source of the Blue Nile. However, in 1770 James Bruce claimed to be the discoverer ...... AS, Langan S, MacAlister C, Langendoen E. 2015. Improving efficacy of.
Bahir Dar University Bahir Dar Institute of Technology Faculty of Civil and Water Resources Engineering
Human impact on Hydro-geomorphology of Gumara River upper Blue Nile basin, Ethiopia By Mengiste Abate Meshesha
A dissertation presented to the School of Research and Graduate Studies of Bahir Dar Institute of Technology, Bahir Dar University in partial fulfillment of the requirements for the Degree of Doctor of Philosophy in Integrated Water Management in the Faculty of Civil and Water Resources Engineering Promotors : Prof. Dr. Tammo S Steenhuis, Cornell University, USA Prof. Dr. Jan Nyssen, Ghent University, Belgium Dr. Michael Mehari Moges, Bahir Dar University, Ethiopia
Dec 2016
DECLARATION
I, the undersigned, certify that research work titled “Human impact on Hydrogeomorphology of Gumara River upper Blue Nile basin, Ethiopia” is my own work. The work has not been presented elsewhere for assessment other than for peer reviewed publication. Where material has been used from other sources, it has been properly acknowledged. Name of the student: Mengiste Abate Meshesha Signature: _______________________ Date of submission: Dec. 2016 Place: Bahir Dar
This thesis has been submitted for examination with my approval as a main promoter. Advisor Name: Prof. Dr. Tammo S Steenhuis Signature _______________________________
©2016 Mengiste Abate Meshesha ALL RIGHTS RESERVED
Bahir Dar Institute of Technology-Bahir Dar University School of research and graduate studies Faculty of Civil and Water Resources Engineering
Human impact on Hydro-geomorphology of Gumara River upper Blue Nile basin, Ethiopia By Mengiste Abate Meshesha
Student Dissertation Advisory Committee approval
Prof. Dr. Tammo S Steenhuis Committee Chair
Prof. Dr. Jan Nyssen Committee member
Dr.Michael Mehari Mogse Committee member
Dr. Temesgen Wondimu Faculty Dean
Dr. Atikilt Abebe Dean of Graduate studies of BiT
________________ Signature
________________ Signature
________________ Signature
________________ Signature
________________ Signature
_____________ Date
_____________ Date
_____________ Date
_____________ Date
_____________ Date
Dec. 2016 Bahir Dar, Ethiopia
Biographical Sketch Mengiste Abate Meshesha was born in 1976 and grew up in a rural area, Jigna Kebel in Dera Woreda, South Gondar Administrative Zone, Amhara National Regional State, Ethiopia. He completed his elementary and high schools’ education at Jigna and Woreta respectively. After he completed high school, he joined Arba Minch Water Technology Institute in 1996 and graduated with Bachelor’s degree in Hydraulic Engineering. After graduation, he worked in the Bureau of Public Works and Urban Development, Amhara National Regional State, Ethiopia in the position of hydraulic engineer. After working three years he was awarded a Nuffic scholarship to study Water Science and Engineering, specialization; Hydraulic Engineering and River Basin Development at UNESCO-IHE in The Netherlands and earned a master’s degree in 2005. After his M.Sc., he joined Bureau of Water Resources Development, Amhara National Regional State, Ethiopia and worked for two years. In 2007, he joined Bahir Dar University in the rank of lecturer. In addition to giving lectures, he served the university in different administrative positions. He started his Ph.D. study in 2013 at the Faculty of Civil and Water Resources Engineering of Bahir Dar University.
Acknowledgments I am deeply indebted to my advisors Prof. Dr. Tammo S Steenhuis at Cornell University, USA, Prof. Dr. Jan Nyssen at Ghent University, Belgium and Dr. Michael Mehari Moges at Bahir Dar University, Ethiopia. Their combined guidance, advice and constructive comments largely enhanced the interpretation of the research findings. Prof. Tammo and Prof. Jan: physical distance was not a barrier for both of you and words couldn’t express your dedication, unreserved support, guidance, encouragement, supervision and scientific advice that have been decisive for the successful completion of this dissertation. In my journey of this PhD field work, I met many local people and got lots of qualitative and quantitative information. Listing all of them would take much space: the farmers in the Gumara delta and in the floodplain and those engaged in sand mining activities are particularly mentioned and highly acknowledged. I also benefited from the support and inspiration of other many people in and outside Bahir Dar University in one way or another and particularly I express my gratitude to Dr. Seifu Admassu, Dr. Eddy J Langendoen, Dr. Atikilt Abebe, Dr. Mulugeta Azeze, Dr. Enyew Adgo, Dn. Molla Feten, and Worku Yazie. The laboratory and fieldwork activities would not have been possible without the support I got from Yimer Degu, Abebech Genetu, Tinsae Abate, Nakachew Assefa, and Muluager Bewket. I am also very grateful to the hydrological directorate of the Ministry of Water, Irrigation and Energy for the supply of information on river cross-sections, streamflow and lake level data. My special gratitude goes to Mr. Gedamu who gave me unique information about the automatic water level recorders installed at Gumara and Abay rivers. This work was funded by Bahir Dar Institute of Technology, Bahir Dar University, Abay basin authority through Tana sub basin organization (Ethiopia), ENTRO, Wase-Tana project (VLIR-UOS) (Belgium) and USAID HED program through the Ethiopian Institute of Water Resources; their support is gratefully acknowledged. The laboratory space and equipment provided by the Faculty of Civil and Water Resources Engineering and
i
Amhara National Regional State Bureau of Public Works and Urban Development is highly acknowledged. My gratitude extended to all academic and administrative staff of Bahir Dar Institute of Technology: staff of Faculty of Civil and Water Resources Engineering and BiT Finance team are particularly mentioned and gratefully acknowledged. There are other energetic colleagues in the integrated water management Ph.D. program, whom should take many thanks: Dr. Temesgen Enku mainly mentioned. Special thank also goes to my parents-in-law: Priest Mandefro Abate (Abuhay) and W/ro Asres Sharew (Emuhay). Their love and prayer that keeps me healthy in all aspects: long live to Aba and Ema. My stay in Addis would have been challenged if Dr. Fenta Mandefro, W/ro Fetlework Bizualem, Dn. Getachew Mandefro (Mamush) and Berihun Asnakew (Beya), had not been there. Your encouragement, inspiration and room provision, that gave me stamina for my work, will not be forgotten. You did great jobs and I engraved on my heart as a legend. Words couldn’t express the endless love, care, concern, and prayer that my father Abate Meshesha and my mother Fenta Alemu have done. I thank for everything you have done for me starting from my childhood to till now and it is always my prayer to have long live. Furthermore, brothers and sisters continuously encouraged me: many thanks. Still there is other crew whom love, care, encouragement, moral support and prayer during difficult moments have always been with me. Most importantly, I would like to forward the leading and loving thanks to my helpful wife Gedam Mandefro for her never ending love, concern, support, encouragement, managing and taking care of the kids with Lemlem. Special thanks go to my lovely kids: Samuael, Tibebe, and Mebatsion. They are always the source of my happiness and pain killer at times of stress. Above all, I thank the Almighty God for his Grace and mercy upon my family and me.
ii
Human impact on Hydro-geomorphology of Gumara River upper Blue Nile basin, Ethiopia By Mengiste Abate Meshesha Bahir Dar University The physical, biological and societal systems are interrelated each other. The societal system over the other systems have brought complex changes. The increasing population density that leads to human-induced land use change, alter fluxes of water and sediment and in turn produces complex landscape changes in physical structure, hydraulics and hydro-geomorphology of river systems. The landscape of upper Blue Nile Basin has been changed due to anthropogenic effects and climate change. Few studies in the upper Blue Nile basin have detailed the response of the rivers to these effects. However, in the international literature this topic has been well addressed and can be applied to Ethiopia in order to understand the past, the present channel dynamics and predicting future channel changes. The objective of this study was to investigate and analyze the response of rivers in Ethiopia to sedimentation, erosion and change of flow regime. We choose the Tana sub-basin because previous data was available. Specifically, this study quantifies the hydro-geomorphology of the 38 km of Gumara River starting from the confluence with Kizin River to Lake Tana using historical images, historical river cross sections, hydrologic, hydraulic, and geotechnical data. In addition, field observations and information obtained from the local people and existing structures were used as data sources. The results show that the changes to Gumara River channel morphology in the alluvial plain is largely controlled by the interplay between discharge regime and sediment loads, which are influenced by human factors and the lake levels. Old maps show that before 1957 when shifting cultivation was practiced the Gumara delta did not expand indicating that the sediment concentrations in the Gumara were small. After 1957 the
iii
lower reach of Gumara near its mouth underwent major morphological change and delta expanded on average by 5 ha/yr and increase in height by 3 cm/yr. After 1957 the Gumara River although the planform changes are very slow, the bed has aggregated greatly with in-channel deposition (3 m in 46 years). When the ratio of curvature radius to channel width is 3.5, the Gumara River at its bends is experiencing a maximum bank retreat of 1.95 m/yr and the nearby land is lost on average by 0.075 ha/km/yr. The raise of the river bed level amplifies the flashy nature of the river, and results in more frequent high floods and sediment concentrations. This will further increase the river bed level and in turn reduces the flood carrying capacity of the channel. This dissertation shows that human factors are the major driver for the increasing sediment loads in the changes that occur in Gumara River. Thus, restoration of rivers in the Fogera Plain to decrease flooding should include effective catchment-based management practices to offset the negative aspects of the increasing population density in the uplands.
