©Haramaya University, 2021 ISSN 1993-8195 (Online), ISSN 1992-0407(Print) East African Journal of Sciences (2021) Volume 15(2): 115-128 Licensed under a Creative Commons *Corresponding Author: tesfayebecon@gmail.com Attribution-NonCommercial 4.0 International License. Multi-criteria based Watershed Prioritization for Soil and Water Conservation: The case of Gotu Watershed, Awash River Basin, Ethiopia Tesfaye Wasihun Abro Department of Geography and Environmental Studies, Dire Dawa University, Dire Dawa, Ethiopia Abstract Background: In Ethiopia, soil and water conservation has often been implemented haphazardly through community mobilization during dry seasons. However, in the process, the question “which part of the watershed should be conserved first considering different criteria and which one should be the last?” is never been answered. Objective: This study was undertaken to prioritize sub-watersheds on the basis of myriad of parameters: morphometric, soil loss, socioeconomic and related criteria for soil and water conservation activities in five catchments in Gotu watershed, Awash River basin, Ethiopia. Materials and Methods: Advanced Space borne Thermal Emission and Reflection Radiometer Digital Elevation Model, Revised Universal Soil Loss Equation (RUSLE), socioeconomic and related data like population density, share of cultivated land, economic status, land pressure, potential labor force for conservation, conservation work performance, and share of unirrigated land were used for prioritization exercise. Results: It was found that Gotu watershed has seven order streams with a mean bifurcation ratio of 2.1; higher stream frequency = 6.4–7.9 streams km–2; low drainage density (0.52–2.85 km km–2) and moderate drainage texture (3.7–5.7); elongated shape (Form factor = 0.16–0.23; elongation ratio = 0.45–0.53; circulatory ratio = 0.17–0.24). Using RUSLE model, the soil loss of the study catchments ranged from 0 t ha–1yr–1in the plain area up to 197.2 t ha–1 yr–1 in the steeper, and fragile part of the watershed which made catchment two an area of severe soil erosion. Considering socioeconomic parameters, catchment five and catchment two have been shown with low and high status, respectively. Therefore, the combined result showed that catchment five that measures about 17.77 km2 out of 160.56 km2 of the total area of the study watershed requires the first priority for soil and water conservation measures. Conclusion: It is concluded that catchments with high soil loss may not usually guarantee primary attention for conservation unless the condition of socioeconomics, morphometry of catchments and related parameters simultaneously contribute to the decision-making process of conservation planners. This implies that land conservation planners should reconsider prioritization criteria of resource flows to soil and water conservation initiatives. Keywords: Morphometric parameter; RUSLE model; Socioeconomics; GIS and RS 1. Introduction A watershed is an area of land and water bounded by a drainage divide within which the surface runoff collects and flows out of the watershed through a single outlet into a larger river or other water bodies (Russo et al., 2008; Rahman et al., 2015). A watershed is an ideal unit that enables interactions among various natural resources, humans, and animals as they all make a unique geo-hydrological entity. Hence, any disruption caused to a watershed is the disturbance of the whole aspects of socioeconomic development and environmental sustainability (Woldeamlak Bewket, 2003). In countries like Ethiopia, where a major part of the population primarily depends on natural resources for livelihoods, the prevailing form of agricultural land use, and the geomorphology of the country makes them susceptible to watershed resources stress. Land degradation in the form of soil erosion (Woldeamlak Bewket, 2003; Temesgen et al., 2017) is a common problem in the highlands of Ethiopia. For example, in Awash River basin alone, up to 259 Mt year– 1 total erosion occurs contributing to high