestation amplifies flood risk and severity in the developing world. Global Change Biology. 13: Brandon, K Ecosystem services from tropical forests: review of current science. Center for Global Development (CGD) Climate and forest paper series #7, working paper 380. p. 86. Branthomme, A National forest inventory field manual template. Forest Resources Assessment Programme, Rome. P. 83. Brown, J.H On the relationship between abundance and distribution of species. American Naturalist. 130: Brown, S Estimating Biomass and Biomass change of Tropical Forest: a Primer. Food and Agroculture Organization (FAO), Forestry Paper 134 Rome, Italy. p. 55. Bryan, J.E., & Shearman, P.L. (Eds.) The state of the forest of Papua New Guinea 2014: measuring change over the period University of Papua New Guinea, Port Moresby. P

207 Bunker, D.E., DeClerck, F.A., Bradford, J.C., Colwell, R., Garden, P., Perfecto, I., Phillips, O., Sankaran, M. & Naeem, S Biodiversity loss and above-ground carbon storage in a tropical forest. Science 301: Burgess, N.D., Balmford, A., Cordeiro, N.J., Fjelda, J., Küper, W., Rahhek, C., Sanderson, E.W., Scharleman, J.P.W., Sommer, J.h., & Williams, P.H Correlations among species distributions, human density and human infrastructure across the high biodiversity tropical mountains of Africa. Biological Conservation. 134: Butler, R.A., & Laurance, W.F New strategies for conserving tropical forests. Trends Ecol. Evol. 23: Cable. S., & Cheek, M The plants of Mount Cameroon. A conservation checklist. Royal Botanic Gardens, Kew. p Campbell, P., Rivera, P., Thomas, D., Bourobou-Bourobou, H., Nzabi, T., Alonso, A., & Dallmeier, F Floristic structure, composition, and diversity in an equatorial forest in Gabon. Bulletin of the Biological Society of Washington. 12: Canadell, J.G & Raupach, M.R Managing forests for climate change mitigation. Science. 320: Carsan, S., Orwa, C., Harwood, C., Kindt, R., Stroebel, A., Neufeldt, H., & Jamnadass, R African Wood density database. World Agroforestry Centre, Nairobi. Kenya. ( Carson, W.P., & Schnitzer S.A. (Eds.) Tropical forest community ecology. Oxford: Wiley- Blackwell. P Central Africa Forestry Commission, (COMIFAC) Annuel Rapport: une dimension régionale pour la conservation et la gestion durable des écosystèmes forestiers d Afrique Centrale. Chandrashekara, U.M., & Ramakrishna, P.S Vegetation and gap dynamics of tropical wet evergreen forest in Western Ghats of Kerala state. Journal of Ecology. 10: Chape, S. et al The World's protected areas: Status, Values and Prospects in the 21 st Century. UNEP-World Conservation Monitoring Centre. Cambridge, UK. Chapman, H.M., Bekker, R., Barnard, J., & Shi, G The botany of Tchabal Mbabo, Cameroon. A contribution towards the Nigeria and Cameroon Transboundary initiative. University of Canterbury/Nigeria montane forest project, Christchurch, New Zealand, p. 1. Chave, J., Andalo, C., Brown, S., Cairns, M.A., Chambers, J.Q., Eamus, D., Fölster, H., Fromard, F., Higuchi, N., Kira, T., Lescure, J-P., Nelson, B.W., Ogawa, H., Puig, H., Riera, B., & Yamakura, T. 180

208 2005. Tree allometry and improved estimation of carbon stocks and balance in tropical forests. Oecologia. 145: Cheek, M., Onana, J.M., & Pollard, B.J The Plants of Mount Oku and the Ijim Ridge, Cameroon. Royal Botanic Garden, Kew. p Cheek, M., Pollard, B.J., Darbyshire, L., Onana, J.M., & Wild, C The plants of Kupe, Mwanenguba, and the Bakossi Mountains, Cameroon, a conservation checklist. Royal Botanic Garden, Kew. p Cheek, M., Harvey, Y., & Onana, J.M The Plants of Dom, Bamenda Highlands, Cameroon: A Conservation Checklist. Royal Botanic Gardens, Kew. p Chuyong, G.B., Newberry, D.M., & Songwe, N.C Litter nutrients and retranslocation in a Central African rainforest dominated by ectomycorrhizal trees. New Phytologist. 148: Chuyong, G.B., Newbery, D.M., & Songwe, N.C Rainfall input, through fall and stem flow of nutrients in a Central African rain forest dominated by ectomycorrhizal trees. Biogeochemistry. 67: Chuyong, B.G., Condit, R., Kenfack, D., Losos, C.E., Sainge, N.M., Songwe, N.C., & Thomas, D.W Korup Forest Dynamics Plot, Cameroon, in Tropical Forest Diversity and Dynamism, in Losos C.E and Leigh E.G. Jr (eds.). University of Chicago Press, Chicago. p Chuyong, G., Kenfack, D., Harms, K., Thomas, D., Condit, R., & Comita, L Habitat specificity and diversity of tree species in an African wet tropical forest. Plant Ecology. 212: Clement, F.E Nature and structure of the climax. Journal of Ecology. 24: Cole, L.E.S., Bhagwat, S.A., & Willis, K.J Recovery and resilience of tropical forests after disturbance. Nature communications. 5: 1-7. Colwell, R. K. & Coddington, J. A Estimating terrestrial biodiversity through extrapolation. Philosophical Transactions of the Royal Society of London. 335: Colwell, R.K EstimateS: Statistical estimation of species richness and shared species from samples. Version Condit, R Tropical Forest Census Plots: Methods and Results from Barro Colorado Island (BCI), Panama and a Comparison with Other Plots. Springer-Verlag, Berlin p Corlett, R.T. & Primack, R.B Tropical rain forests: An ecological and biogeographical comparison, 2nd edition. Wiley, Oxford, UK. 181

209 Cronin, D.T., Libalah, M.B., Bergl, R.A., & Hearn, G.W Biodiversity and Conservation of tropical montane ecosystems in the Gulf of Guinea, West Africa. Arctic, Antarctic, and Alpine Research. 46: Cui, X., Gibbes, C., Southworth & Waylen P Using remote sensing to quantify vegetation change and ecological resilience in a semi-arid system. Land. 2: Daily, G.C Nature's services: societal dependence on natural ecosystems. Island Press. Washington, D.C. Dallmeier, F Long-Term Monitoring of Biological Diversity of Tropical Forest Areas: Methods for the Establishment and Inventory of Permanent Plots. Paris: United Nation Educational Scientific and Cultural Organization (UNESCO). p. 72. Dami, A., Odihi, J.O., & Ayuba, H.K Assessment of land use and land cover change in Kwale, Ndokwa-East local government area, Delta State, Nigeria. Global journal of human-social science (B) Geography, Geo-science, environmental disaster management. 14: 1-9. Dauby, G., Hardy, O.J., Leal, M., Breteler, F., & Stévart, T Drivers of tree diversity in tropical rain forests: new insights from a comparison between littoral and hilly landscapes of Central Africa. Journal of Biogeography. 41: Davies, M.B Quaternary history of deciduous forests of eastern North America and Europe. Annals of Missouri Botanical Garden. 70: Day, M., Baldauf, C., Rutishauser, E., & Sunderland, T.C.H Relationship between tree diversity and above ground biomass in Central Africa rainforests: implications for REDD. Foundation for Environmental Conservation. 41: Demenou, A.P La place de bois de feu dans ub systeme agroforestier. Center for International Forestry Research & International Institute of Tropical Agriculture (CIFOR & IITA) Yaoundé, Cameroon. p. 26. Desalegn, W., & Beierkuhnlein, C Plant species and growth forms richness along altitudinal gradients in the southwest Ethiopian highlands. Journal of Vegetation Science 21: De Wasseige, C., Devers, D., De Marcken, P., Eba'a Atyi, R., Nasi, R., & Mayaux, P Les Forêts du Bassin du Congo - État des Forêts Office des publications de l Union européenne, Luxembourg. Djomo, A.N Ecological management of tropical forests: Implications for climate change and carbon fluxes. PhD thesis, University of Goettingin, Germany. p