iv
Table of contents Acknowledgments ....................................................................................................... i List of Figures .......................................................................................................... viii List of Tables .............................................................................................................. xi List of Acronyms ....................................................................................................... xii List of Symbols ........................................................................................................ xiii Chapter 1 1.1
Introduction .......................................................................................... 1
Background................................................................................................... 1
1.2 Existing knowledge on geomorphological changes of rivers in the upper Blue Nile basin........................................................................................................ 2 1.3
Objective of the thesis .................................................................................. 5
1.4
Thesis outline................................................................................................ 5
References ................................................................................................................ 6 Chapter 2 Long-term landscape changes in the Lake Tana basin as evidenced by delta development and floodplain aggradation, Ethiopia .............. 11 Abstract.................................................................................................................... 11 2.1
Introduction .................................................................................................... 12
2.2
Materials and methods ................................................................................... 15
2.2.1
Description of the study area ................................................................... 15
2.2.2
Collection and preparation of data ........................................................... 16
2.2.3
Mapping of the delta ................................................................................ 17
2.2.4
Land use/cover classification ................................................................... 19
2.3
Results ........................................................................................................... 20
2.3.1
Delta development through time .............................................................. 20
2.3.2
Land use changes ................................................................................... 24
2.3.3
People, delta shape and sediment distribution ........................................ 25
2.4
Discussion...................................................................................................... 28
2.4.1
Development and sediment deposited in delta ........................................ 28 v
2.4.2 Comparison of the delta area, river sediment load and anthropogenic impact 30 2.5
Conclusions ................................................................................................... 31
References .............................................................................................................. 32 Chapter 3 Morphological changes of Gumara River channel over 50 years, upper Blue Nile basin, Ethiopia ............................................................................... 39 Abstract.................................................................................................................... 39 3.1
Introduction .................................................................................................... 40
3.2
Materials and methods ................................................................................... 43
3.2.1
Description of the study area ................................................................... 43
3.2.2
Preparation of data .................................................................................. 45
3.2.3
Mapping of channel boundaries for channel planform change analysis .. 47
3.2.4
Historical changes to river cross-sections ............................................... 48
3.2.5
Additional sources of data ....................................................................... 48
3.3
Results ........................................................................................................... 48
3.3.1
Planform changes.................................................................................... 48
3.3.2
Cross-sectional changes ......................................................................... 53
3.3.3
Flood frequency ....................................................................................... 54
3.3.4
Aggradation of natural levee induced by river bank overtopping ............. 56
3.4
Discussion...................................................................................................... 56
3.4.1
Changes to Gumara River planform ........................................................ 56
3.4.2 Increased sediment load and backwater effect lead to channel sedimentation ....................................................................................................... 58 3.4.3
Impacts of irrigation on river banks .......................................................... 61
3.4.4
Effects of dyke constructions ................................................................... 62
3.4.5
Effects of channel change on flooding ..................................................... 64
3.5
Conclusions ................................................................................................... 65
References .............................................................................................................. 66 vi
Chapter 4 Channel bend analysis of Gumara River based on curvature radius and channel width, Upper Blue Nile basin, Ethiopia .............................................. 74 Abstract.................................................................................................................... 74 4.1
Introduction .................................................................................................... 75
4.2
Materials and Methods ................................................................................... 78
4.2.1
Description of the study area ................................................................... 78
4.2.2
Data collection and preparation ............................................................... 79
4.3
Results ........................................................................................................... 81
4.3.1
Long term change in curvature radius and channel width at the bend .... 81
4.3.2
Channel migration ................................................................................... 84
4.3.3
Grain size distribution of the river banks and bed .................................... 86
4.3.4
Trend of shear stress exerted by the flow in the channel ........................ 88
4.4
Discussion...................................................................................................... 89
4.5
Conclusions ................................................................................................... 91
References .............................................................................................................. 92 Chapter 5 5.1
Conclusions and recommendations ................................................ 96
Overall conclusions ........................................................................................ 96
5.1.1
Long term Landscape changes ............................................................... 96
5.1.2
Morphological changes of Gumara River ................................................ 97
5.1.3
Channel bend analysis ............................................................................ 97
5.2
Recommendations ......................................................................................... 98
5.2.1
Human and natural factors which require further research ...................... 98
5.2.2
Human activities which requires river basin management ...................... 99
Appendix .................................................................................................................. 100
vii
List of Figures Figure 2-1: Left: False-color image of the Gumara delta in the year 2006, indicated by four-point red star in the right map, Right: the Gumara catchment with major watersheds. The upper pilot area for land-use change detection is indicated by the red box with the aerial photograph. The inserted map shows Ethiopia’s major drainage basins, with Blue Nile basin highlighted. ......... 16 Figure 2-2: Images of the Lake Tana shore near the Gumara River outlet between 1881 and 2015. At the mouth of the river, a 2.12 km² delta has been created over 49 years (1957-2006) but the area changes follow an apparent non-linear trend. The red line is lakeshore line in 1957 and used as benchmark. Data sources are given in table 2-1. ................................................................. 20 Figure 2-3: Expansion of the delta of the Gumara River in Lake Tana as a function of lake level (LL) for the period from 1957 to 2015. (a) Growth of the delta in time for lake levels (LL) between 2.14 – 2.68 m (orange symbols) and between 2.68– 3.42 m (blue symbols). Trend lines are shown in the same color as the symbols. (b) Delta area as a function of lake level for 2014 (blue close circles) and 2015 (orange close circles); the trend line is shown in blue. (c) Plot of lake level vs delta area for the period from 1973 to 2015. Symbols are the observations listed in Tables 2-2 and S2-2. For each observation (with the exception of 2014 data that was copied from figure 23b), the function of the lake level and delta area (lines drawn through the symbols) is obtained by assuming that the slope of delta is invariant in time. Data of 1957 are not presented, because lake level was not available. Note that the lake level on the y-axis is the height above 1783.5 m a.s.l. ......... 23 Figure 2-4: Deposited sediment at the apex of the Gumara River delta has buried a 1.5 m high water tank that was constructed in 2006 (photo on May 2015). .... 24 Figure 2-5: Land use dynamics for the three periods (1957, 1984 and 2006) in the upland part of Gumara catchment. a) The aerial photographs and spot image for the qualitative analysis of the land use and b) the classified images for estimation of the magnitude of the dominant changes. ............................ 25 Figure 2-6: a) The Gumara River in the delta at the point of old and new river course junction viewing in downstream direction; emerging poles are part of a barrier aimed at diverting the river flow to areas where sediment deposition is desirable for the local community (photo on 14 May 2015), and (b) satellite image from Google Earth. The arrows (1, 2 and 3) indicate diversions to take sediment laden flow from the river towards the marsh area and the nearby lake shore. Arrow 3 shows the old course (The imagery date is 16 Jan 2014). ................................................................................................................. 26
viii
Figure 2-7: Mean annual Lake Tana levels for the period 1959–2014 (Source: MoWIE). ................................................................................................................. 27 Figure 2-8: Observed sediment concentration for the Gumara River at the Gumara bridge (1983-2008). .................................................................................. 29 Figure 3-1: The Gumara catchment and the study reach. ........................................... 45 Figure 3-2: Superposition of planimetric position of Gumara River for the three periods. Major channel planform changes occurred in Reach 1. ........................... 49 Figure 3-3: The Gumara River outlet at Lake Tana for the three periods (a, b and c). At the mouth of the river (d), a 1.12 km² delta has been created over 27 years (1957-1984) and an additional 1 km² area has been added in the 22 years spanning 1984-2006................................................................................. 50 Figure 3-4: Example of in stream sediment deposition. A 1.74 ha island was observed on the 1957 aerial photograph (a), and after 27 years its area increased by 0.43 ha (b). The extent of the island expanded further and in recent years it became part of the left floodplain (c). ....................................................... 51 Figure 3-5: Changes to Gumara River cross-section at the Gumara bridge (source: MoWIE). ................................................................................................... 54 Figure 3-6: Trend analysis of rainfall and stream flow. The rainfall was recorded at Debre Tabor meteorological station in the upper catchment and the streamflow was measured at Gumara hydrological gauging station near Gumara bridge. Sources: NMA and MoWIE....................................................................... 55 Figure 3-7: Mean annual Lake Tana levels for the period 1959 to 2013 (Source: MoWIE). Starting from 1996 (construction of Chara-Chara weir at the outlet) the lake frequently shows levels higher than any known in historical times. Around 2003 the levels were low in relation to high hydropower production in the dry seasons (which led to an ecological disaster in coastal wetlands). In the rainy season the water levels are much higher than what is represented in this graph with average yearly data. ......................................................... 55 Figure 3-8: Automatic water level recorders installed by USBR in 1959, and photographed in June 2012 along the alluvial Gumara River (a; at 11.838408°N, 37.637086°E) and the bedrock Abay River 4 km downstream from Chara-Chara weir at the outlet from Lake Tana (b; at 11.567543°N, 37.403952°E). Δz indicates the depth of sediment that was accumulated in Gumara’s natural levee over the intermittent 53 years, i.e. 140 cm. ......... 56 Figure 3-9: Effects of direct human intervention on Gumara river banks: (a) remnants of previously densely and high growing Salix subserrata, and (b) sand mining as evidence for deposition of sediment in the river bed, and cause of river bank collapse. .......................................................................................... 59 Figure 3-10: Thickness of deposits in the river bed in the sand mining area. .............. 60 Figure 3-11: Extent of pump irrigation activities along the Gumara River banks. ........ 62
ix
Figure 3-12: Conceptual model of hydro-geomorphic dynamics of Gumara River in the Lake Tana lacustrine plain. Major human impacts that influence the system are written in block letters at the lower, left and right sides of the diagram. Sequence of processes is both from top and bottom of the diagram towards the middle. ................................................................................................ 64 Figure 4-1: Location map of the study site. The blue broken lines are the likely paleo channels at different time. ........................................................................ 78 Figure 4-2: Location map that shows bed and banks material sampling sits. The green dots are the sampling locations and B-23 are typical bend numbers. ...... 81 Figure 4-3:The ratio of channel centre line curvature radius (Rm) to channel width (W) for the bends of Gumara from the confluence with Kizin River to Lake Tana from 1957 to 2014. ................................................................................... 83 Figure 4-4: Frequency distribution of Rm/W for 1957, 1984, 2006 and 2014. .............. 84 Figure 4-5: Rm/W versus channel migration rate for 1957, 1984, 2006 and 2014. The blue fitted line shows the maximum limit (envelope curve) of the 2014 data. ................................................................................................................. 85 Figure 4-6: Rm/W versus migration rate for 2014 data. This graph is used for the determination of the critical range of Rm/W which gives maximum migration rate. .......................................................................................................... 86 Figure 4-7: The grain size distribution and soil classification for three segments of Gumara River between the confluence with Kizin River and Korkua). The sample was take at Left Bank (LB), Right Bank (RB) and at the Bed of the river. ......................................................................................................... 87 Figure 4-8: Diamond plot of the general textural analysis of Gumara River from the confluence with Kizin River to Korkua. ..................................................... 87 Figure 4-9: The trend of the bed shear stress and Manning roughness of Gumara River at the Gumara Bridge. .............................................................................. 88 Figure 4-10: An Example of bank collapse at a location of intensive irrigation activities (a) and at Kizin bend (b) showing that the nearby farm land is lost. ......... 91
x
List of Tables Table 2-1: Summary of maps, images, lake levels, and sediment data and river cross sections available for determining the characteristics of the delta of the Gumara in Lake Tana, Ethiopia. ............................................................... 18 Table 2-2: The area of the Gumara delta and the lake levels from 1957 to 2015. The size of the 1957 delta was taken as the base. .......................................... 21 Table 3-1: Summary of data and their sources. .......................................................... 46 Table 3-2: The study reaches planform characteristics. .............................................. 52 Table 3-3: Description of cross-sectional area, river bed and banks of Gumara River at Gumara Bridge for different periods. ........................................................ 53 Table 4-1: Summary of average channel width, centreline radius of curvature at the bend and sinuosity of 38 km Gumara River channel from the confluence with Kizin River to Lake Tana. .................................................................................. 82 Table 4-2: The hydraulics data of Gumara River at the bridge. Width, depth, velocity, discharge and bed slope are obtained from MoWIE. The others are calculated parameters. ............................................................................. 88
xi
List of Acronyms cm
centimeter
CSA
Central Statistical Agency
DEM
Digital elevation model
ENTRO Eastern Nile Technical Regional Office Eq.
Equation
GCPs
Ground Control Points
ha
hectare
km
kilometer
LB
Left Bank
LL
Lake Level
m
meter
M.Sc.
Master of Science
m a.s.l.
meter above sea level
MoWIE
Ministry of Water Irrigation and Electricity
MoWR
Ministry of Water Resources
RB
Right Bank
SMEC
Snowy Mountains Engineering Corporation
SRTM
Shuttle Radar Topography Mission
St.dev.
Standard deviation
SWAT
Soil and Water Assessment Tool
t
tonne
TaSBO
Tana Sub Basin Organization
USBR
United states Bureau of Reclamation
yr
year
xii
List of Symbols A
Delta area
Fr
Froude number
g
acceleration due to gravity
unit weight of water
h
channel flow depth
M
channel migration rate
n
Manning's roughness
Rm
channel centerline curvature radius
S0
channel bed slope
v
channel flow velocity
W
channel bankfull width
xiii
Chapter 1 1.1
Introduction
Background
Human is the hostile of the natural environment. Many authors have shown that the interference of human on the natural environment influences the nature of the landscape processes and the rate at which it operates; for e.g. Hook, 1994; Antrop, 1998; Hook, 2000; Wood & Handley, 2001; Antrop, 2005; Schneeberger et al., 2007. The increasing extent of human-induced land use change such as, urbanization, agricultural activities, hydraulic structures construction along and/or across the river, gravel and sand mining have brought changes to river systems. For instance, urbanization makes the flow contributing area impervious and as a result it generates high runoff which can have a power of changing the river channel characteristics’ (Hammer, 1972; Hollis, 1975; Urban & Rhoads, 2003; Kiss & Blanka, 2012). The existence of water projects, sand and gravel mining activities, may also lead to increase or decrease the stream transport capacity and thereby aggradation or degradation processes could occur which impacts the planform and shape of the river (Jansen et al., 1979; Kondolf, 1997; Surian, 1999; Kondolf et al., 2002; Surian & Rinaldi, 2003; Li et al., 2014). The understanding of river bank erosion is complicated by the operation of both fluvial and non-fluvial (mass failure) erosion processes in river systems (Lawler, 2005). Different authors reported that meander features (channel center line curvature radius (Rm) and channel bankfull width (W)) play an important role in determining the rate of channel migration; for instance, Hickin (1974), Hickin & Nanson (1975), Nanson & Hickin (1983), Begin (1981), Howard & Knutson (1984), Williams (1986). River meander features are also useful in predicting future changes in fluvial channels and erosion or deposition caused by man-made interferences (Wu,2007). Understanding channel changes from meander features are essential, and the first step to address stability problems is the study of morpho-dynamic processes (Klaassen & Masselink, 1992) and a clear understanding of the relation of channel pattern to river stability is required in the design of channels for minedland reclamation and in the planning for channel modification (Schumm, 1985).
