Effects of integrated land management, landscape position and land-use types on soil physicochemical properties, discharge, species richness and carbon stock in Geda watershed, north Shewa, Ethiopia

Table of Contents

Abstract

Acknowledgement

List of Figures

List of Tables

List of Appendices

List of Acronyms

CHAPTER ONE
1. INTRODUCTION
1.1 Background
1.2 Statement of the problem
1.3 Justification of the study
1.4 Research objectives, hypothesis and research questions
1.4.1 General and specific objectives
1.4.2 Research hypothesis
1.4.3 Research questions
1.5 Structure of the thesis

CHAPTER TWO
2. LITERATURE REVIEW
2.1 Definition and extent of land degradation
2.2 Severity of land degradation in Ethiopia
2.3 Causes of land degradation in Ethiopia
2.4 Impacts of land degradation on the environment and landscape productivity
2.4.1 Impact on the environment
2.4.2 Impact on landscape productivity
2.5 Efforts to rehabilitate degraded lands in Ethiopia
2.5.1 Government initiatives
2.5.2 Ineffectiveness of past interventions
2.5.3 Integrated watershed based approach
2.6 Impacts of conservation measures in degraded land rehabilitation

CHAPTER THREE
3. MATERIALS AND METHODS
3.1 Description of the study area
3.2 Study design
3.3 Determination of selected soil physicochemical properties
3.3.1. Soil sampling
3.3.2 Soil lab analysis
3.3.3 Soil moisture analysis
3.4 Determination of water discharge
3.5 Assessment of plant species richness
3.6 Assessment of carbon stock
3.7 Data Analysis

CHAPTER FOUR
4. RESULTS AND DISCUSSION
4.1 Effects of integrated land management on selected soil properties
4.1.1 Effects of integrated land management on soil physical properties
4.1.2 Effects of integrated land management on soil moisture content
4.1.3 Effects of integrated land management and landscape position on selected soil chemical properties
4.1.4 Effects of land-use and soil depth on selected soil chemical properties
4.2 Effects of intergrated land management on water discharge and irrigation practices
4.2.1 Effects of intergrated land management on water discharge
4.2.2 Effects of integrated land management on irrigation practice
4.3 Effects of integrated land management practice on plant species richness
4.4 Effects of integrated land management on plant biomass production, biomass export and carbon stock
4.4.1 Effects of integrated land management and landscape position on plant biomass production and biomass export
4.4.2 Effects of land-use types on plant biomass production and biomass export
4.4.3 Effects of integrating land management on carbon stock through plant biomass
4.4.4 Effects of integrated land management on soil carbon stock
4.4.5 Effects of integrated land management measure on total carbon stock

CHAPTER FIVE
5. CONCLUSION AND RECOMMENDATIONS
5.1 Conclusion
5.2 Recommendations

References

Appendices

Abstract

Watershed based integrated land management is a recent approach to curb land degradation in Ethiopia and introduced in 2012 in Geda watershed, central highlands of Ethiopia. However, the impacts of the interventions on indicators of some ecosystem services were not assessed. The objectives of this study were to explore the effects of the interventions on soil properties, soil moisture content and water discharge, plant species richness, biomass production and carbon stock by comparing treated site with integrated land management measures and the adjacent untreated site. Samples were collected from treated and untreated sites in the upper and lower landscape positions, from crop land , grazing land and Tree Lucerne plantation based on standard procedures for each objective. The collected data were analyzed following standard statistical procedures with respect to treatment, landscape position, land-use types, and soil depth. M ost of the soil physicochemical properties, soil moisture content, water discharge, plant species richness, biomass production and carbon stocks were significantly (p ≤ 0.001 ) improved in the treated site compared with the untreated one . Clay, total N, available P and soil organic carbon were significantly higher at p = 0.001 and exchangeable K at p = 0.05 in the treated site than the untreated one. This could be due to higher organic matter accumulation and improved vegetation growth as a result of prohibited free grazing, and reduced erosion by the conservation structures. Sand and bulk density were significantly ( p = 0.001) higher in the untreated site that could be attributed to erosion due to absence of conservation measures and compaction by livestock trampling during free grazing practice. Generally,in the treated site, the clay content of the soil was improved by 63.51%, OC by 133%, NPK of the soil by 69.84%, 78.49% and 22.73% respectively. The soil moisture content increased by 14.82 ̶ 19.35% and water discharge increased by 588% over the untreated one; which could be due to reduced runoff and evaporation, improved infiltration and storage processes attributed by the conservation structures and vegetation covers. Besides, total plant richness, regeneration of shrubs and indigenous tree species were increased by 18%(N=27),70.59% (N=12) and 66.67% (N=2) respectively, in the treated site compared to the untreated one. In addition, an average 10.72±0.84 Mg ha-1 additional carbon stock was observed due to the intervention. Prohibited free grazing and land-use change could have contributed to higher species richness, regeneration of indigenous trees and accumulation of higher plant biomass and accumulation of higher carbon stocks. Tree Lucerne plot showed higher biomass production and carbon stock by plant biomass depicting the positive impact of land-use changes . Tree Lucerne plot significantly (p = 0.05 ) improved the upper soil (0 ̶ 15 cm) carbon stock which could be due to its N fixing capacity and fast decomposition of the leaves. However, the lower soil (15 ̶ 30 cm) carbon stock was significantly ( p = 0.001) higher in the crop land of the treated site which could be ascribed to the conservation structures and tillage operations, that conservation structures trap and accumulate transported organic materials from upper landscape; and tillage facilitates aeration and decomposition processes. Planting Tree Lucerne combined with physical structures as a biological measure and on highly degraded part of the landscape resulted higher plant biomass production and carbon stock through plant biomass. Thus, integrating physical, biological and grazing management is a good land-use practice to be expanded to other degraded landscapes.

Key words : Biomass, Carbon stock, integrated land management, Landscape position, Treatment.

Acknowledgement

First of all, I praise the Almighty God, His mother virgin St. Mary and the angel St. Michael for the strength and support throughout my life; formatting me start and finish this study successfully.

I would like to express my sincere gratitude to my Ph.D. supervisors: Dr. Mekuria Argaw, Dr. Lulseged Tamene, and Dr. Kindu Mekonnen for their immense support and guidance. I am deeply indebted to my principal supervisor, Dr. Mekuria Argaw (Associate professor of Environmental Science) for his kind help and friendly treatment, invaluable guidance, and encouragement throughout the process of this work. My appreciation also goes to Dr. Lulseged Tamene for his invaluable time in framing the thesis work from the very beginning and paved a way for me to participate in trainings and workshops at ILRI; and Dr. Kindu Mekonnen for attaching me in Africa RISING project that partly supported my field research and laboratory costs. Africa RISING is a program financed by the United States Agency for International Development (USAID) as part of the United States Government’s Feed the Future Initiative. Africa RISING is aligned with research programs of the CGIAR”. I am also thankful to Dr. John Recha for his meticulous review of my manuscript and encouraged me to submit my works to peer reviewed journals.

I sincerely acknowledge Kesis Dr. Abiyou Tilahun, botanist at Debre Berhan University, for his immense and invaluable help during species identification starting from the field data collection to laboratory identification; without him, identifying 150 plant species would have been too costly. Dr. Alemnew Berhanu, Mr. Getachew Shumneri and Mr. Haileyes Getaneh were exemplary genuine friends who were available all time for kind support from the very beginning to the end of my PhD study; indeed, thank you all!