210 Djomo, A.N., Adamou, I., Saborowski, J., & Gravenhorst, G Allometric equations for biomass estimations in Cameroon and pan moist tropical equations including biomass data from Africa. Forest Ecology and Management. 260: Djomo, A.N Allometric equations: implications for REDD, Carbon stocks and the missing sink. Workshop paper, Yaoundé, Cameroon. Djomo, N.A., Picard, N., Fayolle, A., Henry, M., Ngomanda, A., Ploton, P., McLellan, J., Saborowski, J., Adamou, I., & Lejeune, P Tree allometry for estimation of carbon stocks in African tropical forests. Forestry: an International Journal of forest research. Forestry 2016:1 10. Djuikouo, K.M.N., Doucet, J-L., Nguembou, K.C., Lewis, L.S., & Sonké, B Diversity and aboveground biomass in three tropical forest types in the Dja Biosphere Reserve, Cameroon. African Journal of Ecology. 48: Djuikouo, K.M.N., Peh, H.K.S., Nguembou, K.C., Doucet, J-L., Lewis, L.S., & Sonké, B Stand structure and species co-occurrence in mixed and monodominant Central African tropical forest. Journal of Tropical Ecology. 30: Dkamela, G.P., Mbambu, F.K., Kemen, A., Minnemeyer, S., & Stolle F Voices from the Congo Basin: Incorporating the Perspectives of Local Stakeholders for Improved REDD Design. World Resources Institute. Dowsett-Lemaire, F The flora and phytogeography of the evergreen forests of Malawi. 1: Afromontane and mid-altitude forest. Journal of the National Botanical Garden, Belgium. 59: Durrell, G. 1954, The Bafut Beagles. Pengiun Books. England. Great Britain. p Eba a Atyi, R., Poufoun, J.N., Awono, J-P.M., Manjeli, A.N., & Kankeu, R.S Economic and social importance of fuelwood in Cameroon. International forestry review. 18: Ellatifi, M Morocco. In: Merlo, M. & Croitorou, L. (eds.). Valuing Mediterranean forests. Towards total economic value. CABI Publishing, Wallingford, UK and Cambridge MA. p Ekoungoulou, R., Liu, X.D., Loumeto, J.J., Ifo, S.A., Bocko, Y.E., Koula, F.E., & Niu, S.K Tree Allometry in Tropical Forest of Congo for Carbon Stocks Estimation in Above-Ground Biomass. Open Journal of Forestry. 4: Erdaú, A The Liverwort Flora of Antarctica (2000). Turkey Journal of Botany. 25:167. European Union Institute, Proforest. p. 8. Ewango, C.E.N The liana assemblage of a Congolian rainforest: Diversity,structure and dynamics. Thesis, Wageningen University, Wageningen, Netherland. P

211 Ewango, C.E.N., Bongers, F., Makana, J-R., Poorter, L., & Sosef, M.S.M Structure and Composition of the liana assemblage of a mixed rain forest in the Congo Basin. Plant Ecology and Evolution. 148: Fadrique, B., & Homeier, J Elevation and topography influence community structure, biomass and host tree interactions of lianas in tropical montane forests of southern Ecuador. Journal of Vegetation Science. 27: Fischer, A., Blaschke, M., & Bassler, C Altitudinal gradients in biodiversity research: the state of the art and future perspectives under climate change aspects. Waldokologie, Landschaftsforschung und Naturschutz 11: Foahom, B Biodiversity planning support programme: Integrating biodiversity into forestry sector. Case study: Cameroon. Paper prepared for an international workshop on Integration of biodiversity in national forestry planning programme Bogor, Indonesia.p.23. Foley, J.A., Defries, R., Asner, G.P., Barford, C., Bonan, G., Carpenter, S.R., Chapin, F.S., Coe, M.T., Daily, G.C., Gibbs, H.K., Helkowski, J.H., Holloway, T., Howard, E.A., Kucharik, C.J., Monfreda, C., Patz, J.A., Prentice, C., Ramankutty, N., & Snyder, P.K Global consequences of land use. Science. 309: Food and Agricultural Organization (FAO), State of the world s forests. Food and Agriculture Organization of the United Nations, Rome, Report. p. 24. Food and Agricultural Organization (FAO), Forest resource assessment. Main Report, Forestry Paper 163, Rome 2010, p Food and Agricultural Organization (FAO), State of the world s forests Food and Agriculture Organization of the United Nations, Rome. Report. Fonge, B.A., Tchetcha, D.J., & Nkembi, L Diversity, distribution, and abundance of plants in Lewoh-Lebang in the Lebialem Highlands of southwestern Cameroon. International Journal of Biodiversity. 2013: 1-13 Forboseh, P.F., Sunderland, T.C.H., Comiskey, J.A., & Balinga, M Tree population dynamics of three altitudinal vegetation communities on Mount Cameroon ( ). Journal of Mountain Science. 8: Forestry Ordinance Establishment of the Fungom Native Administration Forest Reserve. Archive Buea. p. 4. Forestry Ordinance, Establishment of the Rumpi Hills Native Authority Forest Reserve. Forestry Archive Buea. p

212 Franke, T Afrothismia saingei (Burmanniaceae), a new myco-heterotrophic plant from Cameroon. Systematics and Geography of Plants. 74: Frodin, D. G Guide to the Standard Floras of the World, 2nd ed. Cambridge University Press. Gentry, A.H Changes in plant community diversity and floristic composition on environmental and geographical gradients. Annals of the Missouri Botanical Garden. 75: 1-34 Gèze, B Géographie physique et Géologie du Cameroon occidental. Muséum National d Histoire Naturelle Paris, T. XVII, 320. Gleason, H.A The individualistic concept of the plant association. American Midland Naturalist. 21: Global Environmental Facility (GEF) Country Portfolio evaluation: Cameroon ( ). Global Observation of Forest Cover and Land Dynamics (GOFC-GOLD), 2008.Reducing greenhouse gas emissions from deforestation and degradation in developing countries: a sourcebook of methods and procedures for monitoring, measuring and reporting, GOFC-GOLD Report version COP13-2. Gonmadje, C.F., Doumenge, C., McKey, D., & Sonké, B Tree diversity and conservation value of Ngovayang s lowland forest, Cameroon. Biodiversity and Conservation. 20: Gopalakrishna, P.S., Kaonga, M.L., Somashekar, R.K., Suresh H.S., & Suresh, R Tree diversity in the tropical dry forest of Bannerghatta National Park in Eastern Ghats, southern India. European Journal of Ecology. 1: Grime, J.P Plant strategies and vegetation processes. New York: John Wiley & Sons. Haggar, J., Medina, B., Aguilar, R.M., & Munoz, C Land use changes on coffee farms in southern Guatemala and its environmental consequences. Environmental Management. 51: Hall, J.B., & Medler, J.A The botanical exploration of the Obudu Plateau area. Nigeria Field 40: Hamilton, A.C A quantitative analysis of altitudinal zonation in Uganda forests. Vegetation. 30: Hammer, Ø., Harper, D.A.T., & Ryan, P.D PAST: Paleontological Statistics Software package for education and data analysis. Palaeontologia Electronica. 4: 1-9. Hansen, M.C., Potapov, P.V., Moore, R., Hancher, M., Turubanova, S. A., Tyukavina, A., Thau, D., Stehman, S. V., Goetz, S. J., Loveland, T. R., Kommareddy, A., Egorov, A., Chini, L.,Justice, C.O., & Townshend, J.R.G High-resolution global maps of 21st-Century forest covers change. Science 342: and Supplementary materials for high-resolution global maps of 21st-Century forest cover change. 185