1
In summary, the complex changes in fluvial processes are the result of changes in the water and sediment regimes induced by the extensive and intensive changes in land use (Poff et al., 2006) and yet, our understanding of the pathways and mechanisms through which land use/cover influences stream conditions is inadequate, particularly for strict management (Allan, 2004).
1.2
Existing knowledge on geomorphological changes of rivers in the upper Blue Nile basin
Studies on the Upper Blue Nile basin including Lake Tana dated back to the sixteen century by travellers and missionaries. As mentioned by Cheesman (1936), one of the Portuguese Jesuit priest Pedro Paez who travelled to the mountain of Sakala with the emperor Susenyous in 1613 was shown the spring and described it as the source of the Blue Nile. However, in 1770 James Bruce claimed to be the discoverer of the source. Stecker (1881); Checchi (1913); Grabham & Black (1925) and Cheesman (1936) described the Blue Nile and Lake Tana through observations and maps. Those maps can be digitized and information’s could be extracted. However, the resolutions of the older maps were so coarse and most of the writings were fairytale. In the late 1950, aerial photographs became available in Ethiopian Mapping Agency which has the required resolution for the analysis of land use/cover change of Upper Blue Nile basin. USBR (1964) conducted a study of the Blue Nile basin that has been a bench mark for the subsequent studies. In this study, many natural resources have been identified and hydrological gauging stations were installed in many tributary rivers of Blue Nile. Until recently, almost all hydrology and sediment studies depend on data collected from these gauging stations. However, there are lots of missing data especially the sediment data. Several studies on hydrology and sediment load have been carried out since 1990’s. The recent studies highlighted that the problem of erosion and sedimentation has become more severe despite governments’ effort in soil and water conservation. The qualitative description and quantitative analysis of previous studies indicated that much of the problem is linked with human induced factors which changes the landscape variables (Conway, 1997; Reid et al., 2000; Bewket, 2002; Nyssen et al., 2004; Dumont, 2009; Steenhuis et al., 2009; Poppe et al., 2013; Steenhuis et al., 2014; Dessie et al., 2015; Tilahun et al., 2013; Ali et al., 2014).
2
Some studies are also available that show the general effects of landscape change on the hydrology, sediment and rainfall-runoff responses of the Lake Tana basin. For instance, Setegn et al. (2008) predicted the hydrology and sediment yield in Lake Tana basin using Soil and Water Assessment Tool (SWAT) and indicated the probability of water shortage in the forthcoming decades due to climate change as well as due to depletion of water resources. Moreover, they indicated that 18.5 % of the basin is erosion potential. From nearby watershed (Debre Mewi which is located 30 km south of Lake Tana), Tebebu et al. (2010) found that gullies were large contributories to sediment load. Gully formation is often attributed to land use changes and agricultural practices at hillslopes including removal of vegetation and creation of surface drainage (Monsieurs et al., 2015). On Gilgel Abay catchment in the Lake Tana basin, in the period of 1973-2005, Rientjes et al. (2011) found that the annual stream flow has decreased by 12% as a result of land use changes (decrease in forest cover due to agricultural expansion) and rainfall (decreasing trend during dry months and increasing trend during wet months (June, July and August)). However, Steenhuis et al. (2014), from 1960 to 2000 for the whole Blue Nile basin, argued that the annual rainfall remained the same but discharge during the rainy phase increased as a result of an increased bare land. Poppe et al. (2013) concluded that in recent years, changes to the Lake Tana coastline are small with the exception of the delta formed by Gilgel Abay River, which is most probably related to human induced land use changes in the catchment. The influences of human on land use changes further explained by Gebremichael et al. (2013) and reported that the vegetation cover has been converted into agriculture and grass lands over wide areas of the Upper Blue Nile basin and the land-use change has caused a significant change of runoff and sediment load during the last four decades. Tilahun et al. (2014) showed that saturated (wetted) and degraded areas contributes high runoff. The floodplain of Lake Tana is acting as a buffer for inflow (water and sediment) to the lake (Dessie et al., 2014). The above studies clearly show that the changed landscape due to human factors and climate change but none of them showed the effects of the changed landscape on the fluvial geomorphological processes of river channels. Given the limitations of the above studies on fluvial processes, however, they confirmed that much sediment is produced from the upstream catchments due to 3
human induced factors. The sediment that is eroded from the upstream catchments has often been deposited in the river channel and at the same time during flooding with lesser depositional height (sediment distribution in wide area) in the alluvial and lacustrine plains too. Sedimentation in the river channel could reduce the flood carrying capacity of the stream channels. Obtaining documented historical information on floods in Ethiopia is difficult. Interviews with local old people and existing local documents have indicated that flooding has been occurring at least since 1964 and the floods of 1988 were particularly mentioned (ENTRO, 2010). For example, in the Fogera plain, east of Lake Tana, in the north central part of the Ethiopian highlands, flooding is increasing, with major occurrences in 1964, 1988, 1993, 1994, 1995 and 2006. One of the causes of flooding of the Fogera plain is believed to be bank spill over from major rivers (Gumara, Ribb, and Megech) which are draining into Lake Tana; particularly the Gumara river channel has a low conveyance capacity (Mekonnen, 2009; ENTRO, 2010). This might have a significant contribution for sediment deposition in the river channel, on the banks and at the lake shore when floods coincide with high lake levels. The problem of sedimentation is evidenced by obstructions at the outlets of the main tributaries (Gumara and Ribb Rivers) to Lake Tana, which led to courses that shifted to another direction (SMEC, 2008). In contrast to the lack of studies in the upper Blue Nile basin concerning the detailed understanding of the responses of the rivers to changes (man-made and natural), the available literature in the international context shows that studying river responses to drivers of changes will give an opportunity to know the dominant processes that make the banks to collapse and the bed to aggrade or degrade. This will be useful to address channel maintenance and restoration work along the river, to understand the present channel dynamics and predicting future channel change. This study has emerged from the above considerations. There is a need for an investigation into how the dynamics of river channels are governed by the anthropogenic, geomorphological, hydrological and climatic characteristics of their watersheds in the Upper Blue Nile Basin. This thesis will attempt to enhance understanding in these areas.
4
1.3
Objective of the thesis
The general objective of this study is to investigate and analyse the response of rivers to sedimentation, erosion and change of flow regime in the upper Blue Nile basin. The study is focused on Gumara River and the specific objectives are: •
to aid the development of river restoration and catchment based management practices from a better understanding of landscape change due to human action;
•
to investigate channel planform and vertical changes and explore the drivers of change;
•
to conceptualize the morphodynamic processes for better understanding of future changes to Gumara River channel;
•
to investigate the distribution and trend of river meander features (scale indicator) using time-sequence maps and
•
1.4
to explain sediment contributions of banks to the river channel and lake.
Thesis outline
This dissertation has five chapters. Chapter 1 is the introduction. Chapter two emphasizes the relationship that exists between lake level fluctuation and delta development at the mouth of Gumara and explains the long- term landscape change of Lake Tana basin. The Morphological changes (planform change and vertical adjustment) of Gumara River channel over 50 years is the third chapter of this research. In chapter four, the past, the present and the future of Gumara River channel characteristics are assessed and explained based on the meander features extracted from historical images and hydraulic and geotechnical data. Chapter 5 summarises main findings and conclusions of this research in the context of the original aims and objectives of the thesis as well as offering suggestions for potential future development of the work. It is my hope that this dissertation will encourage river channel studies in Ethiopia and further scientific understanding of the morphological changes of rivers.
5
References Allan JD. 2004. Landscapes and riverscapes: the influence of land use on stream ecosystems. Annual review of ecology, evolution, and systematics:257-284. Antrop M. 1998. Landscape change: Plan or chaos? Landscape and urban planning 41:155-161. Antrop M. 2005. Why landscapes of the past are important for the future. Landscape and urban planning 70:21-34. Begin Z. 1981. Stream curvature and bank erosion: a model based on the momentum equation. The Journal of Geology:497-504 Bewket W. 2002. Land cover dynamics since the 1950s in Chemoga watershed, Blue Nile basin, Ethiopia. Mountain Research and Development 22:263-269. Checchi M. 1913. Lago Tsana, 1:600 000. Firenze: Studio cartografico G. Giardi. Cheesman RE. 1936. Lake Tana and the Blue Nile: An Abyssinian Quest. London: Macmillan and Co., Limited St. Martin's Street. Conway D. 1997. A water balance model of the Upper Blue Nile in Ethiopia. Hydrological Sciences Journal 42:265-286. Dessie M, Verhoest NE, Teshager A, Pauwels VR, Poesen J, Adgo E, Deckers J, Nyssen J. 2014. Effects of the floodplain on river discharge into Lake Tana (Ethiopia). Journal of Hydrology 519:699-710. Dumont HJ. 2009. A description of the Nile Basin, and a synopsis of its history, ecology, biogeography, hydrology, and natural resources. In: The Nile: Springer. p 1-21. ENTRO. 2010. Flood Risk Mapping Consultancy for Pilot Areas in Ethiopia: Final Report to the Eastern Nile Technical Regional Office (ENTRO). Addis Ababa, Ethiopia: Riverside Technology and its partners. p 1-206. Gebremicael TG, Mohamed YA, Betrie GD, van der Zaag P, Teferi E. 2013. Trend analysis of runoff and sediment fluxes in the Upper Blue Nile basin: A combined analysis of statistical tests, physically-based models and landuse maps. Journal of Hydrology 482:57-68. Grabham G, Black R. 1925. Report of the Mission to Lake Tana, 1920–21. Min. of Public Works, Egypt: Government Press, Cairo. Hammer TR. 1972. Stream channel enlargement due to urbanization. Water Resources Research 8:1530-1540
6
Hickin EJ. 1974. The development of meanders in natural river-channels. American journal of science 274:414-442 Hickin EJ, Nanson GC. 1975. The character of channel migration on the Beatton River, northeast British Columbia, Canada. Geological Society of America Bulletin 86:487-494 Hollis G. 1975. The effect of urbanization on floods of different recurrence interval. Water Resources Research 11:431-435 Hooke, R.L., 2000. On the history of humans as geomorphic agents. Geology, 28: 843-846. Howard AD, Knutson TR. 1984. Sufficient conditions for river meandering: A simulation approach. Water Resources Research 20:1659-1667 Jansen PP, Van Bendegom L, Van Den Berg J, De Vries M, Zanen A. 1979. Principles of river engineering:The non-tidal alluvial river. Pitman: London. Reprinted by Delftse Uitgevers Maatschappij (1994), Delft, The Netherlands. Kiss T, Blanka V. 2012. River channel response to climate-and human-induced hydrological changes: Case study on the meandering Hernád River, Hungary. Geomorphology 175:115-125 Klaassen GJ, Masselink G. 1992. Planform changes of a braided river with fine sand as bed and bank material. In: Fifth Int. Symp. River Sedimentation. Karlsruhe, Germany. p 459-471. Kondolf G, Piégay H, Landon N. 2002. Channel response to increased and decreased bedload supply from land use change: contrasts between two catchments. Geomorphology 45:35-51 Kondolf GM. 1997. Profile: hungry water: effects of dams and gravel mining on river channels. Environmental management 21:533-551 Lawler D. 2005. Defining the moment of erosion: the principle of thermal consonance timing. Earth Surface Processes and Landforms 30:1597-1615 Li W, Wang Z, de Vriend HJ, van Maren D. 2014. Long-Term Effects of Water Diversions on the Longitudinal Flow and Bed Profiles. Hydraulic Engineering 140 Masselink G, Hughes M, Knight J. 2014. Introduction to coastal processes and geomorphology. New York, USA: Routledge.