sedimentation (AwBA, 2017); and the country is losing 2 to 6.75% of agricultural GDP annually from its basins (Sonneveld, 2002). These demands conservation and or development of watershed, for sustainable production of food, fodder, and other agriculture and forest resources for immediate and long-term benefits to the farmers, community, and the basin as a whole. However, watershed development requires financial, human, organizational and other resources which may not be achieved at a time especially in larger watersheds (Mulatie Mekonnen and Assefa Melesse, 2011) making prioritization imperative. Yet, prioritization criteria need to be framed in consideration of the actual reality on the watershed. Because a watershed may be higher in soil loss but the socioeconomic condition of the society may be better off. For example, Sharma and Thakur (2016) found out that micro-watersheds with mailto:tesfayebecon@gmail.com Tesfaye East African Journal of Sciences Volume 15 (2): 115-128 116 severe soil loss were prioritized least based on socioeconomic and demographic prioritization criteria for conservation. Unfortunately, consideration of these variables in soil and water conservation prioritization is lacking in Ethiopia, much emphasis has been given to severity of soil loss as a criterion (Ermias Teferiet al., 2009; Woldeamlak Bewket and Ermias Teferi, 2009; Temesgen Gashaw et al., 2017) and community mobilization during dry season haphazardly. Therefore, this study was intended to prioritize Gotu watershed in consideration of relatively many parameters (Physical, RUSLE indices, socio-eco- demographic and related). The study area was selected for multiple reasons. First the area is prone to soil erosion problems due to overgrazing, over-cultivation, over-exploitation of the vegetation cover; and due to expansion of crop land to steep slopes. Besides, since resource managers have made significant efforts to conserve the area, it is still better to convey prioritization scientifically. The findings of this study could serve as hands on checklist for conservation planners in making knowledge-based decision or for financial and human resource flow in soil and water conservation programs. 2. Material and Methods 2.1. Study Area The study area, Gotu watershed, is part of the Awash River basin and administratively found in North Habru woreda (district), North Wollo Zone of the Amhara National Regional State, Ethiopia. Geographically, it lies between 11°34'44″ and 11°45'4″N latitude, and 39°34'11″ and 39°45'2″E Longitude (Figure 1). The area is 160.56 km2 wide, covering 9% of Habru woreda (1671.83 km2). The present complex topography of the area has been formed by a large-scale tectonic and volcanic activity and covered by Cenozoic volcanic rocks (Mohr 1971). It is composed of a rugged topography and relatively vast plain areas with elevations ranging between 1364 to 3508 meters above sea level. Figure 1. Location map of the study area: bottom left—Awash Basin, top left—Ethiopia watersheds right—Gotu watershed with catchments and elevation information. According to the traditional agro climatic classification, the study area dominantly lies within Dega (2300– 3300masl), weyna dega (1500–2300masl), and kola (1364– 1500masl) zones. Based on a data set from five meteorological stations from the year 2000–2017 the mean annual temperature was 24.5 °C; and characterized by bi-modal pattern of rainfall with a mean annual rainfall of 935.7mm for the years from 2007–2017. The major soil types in the study area are Vertisols, Cambisols, and Regosols (Amhara design and supervision works enterprise, 2011). The study watershed is inhabited by a total population of 48108 with a density of 299 persons/km2, which is far higher than the average for Amhara regional state (108.2 persons km–2) (CSA, 2007). Agriculture is the main economic activity with mixed crop-livestock production on a subsistence level. 