Analysis of Watershed Characteristics and Spatial Soil Loss Estimation: A case Fincha Watershed, Western Ethiopia

Characteristics of watershed were not well investigated as global also in Fincha, Ethiopia. Fincha watershed is the most erosion prone area in the highlands of Ethiopia towards to Fincha dam.Therefore, the aim of this study was toanalysisFincha watershed characteristics based on land use/cover, soil type and slope classication and to identify soil erosionpronearea using Arc-GIS and Arc-SWAT model.Accordingly, nine major land use/covers were identied. Those were:- bush land, dominantly cultivated, moderately cultivated, irrigated land, grass land, water bodies, swamp area, urban and woodland open. From these most part of the catchment was under cultivation with 67.677% of the total area. In the same way, the major soil types identied were: - chromic luvisols, chromic vertisols, dystric cambisols, eutric cambisols, eutric nitosols, eutric regosols, haplic phaeozems, humic cambisols and water. Cambisols were covered the highest percentage, 39.98% area of the catchment.Subsequently, average annual values of estimated soil lossfrom the catchment was231.2 ton/ha.The identied prone erosion areas were sub-basins 1, 8, 11, 14, 17 and 20. It was very critical for environmentalists, hydrologists, agricultural experts, watershed managements and concerned decision makers for sustainable water resource projects and environmental protection system. Therefore, further detail investigation and appropriate watershed management practices should be given for these subbasins.


Introduction
An area that contains a common set of stearms and rivers into a single water bodies is called watershed.
So,characterization of watershed is a necessary and important step in planning and management of a watershed. Miller et al, (2003) stated that, many researchers have used DEM and GIS techniques for watershed delineation and extraction of drainage networks and various topographic characteristics of a watershed.Automatic extraction of important watershed characteristics such as watershed drainage area/boundary, eld and channel slope, aspect, and drainage network using DEM 30 m were most accurately with variations less than 10% (Singh et al., 2005).
Arc GIS was used to extract hydrological characteristics such as ow direction, ow accumulation, ow length, stream networks, and snap pour point of various streams (Fu-quan et al., 2010). Gossa W, (2011), globally soil degradation is affecting 1.9 billion hectares and is increased at a rate of 5 to 7 million ha/yr.
In most parts of Ethiopia the high intensity rainfall occurs when the cultivated land has low cover, which can reduce the impact of the high intensity raindrop and the high runoff which can be slowed by soil cover (White, 2005).Soil erosion and river sedimentation can cause critical environmental, ecological and economic problems worldwide (Yang et al., 2003). Soil erosion from the upstream area of the Blue Nile River Basin and the subsequent sedimentation in its downstream area caused a rapid loss of storage volume due to excessive sedimentation (Betrie GD et al., 2011). Bezuayehu (2006) reported that Fincha watershed is a typical example of many watersheds in the country that had undergone land use change and presently undergoing environmental degradation and causing serious problems. It is one of those highland areas of the country with severe soil erosion Page 3/20 problem draining to the Nile River. Fincha hydropower reservoir is a highland area with a severe soil erosion problem that drains to the Blue Nile River due to increasingly mountainous and steeper slopes area cultivated, in many cases without protective measures against land erosion and degradation (Assefa, 1994).
Soil loss from watershed was increased from time to time due natural and human induce activities.Reduction in storage capacity of rivers and reservoirs become increased due to soil loss and sediment yield from the catchment (Agarwal, et al., 2009). So,characterization of watershed, soil losses, sediment yields,reservoir capacity reduction, disturbance of aquatic ecosystem and watershed managements were main agendas of scienti c researchers.
Since erosion and Sediment yield is one of the most complex natural processes and many factors are involved in it, the full knowledge of the factors in uencing this phenomenon is very di cult (Zahabioun et al., 2010).
Analysis of watershed characteristics, soil erosion, transport and deposition of sediments in reservoirs, irrigation canals and hydropower systems were very essential for land and water management, but these were not studied in-depth in the catchment.
So, watershed characterizations, spatial distributions of soil losses and prone area identi cation of Fincha catchment using GIS and SWAT model was pertinent for best watershed management practices, planning, implementation, decision making, giving solution for future.