7
Mekonnen DF. 2009. Satellite Remote Sensing for Soil Moisture Estimation:Gumara Catchement, Ethiopia. M.Sc. thesis study, International Institute for GeoInformation Science and Earth Observation, Enschede, The Netherlands. Monsieurs E, Poesen J, Dessie M, Adgo E, Verhoest NE, Deckers J, Nyssen J. 2015. Effects of drainage ditches and stone bunds on topographical thresholds for gully head development in North Ethiopia. Geomorphology 234:193-203. Nanson GC, Hickin EJ. 1983. Channel migration and incision on the Beatton River. Journal of Hydraulic Engineering 109:327-337 Nyssen J, Poesen J, Moeyersons J, Deckers J, Haile M, Lang A. 2004. Human impact on the environment in the Ethiopian and Eritrean highlands-a state of the art. Earth- Science Reviews 64:273-320 Poff NL, Bledsoe BP, Cuhaciyan CO. 2006. Hydrologic variation with land use across the contiguous United States: Geomorphic and ecological consequences for stream ecosystems. Geomorphology 79:264-285. Poppe L, Frankl A, Poesen J, Admasu T, Dessie M, Adgo E, Deckers J, Nyssen J. 2013. Geomorphology of the Lake Tana basin, Ethiopia. Journal of Maps 9:431-437. Reid RS, Kruska RL, Muthui N, Taye A, Wotton S, Wilson CJ, Mulatu W. 2000. Landuse and land-cover dynamics in response to changes in climatic, biological and socio-political forces: the case of southwestern Ethiopia. Landscape Ecology 15:339-355 Rientjes T, Haile A, Kebede E, Mannaerts C, Habib E, Steenhuis T. 2011. Changes in land cover, rainfall and stream flow in Upper Gilgel Abbay catchment, Blue Nile basin–Ethiopia. Hydrology and Earth System Sciences 15:1979-1989. Schneeberger N, Bürgi M, Kienast PF. 2007. Rates of landscape change at the northern fringe of the Swiss Alps: historical and recent tendencies. Landscape and Urban Planning 80:127-136. Schumm S. 1985. Patterns of alluvial rivers. Annual Review of Earth and Planetary Sciences 13:5 Setegn SG, Srinivasan R, Dargahi B. 2008. Hydrological modelling in the Lake Tana Basin, Ethiopia using SWAT model. The Open Hydrology Journal 2:49-62.
8
SMEC. 2008. Hydrological study of the Tana-Beles sub basins: Main Report. Australia: Snowy Mountains Engineering Corporation (SMEC) international Pty Ltd. p 1-110. Stecker A. 1881. Der Tana See 1 : 200 000. Mittheilungen Der Afrikanischen Gesellschaft in Deutschland Volume 3. Steenhuis TS, Tilahun SA, Tesemma ZK, Tebebu TY, Moges M, Zimale FA, Worqlul AW, Alemu ML, Ayana EK, Mohamed YA. 2014. Soil Erosion and Discharge in the Blue Nile Basin: Trends and Challenges. In: Assefa M M, Abtew W, Shimelis G S, editors. Nile River Basin. Switzerland: Springer. p 133-147. Strahler AN. 1956. The nature of induced erosion and aggradation. Man's Role in Changing the Face of the Earth 2:621-638 Surian N. 1999. Channel changes due to river regulation: the case of the Piave River, Italy. Earth Surface Processes and Landforms 24:1135-1151 Surian N, Rinaldi M. 2003. Morphological response to river engineering and management in alluvial channels in Italy. Geomorphology 50:307-326 Tebebu T, Abiy A, Zegeye A, Dahlke H, Easton Z, Tilahun S, Collick A, Kidnau S, Moges S, Dadgari F, Steenhuis T. 2010. Surface and subsurface flow effect on permanent gully formation and upland erosion near Lake Tana in the northern highlands of Ethiopia. Hydrology and Earth System Sciences 14:2207–2217. Tilahun SA, Guzman CD, Zegeye AD, Ayana EK, Collick AS, Yitaferu B, Steenhuis TS. 2014. Spatial and Temporal Patterns of Soil Erosion in the Semi-humid Ethiopian Highlands: A Case Study of Debre Mawi Watershed. In: Assefa M M, Wossenu A, Shimelis G, Setegn editors. Nile River Basin. Switzerland: Springer. p 149-163. Urban MA, Rhoads BL. 2003. Catastrophic human-induced change in streamchannel planform and geometry in an agricultural watershed, Illinois, USA. Annals of the Association of American Geographers 93:783-796 USBR. 1964. Land and water resources of the Blue Nile basin. Appendix III. Hydrology. United States Dept. of Interior, Bureau of Reclamation. Williams GP. 1986. River meanders and channel size. Journal of hydrology 88:147164
9
Wood R, Handley J. 2001. Landscape dynamics and the management of change. Landscape research 26:45-54. Wu W. 2007. Computational river dynamics: CRC Press.
10
Chapter 2
Long-term landscape changes in the Lake Tana basin as evidenced by delta development and floodplain aggradation, Ethiopia1
Abstract Landscape response to drivers of change is more visible and noticeable in deltas and floodplains than upstream. Here we address the changes of Lake Tana basin by investigating the delta development of Gumara River and sediment deposition in the Fogera floodplain over a 58-year period when agricultural land expanded and became more intensive and flooding of the alluvial plain became more frequent. Old maps showed that delta formation before the 1950s was minimal indicating that the sediment contributed by the rivers to the lake was small. During the last 58 years, the delta expanded continuously. Considering the same lake level (2.68 m), the delta expanded at an average annual growth of 5 ha and by considering different lake levels and corresponding delta areas, the delta increased in height by an average of 3 cm per year. While the growth of the delta was approximately linear, the sediment concentration in the river doubled in the last thirty years, indicating more efficient sediment trapping in the floodplain that was brought by higher lake levels, rising river bed and farmers intervening with the course of the river near the shore. Unless effective river restoration and catchment based treatment measures are put in place, the capacity of the rivers will further reduce and aggravate the flooding of the floodplain, causing more sediment deposition in the river channel and on the floodplain.
KEY WORDS: East Africa, Blue Nile, human impact, erosion, water management
1
Based on: Abate et al. (2016). Long-term landscape changes in the Lake Tana basin as evidenced by delta development and floodplain aggradation, Ethiopia. Land Degradation & Development. doi: 10.1002/ldr.2648. 11
2.1
Introduction
The complex interaction of society and nature results in reshaping the landscape that often leads to permanent loss of natural resources (Antrop, 1998; Wood & Handley, 2001; Antrop, 2005; Schneeberger et al., 2007). For instance, Marsh (1885) submits the decay of the Roman empire was partly caused by natural causes such as earthquakes that broke cisterns and reservoirs and human induced deforestation that dried up springs and decreased baseflow. Thus, a constant feedback exists between landscape processes and drivers (Lund et al., 2007). For instance, increased population density leads to an increase in bare land and this in turn intensifies runoff and erosion (Dunne et al., 1991; Braud et al., 2001; Nyssen et al., 2004; Steenhuis & Tilahun, 2014) that results in increased sediment deposition downstream in the river channel, floodplain, and lake (Harvey, 1991; Kondolf et al., 2002; Fryirs et al., 2009; Fryirs & Brierley, 2012; Ali et al., 2014). The deposited sediment in the lake is then subject to wave action that changes the patterns of deposition (Crooks, 2004). These changes are particularly pronounced at the shoreline of water bodies. When the supply of upland sediment transported by the river is more rapid than what can be redistributed by shoreline and riverine processes, deltas form (Yang et al., 2003). The shape, pattern and sediment profile of the delta is affected by discharge regime, the sediment load and sediment grain size as well as the water depth, and the relative magnitudes of waves and currents are the factors that influence the shape, pattern and sediment profile of deltas (Thom et al., 1975; Nunn, 1990; Reed, 1995; Masselink et al., 2014). Available experimental hydrological studies in the Lake Tana, or nearby catchments, have shown that saturation excess is the main runoff mechanism and most runoff and sediment are originating from the severely degraded hillsides and the periodically saturated bottom lands (Tilahun et al., 2013; Tilahun et al., 2014; Dagnew et al., 2015). In contrast, catchment simulations using Soil and Water Assessment Tool (SWAT) relate the changing discharge pattern, including the decrease in observed baseflow during the last 20-40 years, to changes in landuse patterns that were derived from 12
satellite imagery (Setegn et al., 2008; Gebremicael et al., 2013). Rientjes et al. (2011) contributed decreases in base flow to landuse change and Enku et al. (2014) to increased use of irrigation. Finally,Tesemma et al. (2010) and Steenhuis et al. (2014) using the Parameter Efficient Distributed model showed that the increase in degraded hillslopes was the cause of the change in flow patterns. The results of experimental studies and various simulation findings may be more similar than they appear because a change in hydrologic regime results in a change in land use (Bayabil et al., 2010). Gully formation is a large source of sediment (Zegeye et al., 2016). In nearby catchments,Tebebu et al. (2010) and Tebebu et al. (2015) argue that the current rapid gully formation in the periodically saturated bottom lands is caused by deforestation of agricultural land in the 1980s. As a consequence, these soils formed a hard pan that caused less baseflow and more direct runoff. Monsieurs et al. (2015) relate gully formation to concentrated flow from traditional drainage furrows in cultivated land. Poppe et al. (2013) concluded that in recent years, changes to the Lake Tana coastline are small with the exception of the delta formed by Gilgel Abay River, which is most probably related to human induced land-use changes in the catchment. The floodplain of Lake Tana is acting as a buffer for inflow (water and sediment) to the lake (Dessie et al., 2014). Abate et al. (2015) investigated planform changes of Gumara River at its mouth, and aggradation at the river bed and in the floodplain, and found these changes were due to increased sediment deposited from 1984 to 2006, which was attributed to lake level rise and human activities in the catchment (expanding agricultural activities including irrigation along the river). Very recent study on Gumara catchment by Scull et al. (2016) highlighted that “while many church forests used to be buffered from intensive agricultural activity today, they find themselves significantly more isolated and vulnerable to edge effects as a result of a general decrease of trees and bushlands surrounding the forests”. The individual studies mentioned above clearly show that the changed landscape in the Lake Tana basin is due to anthropogenic effects (e.g., intensifying agriculture, removal of natural vegetation and decreasing fallow periods), climate change and lake level rise that affected the discharge pattern and increased sediment loads. Similar 13
results have been reported elsewhere, for example by Liu Z. et al. (2014), Zhang et al. (2015) and Weinzierl et al. (2016). The increased sediment loads reduced lake volume and river channel transport capacity and destroyed wetlands and marsh areas. In order to counter the negative effects, appropriate river basin management practices need to be instituted for the upland, the river channel and the lake. Consequently, the objective of this paper is to aid in the development of such management practices by providing a better understanding of the landscape’s response. We will do this by relating landscape changes to delta development of which we have a relatively accurate record. This study particularly builds on the work of Abate et al. (2015) and (Lanckriet et al., 2016). Abate et al. investigated the planform changes of the Gumara River at its mouth (Lake Tana), and qualitatively discribed that the observed planform change was related to lake level rise and changes of upper Gumara catchment. In the Tigray region (Northern Ethiopia), Lanckriet et al. (2016) found sediment in alluvial and lacustrine debris fans could indicate the past condition to land degradation. However, this approach has not been applied to the Lake Tana basin and nearby catchments. In this study, we relate the delta development in Gumara River and fluctuations of lake levels to development in the Gumara cachement over a 58-year period.