213 Harris, N.L., Brown, S., Hagen, S.C., et al Baseline map of carbon emissions from deforestation in tropical regions. Science. 336: Harvey, Y., Pollard, B.J., Darbyshire, L., Onana, J.M., & Cheek, M The plants of Bali Ngemba forest reserve. A conservation checklist. Royal Botanic Gardens, Kew. p Harvey, Y., Tchiengué, B., & Cheek, M The plants of Lebialem Highlands, Cameroon: A conservation checklist. Royal Botanic Gardens, Kew. p Hemp A Ecology of the pteridophytes on the southern slopes of Mt Kilimanjaro-1. Altitudinal distribution. Plant Ecology 159: Henrik, B.H., Moen, J., Virtanen, R., Grytnes J.A., Oksanen, L., & Angerbjörn, A Effects of altitude and topography on species richness of vascular plants, bryophytes and lichens in alpine communities. Journal of Vegetation Science. 17: Hepper, F. N The vegetation and floral of the Vogel Peak massif of the northern Nigeria. Bulletin of the Institut Fondamental d'afrique Noire.. Ser. A. 27: Hepper, F.N Outline to the vegetation and flora of the Mambila Plateau, northern Nigeria. Bulletin of the Institut Fondamental d'afrique Noire, ser. A. 28: Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P. G., & Jarvis, A Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology 25: Hill, M.O TWINSPAN-a Fortran program for arranging multivariate data in an ordered two-way table of classification of individuals and attributes. Cornell University, Ithaca, New York. p. 90. Hill, M.O., & Gauch, H.G Detrended correspondence analysis, an improved ordination technique. Vegetatio. 42: Holmgren, P.K., Holmgren, N.H. & Barnett, L.C Index Herbariorum. 8 th ed. New York Botanical Garden, New York. Hussin, Y.A & Bijker, W Inventory of remote sensing applications in forestry for sustainable management. International archives of photogrammetry and remote sensing. vol. XXXIII, Part B7. Amsterdam. Hutchinson, J., & Dalziel, J.M Flora of west tropical Africa. Volume I, part 1. Whitefriars Press. London, p. 295 Hutchinson J, Dalziel JM Flora of west tropical Africa. Volume I, part 2. Whitefriars Press. London, p

214 Hutchinson J, Dalziel JM Flora of west tropical Africa. Volume II. Whitefriars Press. London, p. 544 Idohou, R., Ariño, A., Assogbadjo, A., Kakai, R.G. & Sinsin, B Diversity of wild palms (Arecaceae) in the Republic of Benin: Finding the gaps in the national inventory combining field and digital accessible knowledge. Biodiversity Informatics. 10: Imani, G., Boyemba, F., Lewis, S., Nabahungu, N.L., Calders, K., Zapfack, L., et al Heightdiameter allometry and above ground biomass in tropical montane forests: Insights from the Albertine Rift in Africa. PLoS ONE. 12: Intergovernmental Panel on Climate Change (IPCC), Chapter 9: Forestry. Climate change 2007: synthesis report. Contribution of Working Groups I, II and III to the fourth assessment report of the Intergovernmental Panel on Climate Change. Geneva, Switzerland: Intergovernmental Panel on Climate Change. Janzen, D.H Tropical dry forests: The most endangered major ecosystem. In E.O. Wilson (Ed.). Biodiversity, National Academic Press, Washington, DC. p Judd, S.W., Campbell, S.C., Kellogg, A.E., & Stevens, F.P Plant systematics: A phylogenetic approach. Sinauer Associates, Sunderland, Massachusetts. p Kapos, V Developing a map of the world s mountain forests. In: Price, M.F. & Butt, N. (eds) Forests in Sustainable Mountain Development (IUFRO Research Series 5). CABI Publishing, Wallingford Oxon. p. 4-9 Keay, R.W.J Trees of Nigeria. Clarendon Press. Oxford, p Kenfack, D Preliminary botanical survey of Mt Nlonako, Makombe, Ebo, and Lake Ossa. Report to the Cross-Sanaga Bioko Coastal Forest Project. Worldwide Fund for Nature-Coastal Forest Program, Cameroon. p. 84. Kenfack, D., Thomas, D.W., Chuyong, G.B., & Condit, R Rarity and abundance in a diverse African Forest. Biodiversity Conservation. 16: Kenfack, D., Zapfack, L., & Sainge, M.N Botanical inventory, biomass estimation and vegetation classification of the Ilor forest block, Southwest Cameroon. A forest inventory commissioned to Proforest by Tropical Plant Exploration Group (TroPEG) Cameroon. Körner, C., Ohsawa, M., Spehn, E., Berge, E., Bugmann, H., Groombridge, B., Hamilton, L., Hofer, T., Ives, J., Jodha, N., Messerli, B., Pratt, J., Price, M., Reasoner, M., Rodgers, A., Thonell, J., Yoshino, M., Baron, J., Barry, R., Blais, J., Bradley, R., Hofstede, R., Kapos, V., Leavitt, P., Monson, R., Nagy, L., Schindler, D.,Vinebrooke, D., Watanabe, T Mountain systems. In: 187

215 Hassan R, Scholes R, Ash N (Eds.) Ecosystems and human well-being: Current state and trends. Island Press, Washington D.C. 1: Kouao, J.K., Kouassi, A.F., Adouyao, C.Y., Bakayoko, A., Josephipou, I., Bogaert, J The present state of botanical knowledge in Côte d Ivoire. Biodiversity Informatics. 10: Kupsch, D., Bobo, K.S., & Waltert, M Biodiversity, carbon stock and market value assessment for the Sustainable Oil (SGSOC) project area, Southwest Region, Cameroon. Report submitted to World Wide Fund for Nature (WWF), Germany. p. 41. Larison, B., Smith, T.B., McNiven, D., Fotso, R., Bruford, M., Holbrook, K., & Lamperti, A Faunal Survey of Selected Montane and Lowland Areas in Cameroon. World wildlife Fund, Cameroon. Laurance, W. F Reflections on the tropical deforestation crisis. Biol. Conservation. 91: Laurance, W. F. & A. Balmford A global map for road building. Nature. 495: Laurance, W.F Emerging threats to tropical forests. Annals of the Missouri Botanical Garden, 100: Le Bras, G.L., Pignal, M., Jeanson, M.L., Muller, S., Aupic, C., Carré, B., Flament, G., Gaudeul, M., Gonçalves, C., Invernón, V.R., Jabbour, F., Lerat, E., Lowry, P.P., Offroy, B., Pimparé, E.P., Poncy, O., Rouhan, G., & Haevermans, T The French Muséum National d Histoire Naturelle vascular plant herbarium collection dataset. Scientific Data. 4: Lee, C-B., & Chun, J-H Environmental drivers of patterns of plant diversity along a wide environmental gradient in korean temperate forests. Forests. 7: 19; doi: /f Letouzey, R Etude phytogéographique du Cameroon. Encyclopedie Biologique. Paris; Lechevalier, 49. p Letouzey, R Carte phytogéographique du Cameroun. vol Institut de la Carte Internationale de la Végétation, Toulouse-France. p Letouzey, R Manual of forest botany, tropical Africa, volume 1. General Botany. Translated by Huggett, R. Technical centre for tropical forestry, Nogent-sur- Marne, France and Ministry of higher education and scientific research, Yaoundé, Cameroon. p Lewis, L.S., Lopez-Gonzalez, G., Sonké, B., Affum-Baffoe, K., Baker, T.R., Ojo, L.O., Phillips, O.L., Reitsma, J., White, L., Comiskey, J., Ewango, C., Feldpausch, T.R., Hamilton, A.C., Gloor, M., Hart, T., Hladik, A., Djuikouo, K.M.N., Jon, L., Lovett, J., Makana, J-R., Malhi, Y., Mbago, F.M., Ndangalasi, H.J., Peacock, J., Peh, K.S-H., Sheil, D., Sunderland, T., Swaine, M.D., Taplin, J., 188