I am very much grateful to the Center for Environmental Science, Addis Ababa University and Debre Berhan University for the permission to join this study program and providing necessary resources to accomplish my thesis research successfully. Further, my appreciation goes to the staff of college of Agriculture and Natural Resource Sciences and the Department of Plant Science, Debre Berhan University, for their kind support during my study. The financial support for this study from the office of Associate Dean for Research and Technology Transfer, Addis Ababa University, and the office of Post Graduate Program, Debre Berhan University, were highly appreciated. My sincere gratitude also goes to Dr. Yedilfana Setarge, Prof. Teshome Soromessa, Dr. Ahmed Hussen, and Dr. Tadesse Alemu for their encouragement to complete seminars, independent study and the thesis work in time and facilitation of presentations. I am grateful to W/ro Kindest Shiferaw and W/ro Eyerusalem Shewarega for their good office facilitation of different matters at the Center for Environmental Science. Further, my appreciation goes to all staff members of the Center for Environmental Science and Ph.D. students enrolled in 2014/15 since we had friendly relationship and encouraging support throughout my study period. I sincerely acknowledge the professional commitments of Dr. Eyasu Elias and Dr. Zenebe Adimassu; thank you so much for your deeds.

I am indebted to Debre Berhan Agricultural Research Center for their kind permission to use their soil laboratory facilities and ILRI for its permission to use the library facility. I also highly appreciate the kind help and encouragement of Dr. Wuletawu Abera (CIAT), Dr. Hailemariam Birkie, Dr. Girma Tadesse, Dr. Almaz Afera, Dr. Gezahegn Degefe, Dr. Nigus Tadese, Dr. Dereje Andargie, Dr. Kasahun Bekele, Dr. Tamirat Cheru, Dr. Wondosen Tena, Mr. Getahun Agumas, W/t Hiwot Zewudie (Debre Berhan University), Dr. Adamu Molla, Mr. Lisanu Getaneh, Mr. Getaneh Shegaw, Mr. Tsegaye Getachew (Debre Berhan Agricultural Research Center), Mr. Talegeta Teketel (North Shewa Zone NRM), W/ro Debrework Ababu, Melake Genet Haile Ayele, W/ro Sisay Aberu, Mr. Girma Tsegaye, W/ro Genet Aragie, Mr. Engida Ababu, Sister Aselef Wolde, Dr. Abebe Tedla, W/ro Tena Manaye, Kesis Hailegiorgis Habtewold, W/ro Marshet Debebe (my neighbors) during my research work. In addition, I would like to thank Mr. Melkamu Dagne (Extension officer of Gudo Beret Kebele) for his generous support, and the farmers at Gudo Beret and Adisgie Kebeles that I collected all my data from their holdings, for their openhearted support. They were highly cooperative and welcoming to me when I repeatedly visited their fields at various times to collect various data. Thank you all so much!

Last but not least, I am grateful to my father Ato Terefe Diressie, my mothers W/ro Yemenzwork Abebe, W/ro Yeshi T/Wold, W/ro Mamitie Jimalu; my brothers Dn. Samuel Alemu, Ato Cherinet Terefe, Ato Getinet Terefe, Dr. Geleta Nigusie, Ato Dagnachew Abebe, Ato Tsegaye Abebe, Ato Endalkachew Abebe, Ato Hailu Tesfu, Abrham Sisay, Chala Sisay, Getinet Hailu; my sisters W/ro Helen Desta, W/ro Kasech Terefe, W/ro Etsegenet Alemu, W/t Hasabe Terefe, W/t Gete Mekonnen, W/t Berhan Hailu, W/ro Wudnesh Abebe, W/ro Adanech Abebe, W/ro Tigist Abebe, W/t Birtukan Abebe, W/ro Helen Hailu, W/t Hana Sisay, W/t Kalkidan Hailu, W/ro Mekdes Eshetu, Hana Eshetu; my brothers in Christ Ato Asmerom G/giorgis, Ermias Asmerom, Ato Negash Worku; my sisters in Christ, W/ro Genet and Dibora Asmerom; my wife W/ro Almaz Abebe; my sons Yosef Hailu and Yosef Samuel; my daughters Edilawit Hailu and Selamawit Hailu whose prayers and support were sources of inspiration and motivation for my work. I highly recognize the time I should give to my family, especially to my children at the age they need it, was paid for the successful end of this study. Thank you for your patience!

There were so many people supporting me during the study and contributed to my success in various ways; though their names are not mentioned here, I highly appreciate all of them for their unreserved support throughout my study life.

Hailu Terefe Diressie

June, 2020

List of Figures

Figure 1. Schematic representation of the links of soil erosion causes

Figure 2. Map of the study area in Geda watershed

Figure 3. Mean annual rainfalls and temperatures in the study area

Figure 4. Schematic representation of the study design

Figure 5. Triangular soil sampling method used in soil sampling

Figure 6. Phase diagram and elements of unsaturated soil

Figure 7. Measuring water flow at the outlets of the sites

Figure 8. Sampling layout for assessing plant species richness

Figure 9. Measuring fresh biomasses for biomass and carbon stock estimation

Figure 10. Drying samples in the oven and measuring the dry mass

Figure 11. Particle size distribution by treatment and landscape position

Figure 12. Particle size distribution by treatment and land-use types.

Figure 13. Partial view of rivers in the treated and untreated sites in the dry months

Figure 14. Amount of mean water discharge in the treated and untreated sites

Figure 15. Number of plant species by growth habits recorded in the study area

Figure 16. Plant biomass retention status of the study sites.

Figure 17. Mean plant biomass in the main rainy season and in the dry seasons.

Figure 18. Expansion of new eucalyptus plantation

Figure 19. Landscape carbon stock in the study area

Figure 20. Integrated land management induced carbon stock

List of Tables

Table 1. Characterization of the treated and untreated sites

Table 2. Effect of integrated land management, landscape position and land-use types on selected soil physical properties

Table 3. Effect of soil depth, landscape position and land-use on soil physical properties

Table 4. Effect of integrated land management and landscape position on soil moisture content

Table 5. Effects of treatment and landscape position on selected soil chemical properties

Table 6. Effects of land-use and soil depth on selected soil chemical properties

Table 7. Dry period water discharge

Table 8. Temporal changes of irrigated land and crop types in the treated site

Table 9. Temporal changes of irrigated land in the untreated site

Table 10. Effects of integrated land management and landscapepe position on plant species richness.

Table 11. Effects of integrated land management, land-use, aspect and conservation types on plant species richness.

Table 12. Numbers of plant species, their families and frequencies recorded in the study area

Table 13. Eight families representing major plant species in the study area

Table 14. Sorensen’s similarity indices for species richness

Table 15. Effects of integrated land management and landscape positions on plant biomass production in the main cropping season and biomass retention in the dry season

Table 16. Correlation between plant biomass production, carbon stock and soil moisture

Table 17. Effects of land-use types on plant biomass production, biomass retention and exported plant biomass

Table 18. The number of plots occupied by new eucalyptus plantation in the study area

Table 19. Effects of integrated land management, landscape position and land-use types on biomass carbon stock

Table 20. Effects of land-use types on plant biomass carbon stock

Table 21. Effects of integrated land management and landscape positions on soil carbon stock

Table 22. Effects of integrated land management, landscape positions and land-use types on total landscape carbon stock

List of Appendices

Appendix 1: List of plant species recorded in the study area

Appendix 2. Lab. analysis results of soil properties

Appendix 3. Lab. analysis results and calculations of plant biomass and carbon stocks

Appendix 4. ANOVA for soil physical properties

Appendix 5. ANOVA for soil moisture analysis

Appendix 6. ANOVA for soil chemical properties

Appendix 7. ANOVA for plant species analysis

Appendix 8. ANOVA for plant biomass, exported biomass and carbon stocks

Appendix 9. Data collection sheet for plant species identification

Appendix 10. Semi-structured interview questions employed in the irrigation sites

Appendix 11. Partial views of the treated site during the rainy and dry seasons.