14
2.2
Materials and methods
2.2.1 Description of the study area The 1,496 km2 catchment of Gumara River is part of the 15,077 km2 Lake Tana basin in the upper Blue Nile basin. The catchment can be divided into four smaller catchments (Figure 2-1): Lecha (29%), Sendega (25.9%), Kinti (13.7%) and Lower Gumara (31.4%). Near to the mouth of Gumara River, there are two massive rock outcrops across the sediment of the lacustrine plain: one to the south (peninsula) and the other to the north (surrounded by land mass) sides of the river. The delta is an agricultural area; the farmers live on an elevated portion and employ recession agriculture. The basin is also rugged. It has 16% steep slopes over 30°; 60% consists of slopes between 8° and 30°; 24% is comprised of slopes less than 8° that are located mainly in the lower part of the watershed near Lake Tana (Dessie et al., 2015). Dominant soil types identified in the upper catchment are Luvisols, Fluvisols, and Leptosols; common in the downstream plains are Vertisols and Fluvisols (Henricksen et al., 1984; Colot, 2012). High runoff response (Dessie et al., 2014) and sediment load (Lemma et al., 2016) of Gumara River occur due to strong degradation of the catchment (Poppe et al., 2013; Steenhuis et al., 2014). Annual rainfall distribution in the catchment is unimodal; the largest floods come in the second part of the rainy season (July and August) when the soils in the catchment become saturated (Liu et al., 2008; Tilahun et al., 2013). The greatest sediment concentrations occur at the beginning of the rainy season (June and July) when the loose plowed soil is easily eroded and transported to the lake (Easton et al., 2010; Lemma et al., 2016; Moges et al., 2016). The lake levels are lowest in June and July after which the level steadily increases and reaches its maximum level in September and October. Average annual lake levels varied by almost 1 m in the period of record from 1959 to present (Figure 2-7). Before the construction of the Chara-Chara weir (before 1995), the lake level fluctuation was fairly regular. After construction of the weir, the levels in Lake Tana could be regulated and were initially held at a higher level than in the past. In 2003, the lake levels were at a record low due to the great demand for electric power. Since 2009/2010, due to 15
the controlled gate operation at Beles hydropower station and a management plan instated by a new agency Tana Sub-Basin Organization (TaSBO) overseeing the operation of the lake, water levels have increased again. Currently, except for a portion of Debre Tabor (population of 87,096 (CSA, 2013)), major cities or towns do not exist in the Gumara catchment. A new road was constructed in 2015 through the entire catchment from east to west (supplementary Figure S2-1). The newly constructed road will accelerate the urbanization and affect the runoff pattern and sediment loss that, in turn, will affect the downstream areas. In addition, diversion structures for irrigation and a multipurpose dam construction project are planned for the future. Past exploitation of water resources has resulted in the river drying up by the end of the dry monsoon phase in the lower part of the Gumara River beginning a few kilometers upstream of the bridge on the road from Bahir Dar to Gondar (supplementary Figure S2-2) to Lake Tana.
Figure 2-1: Left: False-color image of the Gumara delta in the year 2006, indicated by four-point red star in the right map, Right: the Gumara catchment with major watersheds. The upper pilot area for land-use change detection is indicated by the red box with the aerial photograph. The inserted map shows Ethiopia’s major drainage basins, with Blue Nile basin highlighted. 2.2.2 Collection and preparation of data A large set of data consisting of the 1881, 1913,1920-21 and 1936 Lake Tana maps (Stecker, 1881; Checchi, 1913; Grabham & Black, 1925; Cheesman, 1936), aerial photographs (1957 and 1984), SPOT image of 2006, Landsat images of 1973, 1995, 16
2014 and 2015 were collected (Table 2-1). ArcGIS 10.2 tool was used for the data preparation and analysis. The procedure started from scanning, then importing the scanned map, geo-referencing, digitizing the features (that are needed for the study), measuring distances and calculating areas. To understand the effects of the fluctuation of the lake levels on the delta emergence and/or submergence, lake level data from Ministry of Water, Irrigation and Electricity (MoWIE) were collected and used. Assessment of sediment deposits was done by field measurements around permanent installations (stream gauging station (supplementary Figure S2-3), culverts (supplementary Figure S2-4)). Field observations were made in the floodplain, in the channel and in the delta. Unstructured interviews were given during field observation and Google Map analyses was made. All these data were used for interpretation of horizontal and vertical delta development and deposition of sediment in the floodplain. Finally, the available sediment concentrations data (44 data points) of the Gumara River at the bridge over the period from 1983 to 2008 were obtained from MoWIE. 2.2.3 Mapping of the delta The information obtained from the old maps of Lake Tana (1881, 1913, 1920-21 and 1936) were compared with the recent delta development. There was no visible delta at the mouth of the river on the old maps, but the survey procedures were not very accurate and hence, we used the 1957 aerial photograph as a benchmark for mapping of the delta area. Following the lakeshore line at the river mouth, the delta areas were digitized from the orthorectified aerial photographs of 1957 and 1984, SPOT Image of 2006, Landsat images of 1973, 1995, 2014 and 2015 and the delta areas were calculated.
17
Table 2-1: Summary of maps, images, lake levels, and sediment data and river cross sections available for determining the characteristics of the delta of the Gumara in Lake Tana, Ethiopia. Data type
Acquisition date
Resolution
Sources
Lake Tana Maps
1881
Scale 1:200,000
(Stecker, 1881)
1913
Scale 1:600,000
(Checchi, 1913) (Grabham & Black, 1925)
1920-21
Scale 1:500,000
1936
Scale 1:400,000
Dec 11, 1957;
Black and white
Ethiopian
Jan 16,1984
photos.
Mapping Agency
(Cheesman, 1936)
Aerial Photographs
Approx. scale1:50000 Landsat images
Feb 01,1973
60 m by 60 m
USGS
Aug 25, 1984
60 m by 60 m
USGS
Jan 12, 1995
60 m by 60 m
USGS
30 m by 30 m
USGS
30 m by 30 m
USGS
2.5 m by 2.5 m
Airbus Defence
Jan 16, Feb 17, Mar 21, Apr 22, Jun 25, Aug 28, Sep 29, Oct 31 & Dec 18, 2014 Jan 03 & Mar 24, 2015 SPOT Images
Jan 06, 2006
and Space Google Map
2010,2014 and 2015
Gumara River cross
2006
maps.google.be
section at its mouth Tana Lake Level record
1959-2015
daily
Ministry of Water, Irrigation and
at Bahir Dar
Electricity (MoWIE) Sediment concentration
1983-2008
44 samples
data at Gumara Gauging Station
18
2.2.4 Land use/cover classification The main use of aerial photos was to classify forest and non-forest areas (Banko, 1998; Jacob et al., 2015). Based on field observation, and informations from satelite images and Google Maps, one 48.4 km2 pilot area, considered as sediment supply zone that roughly represents the whole catchment, was chosen in the middle portion of the upper catchment of Gumara (Figure 2-1) and information from historical aerial photographs (1957 and 1984) and SPOT images of 2006 were used for the classification process. The pilot area is elongated, across the catchment, in order to occupy a varied set of land uses. Settlement area, bushes, grazing land and agricultural area were considered as one class (non-forest class) and forest considered as second class. The adopted methodology was an automatic classification using image processing software (ENVI 4.2). Visual observations of the black and white aerial photos on the computer screen made it possible to cluster a range of pixel values (0 to 255) interactively by using the density slicing tool in ENVI 4.2 environment. After a number of trials, final classified color map was developed. The procedure of the classification is shown in Annex S2-1. ArcGIS 10.2 tool was also used for the data preparation and analysis. An accuracy assessment of the landuse/cover classification was performed. The overall accuracy is 81% in 1957, 95% in 1984, and 87% in 2006; the kappa value is 62%, 89% and 71% for the same years. All classifications are above the 60% to 80% threshold for kappa value (Stehman, 1996; Viera & Garrett, 2005). Except the 1957 classification, the overall accuracy is above 85%; the common threshold value (Foody, 2002). But Wagner et al. (2013) reported that a value of 79.1% is also good overall accuracy. Therefore, both classifications showed reasonable accuracies, which allow them to be
taken as
substantially good. The relative lower accuracy of the 1957 classification is due to difficult discrimination between pixel values of grassland and forest in these black and white aerial photographs (Annex S2-2 and Supplementary Table S2-1).
19
2.3
Results
2.3.1 Delta development through time Mapping of the Gumara River delta using images from aerial photographs of 1957 and 1984, Landsat of 1973, 1995, 2014 and 2015, and the SPOT image of 2006, show that the Gumara River outlet to Lake Tana is continuously changing (Figure 2-2). Before 1957, a distinct delta did not exist at the mouth of the Gumara. After 1957 a delta start forming that grew steadily as a consequence of the increased sediment concentrations. The size of the emerged delta varies for the same lake level (Table 22) but taking the same lake level (2.68 m) the area from 1984 to 2014 shows that on average it expanded by 5 ha/yr.
Figure 2-2: Images of the Lake Tana shore near the Gumara River outlet between 1881 and 2015. At the mouth of the river, a 2.12 km² delta has been created over 49 years (1957-2006) but the area changes follow an apparent nonlinear trend. The red line is lakeshore line in 1957 and used as benchmark. Data sources are given in table 2-1.
20
Table 2-2: The area of the Gumara delta and the lake levels from 1957 to 2015. The size of the 1957 delta was taken as the base. Intercept of lake level (Eq. 2.2
Day
a The 1957
Month
Year
Delta
Lake
and Figure
area
level
2-3c)
(km2)
(m)
(m)
11
Dec
1957
0
naa
-
1
Feb
1973
0.78
2.14
2.54
16
Jan
1984
1.12
2.32
2.89
25
Aug
1984
0.48
2.68
2.92
12
Jan
1995
1.44
2.61
3.34
6
Jan
2006
2.12
2.5
3.58
16
Jan
2014
0.73
3.41
3.78
22
Apr
2014
2.15
2.68
3.78
25
Jun
2014
2.29
2.61
3.78
3
Jan
2015
0.73
3.42
3.79
24
Mar
2015
2.21
2.66
3.79
lake level is not available (na) and intercept could not be determined. Measurements of lake
level started in 1959.
21
In order to determine the increase in delta area due to sediment deposition, the effect of lake level changes should be taken out. In order to do this analysis, a reasonable assumption is that the slope of the lake bottom near the shore is invariant in time since it is substantially controlled by the friction angle of the saturated sediment. This then implies that there is a linear relationship between lake level and the area of the delta when we assume that delta expands lake wards only. Supplementary Figure S2-6 justifies the linear assumption of the delta front slope. A more precise estimate of the expansion can be obtained by plotting the delta areas (listed in Table 2-2) as a function of time in Figure 2-3a. When doing so, it is evident that the delta area is increasing. The relationship of delta area and lake level can be determined from the repetitive measurements made in 2014 for the lake level and the delta area Supplementary Table S2-2 and Supplementary Figure S2-5. Sediment deposition was likely close to nil during the dry monsoon phase. Using these data points, Figure 2-3b was plotted and the relationship between lake level (LL) in m and delta area (A) in km 2 in 2014 was found as: LL= -0.51A+3.76 (R² = 0.99)
(2.1)
As a check, the observation for 2015 (square) was plotted too and as expected, it is a little above the line due to sediment deposition during the rainy monsoon phase. For remaining years, the slope in Eq. (2.1) of the lake level-area relationship was assumed to remain the same and the intercept (3.76 in Eq. (2.1)) changes through time, with sediment being deposited on the delta. Using the lake level data and delta area, it is possible to calculate the intercept for the various years as: Intercept = 0.51A + LL
(2.2)
22
Figure 2-3: Expansion of the delta of the Gumara River in Lake Tana as a function of lake level (LL) for the period from 1957 to 2015. (a) Growth of the delta in time for lake levels (LL) between 2.14 – 2.68 m (orange symbols) and between 2.68– 3.42 m (blue symbols). Trend lines are shown in the same color as the symbols. (b) Delta area as a function of lake level for 2014 (blue close circles) and 2015 (orange close circles); the trend line is shown in blue. (c) Plot of lake level vs delta area for the period from 1973 to 2015. Symbols are the observations listed in Tables 2-2 and S2-2. For each observation (with the exception of 2014 data that was copied from figure 23b), the function of the lake level and delta area (lines drawn through the symbols) is obtained by assuming that the slope of delta is invariant in time. Data of 1957 are not presented, because lake level was not available. Note that the lake level on the y-axis is the height above 1783.5 m a.s.l.
The intercepts are listed in Table 2-2 and plotted in Figure 2-3c. Although the absolute values have little meaning, the difference between intercepts corresponds to the 23
amount of sediment deposited in the period from 1973 to 2015. We can see that the level is raised by about 1.25 m in 43 years for which lake level data is available. That is about 3 cm per year on average. A nice example of the quantity of deposited sediment is the burying of about 1.5 m height open lined masonry tanks that were built in 2006 at the beginning of the delta; point 1 and 2 in order to store temporally water pumped from Gumara River for a church building purpose (Figure 2-4a). During a field campaign on 14 May 2015, we observed that the sediment deposition was up to top level of tank 2 (Figure 2-4b) and that tank 1 was totally buried by the sediment.