2.2. Input Data and Methodology For the analysis of morphometry of Gotu watershed, ASTER DEM 30 meter resolution (http://earthexplorer.usgs.gov/) was used to delineate and generate the value of different parameters like bifurcation ratio (Rb), stream frequency (Fs), drainage density (Dd), drainage texture (Dt), elongation ratio (Re), circulatory ratio (Rc), form factor (Ff), compactness coefficient (Cc), basin relief (H), relief Tesfaye Multi-criteria Watershed Prioritization for Soil and Water Conservation 117 ratio (Rh), and ruggedness number (Rn). Identification of smaller geohydrological units is needed for more efficient and better targeted resource management programs (Sharma and Thakur, 2016). Based on this, Gotu watershed has been classified in to five catchments using ArcSWAT in ArcGIS 10.4 which were coded as: Catchment one = CI; Catchment two = CII; Catchment three = CIII; Catchment four = CIV and Catchment five = CV. Then, after creating shape file of watersheds, DEM of each catchment was masked and filled the sink; flow accumulation threshold value of 1000 was used and the result stream networks were cross checked with toposheet (1:50,000 scale) of Mersa to have approachable results and generated streams of five catchments (Figure 4). On the other hand, for soil loss estimation using RUSLE model, input data like rainfall data (2008–2017) from national meteorological agency, soil data, ASTER DEM 30 m, and LULC were used. ERDAS 9.2 was used for layer stacking of Landsat 8 bands for LULC classification but other RUSLE indices outputs were generated using ArcGIS 10.4 software. Table 1. Parameters and prioritization scheme as adapted from Sharma and Thakur (2016). Morphometric parameter Definition/formula Priority condition References Bifurcation ratio (Rb) Rb = Nu/Nu+1 Higher the value, More the Priority Schumm (1956) Stream frequency (Fs) Fs = Nu/A Higher the value, More the Priority Horton (1932) Drainage density (Dd) Dd= Lu/A Higher the value, More the Priority Horton (1932) Drainage texture (Dt) Dt = Nu/P Higher the value, More the Priority Horton (1945) Compactness Coefficient (Cc) Cc = 0.2821P/A0.5 Higher the value, More the Priority Horton (1945) Form factor (Ff) Ff=A/Lb2 Lower the value, More the Priority Horton (1932) Elongation ratio (Re) Re=2/Lb*(A/ π)0.5 Lower the value, More the Priority (Schumm ( 1956) Circulatory ratio (Rc) Rc = 4 * π* A/P2 Lower the value, More the Priority Miller (1953) Basin relief (H) HE-LE Higher the value, More the Priority Hadley and Schumm (1961) Relief ratio (Rh) Rh = H/Lb Higher the value, More the Priority Schumm (1956) Ruggedness no. (Rn) Rn = H * D Higher the value, More the Priority Melton (1957) Soil loss A=R.K.L.S.C. P Higher the Average Annual Soil Loss Risk, More the Priority Wischmeier and Smith (1978) Economic demographic parameters* Population density (Pd) Low, medium, high The higher the density, higher the Priority Badar et al., (2013) Average Economic status (ES) Low, medium, high The lower the status, higher the Priority Badar et al., (2013) Cultivated land (CL) Calculated from the generated LULC The lower the percentage, higher the Priority Satellite image analysis (http://earthexplorer.us gs.gov/) Unirrigated land (Unirr.) Very low, moderate, high, very high Higher the unirrigated land, more the Priority Land Pressure (LP) Low, medium, high Higher the pressure, more the Priority Training received (Tr) Number of trained farmers The lower the number in relative to other catchment, the higher the priority Badar et al., (2013) Potential labor force (PLF) Low, medium, high The lower the productive age group, the higher the priority Badar et al., (2013) WSD performance rank (PR) 1st, 2nd, 3rd, 4th, 5th The lower the rank, the more the priority Other parameters Ground cover Low, moderate, high Higher the NDVI, lower the priority Rouse (1973) STI (Flow Acc./22.13)0.6 *(Sin β/0.0806)1.3 Higher the value, more the Priority SPI (Flow Acc. + 1)* (tan β) Higher the value, more the Priority Moore et al., (1991), Florinsky (2012) TWI ln (Facc/ tan β) Lower the value, more the Priority Beven and Kirkby, (1979), Florinsky (2012) Note: A = average annual soil loss potential; R = rainfall erosivity; K = soil erodibility; L = slope length; S = slope steepness; C = land cover management factor; P = conservation practices factors; NDVI = normalized difference vegetation index; WSD = watershed; and PLWSDD = potential labor force for watershed development. *Analyzed on the basis of secondary data from Habru Woreda Agriculture and Rural Development Office and woreda Socioeconomic development and Finance Office (2017). Tesfaye East African Journal of Sciences Volume 15 (2): 115-128 118 Moreover, socio-eco-demographic data regarding the various parameters like population density, economic status, cultivated land, unirrigated land, participation in watershed conservation, training received related to watershed development, potential labor force, and land pressure for all the catchments of study watershed were collected from the Habru Woreda Agricultural Office and personal interviews with natural resource and rural development experts. Furthermore, other parameters such as ground cover condition, sediment transport index (STI), stream power index (SPI) and topographic wetness index (TWI) were also considered for prioritization exercise. Das et al. (2012) indicated that criteria for watershed prioritization are subjective in nature. Therefore, in this study, the basis for assigning weightage to different themes (morphometric, socio-eco-demographic, and mixed- RUSLE factors) was in consideration of which parameter need more attention in relation to prioritization for soil and water conservation. Based on this, soil loss using RUSLE model has given more weight (5) because human interventions (C and P factors) have more influence on soil loss plus physical indices are also embedded in the model. Then, the compound value of all the parameters was generated using the following weighted mean formula. Wx− = W1∗X1+W2∗X2+W3∗X3…WnXn W1+W2+W3…Wn Where, Wx– = Weighted mean; W1 = Weight one; W2 = Weight two ….; and X’s = parameters. Finally, based on average value of these parameters, catchments having the least rating value were assigned the highest priority; the next value was assigned second priority and so on. Figure 2. Flow chart of methodology. 3. Results and Discussion 3.1. Morphometric Analysis Using Strahler (1957) system of stream ordering, Gotu watershed has seven order main streams, each order streams in each catchment having a varying number of streams (Figure 3). The study area has a total of 1127 streams, of which 562 were first order; 278 were second order; 130 were third order; 84 were fourth order; 38 were fifth order; 34 were sixth order; and 1 was seventh order streams with the corresponding stream length (km) of 123.5, 50.4, 30.42, 10.07, 4.68, 4.92 and 5.04, respectively (Table 2). Tesfaye Multi-criteria Watershed Prioritization for Soil and Water Conservation 119 Figure 3. Stream orders of Gotu sub-watersheds produced from ASTER DEM. The study watershed has an area of 160.56 km2 with 675 total streams with a total stream length of 219.12 km. This means that there are about 7.04 streams in each km2 area and an average of 1.36 km stream length per km2 area. Catchment (C) wise, catchment four and catchment two are the highest in terms of stream length with 58.29 km, and 51.35 km, respectively (Table 2). The morphometric parameters (Table 1) have been used for prioritizing watersheds. The linear parameters such as bifurcation ratio, stream frequency, drainage density, and drainage texture have a direct relationship with soil loss; the higher the value, the more is the soil loss. In contrast, shape parameters such as circularity ratio, elongation ratio, form factor, and compactness coefficient have an inverse relationship with erodibility. The intensity of dissection (expressed as bifurcation ratio) influences the discharge and flooding. In the present study, out of the five catchments, only catchment one has bifurcation ratio (Rb) of 3.0 (Table 2) that qualify Horton’s (1945) natural drainage characteristics of Rb value (3.0 to 5.0). Horton (1945) noted that the value of