Appendix 12. Partial views of the untreated site in rainy and dry seasons .

Appendix 13. Soil analysis in the laboratory

Appendix 14. Water collection ditches in Tree Lucerne plantation trapping runoff during the short rainy seasons

Appendix 15. List of Scientific Papers

List of Acronyms

AGB Above Ground Biomass

AI Artificial Insemination

ANOVA Analysis of Variance

CGIAR Consultative Group For International Agricultural Research

CIAT International Center for Tropical Agriculture

DBARC Debre Berhan Agricultural Research Center

FDRE Federal Democratic Republic of Ethiopia

FFW Food for Work

GDP Gross Domestic Product

GIZ Deutsche Gesellschaft für Internationale Zusammenarbeit

IFPRI International Food Policy Research Institute

ILRI International Livestock Research Institute

IPCC Intergovernmental Panel on Climate Change

IRR Internal Rate of Return

MoA Ministry of Agriculture

NGOs Non-Governmental Organizations

NRM Natural Resource Management

PLUP Participatory Land-use Planning

AfricaRISING Africa Research In Sustainable Intensification for the Next Generation

SAS Statistical Analysis System

SCR Silt to Clay Ratio

SLM Sustainable Land Management

SLMP Sustainable Land Management Program

SNNP(R) Southern Nations Nationalities and People (Region)

SOC Soil Organic Carbon

SWC Soil and Water Conservation

TLU Tropical Livestock Unit

TN Total Nitrogen

USAID United States Agency for International Development

USEPA United States Environmental Protection Agency

WFP World Food Program

CHAPTER ONE

1. INTRODUCTION

1.1 Background

The highlands of Ethiopia experienced severe land degradation through soil erosion by water due to proximate causes such as deforestation, overgrazing, over-cultivation, and underlying causes such as population pressure, steepness of the topography, poor farming practices, poverty and tenure insecurity (Gideon, 2004; Haile et al., 2006; Mushir, 2013; Gashaw et al., 2014;2015). Most of the mountainous landscapes of the country have been cultivated for decades without adequate use of soil and water conservation measures to minimize soil erosion by water. As a result, soil erosion by water affected about 50% of the Ethiopian highlands (Asfaw and Neka, 2017; Ebabu et al., 2019). Each year, about 1900 million tons of soil, equivalent to an average net of 100 tons ha-1 year-1 soil eroded and annual soil loss from farmlands reached 200-300 tons ha-1year-1 which is equivalent to 8 mm soil depth (FAO, 1986; Tilahun et al., 2018).

In monetary terms, Ethiopia loses US$ 1 to 2 billion year–1 (Dessalegn et al. 2015). Ayalew (2011) reported 17% of the potential annual agricultural GDP has been lost due to physical and biological soil degradation. It is estimated that the cost of land degradation in Ethiopia reached 23% of the country’s GDP (Kirui and Mirzabaev, 2015) . In addition, soil degradation brings about indirect costs such as loss of environmental services, silting of dams and river beds and reduced groundwater.

In the highlands of Ethiopia, soil erosion by water reduced soil fertility and water availability for agricultural and non-agricultural activities. Unpredictable and uneven distribution of rainfall coupled with lack of adequate water storage capacity increased water demand for agriculture, livestock and human use (Jemberu, 2018). Land-use changes due to soil erosion and human activity in these highlands also contributed to nutrient loss, reduced water holding capacity, reduced base flows and reduced landscape productivity (Yoseph et al., 2017).

Apart from soil erosion, loses in plant species richness and diversity are also increasing ecological problems in the highlands of Ethiopia (Haile et al., 2017; Kidane et al., 2019). Reduction of plant species richness and diversity further affects ecosystem sustainability through the negative effects on the food web connectivity and energy transfer among interacting species (Adu et al., 2017). Evidences showed that erosion controlling measures enhanced vegetation composition and diversity (Mekuria et al., 2011; Damtie, 2017).

Considering the severity of soil erosion and its associated consequences, conservation and restoration measures are essential to improve soil fertility, enhance primary production and nutrient cycling, and preserve biodiversity at farm and landscape levels (Adimassu et al., 2017; Damtie, 2017). Conservation measures can change the physical conditions of soils such as soil organic matter, soil structure, soil water holding capacity, soil bulk density, soil porosity, soil pH and its workability (Mulugeta and Karl, 2010).

Thus, the government of Ethiopian launched massive rehabilitation programs starting from mid–1970s (FAO, 1986; Ebabu et al., 2017) with an estimated investment of more than 1 billion US dollars during the years between 1974 to1991 (Adimassu et al., 2017). Despite the widespread effort to restore degraded areas in Ethiopia, success of the conservation measures were generally limited (Aune et al., 2002; Gashaw, 2015; Tiki et al., 2016). For example, from the rehabilitation works done between 1976 and 1990, only 30% of soil bunds, 25% of the stone bunds, 60% of hillside terraces, 22% of land planted with trees, and 7% of the reserve areas survived in the year 1990 (USAID, 2000). Some of the reasons mentioned for the failure were the ‘top-down’ approach that the government followed without involving local communities and stakeholders in planning and implementation (Tiki et al., 2016), limitations of technologies in benefiting smallholder farmers, poor implementation and maintenance of structures (Shiferaw and Holdern, 2016), lack of understanding of nature’s functioning and processes (Keesstra et al, 2018).

Thus, there was a change in approach towards integrated, participatory and watershed-based interventions as of 1980s (Haregeweyn et al., 2012; GIZ, 2015; Gashaw, 2015; Ebabu et al., 2017; Gashaw et al., 2017); and it has been extensively implemented by the government and NGOs. The basis of the integrated watershed-based intervention approach was that the local communities would take part as major actors in the process of planning and implementation of rehabilitation activities (Haregeweyn et al., 2012). Accordingly, recent efforts attempted to follow integrated and participatory approaches in their endeavor to restore degraded areas across different watersheds in Ethiopia (Mekuria et al., 2011; Haregeweyn et al., 2012; Ebabu et al., 2017). Integrated and watershed based land management practices include the construction of soil and or stone bunds, terraces, trenches, cut off drains, drainage channels, check dams, fanya juu, planting of shrubs or trees, establishing area exclosure, combinations of structural and vegetative measures, and reduction of household livestock numbers (IFPRI, 2009; Sultan et al., 2017; Ebabu et al., 2019).

Various NGOs have been supporting the government of Ethiopia in its effort to enhance overall system productivity and improve ecosystem services through integrated watershed management practices. In order to tackle the problem of limited data availability related to the performances of integrated and watershed based land management practices and define methods for scaling, some projects a establishing learning watersheds where integrated land management practices are co-implemented and evidences are generated for awareness creation and further scaling. The Africa RISING (Research In Sustainable Intensification for the Next Generation) program in the highlands of Ethiopia is one such projects promoting learning watersheds where integrated sustainable intensification (SI) and natural resources management (NRM) can be undertaken (Mekonnen, 2018). The program has six learning watersheds in which technologies, methods, tools, frameworks are tested before scaled to other areas. One such learning watershed is the Geda watershed of the Amhara region, north Shewa zone, characterized by a mixed crop-livestock farming system of the central highlands. This study assessed the effects of integrated land management practices on selected soil physicochemical properties, species richness, and water flow and carbon stock considering landscape position, soil depth and land-use type. This helps to understand the benefits of those practices in terms of livelihood improvement and indicators of ecosystem services.