Figure 2-4: Deposited sediment at the apex of the Gumara River delta has buried a 1.5 m high water tank that was constructed in 2006 (photo on May 2015). 2.3.2 Land use changes The landscape change dynamics of the Gumara catchment has been analyzed by considering an area of 48.4 km2 in the upper part of the catchment for the years 1957, 1984, and 2006. Field observations that we made in the catchment, information obtained from old aerial photographs, satellite images, and Google maps showed that the major portions of the non-forest class were agriculture and rural settlements. The results show that there was a significant agricultural development from 1957 to 2006 and as a consequence, the forest coverage decreased from 34.5 Km2 (in 1957) to 24.7 Km2 (in 1984) to 12.5 Km2 (in 2006) (Figure 2-5 and supplementary Table S2-5).
24
Figure 2-5: Land use dynamics for the three periods (1957, 1984 and 2006) in the upland part of Gumara catchment. a) The aerial photographs and spot image for the qualitative analysis of the land use and b) the classified images for estimation of the magnitude of the dominant changes. 2.3.3 People, delta shape and sediment distribution In the recent times, the shape of the delta and the distribution of sediments in the delta are influenced by humans who practice agriculture along the lake shore. Since the beginning of the 1980s, the river has changed its course at its mouth to the south-west direction. The farmers are attempting to reactivate the old channel course by constructing traditional diversion structures (Figure 2-6a) by putting wooden barriers across the river. Starting at the river entry point to the delta, they have also been trying
25
to obtain additional land by breaching the levee and digging small ditches towards the marsh area at the lake shore (Figure 2-6b).
Figure 2-6: a) The Gumara River in the delta at the point of old and new river course junction viewing in downstream direction; emerging poles are part of a barrier aimed at diverting the river flow to areas where sediment deposition is desirable for the local community (photo on 14 May 2015), and (b) satellite image from Google Earth. The arrows (1, 2 and 3) indicate diversions to take sediment laden flow from the river towards the marsh area and the nearby lake shore. Arrow 3 shows the old course (The imagery date is 16 Jan 2014). The farmers act on the flow of the Gumara River and probably changed the shape of the delta and the sediment distribution pattern. Due to the farmers’ influence, more sediment is deposited near the lake shore than at the westernmost river outlet to the lake and it is not unlikely that the farmers’ actions are a major factor in changing the shape of the delta and the sediment distribution pattern. This may be one of the reasons why in the 2015 satellite imagery at the river mouth the planform has shown a little shift towards a northwest direction.
26
Figure 2-7: Mean annual Lake Tana levels for the period 1959–2014 (Source: MoWIE).
27
2.4
Discussion
2.4.1 Development and sediment deposited in delta The high runoff response and sediment load of Gumara River are reported to occur due to strong degradation of the catchment. Steenhuis & Tilahun (2014) found that the sediment concentration at Gumara gauging station has increased by a factor of two between 1981 and 2005. The high sediment concentrations in the period after 1980 were confirmed in the Ethiopian highlands by (Conway, 1997; Bewket, 2002; Nyssen et al., 2008) and (Steenhuis et al., 2014). Easton et al. (2010) estimated a soil loss of 84 t/ha/yr based on limited data for Gumara and Tenaw & Awulachew (2009) estimated 22 t/ha/yr, while Zimale et al. (2016) indicated a loss of 49 t/ha/yr at the gauging station. Only Hanibal et al. (2016) showed relatively a value below 10 t/ha/yr at the mouth of Gumara based on 500 samples whereas others are using very few samples (not more than 44 samples from 1983 to 2008). Perhaps these large differences in soil loss values calculated by SWAT and other models are related to the difficulty to assess the delivery ratio and the antecedent soil moisture conditions. These suggest that: (1) only good field measuring campaigns can return reliable values of sediment yield to the lake; and (2) the causes of the differences need to be investigated. The increased sediment supply from the catchment to the river is most likely associated with population growth (from 22.2 million in 1960 to 95.5 million in 2014 (Supplementary Table S2-4, (Trading-Economics, 2016)) in the country, Ethiopia, and from approx. 277,500 in 2007 to 349,100 in 2014 in the Gumara watershed (Table 21, (CSA, 2008, 2013)) leading to extensification of agriculture, which in turn influenced negatively the forest coverage (Lambin et al., 2001; Jacob et al., 2015). Though the drought in the 1980s could have an effect to decrease the forest coverage, the continuous population growth (Supplementary Table S2-4 and Table 2-1) might take the greater attribution to the declined forest in Gumara watershed (Figure 2-5). The agricultural intensification by land clearing and deforestation has resulted in increasing the area of degraded soil. This suggests high runoff response (Dessie et al., 2014; Tebebu et al., 2015) and sediment load (Hanibal et al., 2016; Zimale et al., 2016) of Gumara River to the delta and the floodplain . Forest clearing was occurring 28
also in earlier times, but cropping periods were followed by fallow periods where the land could restore itself. Less than a century ago, as much as two third of the cultivable land was fallowed in the area (Grabham & Black, 1925). However, with the increasing population pressure, the re-growth did not occur and the continuous cultivation caused a loss of organic matter of the soil that has a binding effect (Tebebu et al., 2015). This has led to more bare and plowed soil and decrease the aggregate stability of soil. This, in turn, resulted in higher sediment concentrations in streams and rivers. All that has led to greater sediment transport by Gumara River towards the lake as shown by Steenhuis &
Tilahun (2014) and the sediment deposition became visible as an
emerged delta at the mouth of the Gumara River. The old maps hinted that, before 1957, there was an equilibrium between two processes: sediment input to the delta by the river, and shoreline erosion by wave action. But after 1957 the balance was interrupted (higher sediment supplied to the delta by the river than shoreline erosion by the wave action) and sediment deposition increased the size of the emerged delta. (Table 2-2 and Figure 2-8).
Figure 2-8: Observed sediment concentration for the Gumara River at the Gumara bridge (1983-2008). 29
2.4.2 Comparison of the delta area, river sediment load and anthropogenic impact Figure 2-3 and Figure 2-5 explain the relationship between delta area with that of lake level fluctuation and land use changes. Area of the delta increases by an average rate of 5 ha yr−1) was observed between 1984 and 2014 for the same lake level and depth of sediment deposition on the delta is about 3 cm yr−1 on average (Table 2-2). This is in line with the land use change in which large amounts of deforestation took place (Figure 2-5 and Supplementary Table S2-5) and this is supported by Jacob et al. (2015) in Tigray region in the same time period. In addition, the rise in lake level increase backwater effects (Jansen et al., 1979) which decreases the inflow velocity or the slope of water profile far upstream in the river channel and gives much time for the deposition of sediment in the channel and in the floodplain as well (Abate et al., 2015). Based on the data in Figure 2-7 this sediment deposition was affected systematically by the operation of the power plants. Unlike the findings by Ligdi et al. (2010) concerning possible increasing sediment input to Lake Tana, the increase rate of the delta area has slightly slowed down after 2006 compared with the 20 year period before; this is likely related to the higher frequency of overbank flooding and concomitant sediment deposition higher up in the floodplain. Similar results were reported for the River Ob (in Siberia) which drains to Arctic Ocean (Walling, 2008). During the last 50 years, the bed level of the Gumara River has been increasing by 6.33 cm/yr (Abate et al., 2015) while the banks’ level grew by only half of that. Consequently, the conveyance (the flow carrying) capacity of the river is decreasing, making flooding much more common than in earlier times as reported by Dessie et al. (2014) and Abate et al. (2015). In addition, farmers are actively trying to divert the sediment to lateral areas (not frontal) of the delta. Both factors explain that the delta is not growing faster while the sediment load from the watershed is increasing. A gradual increase in sediment concentration with time (Figure 2-8) after deforestation in 1984 (Figure 2-5) shows that the cutting of the forest set up a chain of events that start with more bare and plowed soil, loss of organic matter that led to aggregate instability, increase in sediment concentration and finally reduced infiltration capacity. 30
The question is whether the flooding of the Gumara is detrimental or should be encouraged. There are no villages or neighborhood in the plain and the rice grown in the plain during the rainy phase can withstand the floods. Farmers are actively breaching the dikes and redirecting the water and see the sediment as a resource (Figure 2-6) and local inhabitants did not allow papyrus to grow up in the newly deposited sediment at the lake fringe, but rather established farmlands there. Despite the positive aspects, sedimentation reduces channel water depth and storage capacity of the lakes, destroys wetlands and marsh areas. One can argue that in recent years the Fogera floodplain has become a very productive area because of the sediment additions that have increased the height of wetlands and marsh areas and can be used during the monsoon dry phase. Moreover, the addition of sediment allows the farmers in the floodplain to harvest crops two times in a year sustainable without fear for decreasing yield. As shown by the linear increase of the delta, while the sediment concentration in the rivers has doubled, the delta floodplain captures the sediment before it enters the lake. Therefore, further investigation is required for the impact of flood on the future of Gumara floodplain.
2.5
Conclusions
Using early maps, aerial photographs, satellite images and lake level records, changes in the catchment and hydro-geomorphology of the Gumara River were linked both with delta development and floodplain deposition. Old maps show that before 1957 when cropland was frequently fallowed, the delta did not expand indicating that the sediment concentrations were small. With the continuous increase in population, the land became cultivated intensively in the early nineteen-eighty’s and the delta started expanding. While the sediment loads increased exponentially after 1980, the delta area grew linearly indicating that more sediment was being trapped before entering to the lake by the floodplain. Since increasing population is the major driver for the greater sediment loads, effective catchment-based management practices should therefore offset the negative aspects of the increasing population density.
31
References
Abate M, Nyssen J, Steenhuis TS, Moges MM, Tilahun SA, Enku T, Adgo E. 2015. Morphological changes of Gumara River channel over 50 years, Upper Blue Nile basin, Ethiopia. Journal of Hydrology 525:152-164. Ali Y, Crosato A, Mohamed YA, Abdalla S, Wright N. 2014. Sediment balances in the Blue Nile River Basin. International Journal of Sediment Research 29:316-328. Antrop M. 1998. Landscape change: Plan or chaos? Landscape and urban planning 41:155-161. Antrop M. 2005. Why landscapes of the past are important for the future. Landscape and urban planning 70:21-34. Banko G. 1998. A review of assessing the accuracy of classifications of remotely sensed data and of methods including remote sensing data in forest inventory. IIASA Interim Report . IIASA, Laxenburg, Austria, IR-98-081. p 42. Bayabil HK, Tilahun SA, Collick AS, Yitaferu B, Steenhuis TS. 2010. Are runoff processes ecologically or topographically driven in the (sub) humid Ethiopian highlands? The case of the Maybar watershed. Ecohydrology 3:457-466. Bewket W. 2002. Land cover dynamics since the 1950s in Chemoga watershed, Blue Nile basin, Ethiopia. Mountain Research and Development 22:263-269. Bielli C, Gezu B, Amare I, Orasi A. 2001. Population growth and environment in Ethiopia. Roma, Italy: Institute for Population Research – National Research Council (Irp-Cnr) and Addis Ababa, Ethiopia: Central Statistical Agency. p 73. Braud I, Vich A, Zuluaga J, Fornero L, Pedrani A. 2001. Vegetation influence on runoff and sediment yield in the Andes region: observation and modelling. Journal of Hydrology 254:124-144. Checchi M. 1913. Lago Tsana, 1:600 000. Firenze: Studio cartografico G. Giardi. Cheesman RE. 1936. Lake Tana and the Blue Nile: An Abyssinian Quest. London: Macmillan and Co., Limited St. Martin's Street. p 406. Colot C. 2012. Soil-landscape relation at regional scale in Lake Tana basin. MSc thesis, Department of Soil and Water Management, KULeuven.