216 Taylor, D., Sean, C.T., Votere, R., & Hannsjörg, W Increasing carbon storage in intact African tropical forests. Nature. 457: Lewis, S.L., Sonké, B., Sunderland, T., Begne, S.K., Lopez-Gonzalez, G., Van der Heijden, G.M.F., Philips, O.L., Affum-Baffoe, K., Baker, T.R., Banin, L., Bastin, J-F., Beeckman, H., Boeckx, P., Bogaert, J., De Cannière, C., Chezeaux, E., Clark, C.J., Collins, M., Djagbletey, G., Djuikouo, M.N.K., Droissart, V., Doucet, J-L., Ewango, C.E.N., Fauset, S., Feldpausch, T.R., Foli, E.G., Gillet, J-F., Hamilton, A.C., Harris, D.J., Hart, T.B., de Haulleville, T., Hladik, A., Hufkens, K., Huygens, D., Jeanmart, P., Jeffery, K.J., Kearsley, E., Leal, M.E., Lloyd, J., Lovett, J.C., Makana, J-R., Malhi, Y., Marshall, A.R., Ojo, L., Peh, K.S-H., Pickavance, G., Poulsen, J.R., Reitsma, J.M., Sheil, D., Simo, M., Steppe, K., Taedoumg, H.E., Talbot, J., Taplin, J.R.D., Taylor, D., Thomas, S.C., Toirambe, B., Verbeeck, H., Vleminckx, J., White, L.J.T., Willcock, S., Woell, H., & Zemagho, L Above ground biomass and structure of 260 African tropical forests. Philosophical Transaction of the Royal Society Biology. 368: Locatelli, B Ecosystem services and climate change. In: Routledge handbook of ecosystem services. M. Potschin, R. Haines Young, R. Fish and R.K. Turner (eds). Routledge, London and New York. p Losi, C.J., Siccama, T.G., Condit, R., & Morales, J.E Analysis of Alternatives methods for estimating carbon stock in young tropical plantations. Forest Ecology and Management. 184: Lovett, J.C Elevational and latitudinal changes in tree associations and Diversity in the eastern arc mountains of Tanzania. Journal of Tropical Ecology. 12: Lovett, J.C., Marshall, A.R., & Carr J Changes in tropical forest vegetation along an altitudinal gradient in the Udzungwa Mountains National Park. African Journal of Ecology 44: Lugo, A.E Estimating reductions in the diversity of tropical forest species. In Wilson, E.O. & Frances, M.P. (eds). Biodiversity. National Academy press. Washington, D.C. USA. p Lutz, J.A., Larson, A.J., Swanson, M.E., & Freund, J.A Ecological Importance of large-diameter trees in a temperate mixed-conifer forest. PLoS ONE. 7: MacArthur, R.H Population ecology of some warblers of northeastern coniferous forest. Ecology. 39: Mahmood, R., Pielke, R.A., Hubbard, K.G., Yogi, D.N., Bonan, G., Lawrence, P., Mcnider, R., Mcalpine, C., Etter, A., Gameda, S., Qian, B., Carleton, A., Beltran-Przekurat, A., Chase, T., Quintanar, A.I., Adegoke, J.O., Arambu, S.V., Conner, G., Asefi, S., Sertel, E., Legates, D.R., Wu, Y., Hale, R., 189

217 Frauenfeld, O.W., Watts, A., Shepherd, M.H., Mitra, C., Anantharaj, V.G., Fall, S., Lund, R., Treviño, A., Blanken, P., Du,J., Chang, H-I., Leeper, R., Nair, U.S., Dobler, S., Deo, R., & Syktus, J Impact of land use and land cover change on change and future research priorities. American meteorological society. p Maisels, F.M., & Forboseh, P Vegetation survey. Report to Kilum-Ijim Forest Project. Oku, Cameroon. p. 51. Maisels, F.M., Cheek, M., & Wild, C Rare plants on Mt Oku summit, Cameroon. Oryx. 34: Makana, J-R Canopy (Aerial) carbon stocks measurement in Congo Basin Forest/Estimation des stocks de carbone aérien dans les forêts du Bassin du Congo: Cas de parcelles permanents de l Ituri et de la Salonga en RDC. In Monitoring forest carbon stocks and fluxes in the Congo Basin. Conference Report. Central Africa Forest Commission (COMIFAC). Brazzaville, Republlic of Congo. p Marchese, C Biodiversity hotspots: A shortcut for a more complicated concept. Global Ecology and Conservation 3: Marshall, C.A.M., & Hawthorne, W.D Important plants of northern Nimba County, Liberia: A guide to the most useful, rare or ecologically important species, with Mano names and uses. Oxford Forestry Institute, Oxford, United Kingdom. Marzoli, A., Piccirillo, E. M., Renne, P. R., Bellieni, G., Iacumin, M., Nyobe, J. B., & Tongwa, A. T The Cameroon Volcanic Line revisited: Petrogenesis of continental basaltic magmas from lithospheric and asthenospheric mantle sources. Journal of Petrology. 41: Mbenkum, T. F., & C. F. Fisey Ethnobotanical Survey of Kilum Mountain Forest. World Wildlife Fund, Yaoundé, Cameroon. McCain, C.M., & Grytnes, J-A Elevational Gradients in Species Richness. In: Encyclopaedia of Life Sciences (ELS).John Wiley & Sons, Ltd: Chichester McCune, B., & Mefford, M.J PC-ORD Multivariate Analysis of Ecological Data. Version MjM Software Design, Gleneden Beach, Oregon USA. p. 28. Memiaghe, H.R., Lutz, J.A., Korte, L., Alonso, A., & Kenfack, D Ecological importance of small diameter trees to the structure, diversity and biomass of a tropical evergreen forest at Rabi, Gabon. PloS ONE. 11: Meyer, W.B., & Turner II, B.L Changes in land use and land cover: A global perspective. Cambridge. Cambridge University press. p