1.2 Statement of the problem

Land degradation is severe in Geda watershed due to the steepness of the topography, traditional farming practices, overgrazing, and deforestation. To halt the problem and foster rehabilitation works, integrated land management practices having several intervention packages (physical, biological measures, restriction of free grazing) were introduced. Nevertheless, studies on the effects of such practices are limited and results are inconsistent. Moreover, best land management practices were not documented and identified. Hence, a comparative study is was essential to examine and document the effects of the practices and to recommend the best ones to expand to other degraded areas through scaling.

1.3 Justification of the study

The dominant farming practice in Geda watershed is a mixed crop-livestock system where crop production and livestock husbandry compete for spaces and resources incurring over exploitation of natural resources which further aggravate land degradation. To improve the ecosystem's productivity and adequately support both the crop and livestock production systems, the Amhara Region has introduced various integrated land management practices in Geda watershed since 2012. The introduced practices combined physical and biological measures and prohibitions of free grazing that could affect the soil physicochemical properties, discharge capacity, resources remaining in the landscape and leaving from the landscape; which intern affects species richness, and carbon stocks of the landscape. Africa RISING project in collaboration with CIAT, ILRI and other CGIAR centers and national partners started working in Geda watershed a few years ago to generate evidences on the impacts of the development interventions on controlling soil erosion, runoff and improve crop yield, and support government NRM initiatives (Mekonnen, 2018). In the country, researches on impacts of NRM interventions focused mainly on plot/farm levels and studies on watershed-level impacts are limited. Very few researchers such as Mulugeta and Karl (2010), Wolka et al. (2011), Hishe et al. (2017b), and Ademe et al. (2017) studied the impacts of SWC practices on soil physicochemical characteristics in northern and southern parts of Ethiopia highlands. Since the setup and agro-ecology of the central highlands are different from the northern and southern parts of the country and the current intervention integrated various rehabilitation measures than the classical SWC practices, watershed level comparative study is essential. Thus, exploring watershed level impacts of integrated land management practices on soil physicochemical characteristics, water discharge capacity, species richness, and carbon stock is of paramount importance. Pieces of evidence generated on these basic ecosystem indicators (Wubet et al., 2013) would support policymakers for their decision to upscale best practices and improve the rehabilitation measures in the central highlands of Ethiopia as well as in various parts of the country.

1.4 Research objectives, hypothesis and research questions

1.4.1 General and specific objectives

This research is aimed at exploring the changes in indicators of ecosystem services associated with integrated land management practices and generating information and data from agricultural landscapes. The specific objectives are to:

- evaluate changes in selected soil physicochemical properties of the treated site taking the neighboring control site as a base
- quantify the change in water discharge due to integrated land management practices
- assess plant species richness in the watershed and compute changes due to integrated land management practices
- determine the plant biomass production and carbon stock of the watershed associated with integrated land management practices

1.4.2 Research hypothesis

We hypothesis that the treated and the adjacent control sites had at least similar conditions before the introduction of integrated land management practices in the treated site. Yet, it is common that the community and extension experts decide implementation of conservation practices at highly degraded landscapes that prove low biomass productivity. Thus, the treated site might be highly degraded compared to the untreated site in the beginning; if not, they might have at least similar conditions. Then, the following null hypotheses were tested in the course of the study.

- Integrated land management practices do not improve the soil physical and chemical properties in Geda watershed.
- There is no difference in stream flow between the treated and untreated sites.
- Integrated land management practices do not improve plant species richness in Geda watershed.
- There is no difference in plant biomass production and carbon stock between the treated and untreated sites in Geda Watershed.

1.4.3 Research questions

To achieve the stated objectives and to test the research hypothesis, the following questions were designed and addressed.

- What are the effects of integrated land management practices in soil physical and chemical characteristics in different landscape position, land-use type and soil depth?
- How do integrated land management practices improve water discharges in Geda watershed?
- What is the contribution of integrated land management practices in plant species richness under different landscape position and land-use types?
- How do integrated land management practice influence plant biomass production and carbon stocks under different landscape position and land-use types?

1.5 Structure of the thesis

This thesis is organized in five chapters. The first chapter provides general background information followed by the research problem, justification of the study, research objectives, hypotheses and research questions. The second chapter is a review of relevant literatures that gives existing evidences on the severity of land degradation, rehabilitation efforts and outcomes of rehabilitation works in Ethiopia, and the third chapter is the materials and methods section that begins with a description of the study area and explanations the research methods. Chapter four presents results and discussion of each research objective which are published in or submitted to peer-reviewed scientific journals and manuscripts under preparation. Chapter five provides the conclusions and recommendations of the research.

CHAPTER TWO

2. LITERATURE REVIEW

2.1 Definition and extent of land degradation

Land degradation is defined as a persistent decline in the productivity of land and its ecosystems (Hurni et al., 2010). Land degradation involves a natural process through the process of geological evolution (FAO, 1986) and accelerated by anthropogenic activities mainly due to deforestation, raising crops and livestock, mining and construction (FAO, 1986; Wolka et al., 2011; Gashaw, 2014; Berhanu et al., 2016). Interaction of the natural and social interlocking systems determine resource management situations that further threatens the long-term growth of agricultural productivity, food security, and the quality of life (Amare et al., 2013a; Berhanu et al., 2016). Both natural and anthropogenic causes determine the occurrence and spatial dynamics of land degradation (Kumar and Das, 2014).

Globally, nearly five billion ha (about 43% of Earth's vegetated surface) has been degraded due to deterioration of dryland vegetation and tropical moist forests; land degradation in the tropics amounts to 2.1 billion ha (Gebretsadik, 2013; Gashaw, 2015). Currently, the rate of global land degradation is 10 to 12 million ha year–1 (Thomas et al., 2018). The problem of land degradation has been undermining agricultural development and hinders environmental sustainability in many countries including Ethiopia (Berhanu et al., 2016; Ademe et al., 2017).

2.2 Severity of land degradation in Ethiopia

Land degradation mainly soil erosion by water is a primary environmental problem of Ethiopia (FAO, 1986; Atnafe et al., 2015; Berhanu et al., 2016; Dabi et al., 2017). Its impact is very serious in Ethiopia as compared to other countries in the world (Gashaw et al., 2015; Berhanu et al., 2016; Dabi et al., 2017). The Tigray and Amhara regions are the most seriously affected parts of the country (FDRE, 2015).

In Ethiopia, 27 million ha was extensively eroded, 14 million ha was seriously eroded and over two million ha became beyond reclamation in mid-1980s (FAO, 1986; Holden et al., 2005; Haile et al., 2006; GIZ, 2015; Gashaw, 2015). The total land degraded in Ethiopia between 1981 and 2008 is estimated to be 297,000 km2 (Dabi et al., 2017). Soil erosion in Ethiopia ranges 3.4 to 84.5 tons ha-1year-1 with a mean value of 42 tons ha-1year-1 (Haile et al., 2006; Hurni et al, 2015; Gashaw, 2015; Dabi et al., 2017). Still, in the highlands of Ethiopia, estimates reach up to 130-300 tons ha-1 year-1on croplands depending on the landscape and vegetation cover conditions (FAO, 1986; GIZ, 2015; Hurni et al, 2015; Gashaw, 2015). Gashaw et al. (2017) also reported annual soil loss of 237 tons ha−1 in the steep landscape areas of Geleda watershed.

In Ethiopia, soil erosion by water is severe on agricultural lands for it is concentrated on steep landscapes > 20% (FAO, 1986) and predominantly rain-fed (Hurni et al., 2015; Tiki et al., 2016). The northeastern parts of the country have been affected by soil erosion due to long time agricultural practices; however, currently, the western parts of the highlands are experiencing highest soil erosion rates (Mulugeta and Karl, 2010; Hurni et al., 2015). Agriculture is under continuous threat of soil erosion and nutrient depletion in the Ethiopia (FAO, 1986; Mulugeta and Karl, 2010; Adimassu et al., 2012; Gashaw et al., 2014). This seriously affects the country’s economy and livelihood as a whole, since the country is highly dependent on the agricultural sector (FAO, 1986).