stream frequency depends on the total number of streams and the corresponding basin area. In the present study, stream frequency varies from 6.4 to 7.9 streams per square kilometer. In general, the higher the drainage density, the less the density of vegetation cover (Table 3 and Figure 6) and the more impermeable soil and rock surface which lets the movement of overland flow of runoff and hence reflecting the presence of enhanced soil erosion. The drainage density varies from 0.52 to 2.85 km km–2 reflecting comparatively high permeable surface and lower soil erosion rate. 3.2. Estimating Soil Loss The soil loss in Gotu sub-watershed was estimated using RUSLE model–an efficient and quite reliable method of predicting soil loss as it considers both the physical and anthropogenic factors which are responsible for causing soil erosion (Wischmeier and Smith, 1978). Catchment wise, a moderate soil loss was observed in three catchments (CV, CIII and CIV) while the remaining two catchments experienced a severe soil erosion (Figure 5). Tesfaye East African Journal of Sciences Volume 15 (2): 115-128 120 Table 2. Linear aspects of morphometrical parameters. Catchment Parameter Stream order 1st 2nd 3rd 4th 5th 6th 7th 1 No. of streams 134 62 31 34 5 Stream Length/km 18.8 11.52 1.49 5.75 0.12 Ave. length/km 0.99 0.96 0.37 1.15 0.12 Basin Length (km) 11.24 Mean Rb 3 2 No. of streams 133 68 33 26 Stream Length/km 30.45 7.69 11.15 2.06 - Ave. length/km 1.12 0.64 0.85 1.03 - Basin Length (km) 13.02 Mean Rb 1.7 3 No. of streams 125 65 32 15 13 1 Stream Length/km 13.61 2.8 0.13 0.012 4.56 0.35 Ave. length/km 1.04 0.93 0.07 0.012 0.41 Basin Length (km) 10.49 Mean Rb 1.8 4 No. of streams 105 55 23 9 13 11 Stream Length/km 32.28 18.54 7.47 - - 4.92 Ave. length/km 1.19 1.09 0.93 - - Basin Length (km) 13.88 Mean Rb 1.7 5 No. of streams 65 28 11 7 23 Stream Length/km 28.41 9.85 10.18 2.25 - 4.69 Ave. length/km 1.05 0.98 0.78 2.25 - Basin Length (km) 14.34 Mean Rb 2.4 Tesfaye Multi-criteria Watershed Prioritization for Soil and Water Conservation 121 Figure 4. Streams in Gotu catchments derived from ASTER DEM. Tesfaye East African Journal of Sciences Volume 15 (2): 115-128 122 Table 3. Morphometric parameters. Catchment A (km2) P (km) Fs Dd Dt Ff Re Rc Cc H (km) Rh Rn CI 40.08 46.07 6.6 0.94 5.7 0.23 0.53 0.23 2.05 1.88 0.14 2.41 CII 27.41 42.71 7.9 1.87 5.1 0.21 0.52 0.19 2.30 0.7 0.060 0.96 CIII 40.44 45.69 6.4 0.52 5.6 0.21 0.50 0.24 2.02 2.02 0.145 2.9 CIV 34.9 45.88 7.1 1.67 5.4 0.17 0.46 0.20 2.19 2.19 0.063 1.32 CV 17.77 35.44 7.5 2.85 3.7 0.16 0.45 0.17 2.37 0.61 0.05 0.73 Note: A = Area; P = Perimeter; Fs = Stream frequency; Dd = Drainage density; Dt = Drainage texture; Ff = Form factor; Re = Elongation ratio; Rc = Circulatory ratio; Cc = Constant of channel maintenance; H = Basin relief; Rh = Relief ratio; and Rn = Ruggedness number. The later may be attributed to the physical relief (average of H, Rh and Rn is highest) (Table 3) which was consistent with previous studies by Woldeamlak Bewket and Ermias Teferi (2009); Mulatu Mekonnen and Assefa Melese (2011); Temesgen et al. (2017); Birhan Asmame and Assefa Abegaz (2017), and Gezahegn et al. (2018). However, CIII showed higher physical relief but moderate soil loss. This may be due to better land use and land cover conditions (Figure 5). Figure 5. Soil loss computed using RUSLE model. Tesfaye Multi-criteria Watershed Prioritization for Soil and Water Conservation 123 Table 4. Soil loss (t ha–1year–1) of each catchment. Catchment Soil loss (t ha–1yr-1) Area (km2) CI <3.5 19.0 3.6–22 7.0 22.7–69.9 14.08 CII <10.8 13.6 10.9–64.2 6.0 64.3–197 7.81 CIII <2.09 19.0 2.09–7.25 13.4 7.26–19.07 8.04 CIV <2.7 21.8 2.7–10.2 4.6 10.3–32 8.5 CV <0.58 10.1 0.58–2.19 6.2 2.2–7.79 1.4 Note: CI = Catchment one; CII = Catchment two; CIII = Catchment three; CIV = Catchment four; and CV = Catchment five. 3.3. Socio-economic and Demographic Analysis The socioeconomic and demographic data (Table 5) revealed that catchment four and catchment five were the highest and the lowest in terms