218 Millennium Ecosystem Assessment (MEA), Ecosystems and human well-being: Biodiversity synthesis. World Resources Institute, Washington, DC Ministry of the Environment and Forestry (MINEF), A compendium of official instruments on forest and wildlife management in Cameroon: Legal regulations on forestry and wildlife. Law number of 20 January Yaoundé, Republic of Cameroon. p Mittermeier, R.A., Myers, N., Mittermeier, C.G., & Robles-Gil, P Hotspots: Earth's Biologically Richest and Most Endangered Terrestrial Ecoregions. CEMEX, SA, Agrupación Sierra Madre, Mexico City, Science. Molles, M.C Ecology: concepts and applications, fourth edition. McGraw hills, New York. p.604. Montagnini, F & Jordan, C Tropical Forest Ecology: The Basis for Conservation and Management. Heidelberg: Springer. Morgan, B.J., Adeleke, A., Bassey, T., Bergl, R., Dunn, A., Fotso, R., Gadsby, E., Gonder, K., Greengrass, E., Koulagna, D.K., Mbah, G., Nicholas, A., Oates, J., Omeni, F., Saidu, Y., Sommer, V., Sunderland-Groves, J., Tiebou, J., & Williamson, E Regional Action Plan for the Conservation of the Nigeria-Cameroon Chimpanzee (Pan troglodytes ellioti). IUCN/SSC Primate Specialist Group and Zoological Society of San Diego, CA, USA, p. 48. Murphy, P.G., Lugo, A.E Ecology of Tropical Dry Forest. Annual Review of Ecology and Systematics. 17: Myers, N., Mittermeier, R.A., Mittermeier, C.G., da Fonseca, G.A.B., & Kent, J Biodiversity hotspots for conservation priorities. Nature 403: Myneni, R.B., Dong, J., Tucker, C.J., Kaufmann, R.K., Kauppi, P.E., Liski, J., Zhou, L., Alexeyev, V., & Hughes, M.K A large carbon sink in the woody biomass of Northern forests. Proceedings of the National Academy of Sciences of the United States of America. 98: Neba, A Modern Geography of the Republic of Cameroon, 3 rd edition. Neba Publishers. Bamenda, Cameroon. p Neelo, J., Teketay, D., Kashe, K., & Masamba, W Stand structure, diversity and regeneration status of woody species in open and enclosed dry woodland sites around Molapo farming area of the Okavango delta, northeastern Botswana. Open Journal of Forestr 5: Nembot, T.F., & Tchanou, Z La Gestion des ecosystems forestriers du Cameroun a l aube de l an (2000). vol. 2. Monographies des sites critiques. International Union for Conservation of Nature, Yaounde, Cameroun. p

219 Newbery, D.M., & Gartlan, J.S A structural analysis of rain forest in Korup and Douala-Edea, Cameroon. Proceedings of the Royal Society of Edinburgh. 104B: Newsham, K.K The biology and ecology of the liverwort Cephaloziella varians in Antarctica. Antarctic Science. 22: Ngwa, J.A A New Geography of Cameroon. Longman Group Limited, Essex, United Kingdom. p Nono, A., Njonfang, E., Dongmo, A.k., Nkouathio, D.G., & Tchoua, F.M Pyroclastic deposits of the Bambouto volcano (Cameroon Line, Central Africa): evidence of an initial strombolian phase. Journal of African Earth Sciences. 39: Norris, K., Asase, A., Collen, B., Gockowksi, J., Mason, J., Phalan, B., & Wade, A Biodiversity in a forest-agriculture mosaic-the changing face of West African rainforests. Biodiversity conservation. 143: Noumi, E Floristic inventory of woody species in the Manengouba Mountains Forest Cameroon. Journal of Biology and Life Science. 4: Oates, J.F., Bergl, R.A., & Linder, J.M Africa s Gulf of Guinea forests: Biodiversity patterns and conservation priorities. Advances in Applied Conservation Biology (6). Conservation International, Washington, DC. p. 90. Olivry, J.C Fleuves et riviéres du Cameroun Collection Monographies Hydrologiques d ORSTOM 9. Mesres-Orstom, Paris. p Onana, J. M. 2010a. Endémisme floristique du Cameroun: Inventaire systématique et conservation de la biodiversité. In: X. M. van der Burg, L. J. G. van der Maesen, and J. M. Onana (eds.) Systematics and Conservation of African Plants. Proceedings of the 18th AETFAT Congress, Yaounde, Cameroon. Royal Botanic Gardens, Kew, United Kingdom. p Onana, J. M. 2010b. Etat de connaissance de la flore du Cameroun. In: X. M. van der Burg, L. J. G. van der Maesen, and J. M. Onana (eds.) Systematics and Conservation of African Plants. Proceedings of the 18th AETFAT Congress, Yaounde, Cameroon. Royal Botanic Gardens, Kew, United Kingdom. p Onana, J.M The vascular plants of Cameroon. A taxonomic checklist with IUCN assessments. Flore du Cameroun 39. IRAD-National Herbarium of Cameroon, Yaoundé. p Onana, J.-M., & Cheek, M Red Data Book of the flowering plants of Cameroon: IUCN global assessments. Royal Botanic Gardens, Kew. p

220 Onana, J-M Environnement biophysique, In : Onana JM (ed.). Synopsis des espèces végétales, vasculaires endémiques et rares du Cameroun. Checkliste pour la conservation et la gestion durable de la biodiversité. Flore du Cameroun 40. Ministère de la Recherche Scientifique et de l Innovation, Yaoundé. p Onana, J.M The World Flora online 2020 project: Will Cameroon come up to the expectation. Rodriguésia. 66: Osborne, P Tropical Ecosystems and ecological concepts. Cambridge, UK. p Ostertag, R., Inman-Narahari, F., Cordell, S., Giardina, C.P., & Sack, L Forest structure in lowdiversity tropical forests: A study of Hawaiian wet and dry forests. PLoS ONE. 9: Pan, Y., Birdsey, R.A., Fang, J., Houghton, R.A., Kauppi, P.E., Kurz, W.A., Phillips, O.L., Shvidenko, A., Lewis, S.L., Canadell, J.G., Ciais, P., Jackson, R.B., Pacala, S., McGuire, A.D., Piao, S., Rautiainen, A., Sitch, S., & Hayes, D A large and persistent carbon sinks in the world s forests. Science. 333: Parren, M.P.E Lianas and logging in West Africa. Thesis, Wageningen University, Wageningen, Netherland. p Parthasarathy, N., & Karthikeyan, R Plant biodiversity and conservation of two tropical dry evergreen on the Coromandel Coast, South India. Biodiversity Conservation. 6: Philips, O.L., & Gentry, A.H Increasing turnover through time in tropical forests. Science. 263: Plumptre, A.J., Davenport, T., Behangana, M., Kityo, R., Ndomba, E., Nkuutu, D., Owiuji, I., Eilu, G., Ssegawa, P., Ewango, C., Meirte, D., Kahindo, C., Herremans, m., Peterhans, J.K., Pilgrim J.D., Wilson, M., Languy, M., & Moyer, D The Biodiversity of the Albertine Rift. Biological Conservation 134: Poorter, L., Bongers, F., & Lemmens, R.H.M.J West African Forest. In Poorter, L., Bongers, F., Kouamé, F.Y.N., Hawthorne, W.D. (ed.). Biodiversity of West African Forest. An Ecological Atlas of woody plant species. CAB International, Cambridge, USA. p Putzke, J., Athanásio, C. G., de Albuquerque, M.P., Victoria, F.C., Pereira, A.B Comparative study of moss diversity in South Shetland Islands and in the Antarctic Peninsula. Revista Chilena de Historia Natural. 88: 1-6. Ramankutty, N., Evan, A.T., Monfreda, C., & Foley, J.A Farming the planet: 1. Geographic distribution of global agricultural lands in the year Global Biogeochemical Cycles 22: R Foundation for Statistical Computing R, version