Although the government of Ethiopia took land degradation as a serious case and invested a lot of efforts in land rehabilitation and reclamation initiatives (Gebremichael et al., 2005; Amare et al., 2013a), it is still at a severe stage and it becomes the root cause of poverty with considerable negative impacts on the national economy (Mulugeta and Karl, 2010; FDRE, 2015; Gashaw, 2015).

2.3 Causes of land degradation in Ethiopia

The causes of land degradation in the country are multiple and interacting forces (Gashaw, 2015) attributed to a combination of biophysical, social, economic and political factors (Haile et al., 2006; Gashaw, 2015; FDRE, 2015). Rapid population growth, cultivation on steep landscapes, clearing of vegetation and overgrazing are the main factors accelerating soil erosion (Ayalew, 2011; Masebo et al., 2014; Gashaw, 2014, 2015; Atnafe et al ., 2015). Accelerated soil erosion by water depends on rainfall erosivity; soil erodibility; landscape and land-use types. Land-use is the most important factor of soil erosion followed by landscape, soil erodibility and rainfall erosivity (FAO, 1986). Further, various authors (Haile et al., 2006; Gashaw, 2014; Atnafe et al ., 2015; Shiferaw and Holden, 2016) considered subsistence agriculture, poverty, and illiteracy as important causes of land degradation. The use of wood biomass for fuel and encroachment of forests are also causes of land degradation (Bojo and Cassels, 1995; Ayalew, 2011; Gashaw, 2015; FDRE, 2015). Further, weak extension services and weak management of public lands are reported as causes for land degradation in Ethiopia (Mulugeta and Karl, 2010; Ayalew, 2011; Masebo et al., 2014).

Thus the causes of land degradation are multiple and intermingling (Gashaw, 2015); but grouped in to direct and indirect (Bojo and Cassels, 1995; Gashaw, 2015). The direct causes are forest clearance, poor cultivation practices, burning of dung, removal of crop residues, low vegetative cover of croplands, unbalanced crop and livestock production, and extensive use of charcoal. Whereas, the indirect causes include poverty, tenure insecurity, economic policies, and population pressure. Thus, the above causes of land degradation can be summarized into proximate causes, underlying causes and policy implications (Fig. 1).

Abbildung in dieser Leseprobe nicht enthalten

Figure 1. Schematic representation of the links of soil erosion causes (Source: Fitsum et al., 1999)

Population pressure, both human and livestock, is exceptionally severe in the highlands of Ethiopia (Keesstra et al, 2018). The highlands cover about half of the country’s territory, over 95% of regularly cropped land, support about 88% of the population, two-thirds of the livestock, and over 90% of the national economic activity (FAO, 1986) The average population concentrations of the area was reported as 144 person km-2 (Sonneveld and Keyzer, 2003) but in some areas population density exceeds 300 person km-2 (Ebabu et al., 2017). The livestock density in the highlands is also high (160 tropical livestock units (TLUs) km-2) as compared to the recommended densities of 19 - 42 TLU km-2 (Bojo and Cassels, 1995; Sonneveld and Keyzer, 2003).

2.4 Impacts of land degradation on the environment and landscape productivity

2.4.1 Impact on the environment

Ethiopia loses 30,000 ha of land year-1, 1 billion tons of topsoil year-1 and significant nutrient depletions due to water erosion (Berry, 2003). Haileslassie et al., (2006) estimated the national level nutrient depletion rate due to soil erosion as 122kg ha-1 year-1N 13 kg ha-1 year-1 P and 82 kg ha-1 year-1 K. In addition, According to Wolka et al. (2011), soil erosion has a sorting action by its nature. It removes large proportions of the clay and humus from the soil, leaving behind the less productive coarse sand, gravel, and even stones in some cases. The removal of this organic matter affects soil quality such as texture, structure, nutrient availability and biological activity and makes the soil more susceptible to further erosion (Yoseph et al., 2017) and less productive.

Generally, soil erosion in Ethiopia has brought a continuous decline in land productivity (Adimassu et al., 2012; Masebo et al., 2014) and affected the economy of the country (Ayalew, 2011; Dagnew et al., 2015). Thus, careful management of watersheds is a core element of good agricultural and forest restoration to minimize land degradation, stabilize water/stream flows, improve groundwater recharge and reduce sediment load (GIZ, 2015).

2.4.2 Impact on landscape productivity

Soil erosion affects half of the agricultural land in the country (Dagnew et al., 2015) and incurs production loses due to physical, chemical and biological deterioration of the soil (Elias, 2002; Ademe et al., 2017). Soil erosion removes chemically active parts of the soil such as organic matter and clay fractions that make the soil more productive. Furthermore, it deteriorates soil structure and moisture-holding capacity through lowering soil depth, increasing bulk density, forming soil crust, and reducing water infiltration (Wubet et al., 2013).

Soil degradation is the most serious limiting factor for crop production in Ethiopia. It includes soil erosion, chemical degradation, physical and biological deterioration of the soil that negatively affect crop production (Elias, 2002). Ayalew (2011) reported a loss of 17% of the potential annual agricultural GDP due to physical and biological soil degradation. Furthermore, Sonneveld and Keyzer (2003) predicted the reduction of land potential production by 30% in 2030, unless management interventions are not implemented. This reduces the annual value-added per capita in the agricultural sector by USD 162 in 2030. i.e. below the poverty line as defined by the World Bank (income of less than one USD day-1). Besides, food availability per capita drops from 1971 to 686 Kcal day-1. This is far below the World Health Organization threshold of the minimum: 2600 Kcal day-1 for adults and 1600 Kcal day-1 for children). Irreversible changes in soil productivity due to soil erosion coupled with population pressure is leading to land scarcity (Ebabu et al., 2017), conflict, violence, drought, food scarcity and insecurity in Ethiopia (Adimassu et al., 2017).

2.5 Efforts to rehabilitate degraded lands in Ethiopia

2.5.1 Government initiatives

In order to tackle the outbreak of the 1973/74 drought and halt land degradation and its associated impacts, the government of Ethiopia launched massive rehabilitation programs starting from the mid-1970s (FAO, 1986; USAID, 2000; Haile et al ., 2006; Ebabu et al., 2017; Adimassu et al., 2017). As a first initiative of SWC investment, the government established SWC division within the Ministry of Agriculture (MoA) (Amare et al., 2013a; Adimassu et al., 2017). In the beginning, SWC investment started in drought-prone areas using a food-for-work payment mainly funded by the World Bank, World Food Program (WFP) and the Food and Agricultural Organization (IFPRI, 2009; Atnafe et al ., 2015; Adimassu et al., 2017).

SWC practices are categorized into physical, biological and agronomic; but sometimes there is an overlap in these categories. For example, grass strip is a biological SWC practice, by definition, but it has also the role of physical SWC practices (Adimassu et al., 2017). Nevertheless, widely implemented SWC practices are mostly physical structures which include stone bunds, soil bunds (level/graded), fanya juu (level/graded) (IFPRI, 2009; Adimassu et al., 2017). Biological SWC practices include maintaining natural vegetation and tree plantation in area closures, plantation of valley bottoms, and stabilization of physical structures through vegetation such as grass strips, vetiver grass, elephant grass, and so on (Berhanu et al., 2016). Agronomic SWC are also among the intervention options practiced in Ethiopia. These include minimum tillage, tied-ridging, application of compost, farmyard manure, mulching and so on (Adimassu et al., 2017). However, physical SWC practices are the widely practiced conservation options to curb soil erosion in Ethiopia (Wolka et al., 2011).