of population density and economic status, respectively. The higher population density is attributed to the presence of congested town (Girana) largest market center pulling the surrounding population, suitable climate, and much irrigable land and thus the population has relatively better economic status. Conversely, the lowest population density and economic status is found in the eastern part of the watershed i.e. catchment five which lies near to Afar region in an area of hot temperature and low rainfall amount, low irrigation access, and relatively steep terrain. According to a report of Habru Woreda Agriculture Office (2017), potential labor force for working to develop the watershed was identified from all 35 rural kebeles in the woreda, and training was offered to farmers in relation with watershed development activities. Based on this, the greater the number of trained farmers in a catchment, the lower the priority for soil and water conservation expecting that the trained farmers may mobilize other farmers for participation in conservation activities. Table 5. Socioeconomic and demographic data with respect to different catchments. Criterion Catchment CI CII CIII CIV CV Population density 224 273 280 296 199 Economic status 4th 2nd 3rd 1st 5th Potential labor force 1820 3206 3102 3312 1815 Cultivated land (%)* 38.9 47.8 30.9 42.7 43.6 Unirrigated land Low Medium High Very low Very high Land Pressure High VH Low Low High Number of farmers received training 305 313 281 432 190 WSD conservation performance rank 8 1 12 16 22 Note: CI = Catchment one; CII = Catchment two; CIII = three; CIV=Catchment four; and CV= Catchment five. Analyzed on the basis of secondary data from Habru Woreda Agriculture and Rural Development Office and woreda socioeconomic development and Finance Office (2017). *Analyzed from Landsat 8 image. Thus, the results in Table 5 showed that catchment five has lower number of trained farmers (190) relative to other catchments. Concerning cultivated land, the analyzed image of Landsat 8, the analysis showed that about 64.0161 km2 (39.87%) area of total land is under cultivation in the five catchments of Gotu watershed (Figure 6). Catchment wise, almost half (47.8%) of the total area is under crop cultivation in CII. It observed that crop cultivation is a dominant activity and has a significant share as an economic pursuit in the remaining catchments. The woreda agriculture and rural development office (2017) report revealed that watershed development work performance of each kebele is usually evaluated quarterly by experts using field observations in each Tesfaye East African Journal of Sciences Volume 15 (2): 115-128 124 catchment area and by releasing performance ranking. Accordingly, population in catchment two and catchment one has a better record in watershed development activities and catchment five with the lowest performance (Table 5). Catchment two and catchment one are the leading ones in terms of pressure exerted on the land due to the presence of cross over main asphalt road in these catchments. Farmers usually sell fuelwood collected from the available bush lands in the highlands to make ends meet. Figure 6. Land use land cover of Gotu catchments derived from Landsat 8. 3.4. Other Parameters Sediment transport index value ranged from 1.88 in CV to 5.22 in CIII. The highest value was associated with steep slope and ridges which may contribute to sediment transportation and consequent soil erosion and degradation. The sediment transport index result has relative similarity with the relief parameters i.e. the higher sediment transport index value is found in areas with steep slopes (Tables 3 and 6). Table 6. Other parameters (derived from DEM and Landsat-8 OLI). Parameter Catchment CI CII CIII CIV CV Sediment transport index 4.86 2.03 5.22 2.30 1.88 Stream power index 31.58 26.78 35.71 14.07 64.91 Topographic wetness index 15.53 16.93 15.09 15.68 18.04 NDVI as a proxy for ground cover condition 0.37 0.43 0.38 0.35 0.45 Note: CI = Catchment one; CII = Catchment two; CIII = three; CIV = Catchment four; and CV = Catchment five. Similarly, high stream power index values represent areas on the landscape where high slopes and flow accumulations exist and thus areas where flows can concentrate with erosive potential. The higher the power of stream, the greater the probability of vulnerable topsoil washing down and contributing strongly to the land degradation process, and then transporting soil material and sediment to the plain areas. So, in the present study, the western and eastern parts of the watershed (CV = 64.91) have high stream power which may be associated with more inflow of water from the upper areas (increase in the stream orders i.e. seventh order stream is entirely found in CV). This implies that more attention should be given to soil conservation due to high probability of erosion. Topographic wetness index, on the other hand, was used to show wetness of an area and related to slope of the area i.e. water tend to accumulate in flat area than steep slope area and hence the more the topographic wetness index, the flatter the slope and hence more flow accumulation and wetness (Beven and Kirkby, 1979). Tesfaye Multi-criteria Watershed Prioritization for Soil and Water Conservation 125 3.5. Prioritization of Catchments All catchments in Gotu watershed have been prioritized by considering the weighted average of different parameters–morphometric, soil loss, socio-eco- demographic and others (Table 7). The value of the comprehensive weightage of all the parameters varied from 2.71 (lowest) to 3.3 (highest). Out of the five catchments, CV was assigned as high priority while CIV as low priority. Catchments with high priority indicate the need of soil and water conservation attention to them before other catchments. The least prioritized catchment, i.e., CIV has better socio-economic status, and lower aggregate morphometric ranking with medium soil loss priority (Table 7) whereas CV– given high priority which was characterized as lower relative soil loss and lower socio-eco-demographic condition. However, in terms of individual parameters, catchments with high priority in soil loss ranked the second least in terms of priority in aggregate morphometric indices. But Hlaing et al. (2008) has found sub watersheds with high priority in soil erosion also ranked in the morphometric parameters. Table 7. Comprehensive ranking of catchments. Aspect Parameter s Weight (W) Ranking C I W*C I CI I W*CI I CII I W*CI II CI V W*CI V C V W*C V Morphometric Rb 3 1 3 4 12 3 9 4 12 2 6 Fs 3 4 12 1 3 5 15 3 9 2 6 Dd 3 4 12 2 6 5 15 3 9 1 3 Dt 3 1 3 4 12 2 6 3 9 5 15 Ff 3 5 15 3 9 3 9 2 6 1 3 Re 3 5 15 4 12 3 9 2 6 1 3 Rc 3 4 12 2 6 5 15 3 9 1 3 Cc 3 2 6 4 12 1 3 3 9 5 15 H 3 4 12 3 9 1 3 2 6 5 15 Rh 3 3 9 1 3 4 12 2 6 5 15 Rn 3 3 9 1 3 4 12 2 6 5 15 Soil Soil loss 5 2 10 1 5 4 20 3 15 5 25 Socio-eco- demographic Pd 4 4 16 3 12 2 8 1 4 5 20 Es 4 2 8 4 16 3 12 5 20 1 4 Cl 4 2 8 5 20 1 4 3 12 4 16 Unirr 4 4 16 3 12 2 8 5 20 1 4 LP 4 2 8 1 4 4 16 4 16 2 8 PLF 4 2 8 4 16 3 12 5 20 1 4 WPR 4 4 16 5 20 3 12 2 8 1 4 Tr 4 3 12 4 16 2 8 5 20 1 4 Others NDVI 4 4 16 2 8 3 12 5 20 1 4 STI 3 2 6 4 12 1 3 3 9 5 15 SPI 3 3 9 4 12 2 6 5 15 1 3 TWI 3 2 6 4 12 1 3 3 9 5 15 Sum 83 247 252 232 275 225 WX- 2.9 3.0 2.79 3.3 2.71 Rank 3rd 4th 2nd 5th 1st Note: Pd = Population density; Es = Economic status; Cl = Cultivated land; Unirr = Un-irrigated land; LP = Land pressure; PLF = Potential labor force; WPR = Watershed conservation activities performance rank; and Tr = Training received related to conservation. In the present study, it was also observed that catchments with high soil loss, low economic status, high population density, low potential labor force, high unirrigated land, high land pressure, and low watershed conservation performance were prioritized first against catchments with lower soil loss and better socioeconomic condition. This result is consistent with the findings of Sharma and Thakur (2016). In their finding micro watersheds with high soil loss risk, high population size, low agricultural