Study Area
The study area is situated in Abay/Nile River basin which is located in the Horro Guduru Wollegga Zone, Oromia National Regional State, Ethiopia. Geographically located between 9°9′53″ N to 10°1′00″ N latitudes and 37°00′25″ E to 37°33′17″ E longitudesas shown in Fig. 1. The altitude in the catchment ranges from 902 m in the lowlands up to 3,171 m amsl in the highlands. The climate of the Fincha watershed is tropical highland monsoon with an average annual rainfall of 1763.6 mm and mean monthly temperature of the area varies from 14.6 to 17.7 °C. Most of the rain falls during the months of June to September with peaks occurring during July to August and it is virtually somewhat dry from November through to April.The total area of watershed was 2,619 km 2 and six districts-namely, Jimma Geneti, Horro, AbbayChomen, AbaboGuduru, Guduru, and Jimma Rare. The watershed is bordered on the north by the Blue Nile River (also called Abbay River in Ethiopia), on the east by the Guder River Basin, on the south by Awash River Basin, and on the west by Diddessa River Basin.

Materials And Methods
Majority of data used were secondary data that includes meteorological, spatial and hydrological data in general. Those, data were collected from different organizations like Ethiopia National Metrological Agency and from Ministry of Water, Irrigation and Electricity. Also eld observation was done for locating the outlet and visualizing the catchment.The main tools that have been used in this study for data collection, preparation and analysiswere: Arc VIEW GIS10.4, Arc SWAT2012, different maps, andMicrosoft excel. ArcGIS10.4 was used for creating maps, managing geographic information in a database and execution of GIS processing tools (such as clipping, overlay, and spatial analysis). SWAT2012 model was used for setting up the study project, delineating the study area, analysing HRU, writing all input tables, editing inputs and simulating all inputs.
3.1. Model inputs and their compatibility for the model Three main input data were 1) metrological data 2) spatial data

Watershed Delineation process
The Arc SWAT requires model setup and parameterization of the model. Watershed was automatically delineated using 30 m x 30 m DEM.
Automatically delineated the watershed was classi ed into several hydrologically connected subwatersheds. After the DEM grid loaded and the stream networks superimposed, the DEM map grid was prepared to remove the non-draining zones. The initial stream network and sub-basin outlets also de ned based on drainage area threshold approach. Sub-basin parameters were calculated in order to estimate the basic watershed characteristics from the DEM and 21 sub-basins with different HRU were identi ed. General procedure of watershed delineation was presented as Fig. 4 below.

HRUs analysis
It includes requires land use/cover, soil and slope layers and their threshold inputs. HRU is de ned as a lumped land area comprised of a uniform land use land cover, soil type and uniform slope of the watershed. Land use/cover, soil type, runoff, soil loss and sediment yield could be predicted separately from each HRU.It increases the accuracy in prediction that provides a much better physical description of the water balance rather than prediction from the whole catchment at once.

Land use/land cover analysis
The LULC is one of the most important watershed characteristics and factors that affect soil erosion, runoff, sediment and evapo-transpiration in a watershed during simulation (Neitsch et al., 2005). For LULC analysis of the catchment, the land use map Ethiopia and Fincha basin were overlaid and clipped. There were nine major land use land cover classes were identi ed in Fincha watershed. Those were:-bush land, dominantly cultivated, moderately cultivated, irrigated land, grass land, water bodies, swampy area, urban and woodland that presented in Table 3 and Fig. 5 respectively.Bush land occupies about 2.13% of total area. These were type of vegetation which is either a remainder of natural vegetation of the land or, if altered, was still representative of the structure and oristic of natural vegetation. Part of the watershed covered by grasslands was 10.64%. Are types of vegetation dominated by grasses rather than large shrubs or trees. As degradation of grassland increase, soil erosion also increased. Water bodies occupied 11.42% of total area. Due to poor watershed management, quality of these water bodies decreased and aquatic ecosystems was disturbed. Similarly, swamp area, urban and woodland open covered 2.533%, 1.54% and 4.06% respectively. About 67.677% of the total catchment was under human induces, cultivation practices. These indicate thatmajority of the catchment was very sensitive to runoff, soil erosion/sediment yield due to cultivation practices.Therefore, best watershed management practices should be applied to protect this watershed by concerned stakeholders.