According to USAID (2000), between 1976 and 1990, the conservation structures in the country were 71,000 ha of soil and stone bunds, 233,000 ha of hillside terraces for afforestation, 12,000 km of check dams in gullied lands, 390,000 ha of closed areas for natural regeneration, 448,000 ha of land planted with different tree species, and 526,425 ha of bench terrace interventions. Berry (2003) reported that between 1976 and 1985, Ethiopia constructed some 600,000 kilometers of soil and stone bunds on cultivated land, about half a million kilometers of hillside terraces, 500 million tree seedlings were planted, and 80,000 ha were closed off for natural regeneration. Furthermore, Wubet et al. (2013) reported the construction of 800,000 km of soil and stone bunds on cultivated lands; 600,000 km of hillside terraces and 80,000 hectares were closed for regeneration and for afforestation on steep landscapes between 1976 and 1988, which was funded by food-for-work (FFW) programs of WFP.

Furthermore, various campaigns have been carried out since the early 1980s to build terraces on farmlands and steep areas with an emphasis on structural technologies over the vegetative measures (GIZ, 2015; Gashaw, 2015). The dominant SWC structures implemented in Tigray were soil and stone bunds with soil bunds covering 63%; in Amhara, stone bunds and waterways, in Oromia, soil bunds and waterways, in Benishangul Gumuz, waterways accounting 55%; and in SNNPR, tree plantations (IFPRI, 2009).

2.5.2 Ineffectiveness of past interventions

Despite tremendous efforts made to expand SWC practices in the earlier times, achievements did not match the vast needs of the country and weaknesses and in effectiveness of the SWC were more prominent. For example, conservation measures had covered only 1% of the highlands during mid-1980s (Bojo and Cassels, 1995). Many of the physical installations were based on simplistic rules of thumb making them less adapted to local conditions. Maintenance was lacking, survival rate of tree seedlings has been low, destruction of bunds and trees for short term benefit were practiced during political instability (Bojo and Cassels, 1995). Evaluation of success rates of the earlier conservation measures in 1990 revealed that only 30% of soil bunds, 25% of the stone bunds, 60% of hillside terraces, 22% planted trees, and 7% of the reserve areas survived (USAID, 2000).

According to Kassie et al. (2011) and Atnafe et al . (2015) adoptions of SWC technologies by smallholder farmers were low. The identified reasons for the low success rates were: a) lack of integrating biophysical measures and indigenous knowledge, b) due to the negative impacts of incentives (e.g. food for work), c) inappropriate perception of farmers, d) socio-economic reasons, e) lack of adequate design, f) inappropriate land-use, g) lack of strong maintenance of structures, h) lack of adequate monitoring and evaluation and i) lack of farmers’ participation in decision making at all stages j) limitations of technologies in benefiting smallholder farmers (Zegeye et al., 2010; Tiki et al., 2016; Shiferaw and Holdern, 2016; Tiki et al., 2016; Dabi et al., 2017). Furthermore, the study of Atnafe et al. (2015) in Goromti watershed, west Ethiopia revealed as landscape of the area, tenure status, contacts with extension workers, training situations, age and family size determined adoption of SWC technologies. In addition, lack of understanding of nature’s functioning and processes by conservation experts determine the effectiveness of conservation measures (Keesstra et al., 2018)

2.5.3 Integrated watershed based approach

Degraded ecosystem is assisted to recover to the best possible level through integrating various compatible structural, biological and cultural methods (GIZ, 2015). Thus, starting the 1980s, formal and planned watershed development approaches by which local people play sufficient roles was introduced with a primary purpose of natural resource conservation and enhancing agricultural productivity through which livelihood improvement can be achieved (Haregeweyn et al., 2012; Tiki et al., 2016; Dabi et al., 2017). In the beginning, the approach puts a planning unit of developing large watersheds of 30-40,000ha (GIZ, 2015; Gashaw, 2015). However, large scale watershed developments were not satisfactory due to unmanageable planning units, lack of effective community participation, limitations in responsibilities for assets created (GIZ, 2015). Consequently, pilot watershed planning approaches based on smaller units and on a bottom-up basis were introduced in 1988-91 (GIZ, 2015).

Following the introduction of watershed development approach, MoA and United Nations World Food Program (WFP) staff developed participatory and community-based watershed planning guidelines called Local-Level Participatory Planning Approach (LLPPA) with a practical focus for development agents. The emphasis was on the integrated natural-resource management (NRM) interventions, productivity-intensification measures and small-scale community infrastructure such as water ponds and feeder roads (GIZ, 2015; Dabi et al., 2017). Watershed based natural resource management approaches across a range of hierarchies, from small catchments to larger streams better address upstream and downstream effects and interactions that exist among components of the natural system (USAID, 2000; Dabi et al., 2017).

Various programs such as Participatory Land-use Planning (PLUP), Sustainable Utilization of Natural Resources for Improved Food Security (SUN), Sustainable Land Management (SLM) incorporated best experiences of natural resource management gained from watershed approaches (GIZ, 2015). In Ethiopia, Sustainable Land Management Project (SLMP) was launched in 2008 and the first phase of the project finalized in September 2013 (FDRE, 2013; Adimassu, 2017). SLM practices include construction of soil and or stone bunds, terraces, trenches, cut of drains, drainage canals, check dams, fanya juu, planting of shrub or tree species, establishing area exclosure, combinations of structural and vegetative measures, and reduction of household livestock numbers (Ebabu et al., 2019). SLM practices are designed to increase agricultural productivity, improve ecosystem functions and enhance resilience to adverse environmental impacts (Liniger et al., 2011; Ademe et al., 2017).

In Ethiopia, SLM project has successfully introduced land management practices and rehabilitated thousands of hectares of degraded lands using physical and biological measures (FDRE, 2013). Currently the second phase of the project has been under implementation since September 2013 (FDRE, 2013). The focus has also included livelihood improving options to increase economic gains and promote adoption.

The SLM project presented different components and sub-components of natural resource restoration activities such as (FDRE, 2013):

1) Integrated Watershed and Landscape Management: intended to support scaling up and adoption of appropriate sustainable land and water management technologies and practices by small-holder farmers and communities in selected watersheds. This was planned to be achieved through different sub components:

a) Sustainable Natural Resource Management in Public Lands focuses on Afforestation and Reforestation of degraded communal land (hillside communal land treatment and management including woodlot establishment, gully rehabilitation using biophysical measures and seedling production); crop production to increase productivity and carbon sequestration (treatment of farmland < 30% landscape with suitable bio-physical measures, > 30% with suitable bio-physical measures, applying conservation agriculture, agro-forestry, and so on), improving livestock production/productivity and reducing carbon emission through promoting fodder or forage production, improve breed for stock reduction, improved poultry breed, improved beekeeping activities, artificial insemination (AI) service and cattle crush. Climate Resilience Building and Increasing Water Availability focused on supporting small scale irrigation, potable water supply - hand dug well and spring development, renewable energy potential for the rural setting.
b) Homestead and Farmland Development focuses on construction of water harvesting structures with water efficient irrigation methods, homestead development by promoting high value crops and multi- purpose fruit trees and forage tree planting, livestock improvement (e.g. small ruminant fattening, promotion of beekeeping and honey production, etc.), promoting bio-fuel/biomass, biogas energy, promotion of fuel saving and efficient technologies, and feeder road construction.

2) Rural Land Administration, Certification and Land-use: to enhance the tenure security of smallholder farmers in order to increase their motivation to adopt sustainable land management practices on communal and individual lands.