32
Conway D. 1997. A water balance model of the Upper Blue Nile in Ethiopia. Hydrological Sciences Journal 42:265-286. Crooks S. 2004. The effect of sea‐level rise on coastal geomorphology. Ibis 146:1820. CSA. 2008. Summary and Statistical Report of 2007 Population and Housing Census. Addis Ababa: Federal Democratic Republic of Ethiopia Central Statistical Agency (CSA). p 113. CSA. 2013. Population Projection of Ethiopia for All Regions At Wereda Level from 2014 – 2017. Addis Ababa: Federal Democratic Republic of Ethiopia Central Statistical Agency (CSA). p 118. Dagnew DC, Guzman CD, Zegeye AD, Tibebu TY, Getaneh M, Abate S, Zemale FA, Ayana EK, Tilahun SA, Steenhuis TS. 2015. Impact of conservation practices on runoff and soil loss in the sub-humid Ethiopian Highlands: The Debre Mawi watershed. Journal of Hydrology and Hydromechanics 63:214-223. Dessie M, Verhoest NE, Pauwels VR, Adgo E, Deckers J, Poesen J, Nyssen J. 2015. Water balance of a lake with floodplain buffering: Lake Tana, Blue Nile Basin, Ethiopia. Journal of Hydrology 522:174-186. Dessie M, Verhoest NE, Teshager A, Pauwels VR, Poesen J, Adgo E, Deckers J, Nyssen J. 2014. Effects of the floodplain on river discharge into Lake Tana (Ethiopia). Journal of Hydrology 519:699-710. Dunne T, Zhang W, Aubry BF. 1991. Effects of rainfall, vegetation, and microtopography on infiltration and runoff. Water Resources Research 27:2271-2285. Easton Z, Fuka D, White E, Collick A, Biruk Ashagre B, McCartney M, Awulachew S, Ahmed A, Steenhuis T. 2010. A multi basin SWAT model analysis of runoff and sedimentation in the Blue Nile, Ethiopia. Hydrology and Earth System Sciences 14:1827-1841. Enku T, Tadesse A, Yilak D, Gessesse A, Addisie M, Abate M, Zimale F, Moges M, Tilahun S, Steenhuis T. 2014. Biohydrology of low flows in the humid Ethiopian highlands: The Gilgel Abay catchment. Biologia 69:1502-1509. Foody GM. 2002. Status of land cover classification accuracy assessment. Remote sensing of environment 80:185-201. 33
Fryirs K, Spink A, Brierley G. 2009. Post‐European settlement response gradients of river sensitivity and recovery across the upper Hunter catchment, Australia. Earth Surface Processes and Landforms 34:897-918. Fryirs KA, Brierley GJ. 2012. Geomorphic analysis of river systems: an approach to reading the landscape. UK: John Wiley & Sons. p 360. Gebremicael TG, Mohamed YA, Betrie GD, van der Zaag P, Teferi E. 2013. Trend analysis of runoff and sediment fluxes in the Upper Blue Nile basin: A combined analysis of statistical tests, physically-based models and landuse maps. Journal of Hydrology 482:57-68. Grabham G, Black R. 1925. Report of the Mission to Lake Tana, 1920–21. Min. of Public Works, Egypt: Government Press, Cairo. p 207. Haack B. 1996. Monitoring wetland changes with remote sensing: an East African example. Environmental Management 20:411-419. Harvey A. 1991. The influence of sediment supply on the channel morphology of upland streams: Howgill Fells, Northwest England. Earth Surface Processes and Landforms 16:675-684. Henricksen BL, Ross S, Sultan T, Wijntje-Bruggeman HY. 1984. Assistance to Land Use Planning Ethiopia, Geomorphology and Soils. AGOA: ETH/ 78/003 Field Document 3. Food and Agriculture Organization of the United Nations, Rome. Jacob M, Frankl A, Beeckman H, Mesfin G, Hendrickx M, Guyassa E, Nyssen J. 2015. North Ethiopian Afro‐Alpine Tree Line Dynamics and Forest‐Cover Change Since the Early 20th Century. Land Degradation & Development 26:654-664. Jansen PP, Van Bendegom L, Van Den Berg J, De Vries M, Zanen A. 1979. Principles of river engineering:The non-tidal alluvial river. Pitman: London. Re-printed by Delftse Uitgevers Maatschappij (1994), Delft, The Netherlands. p 509. Kondolf G, Piégay H, Landon N. 2002. Channel response to increased and decreased bedload supply from land use change: contrasts between two catchments. Geomorphology 45:35-51. Lambin EF, Turner BL, Geist HJ, Agbola SB, Angelsen A, Bruce JW, Coomes OT, Dirzo R, Fischer G, Folke C. 2001. The causes of land-use and land-cover change: moving beyond the myths. Global environmental change 11:261-269.
34
Lanckriet S, Asfaha T, Frankl A, Zenebe A, Nyssen J. 2016. Sediment in Alluvial and Lacustrine Debris Fans as an Indicator for Land Degradation Around Lake Ashenge (Ethiopia). Land Degradation and Development 27 (2):258-269. Lemma H, Admasu T, Dessie M, Fentie D, Poesen J, Lanckriet S, Adgo E, Nyssen J. 2016. Sediment budget including the role of floodplains: the case of Lake Tana Basin (Ethiopia). In: EGU General Assembly 2016, Conference Abstracts. 1722 April, 2016 in Vienna, Austria. p 4751. Ligdi EE, El Kahloun M, Meire P. 2010. Ecohydrological status of Lake Tana—a shallow highland lake in the Blue Nile (Abbay) basin in Ethiopia: review. Ecohydrology & Hydrobiology 10:109-122. Liu BM, Collick AS, Zeleke G, Adgo E, Easton ZM, Steenhuis TS. 2008. Rainfalldischarge relationships for a monsoonal climate in the Ethiopian highlands. Hydrological processes 22:1059-1067. Liu Z., Yao Z., Huang H., Wu S., Liu G. . 2014. Land use and climate changes and their impacts on runoff in the Yarlung Zangbo river basin, China. Land Degradation and Development 25:203-215. Lund JR, Hanak E, Fleenor W, Howitt R, Mount J, Moyle P. 2007. Envisioning futures for the Sacramento-San Joaquin delta. San Francisco, USA: Public Policy Institute of California. p 324. Marsh GP. 1885. The earth as modified by human action. New York, USA: C. Scribner's sons. p 629. Masselink G, Hughes M, Knight J. 2014. Introduction to coastal processes and geomorphology, Second Edition ed. New York, USA: Routledge. p 413. Moges MA, Zemale FA, Alemu ML, Ayele GK, Dagnew DC, Tilahun SA, Steenhuis TS. 2016. Sediment concentration rating curves for a monsoonal climate: upper Blue Nile. SOIL 2:337-349. Monsieurs E, Poesen J, Dessie M, Adgo E, Verhoest NE, Deckers J, Nyssen J. 2015. Effects of drainage ditches and stone bunds on topographical thresholds for gully head development in North Ethiopia. Geomorphology 234:193-203. Nunn PD. 1990. Coastal processes and landforms of Fiji: Their bearing on Holocene sea-level changes in the south and west Pacific. Journal of Coastal Research:279-310. 35
Nyssen J, Poesen J, Moeyersons J, Deckers J, Haile M, Lang A. 2004. Human impact on the environment in the Ethiopian and Eritrean highlands-a state of the art. Earth- Science Reviews 64:273-320 Nyssen J, Poesen J, Moeyersons J, Haile M, Deckers J. 2008. Dynamics of soil erosion rates and controlling factors in the Northern Ethiopian Highlands– towards a sediment budget. Earth Surface Processes and Landforms 33:695711. Poppe L, Frankl A, Poesen J, Admasu T, Dessie M, Adgo E, Deckers J, Nyssen J. 2013. Geomorphology of the Lake Tana basin, Ethiopia. Journal of Maps 9:431437. Reed DJ. 1995. The response of coastal marshes to sea‐level rise: Survival or submergence? Earth Surface processes and landforms 20:39-48. Rientjes T, Haile A, Kebede E, Mannaerts C, Habib E, Steenhuis T. 2011. Changes in land cover, rainfall and stream flow in Upper Gilgel Abbay catchment, Blue Nile basin–Ethiopia. Hydrology and Earth System Sciences 15:1979-1989. Schneeberger N, Bürgi M, Kienast PF. 2007. Rates of landscape change at the northern fringe of the Swiss Alps: historical and recent tendencies. Landscape and Urban Planning 80:127-136. Scull P, Cardelús C, Klepeis P, Woods C, Frankl A, Nyssen J. 2016. The Resilience of Ethiopian Church Forests: Interpreting Aerial Photographs, 1938‐2015. Land Degradation & Development. DOI: 10.1002/ldr.2633 Setegn SG, Srinivasan R, Dargahi B. 2008. Hydrological modelling in the Lake Tana Basin, Ethiopia using SWAT model. The Open Hydrology Journal 2:49-62. Stecker A. 1881. Der Tana See 1 : 200 000. Mittheilungen Der Afrikanischen Gesellschaft in Deutschland Volume 3. Steenhuis T, Tilahun S. 2014. Changes in Climate and Landscape in the Blue Nile Basin: What it Means Downstream? In: De Châtel F, Holst Warhaft G, Steenhuis TS, editors. Water Scarcity,
Security and democracy: a
Mediterranean Mosaic. Athens, Greece, and Ithaca, NY.: Global Water Partnership Mediterranean, Cornell University and the Atkinson Center for a Sustainable Future. p 146-151.
36
Steenhuis TS, Tilahun SA, Tesemma ZK, Tebebu TY, Moges M, Zimale FA, Worqlul AW, Alemu ML, Ayana EK, Mohamed YA. 2014. Soil Erosion and Discharge in the Blue Nile Basin: Trends and Challenges. In: Assefa M M, Abtew W, Shimelis G S, editors. Nile River Basin. Switzerland: Springer. p 133-147. Stehman S. 1996. Estimating the kappa coefficient and its variance under stratified random sampling. Photogrammetric Engineering and Remote Sensing 62:401407. Tebebu T, Abiy A, Zegeye A, Dahlke H, Easton Z, Tilahun S, Collick A, Kidnau S, Moges S, Dadgari F, Steenhuis T. 2010. Surface and subsurface flow effect on permanent gully formation and upland erosion near Lake Tana in the northern highlands of Ethiopia. Hydrology and Earth System Sciences 14:2207–2217. Tebebu TY, Steenhuis TS, Dagnew DC, Guzman CD, Bayabil HK, Zegeye AD, Collick AS, Langan S, MacAlister C, Langendoen E. 2015. Improving efficacy of landscape interventions in the (sub) humid Ethiopian highlands by improved understanding of runoff processes. Frontiers in Earth Science 3:49. Tenaw M, Awulachew SB. 2009. Soil and Water Assessment Tool (SWAT)-Based Runoff and Sediment Yield Modeling: A Case of the Gumera Watershed in Lake Tana Subbasin. In: Improved Water and Land Management in the Ethiopian Highlands: Its Impact on Downstream Stakeholders Dependent on the Blue Nile. February 5-6, 2009. Addis Ababa, Ethiopia. p 100-111. Tesemma ZK, Mohamed YA, Steenhuis TS. 2010. Trends in rainfall and runoff in the Blue Nile Basin: 1964–2003. Hydrological Processes 24:3747-3758. Thom BG, Wright L, Coleman JM. 1975. Mangrove ecology and deltaic-estuarine geomorphology: Cambridge Gulf-Ord River, Western Australia. The Journal of Ecology:203-232. Tilahun SA, Guzman CD, Zegeye AD, Ayana EK, Collick AS, Yitaferu B, Steenhuis TS. 2014. Spatial and Temporal Patterns of Soil Erosion in the Semi-humid Ethiopian Highlands: A Case Study of Debre Mawi Watershed. In: Assefa M M, Wossenu A, Shimelis G, Setegn editors. Nile River Basin. Switzerland: Springer. p 149-163.