221 Raven, P. H. & Williams, T. (editors) Nature and Human Society: The Quest for a Sustainable World. National Academy Press, Washington, D.C. Reich, P Taking stock of forest carbon. Nature Climate change. 1: Republic of Cameroon, National Biodiversity Strategy and Action Plan-Version II-Ministry of Environment, Protection of Nature and Sustainable Development (MINEPDED).p Richards, P.W Ecological notes on West African Vegetation III. The upland forests of Cameroon Mountains. Journal of Ecology. 51: Richter, M Tropical Mountains forests-distribution and general features. Biodiversity and Ecology Series. 2: Ricketts, T.H Tropical forest fragments enhance pollinator activity in nearby coffee crops. Conservation Biology. 18: Risser,P.G Diversity in and among grassland. In Wilson, E.O. & Frances, M.P. (eds). Biodiversity. National Academy press. Washington, D.C. USA. p Rogers, J The effectiveness of protected areas in Central Africa: A remotely sensed measure of deforestation and access. Thesis submitted to the Graduate School of Arts and Sciences. Columbia University. p Sainge, M.N., Franke, T., & Agerer, R Afrothismia korupensis (Burmanniaceae, tribe Thismieae) from Korup National Park, Cameroon. Wildenowia 35: Sainge, M.N., Franke, T., Merckx, V., & Onana, J-M Distribution of myco-heterotrophic (saprophytic) plants of Cameroon. In: X. van der Burgt, J. van der Maesen & J-M. Onana (eds), Systematics and conservation of African plants, Royal Botanic Gardens, Kew. p Sainge, M.N Systematics and Ecology of Thismiaceae of Cameroon, Master of Science Thesis, University of Buea, Cameroon. p Sainge, M.N., Lyonga, M.N., Libalah, M.B., Achah, R.A., Fon, J.N., & Kenfack, D Biodiversity Assessment and Conservation Status of Plants in the Mbembe Forest Reserve, Donga-Mantung Division, North West Region, Cameroon. Tropical Plant Exploration Group (TroPEG) Cameroon. Report to Rufford Small Grant Foundation. Sainge, M.N., Kenfack, D., & Chuyong, G.B Two new species of Afrothismia (Thismiaceae) from southern Cameroon. Kew Bulletin. 68: 1-6. Sainge, M.N., Lyonga, M.N., Libalah, M.B., Achah, R.A., Fon, J.N., Kenfack, D., & Moudingo, J-H. E Biodiversity Assessment and Conservation Status of Plants in the Mbembe Forest Reserve, 194

222 Donga-Mantung Division, North West Region, Cameroon. Vol. II, ISBN: Tropical Plant Exploration Group (TroPEG) Cameroon. p Sainge, M.N., & Cooper, J.C Reconnaissance survey of the biodiversity of the Cameroon Mountains. Cameroon Mountains research project. Tropical Plant Exploration Group (TroPEG), Cameroon. p. 55. Sainge, M.N Patterns of distribution and endemism of plants in the Camern Mountains: A case study of protected areas in Cameroon: Rumpi Hills Forest Reserve (RHFR) and the Kimbi Fungom National Park (KFNP). Tropical Plant Exploration Group (TroPEG), Cameroon. Report o Rufford Small Grant Foundation. p Sainge, M.N., Onana, J-M., Nchu, F., Kenfack, D., Peterson, A.T Botanical sampling gaps across the Cameroon Mountains. Biodiversity Informatics 12: Sonwa, D.J., Walker, S., Nasi, R., & Kanninen, M Potential synergies of the main current forestry efforts and climate change mitigation in Central Africa. Sustainability Science. 6: Sousa-Baena, M.S., Garcia, L.C., Peterson. A.T Completeness of digital accessible knowledge of the plants of Brazil and priorities for survey and inventory. Diversity and Distributions 20: Stattersfield, A.J., Crosby, M.J., Long, A.J., & Wege, D.C Endemic bird areas of the world: priorities for biodiversity conservation. BirdLife Conservation Series No. 7., BirdLife International, Cambridge, UK. Strassburg, B.B.N., Kelly, A., Balmford, A., Davies, R. G., Gibbs, H.K., Lovett, A., Miles, L., Orme, C.D.L., Price, J., Turner, R.K., & Rodrigues, A.S.L Global congruence of carbon storage and biodiversity in terrestrial ecosystems. Conservation Letters. 3: Sunderland, T.C.H., Comiskey, J.A., Besong, S., Mboh, H., Fonwebon, J., & Dione, M.A Vegetation Assessment of Takamanda Forest Reserve, Cameroon. In: Comiskey, J.A., Sunderland, T.C.H., & Sunderland-Groves, J.L. (eds., 2003). Takamanda: the Biodiversity of an African rainforest, Smithsonian Institution/Man and the Biosphere #8, Smithsonian Institution, Washington, D.C. p Swanson, T Global action for biodiversity: An international framework for implementing the convention on biological diversity. Earthscan publication limited, Cambridge, UK. p.191. Tabue, M.R.B., Zapfack, L., Noiha, N.V., Nyeck, B., Meyan-Ya, D.R.G., Ngoma, L.R., Kabelong, B.L- P., Chimi, D.C Plant diversity and carbon storage assessment in an African protected forest: 195

223 a Case of the eastern part of the Dja Wildlife reserve in Cameroon. Journal of Plant Sciences. 4: Tame, S., & Asonganyi, J Vegetation survey of the Ijim Mountain forests, Northwest Province, Cameroon. Report, Birdlife International, Yaounde, Cameroon. Appendix 2. p. 25. Tchetcha, D.J Plant Diversity and Ethnobotanical Investigations of Selected Sites in Lebialem Highlands, South West Region Cameroon. Master Thesis, University of Buea, Buea, Cameroon. Tchiengué, B Etude Ecologique et Floristique de la vegetation d un massif de la ligne du Cameroun: Le Mont Koupe. Thèse, Maitre és Sciences, Université de Yaounde 1. p Tchouto, M.G.P The vegetation of the proposed Etinde rain forest Reserve, Mount Cameroon and its conservation. M.Sc. Thesis, University of Edinburgh. Tchouto, P., & Ebwekoh, M.O A participatory Rapid Biodiversity Survey of the Muanemguba Mountain Forest. Final report to Centre for Environment and Rural Transformation (CERUT), Limbe, Cameroon. p. 54. Tchouto, M.G.P., Edwards, I., Cheek, M., Ndam, N., & Acworth, J Mount Cameroon cloud forest. In Timberlake, J. & Kativu, S. (eds.). African Plants: Biodiversity, Taxonomy and Uses. Royal Botanic Gardens, Kew. p Tchouto, M.G.P Plant diversity in a Central Africa rain forest: implications for biodiversity conservation in Cameroon. Ph.D thesis, Department of Plant Sciences, Biosystematics Group, Wageningen University, the Netherlands. p Teillet, P.M., Barker, J.L., Markham, B.L., Irish, R.R., Fedosejevs, G., Storey, J.C Radismetric cross-calibration of the Landsat-7 ETM+ and Landsat-5 TM sensors based on tandem data sets. Remote sensing of Environment. 78: Terborgh, J Diversity and the tropical rain forest. Scientific America Library. P Thomas, D.W. 1986a. Provisional Species List for Mount Oku Flora. In H. L. McLeod, H.L. (1986). The Conservation of Oku Mountain Forest, Cameroon. Study Report No. 15. International Council of Bird Preservation. Cambridge, United Kingdom. Thomas, D.W. 1986b. Vegetation in the montane forest of Cameroon. In S.N. Stuart (ed.), Conservation of Comeroon Montane Forest. International Council for Bird Preservation, Cambridge. p Thomas, D.W., & Cheek, M Vegetation and plant species in the proposed Etinde Reserve. Report to Government of Cameroon from Overseas Development Authority. Royal Botanic Gardens, Kew. p