2.6 Impacts of conservation measures in degraded land rehabilitation

A number of researches were done at various times on the effectiveness and impacts of SWC practices on the environment, land productivity and livelihood improvements. Adimassu et al. (2017) reviewed and synthesized the impacts of SWC on crop yield, runoff, and soil and nutrient loses in Ethiopia. According to their report, most physical SWC practices such as soil bunds and stone bunds were very effective in reducing runoff, soil erosion and nutrient depletion.

However, SWC practices showed site-specific impacts on crop yield. For instance, soil and stone bunds increased crop yield up to 10% in Tigray, while this reduced crop yields up to 7% in other parts of the country mainly due to the reduction of the effective cultivable area (Adimassu et al., 2017; Shiferaw and Holden, 2016). Conservation measures occupy considerable space; for example, grass strips, bench terraces and fanya juu occupy 1-15%, 5-42% and 8-40% of cultivable lands, respectively (Dabi et al., 2017). Adimassu et al. (2017) suggested that the reduced areas by SWC structures can be compensated by growing high value trees/fodders for livestock feed. They further reported as agronomic SWC practices such as compost, farmyard manure, tied-ridging, minimum tillage and mulching are best alternatives to increase crop yields while conserving soil and water.

In low rainfall areas of Amhara and Tigray, Ethiopia, higher crop yields of an average 42 % and 23% respectively, were obtained from plots with stone terraces (Kassie et al ., 2011). This is due to the moisture conserving property of the stone bunds in drier areas (Wolka et al., 2011). According to most of the research findings, introduction of stone bunds on cropland reduces soil loss (annual average of 61–68%) and the soil is deposited behind the bunds (Nyssen et al., 2007) having improved physicochemical characteristics (Bulk density, SOM, TN, pH, K+, available P, SOC, clay and CEC (Mulugeta and Karl, 2010; Ademe et al., 2017). For instance, Wolka et al. (2011) found higher values of clay and silt fractions on terraced plots than on non-terraced ones. Agroforestry based SWC showed higher organic matter content at treated plots in Tembaro district, southern Ethiopia (Masebo et al., 2014).

SWC measures also play a significant role in reducing runoff and on-site sediment deposition. Sultan et al. (2017) found a runoff reduction of 49% through the combination of soil bunds with vegetation on cultivated lands while the use of trenches across the landscape on non-cultivated plots reduced runoff by 65%. According to Adimassu et al. (2012), soil bunds reduced average annual runoff by 28% and average annual soil loss by 47% (39% in Tigray and 50% in Anjeni). The average annual runoff reduction due to soil bunds at Galessa Watershed, was 28% (Adimassu et al., 2012) and at Enabered watershed, was 27 % (Haregeweyn et al., 2012). Taye, et al. (2013) and Masebo et al. (2014) also generally agree as SWC practices are effective in reducing runoff and increasing soil moisture content and base flow.

Furthermore, integrated watershed management at Enabered watershed (Tigray) decreased runoff by 27 % and soil loss due to sheet and rill erosion by 89 % and reclaimed gully channels and converted them for agricultural purposes (Haregeweyn et al., 2012). Integration of terraces and infiltration furrows reduced runoffs and sediment concentration at Debre Mewi watershed (Dagnew et al., 2015). The study of Wubet et al. (2013) at Anjeni watershed, northwest Ethiopia, also reported that SWC structures improved soil quality and land suitability to crop production. According to their study, lands that were moderately (S2) and marginally (S3) suitable for major crops of the watershed such as tef, barley, wheat, and maize in 1984 and 1997 were improved to highly suitable (S1) for wheat and tef, and a large proportion of the remaining area was changed to moderately suitable class (S2) for barley and maize in 2010.

Subhatu et al., (2017) studied on-site sediment deposition and net soil loss in terraced crop lands at Minchet catchment in the sub-humid Ethiopian highlands and showed that narrow terrace spacing (<13m) had more sediment deposition than wider spaced terraces due to off-site removal of sediments through waterways in wider catchments. Furthermore, they reported that fields with higher landscapes produced more sediment yield than gentle landscapes; and found out average soil loss by water from 31 to 37 t ha-1year-1. However, if terraces are constructed on crop lands, 54–74% of soil loss was deposited there.

In addition, catchment management has resulted in higher infiltration rate and a reduction of direct runoff volume by 81% (Nyssen et al., 2010). The yearly rise in water table after the onset of the rains (WT) relative to the water surplus (WS) over the same period increased between 2002-2003 (WT/WS = 3.4) and in 2006 (WT/WS >11.1).

An area exclosure also showed a significant positive impact on reducing soil loses and improving soil fertility. Descheemaeker et al. (2006) reported a mean sediment deposition of 26 to 123 Mg ha-1 year-1 and development of Phaeozems (dark soils rich in organic matter on area exclosure), Further, they noted that area exclosure increased regeneration of natural vegetation; stabilized the land and reinstalled microclimate, reduced runoff, sheet and rill erosion. The exclosure showed 47% reduction of soil erosion compared to the grazing areas (Mekuria et al. (2009).

An area enclosures of about 20-30 years enhanced vegetation regrowth, increased biomass production that covered up to 15 % of the land in several districts of Tigray region (Nyssen et al., 2008). Increased vegetation cover decrease downstream sediment deposition and flooding, provides ecosystem services such as growth of grass and trees, increased firewood production, improved wildlife habitat and enhanced biodiversity which further contribute to climate regulation, drought mitigation, and carbon sequestration (Nyssen et al., 2008; Hishe et al. (2017a). Exclosure improve environmental resources on degraded and generally open access lands so that natural regeneration of plant species is conditioned (Mekuria et al., 2009).

A recent meta-analysis by Abera et al. (2019) revealed that integrated conservation measures resulted higher positive impacts compared to a single practice. According to Abera et al. (2019), combination of bunds and biological interventions and conservation agriculture showed 170% and 18% mean effects on agricultural productivity, respectively. Bunds supported by biological measures increased soil organic carbon by 139% while exclosure increased by 90%. However, monolithic measures of biological intervention and terracing (fanya juu) revealed negative effects on agricultural productivity. Thus, low adoption rate and weaknesses of the conservation measures in the earlier times could be among others-due to lack of integrating different compatible measures. Therefore, according to Adimassu et al. (2017) and Abera et al. (2019), it is critically important to integrate different practices appropriately in order to rehabilitate degraded landscapes and enhance the provisioning and regulating ecosystem services.

CHAPTER THREE

3. MATERIALS AND METHODS

3.1 Description of the study area

The study was conducted in Geda watershed, located in the North Shewa Zone of Amhara National Regional State in central Ethiopia. The watershed is at the upper part of the Blue Nile basin in the eastern escarpment of the Ethiopian highlands. It is situated at 165 km north of Addis Ababa (the capital of Ethiopia), on the way from Addis Ababa to Dessie, and 35 km to the north east of Debre Berhan town (the capital of North Shewa Zone). Specifically, the study site is located between 39040'30" to 39041'30" East longitude and 9048'30" to 9049'30" North latitude in the Blue Nile basin (Fig. 2).

Abbildung in dieser Leseprobe nicht enthalten

Figure 2. Map of the study area in Geda watershed

Geda watershed has a total area of 1056 ha; with more than 66% cultivable, 25% grazing and 6% woodlot (Tamene, 2017). The altitude range of the watershed is between 2700 and 3500 masl. However, the study was conducted in part of the watershed with in the altitude range of 2947 to 3156 masl; covering about 365 ha of land; of which about 202.45 ha was treated by integrated land management technologies and about162.95 ha was under the conventional practice.