37
Tilahun SA, Mukundan R, Demisse BA, Engda TA, Guzman CD, Tarakegn BC, Easton ZM, Collick AS, Zegeye AD, Schneiderman EM, Parlange J-Y, S. S. 2013. A saturation Excess Erosion Model. Transactions of the ASABE 56:681-695. Viera AJ, Garrett JM. 2005. Understanding interobserver agreement: the kappa statistic. Fam Med 37:360-363. Wagner P, Kumar S, Schneider K. 2013. An assessment of land use change impacts on the water resources of the Mula and Mutha Rivers catchment upstream of Pune, India. Hydrology and Earth System Sciences 17:2233-2246. Walling D. 2008. Studying the impact of global change on erosion and sediment dynamics: current progress and future challenges. In: Discussion Document for the ISI Workshop. November 6, 2008. IRTCES, Beijing. Weinzierl T, Wehberg J, Böhner J, Conrad O. 2016. Spatial Assessment of Land Degradation Risk for the Okavango River Catchment, Southern Africa. Land Degradation & Development 27:281-294. Wood R, Handley J. 2001. Landscape dynamics and the management of change. Landscape research 26:45-54. Yang S, Belkin I, Belkina A, Zhao Q, Zhu J, Ding P. 2003. Delta response to decline in sediment supply from the Yangtze River: evidence of the recent four decades and expectations for the next half-century. Estuarine, Coastal and Shelf Science 57:689-699. Zegeye AD, Langendoen EJ, Stoof CR, Tilahun SA, Dagnew DC, Zimale FA, Guzman CD, Yitaferu B, Steenhuis TS. 2016. Morphological dynamics of gully systems in the sub-humid Ethiopian Highlands: The Debre Mawi watershed. Soils 2:443458. Zhang F, Tiyip T, Feng ZD, Kung HT, Johnson VC, Ding JL, Tashpolat N, Sawut M, Gui DW. 2015. Spatio-Temporal patterns of land use/cover changes over the past 20 years in the middle reaches of the Tarim River, Xinjiang, China. Land Degradation and Development 26 284-299. Zimale FA, Moges MA, Alemu ML, Ayana EK, Demissie SS, Tilahun SA, Steenhuis TS. 2016. Calculating the sediment budget of a tropical lake in the Blue Nile basin: Lake Tana. SOIL.
38
Chapter 3
Morphological changes of Gumara River channel over 50 years, upper Blue Nile basin, Ethiopia2
Abstract In response to anthropogenic disturbances, alluvial rivers adjust their geometry. The alluvial river channels in the upper Blue Nile basin have been disturbed by humaninduced factors since a longtime. This paper examines channel adjustment along a 38-km stretch of the Gumara River which drains towards Lake Tana and then to the Blue Nile. Over a 50 years’ period, agriculture developed rapidly in the catchment and flooding of the alluvial plain has become more frequent in recent times. The objectives of this study were to document the changes in channel planform and cross-section of the Gumara River and to investigate whether the changes could have contributed to the frequent flooding or vice versa. Two sets of aerial photographs (1957 and 1980) were scanned, and then orthorectified. Recent channel planform information was extracted from SPOT images of 2006 and Google Earth. Channel planform and bed morphology (vertical changes) were determined for these nearly 50 years’ period. The vertical changes were determined based on aggradation along a permanent structure, historic information on river cross-sections at a hydrological gauging station, and field observations. The results indicate that the lower reach of Gumara near its mouth has undergone major planform changes. A delta of 1.12 km2 was created between 1957 and 1980 and an additional 1.00 km2 of land has been added between 1980 and 2006. The sinuosity of the river changed only slightly: negatively (-1.1% i.e. meandering decreased) for the period from 1957 to 1980 and positively (+3.0%) for the period 1980-2006. Comparison of cross-sections at the hydrological gauging station showed that the deepest point in the river bed aggraded by 2.91 m for the period 1963-2009. The importance of sediment deposition in the stream and on its banks is related to land degradation in the upper catchment, and to artificial rising of Lake Tana level that creates a backwater effect and sediment deposition in Gumara River. Direct anthropogenic impacts (irrigation activities and building of dykes along the river banks) have contributed to the huge deposition in the river bed. Where the abstraction of water for irrigation is intensive, seepage water through the banks has contributed to river bank failure. In general, this study showed that changes to the planform at the mouth of the river and to the riverbed level are substantial. Moreover, the study indicated that the flood carrying capacity of the Gumara river channel has diminished in recent times. KEYWORDS: Channel planform; Sediment plug; Backwater effect; Human impact; Meander; Sinuosity
2
Based on Abate et al. (2015). Morphological changes of Gumara River channel over 50 years, Upper Blue Nile basin, Ethiopia. Journal of Hydrology 525:152-164. 39
3.1
Introduction
Many authors showed that the interference of human on the natural environment influences the nature of the landscape processes and the rate at which it operates. The increasing extent of the human disturbances such as land use changes, urbanization, channelization, hydraulic structures construction along and/or across the river gravel and sand mining have brought changes to river systems (Strahler, 1956; Wolman, 1967; Hammer, 1972; Graf, 1975; Hollis, 1975; Jansen et al., 1979b; Simons, 1979; Williams & Wolman, 1984; Kondolf & Swanson, 1993; Kondolf, 1997; Surian, 1999; Reid et al., 2000; Kondolf et al., 2002; Surian & Rinaldi, 2003; Urban & Rhoads, 2003; Zhe-ren, 2003; Julien et al., 2005; Vanacker et al., 2005; Galster et al., 2006; Clark & Wynn, 2007; Conesa‐García et al., 2007; Kondolf et al., 2007; Slaymaker, 2010; Kiss & Blanka, 2012; Li et al., 2014). For instance, urbanization makes the flow contributing area impervious and as a result it generates high runoff which can have a power of changing the river channel characteristics’. The existence of water projects, sand and gravel mining activities, may also lead to increase or decrease the stream transport capacity and thereby aggradation or degradation processes could occur which impacts the planform and shape of the river. There are number of dependent (sediment supply, stream discharge and vegetation) and independent (geology, climate, human and time) landscape variables (Hogan & Luzi, 2010) which have a decisive role in the fluvial processes. Channel morphology is the result of the combined influence of the dependent landscape variables, and the channel responds to changes in these variables by adjustments in one or many of the dependent channel variables (width, depth, bed slope, grain size, bedforms, sinuosity, scour depth) (Hogan & Luzi, 2010; Ibisate et al., 2011). In response to changes in streamflow and sediment load (dependent landscape variables), alluvial rivers are dynamic systems that adjust their geometry (Lane, 1955a; Leopold & Wolman, 1957; Heede, 1980; Schumm, 1985; Simon et al., 2002; Heitmuller & Greene, 2009; Dust & Wohl, 2012). The width of the channel, for instance, changes significantly by increased and decreased bedload supply from land use change (Liébault & Piégay, 2001; Kondolf et al., 2002; Liébault & Piégay, 2002). Alluvial river beds are subjected to morphological changes during flood events with significant implications for the water 40
level (Neuhold et al., 2009). These changes may be the result of altered catchment hydrology or sediment supply, construction of riparian training structures, or channel alterations to support navigation or environmental needs (Scott & Jia, 2005; Blöschl et al., 2007). The stream corridors strongly depend on the river morphological processes and hydrological conditions, as well as on the catchment geology, topography, climatology, vegetation cover, land use practices and human interventions (Globevnik, 1998). During the preparation of flood hazard maps, it is generally assumed that the river morphology will have changed neither during flood events nor by long-term erosion or deposition. However, quantitatively and qualitatively observed morphological developments during and after flood events indicate changes in river bed elevation due to sediment transport, log jams or rock jams (Neuhold et al., 2009). The morphology of an alluvial river channel is the result of net sediment entrainment and deposition in the river (Russell, 1954; Strahler, 1956; van Rijn, 1984; Gaeuman et al., 2005; Paola & Voller, 2005; Church, 2006; Conesa‐García et al., 2007). The sediment, either from the catchment or from the channel bed and banks, might have huge contribution to aggradation in one location and degradation in another location. Erosion of stream banks and shorelines by piping, for instance, has very significant impacts on bank and shore stability (Hooke, 1979; Odgaard, 1987; Hagerty, 1991; Julian & Torres, 2006; Bartley et al., 2008; Langendoen & Simon, 2008). Therefore, understanding the stability of rivers is essential, and the first step to address stability problems is the study of morphodynamic processes (Klaassen & Masselink, 1992) and a clear understanding of the relation of channel pattern to river stability is required in the design of channels for mined-land reclamation and in the planning for channel modification (Schumm, 1985). In geological times, the landscape of the upper Blue Nile basin, Ethiopia, has been modified in fundamental ways by the processes of erosion, transport and deposition of sediments (Conway, 1997; Bewket, 2002; Dumont, 2009). The sediment that is eroded from the upstream catchments has often been deposited in the river channel and at the same time during flooding with lesser depositional height (sediment distribution in wide area) in the alluvial and lacustrine plains too. Sedimentation in the river channel could reduce the flood carrying capacity of the stream channels. This might be due to changes of river morphology or as a result of land use and land cover 41
changes (Poppe et al., 2013). Though obtaining documented historical information on floods in Ethiopia is hardly possible, interviews with local old people and existing local documents have indicated that flooding has been documented since 1964 and the floods of 1988 were particularly mentioned (ENTRO, 2010). For example, in the Fogera plain, flooding is increasing, with major occurrences in 1964, 1988, 1993, 1994, 1995 and 2006. The Fogera plain is located at the east of Lake Tana, in the north central part of the Ethiopian highlands. One of the causes of flooding of the Fogera plain is believed to be bank spillover from major rivers (Gumara, Ribb, and Megech) which are draining into Lake Tana; particularly the Gumara river channel has a low conveyance capacity (Mekonnen, 2009; ENTRO, 2010). This might have a significant contribution for sediment deposition in the river channel, on the banks and at the lake shore when floods coincide with high lake levels. The problem of sedimentation is evidenced by obstructions at the outlets of the main tributaries (Gumara and Ribb Rivers) to Lake Tana, which led to courses that shifted to another direction (SMEC, 2008). In the Northern Ethiopian highlands, particularly human activities (removal of natural and semi-natural vegetation) have led to an overall increase in erosion process intensity (Dejene, 1990; Zeleke & Hurni, 2001; Hurni et al., 2005; Nyssen et al., 2008) despite recent positive impacts of targeted interventions by the society (Nyssen et al., 2007). Though quite a number of studies have been carried out in Lake Tana basin, the effects of the developmental activities and other natural and man-made phenomena on the morphology (channel adjustment) of the basin’s tributary rivers has not yet been addressed. In contrast to the lack of studies in the upper Blue Nile basin regarding the detailed understanding and documentation of the responses of the rivers to changes (man-made and natural), the available literature in the international context shows that studying river responses to changes will give an opportunity to know the dominant processes that make the banks to collapse and the bed to aggrade or degrade. This will be useful to address channel maintenance and restoration work along the river, to understand the present channel dynamics and predicting future channel evolution as these are a key steps in reducing flooding. Therefore, the general objective of this study is to investigate and analyze the response of rivers to sedimentation, erosion and change of flow regime in the upper Blue Nile basin. The study was focused on 42
Gumara River and specifically the objectives were: to investigate the channel planform and vertical changes; to explore the drivers for these channel morphological changes; and to conceptualize the morphodynamic processes at stake so as to understand future changes to Gumara River channel.
3.2
Materials and methods
3.2.1 Description of the study area The Gumara catchment is located within Lake Tana basin in Ethiopia (Figure 3-1). River Gumara originates from the Guna Mountains south and east of Debre Tabor at an altitude of approx. 3250 m a.s.l. The river flows westwards for 132.5 km until it reaches Lake Tana. The level of this shallow lake, the largest of Ethiopia, is regulated since 1995-1997 by a flow regulation structure just at its outlet to the Blue Nile River, the Chara-Chara weir which increases the lake level by up to 2 m, particularly in the rainy season. A stakeholder analysis (McCartney et al., 2010) has investigated the consequences for downstream users along the Blue Nile (irrigation, hydropower, tourism) and for farmers and fishers on the lake shore; however the consequences of the backwater effect along major rivers in the lacustrine plain was not investigated. The Gumara catchment covers a total area of about 1496 km2. There are many small intermittent and perennial rivers and springs in the catchment, which flow into the main stem, Gumara River. The catchment consists of rugged and undulating topography, which varies from 1788 m a.s.l. up to 3750 m a.s.l. The area has steep slopes (frequently >25%) in the high mountainous region in the east, and gentle slopes (