224 Thomas, D.W., & Achoundong, G Montane Forest of Western Africa. In J. H. Seyani and A.C. Chikuni, Proc. XIII th plenary meeting AETFAT, Malawi. 2: Thomas, D.W., & Thomas, J Tchabal Mbabo Botanical Survey. World Wide Fund for Nature (WWF). Yaoundé, Cameroon. p. 90. Thomas, D.W Botanical Survey of the Rumpi Hills and Nta Ali with special focus on the submontane zone above 1,000 m elevation. Final report to German Technical Service, Korup Project, Mundemba, Cameroon. p. 95. Thomas, D.W Botanical inventory of Ejagham forest reserve Cameroon. Final report to Korup Project, Mundemba, Cameroon. p. 92. Thomas, D.W., Kenfack, D., Chuyong, G.B., Sainge, M.N., Losos, E., Condit, R.S., & Songwe, N.C Tree species of south western Cameroon: Tree distribution maps, diameter tables, and species documentation of the 50-hectare Korup Forest Dynamics Plot. Center for Tropical Forest Science of the Smithsonian Tropical Research Institute, Washington, D.C. p Thomas, D.W., & Chuyong, G.B The establishment of long-term forest monitoring plots in Southeast Cameroon. Report to Central African Regional Programme for Environment (CARPE). p 38. Thomas, D., Burnham, R.J., Chuyong, G., Kenfack, D., & Sainge, N.M. Liana abundance anddiversity in Cameroon s Korup National Park. In Patterns of liana demography and distribution:from local to global (eds) Stefan A. Schnitzer, Frans Bongers, Robyn J. Burnham. And Francis E. Putz. Ecology of lianas JohnWiley & Sons. Thompson, I.D., Mackey, B., McNulty, S., & Mosseler, A Forest resilience, biodiversity, and climate change. A synthesis of the biodiversity/resilience/stability relationship in forest ecosystems. Montreal: Secretariat of the Convention on Biological Diversity. Technical Series no. 43. p. 67. Tim, C The Convention on Biological Diversity and the Post-2010 target, Green week, Brussels p. 16. Tsafack, J-P.F., Wanji, P., Bardintzeff, J-M., Bellon, H., & Guilou, H The Mount Cameroon Stratovolcano (Cameroon Volcanic Line, Central Africa): Petrology, geochemistry, isotope and age data. Geochemistry, Mineralogy and Petrology. 47: Turner, B.L., Meyer, W.B., & Skole, D.L Global land use and land cover change: Towards an integrated study. Allen press on behalf of Royal Swedish Academy of science. 23:

225 Turner, W.R., Oppenheimer, M., & Wilcove, D.S A force to fight global warming. Natre. 428: United Nations Environmental Program (UNEP), Africa: Atlas of Our Changing Environment. United Nations Framework Convention on Climate Change (UNFCCC), Views on outstanding methodological issues related to policy approaches and positive incentives to reduce emissions from deforestation and forest degradation in developing countries. Advanced version. Bonn. United Nations Environmental Program/Convention for Biological Diversity (UNEP/CBD), Global strategy for plant conservation for a World Flora online by United Nations Convention on climate change (UNFCCC), Conference of Parties 21. Paris. p. 31. Uprety, D.C., Dhar, S., Hongmin, D., et al Technology for climate change mitigation: Agriculture sector. UNEP Centre on energy, climate and sustainable development, Copenhagen Vallèrié, M Notice explicative No 45. Carte pédologique du Cameroun occidental à 1/ Office de la Recherché Scientifique et Technique Outré-mer, Paris. p. 59. Van der Werf, G.R., Morton, D.C., Defries, R.S., et al Carbon dioxide emissions from forest loss. Natre Geoscience. 2: Vivien, J., & Fare, J.J Arbres des forêts denses d Afrique Centrale. Ministère de la Coopération. Agence de Coopération Culturelle et Technique. Paris, p White, F The Vegetation map of Africa: A descriptive memoir, United Nation Educational, Scientific, and Cultural Organization (UNESCO), Paris. 20. p Whittaker, R.H Vegetation of the Great Smoky Mountains. Ecological monographs. 26: Whittaker, R.H Vegetation of the Siskiyou Mountains, Oregon and California. Ecological Monographs. 30: Whittaker, R.H Dominance and diversity in land plants communities. Science. 147: World Resource Institute (WRI) & International Union for Conservation of Nature (IUCN), Climate, Biodiversity, and Forests: Issues and opportunities emerging from the Kyoto protocol. Washington DC, USA. p. 36. Yerima, B.P.K., & Ranst, E.V Major soil classification systems used in the tropics: soils of Cameroon: distribution, genesis, characteristics, management and utilization. Oxford, Trafford. p Zanne, A.E., Lopez-Gonzaley, G., Coomes, D.A., Ilic, J., Jansen, S., Lewis, S.L., Miller, R.B., Swenson, N.G., Wiemann, M.C., & Chave, J Global wood density database. (Dryad. Identifier

226 Zhong, L., Chang-Yang, C., Lu, P., Gu, X., Lei, Z., Cai, Y., Zheng, F., Sun, F., Yu, M Community structure and species composition of the secondary evergreen broad-leaved forest: The analyses for a 9 ha forest dynamics plot in Wuyanling Nature Reserve, Zhejiang Province, East China. Biodiversity Science. 23: Zuidema, P.A., Baker, P.J., Groenendijk, P., Schippers, P., Sleen, P.V., Vlam, M., & Sterck, F Tropical forest and global change: filling knowledge gaps. Trends in Plant Science. 18:

227 SHORT BIOGRAPHY Moses Nsanyi SAINGE was born on the 18 of September 1974 in Limbe, Cameroon. He attended the then Saint James College, Ndop between 1985 and 1990, Success Evening School, Tiko between 1991 and 1992, and Government High School, Mundemba between 1992 and In 2004 he was admitted at the University of Buea to read Botany between 2004 and 2007, He returned to the University of Buea between 2009 and 2012 where he obtained an MSc in Botany at the Department of Botany and Plant Physiology. In 2014 he was granted admission to do a PhD program at the Cape Peninsula University of Technology in Cape Town, South Africa. He carried out his research work on the Cameroon Mountains specifically on Rumpi Hills Forest Reserve and the Kimbi Fungom National Park, Cameroon from February to November, Between 1997 and 2004, he assisted in the establishment of the 50 ha permanent plot in central Korup National Park and the establishment of 18 ha of lianas in different capacities: Team Leader, Camp Manager, and Field Manager. Between 2008 and 2009, he led the first re-census of the 50 ha plot in Korup National Park. Between 2012 and 2013, he led the second re-census of trees and first re-census of lianas in the 50 ha plot. Between 2010 and 2015 he established and monitored six hectare plots within 2 km apart from the 50 ha plot for a collaborative research between Center for Tropical Forest Science (CTFS) and the Tropical Ecology and Monitoring Network (TEAM). He served the Center for Tropical Forest Science in Korup National Park, Cameroon for twenty years ( ). His long experience in studying the Flora of Tropical Africa has earned him the Africa specialist for Mycoheterotrophic plants. A study he has been working on for over fifteen years and has discovered and described about seven species new to science (Afrothismia saingei Franke, A. hydra Sainge & Franke, A. korupensis Sainge, & Franke, A. foertheriana Franke, Sainge & Agerer, A. fungiformis Sainge & Kenfack, A. pusilla Sainge & Kenfack and A. sp. nov.) and a new genus for Cameroon (Kihansia jengiensis Sainge & Kenfack). Beside this, one of the species was named after him (Afrothismia saingei Franke). He s much interested and specialized in plant taxonomy, tropical ecology, and biodiversity informatics. He is a co-founder of Tropical Plant Exploration Group (TroPEG) Cameroon. A Non-Governmental Organization where he and his colleagues are trying to understand the forest structure, composition, diversity, carbon stock, and the ecological, livelihood, and ethnobotanical aspect of the Cameroon Mountains through the Cameroon Mountains Biodiversity Project. Between 2011 and 2016, he and his colleagues of TroPEG Cameroon have established 70 permanent 1-ha plots across the Cameroon Mountains: 4 ha at the Mbembe Forest Resreve, 17 ha at the Kimbi Fungom National Park, 12 ha at the Bakossi National Park, 12 ha at Mt Nlonako and 25 ha at the Rumpi Hills Forerst Reserve. In 2013, he was involved with the Biodiversity Informatics 200

228 Training Curriculum (BITC) team headed by Prof. A. Townsend Peterson of the Biodiversity Institute, University of Kansas, USA and is currently a Biodiversity Informatics trainer. Moses Sainge is married to Benedicta Jailughe and has four kids: Elizabeth Sainge Saingi, Thelma Sainge Saingi, Morgan Sainge Saingi, and Carlson Sainge Saingi. LIST OF PUBLICATIONS Sainge, M.N., Chuyong, G.B., Peterson, A.T Endemism and geographic distribution of African Thismaiceae. Plant Ecology and Evolution 150: Sainge, M.N., Onana, J-M., Nchu, F., Kenfack, D., Peterson, A.T Botanical sampling gaps across the Cameroon Mountains. Biodiversity Informatics 12: Beenken, L., Sainge, M.N., Kocyan, A Lactarius megalopterus, a new angiocarpous species from a tropical rainforest in Central Africa, shows adaptations to endozoochorous spore dispersal. Mycological Progress. 58: Ewango, C.E.N., Thomas, D.W., Kenfack, D., Sainge, M.N., & Burgt, X.M. van der Gambeya korupensis (Sapotaceae: Chrysophylloideae), a new rain forest tree species from the Southwest Region in Cameroon. Kew Bulletin 1-7. Sainge, M.N., Kenfack, D A new species of Triuridaceae from southeastern Cameroon, Kihansia jengiensis Sainge & Kenfack. Kew Bulletin 70(7): 1-5. Kenfack, D., Gereau, R.E., Thomas, D.W., and Sainge, M.N Revision of Crotonogynopsis (Euphorbiaceae) with two new species from Cameroon and Tanzania. Novon. Sainge, M.N., Kenfack, D., & Chuyong, B.G Two new species of Afrothismia (Thismiaceae) from southern, Cameroon. Kew Bulletin, 68(4): Mennes, C.B., Smets, F. E., Sainge, M.N., & Merckx, S.F.T.V New insights in the long debated evolutionary history of Triuridaceae (Pandanales). Molecular Phylogenetics and Evolution. Waltert M, Bobo KS, Kaupa S, Montoya ML, Sainge, M.N., and Fermon, H Assessing Conservation Values: Biodiversity and Endemicity in Tropical Land Use Systems. PLoS ONE 6(1): 1 7. DeWalt, S. J., Schnitzer, S. A., Chave, J., Bongers, F., Burnham, R. J., Cai, Z., Chuyong, G. B., Clark, D. B., Ewango,C. E. N., Gerwing, J.J., Gortaire, E., Hart, T., Guillermo, I. K. I., Kenfack, D., Manuel, J. M., Makana, J. M., Martinez-Ramos, M., Mascaro, J., Sainge, M.N., Muller-Landau, H. C., Parren, M. P. E., Parthasarathy, N., Pérez-Salicrup, D. R., Putz, F. E., Romero-Saltos, H., 201

229 and Thomas, D. W Annual rainfall and seasonality predict pan-tropical patterns of liana density and basal area. Biotropica 42: Sainge, M.N., Franke, T., Merckx, V., and Onana, J.-M Distribution of myco-heterotrophic (saprophytic) plants of Cameroon. In: X. van der Burgt, J. van der Maesen & J.-M. Onana (eds), Systematics and conservation of African plants, pp Royal Botanic Gardens, Kew. Merckx, V., Chatrou, L.W., Lemaire, B., Sainge, M.N., Huysmans, S., and Smets, E. F Diversification on myco-heterotrophic angiosperms: evidence from Burmanniaceae. BMC Evolutionary Biology, 8:178. Imhof, S., and Sainge, M.N Ontogeny of the myco-heterotrophic species Afrothismia hydra (Burmanniaceae). Botanical Journal of the Linnean Society, 157: Kenfack, D. Sainge, M.N., and Thomas, D. W A new species of Cassipourea korupensis (Rhizophoraceae) from western Cameroon. Novon 16(1) Sainge, M.N., Franke, T., and Agerer, R Afrothismia korupensis (Burmanniaceae, tribe Thismieae) from Korup National Park, Cameroon Wildenowia 35: LIST OF REPORTS Sainge, M.N Patterns of distribution and Endemism of Plants in the Cameroon Mountains. A case study of Protected Areas in Cameroon: Rumpi Hills Forest Reserve (RHFR) and the Kimbi Fungom National Park (KFNP). Tropical Plant Exploration Group (TroPEG) Cameroon. Technical Report submitted to the Rufford Small Grant Foundation, UK. Sainge, M.N., Libalah, M. B., Lyonga, M. N., Arifique, R. A., Fon, J. N., and Kenfack, D Report on Biodiversity Assessment, and Conservation Status of Plants in the Mbembe Forest Reserve. Donga Mantung, Northwest Region (NWR), Cameroon. Rufford Small Grants Foundation, UK. BOOKS Sainge, M.N., Lyonga, M. N., Libalah, M. B., Achah, R. A., Fon, J. N., Kenfack, D. and Moudingo, J- Hude E Biodiversity Assessment and Conservation Status of Plants in the Mbembe Forest Reserve. Donga Mantung, Northwest Region (NWR), Cameroon. Vol. II, 140pp, ISBN: , Tropical Plant Exploration Group (TroPEG) Cameroon. 202

230 PLOT DATA 203

231 Plot Habitat data form 204

232 Plot Main Data Form 205

233 Observational Data Form 206

234 PERMITS Ministry of Scientific Research and Innovation (MINRESI) 207

235 Ministry of Forestry and Wildlife (MINFOF) 208

236 Letter of Authorization from Fon of Esu 209

237 Photographs of study sites 210