The total annual rainfall of the watershed ranges 1225.04 ̶ 2061.3mm. The area receives an average annual rainfall of 1632.42 mm. The annual minimum temperature of the study watershed was within a range of 2.46oC to 8.49oC. The maximum annual temperature ranged from 16.64 to 19.21oC (Fig. 3).

Abbildung in dieser Leseprobe nicht enthalten

Figure 3. Mean annual rainfalls and temperatures in the study area (Data source: DBARC); RF: Rain fall; Max. temp: Maximum temperature; Min. temp.: Minimum temperature

Based on the slope classification of FAO (2006), the dominant topography of the study area ranges from gently sloping (2-5%) covering 34% of the area to strongly sloping (5-10%) covering 34.11% of the area (own data). Geda watershed is characterized by volcanic rocks such as rhyolites, trachites, tuffs and basalts. Major soils of the area are Leptosol on steep landscape, Fluvisol at the Valley bottoms, and Regosol at eroded parts (Ashagrie, 2009; Amare et al., 2013b). Major farming system in Geda watershed is a mixed crop-livestock system. Cultivated land in the watershed covers more than 66%, grazing land 25% and wood lot 6% (Tamene, 2017). Farmers cultivate cereals such as barley and wheat; pulses such as faba bean and field peas; and rarely lentil and linseeds (Ashagrie, 2009). Sheep, poultry and cattle are major livestock types of the watershed (Ellis-Jones et al., 2013).

Recently, the population pressure and continuous cropping with less or no fallow periods is putting less regeneration chance of the landscape (Kuria et al., 2014). The fertile top soils are removed by surface runoff (Ashagrie, 2009). Farmers have limited income and are unable to buy artificial fertilizers to improve the productivity of the land and slow the process of degradation. Furthermore, dungs (natural fertilizers) are used for cooking fuel instead of adding to the soil to improve the soil fertility (Ashagrie, 2009). With this context Geda watershed was designed to be a learning watershed for an integrated land management interventions in the crop-livestock mixed farming system of the central highlands.

The land management intervention covered more than 80 km soil bund with trenches, 71 m3 of gabion check- dams, 730 m3 wooden check dams, 19 percolation pits and planting Tree Lucerne on highly degraded plots (Tamene, 2017). Currently the adjacent untreated site is under intensive crop production growing mainly barley (Hordeum vulgare), wheat (Triticum aestivum), faba bean (Vicia faba), and field pea (Pisum sativum) during the main cropping season (June to September). After the crop is harvested in mid- October, livestock start grazing on stubbles and plot margins. Thus, after harvest, each farmer transports the produce and hay to homesteads as quickly as possible in order to protect from livestock damage in the field when free gazing starts. Then, the area is left for free grazing from November to mid-June until tillage for the next season’s crop planting is taking place. Cattle, sheep, donkeys and horses are the major livestock groups that freely graze for about eight months, leaving hardly any soil cover in the landscape; and the cycle continues each year.

3.2 Study design

Systematic sampling method (Pearson et al., 2005) was employed in the main and the dry seasons of 2018/2019 to gather data from the treated and untreated sites, upper and lower landscape positions and different land-use types at each landscape position (Table 1; Fig. 4). The landscape positions of the sites were classified into upper (3031–3156 masl) and lower (2947–3024 masl). Since there was a village in between the upper and lower landscape positions (3024-3057 masl) in the treated site that live following the conventional land-use practices, we skip collecting data from this landscape range. The land-uses in the treated site were crop land and grazing land in the upper landscape and crop land, grazing land and Tree Lucerne plantation in the lower landscape position. Land-use types in the untreated site were crop land and grazing land in both landscape positions. Similar land-uses and landscape positions were purposely selected to collect representative data for all parameters. Since Tree Lucerne plot is found only in the treated site, data from this plot was compared across all land-uses and the grazing land in the untreated site when necessary.

The main land management measures practiced in the crop lands were soil/stone bunds supported by Phalaris (mixture of Phalaris acquatica and Phalaris arundinacea) and Tree Lucerne (Chamaecytisus palmensis); whereas, soil bunds and water collecting ditches were the dominant land management measures in the grazing lands, and the Tree Lucerne plantation has water collecting ditches and soil bunds. Percolation pits in different location and check-dams at river banks were also done. Free grazing was prohibited at both landscape positions and land-uses in the treated site. The landscape position, land-use types and major conservation measures are summarized in Table 1 below.

Table 1. Characterization of the treated and untreated sites

Abbildung in dieser Leseprobe nicht enthalten

* There is a village practicing the conventional land-use practices between 3024-3057 masl in the treated site. Thus, we didn’t collect data from this elevation range.

Abbildung in dieser Leseprobe nicht enthalten

Figure 4. Schematic representation of the study design

* Two soil depths (0 ̶ 15cm and 15 ̶ 30 for analysis of soil properties and 0 ̶ 20 cm and 20 ̶ 40 cm for moisture analysis) were used at each land-use type; ILM = Integrated Land Management.

3.3 Determination of selected soil physicochemical properties

3.3.1. Soil sampling

Judgment sampling method (USEPA, 2002; Wolka et al., 2011; Mekuria, 2018) was followed to locate a representative sampling plot for both treated and untreated sites, landscape positions and land–use types at two different soil depths (0–15 cm and 15–30 cm). Although the introduced land-use practices are expected to change the top soil at this younger age, data collection included 15 ̶ 30 cm depths to compare changes, if any, in the soil physicochemical properties. In order to maintain homogeneity of crop lands, fields planted with cereals such as barley and wheat were purposely selected. Soil samples were collected using an Edelman auger (Tor-Gunnar et al., 1999) following a triangular sampling pattern (Wilke, 2005) (Fig. 5).

Abbildung in dieser Leseprobe nicht enthalten

Figure 5. Triangular soil sampling method where sampling points in the parallel rows are staggered and form a triangular grid (Source: Wilke, 2005)

Due to high cost of laboratory analysis we limited the samples into two comparable land-uses in the treated and untreated sites. Thus, a total of 48 samples were collected (two sites × two landscape positions × two land-uses × two depths × three replications). Samples were collected at five sampling points per plot and mixed to make a composite sample (Paetz and Wilke, 2005; Wolka et al., 2011). After thoroughly mixing point samples, about 1 kg of composite sample was packed in a plastic bag and taken to the laboratory for analysis. For determination of bulk density, undisturbed soil samples were taken using core sampler (Masebo et al., 2014) having 5 cm height and 5 cm diameter (Volume = 98.125 cm3).

3.3.2 Soil lab analysis

The collected soil samples were analyzed at the laboratory of Debre Berhan Agricultural Research Center. In the laboratory, the samples were air dried, crushed and sieved by 2 mm mesh sieve and passed through 0.5 mm diameter sieve to prepare for the following analysis. Particle size analysis (% sand, clay, silt) was carried out using the hydrometer procedure (van Reeuwijk, 2002). The soil reaction (pH) was measured using pH meter in a 1:2.5 soil/water suspension (Torn–Gunnar et al., 1999; Descheemaeker et al., 2006). Exchangeable cation (Na+ and K+) were analyzed by adding 1 M ammonium acetate solution at pH 7 (Rowell, 1994; Haldar and Sakar, 2005). Available P was determined following Olsen’s extraction method (Olsen et al., 1954). Total nitrogen (TN) was determined following Kjeldahl procedure as described in Wilke (2005). Organic carbon (OC) was determined based on the Walkley–Black (Torn-Gunnar et al., 1999) rapid titration method. Bulk density of the soils was determined by dividing the mass of an oven dried soil at 105°C for 24 hours by the volume of the core sampler (Wilke, 2005; Masebo et al., 2014; Tiki et al., 2016; Alemayehu and Fisseha, 2019).

[...]