Soil type analysis
Arc SWAT model require soil data to provide both the distribution of the soil type in the watershed and the various parameters describing the soils hydrological and textural properties. The soil textural and physicochemical properties include soil texture, available water content, hydraulic conductivity, bulk density and organic carbon content for each soil type.Fincha basin was overlaid with Ethio -soil map and clipped.
Clipped Fincha soil map was projected to UTM zone 37, projection area of Ethiopia and nine major soil types were obtained. These soil categories and spatial distribution of the catchment was presented in Table 4 and Fig. 7 respectively.
These major soils were: chromic luvisols, chromic vertisols, dystric cambisols, eutric cambisols, eutric nitosols, eutric regosols, haplic phaeozems, humic cambisols and water. From these, soil type that covers most part of the watershed were dystric and eutric cambisols with 39.98% of a total area.As of the soil properties, cambisols show variations in colors, structure and carbon. They have medium to ne textured materials that result from the decomposition of a wide range of rocks. Cambisols exist on a level to mountainous topography of all-weather climates.
They are favorable for a wide range of vegetation growth and have good production potential for agricultural activities. Under uncontrolled agricultural activities and poor land management, the soil was exposed to a high potential of erosion occurrence and sediment yield. The second soil class was eutric regosols and eutric nitosols with 21.65% of total area of the watershed. Regosols are very weakly developed mineral soils in unconsolidated materials that have only an ochric surface horizon and that are very deep. They are extensive in eroding lands, in particular in arid and semi-arid areas and in mountain regions.
Nitosols mostly found at higher altitudes. They are inherently fertile of the tropical soils because of their high nutrient contents and deep, permeable structure and exploited widely for plantation agriculture.
The third category was chromic vertisols and chromic luvisols that cover20.27% of the watershed.The reference soil group of Luvisols holds soils whose dominant characteristics is a marked textural differentiation within the soil pro le, with the surface horizon being depleted of clay and accumulation of clay in a subsurface argic horizon. Most chromic luvisols were well drained but shallow groundwater may occur in luvisols in depression area. From soil properties these soils have high activity clays throughout and lack the abrupt textural change of planosols. Luvisols are fertile soils and suitable for a wide range of agricultural use when in controlled system. So, under poor watershed management clay will be washed down from the surface soil to accumulation horizon at lower depth.

Slope analysis
Slope was another watershed characteristics and critical for watershed management. As the slope of the catchment became steep, runoff and soil erosion were increased. Example, sub-watershed with steep slope might contribute more soil losses, sediment yield to water bodies than with at landscapes.
The slope of the catchment was classi ed into four multiple classes0-3%, 3-6%, 6-9% and above 9% based on the DEM data used during the watershed delineation and the topography of the catchment and shown in Fig. 8.
General at slope covers about 56.43%, moderate slope 28.54% and steep slope 15.03% of the total catchment. These indicate that of the catchment area had high general at. The assessment of the spatial variability of sediment yield at HRU and sub basin level is useful for catchment management planning and reservoir protection. At sub basin level the highest and lowest average sediment yield was from sub basin 11 and sub basin 18 with values of582.022ton/ha and 0 ton/ha respectively. This indicates sub basin 11 was under extreme soil erosion and the most prone area of Fincha watershed at sub basin level from a total of 21 sub basins.So,priorityof watershed management practices at subbasin level should start from this subbasin and proceed to the rest. The spatial prone area in relation to soil erosion of Fincha watershed was shown in Fig. 9.
Soil erosion prone area was classi ed based on the classi cations of erosion rates in the Ethiopian high lands. It was classi ed into four main classes in this study as very high, high, medium and low. Very high from sub-basin 1, 8, 11, 14, 17 and 20that covers an area of 27.84%. High erosion level from sub-basin 3, 7, 9, 12, 13, 15 and16 with covered area of 36.03%. Medium erosion level from sub-basin 4, 5, 6,10 and covers area of 24.24%. The last classi cation as low was from sub-basin 2, 18, 19 21 and covers an area of 11.9%.The average annual sediment yield total sub basins is 231.2ton/ha.The statistical data of the soil loss class and corresponding area ratios were presented in Table 6. Declarations 6.1 Availability of data and materials All data generated or analysed during this study are included in this manuscript and it can be found from outhors if required latter.

Competing interests
As authors, we have no competing interests to disclosure our submitted article in this section.

Funding
Publication costs for Sustainable Environment Research will be covered by the Chinese Institute of Environmental Engineering so authors do not need to pay an article-processing charge.
6.4 Authors' contributions SK literarure review, data collection, analyzed and interpreted the required data accordingly regarding with LULC analysis, soil type analysis and soil loss estimation and major contributor in writing the manuscript.
FF performed on check up and comments, writing the manuscript. Finally, all authors read and approved the nal manuscript.

Acknowledgements
First and above all praise to be Almighty God, for endless love and for He gives me healthy mind to think and to do my works and great thanks be to Him forever.
My special and deep thanks goes to my parents for their encouragement, love and advices.
I heartedly gratify my lovely wife Derartu Bultum, who has took double responsibility of handling issues of family and successful accomplishment of my vision.
Also, I would like apreatiate Sustainable Environment Research for giving me such opportunity to publish my original paper