land, high relief ratio, lower economic status was given high priority for soil and water conservation planning while watersheds with low lying area, less areas under steep slope, and low drainage density were assigned to low priority. Similarly, Vittala et al. (2008) found out that sub watersheds having population with low economic status were given high priority while sub watersheds with better natural resource base were given prioritized least. The aggregate results of prioritization may have their own limitation and, in some cases, inconsistent results may be observed because of the subjective nature of weight assignment to the parameters. Therefore, the author acknowledges that the methodology has inherent Tesfaye East African Journal of Sciences Volume 15 (2): 115-128 126 limitations and may result in assignment of erroneous weight to the parameters. Again, this study was not able to include other important variables affecting prioritization for soil and water conservation. Finally, there may also be some overlapping parameters. 4. Conclusion This study has demonstrated that Gotu watershed has seven order streams with a mean Rb of 2.1; low drainage density (0.52–2.85 km km–2) and moderate drainage texture (3.7–5.7); elongated shape (Ff = 0.16–0.23; Re = 0.45–0.53) and 74.81% of Gotu watershed have a less than 0.145 and 2.9 of relief ratio and ruggedness number, respectively. These proved that the watershed has relatively less dissected terrain features; low and extended flood flow and dominance of low sloped terrain. Soil erosion was very severe (64.2–197.2 t ha– 1yr–1) in CII as per RUSLE model. The results of socioeconomic and demographic characteristics, on the other hand, showed that out of the five catchments of Gotu watershed, CV was characterized by low economic status, labor force, watershed work and related training participation which made this catchment feasible for primary conservation. Finally, the compound results have revealed that relatively CV having an area of 17.77 km2 (11.06%) of Gotu watershed, should get the first priority for soil and water conservation. The results of this study could be useful for watershed planners and managers towards implementing various water and soil conservation measures in the study area. It is deduced that catchments with high soil loss may not usually guarantee primary attention for conservation unless the condition of socioeconomics, morphometry of catchments and related parameters are considered simultaneously and contribute to the decision making process of conservation planners. Similar studies should be made to understand the condition of different watersheds in Ethiopian highlands versus resource flow for soil and water conservation work. 5. Acknowledgments The author thanks the Amhara National Regional State – Bureau of Agriculture and Rural Development, Ethiopian Metrological Agency, and Habru Woreda Agricultural Bureau for providing soil, meteorological, and annual report data, respectively. Also, the author thanks the United States (USA) geological survey for providing remote sensing data free of charge. 6. References AwBA (Awash Basin Authority). 2017. Awash River Basin Integrated Participatory Watershed Management Strategic and Detail Action plan. Awash Basin Authority, Addis Ababa, Ethiopia. Pp. 147. Ayele Almaw, Hiroshi, Y., Katsuyuki, S., Nigussie Haregewein and Kifle Weldaregay. 2017. Quantitative analysis and implications of drainage morphometry of the Agula watershed in the semi-arid northern Ethiopia. Applied Water Science, 7(7): 3825–3840. Badar, B.K, Romshoo, S.A. and Khan, M.A. 2013. Integrating biophysical and socioeconomic information for prioritizing watersheds in a Kashmir Himalayan lake: A remote sensing and GIS approach. Environmental monitoring and assessment, 185(8): 6419–45. Doi10.1007/s10661-012-3035-9. Beven, K.J. and Kirkby, M.J. 1979. A physically based, variable contributing area model of basin hydrology. Hydrolological Sciences Journal, 24: 43– 69. Birhan Asmame and Assefa Abegaz. 2017. Estimating soil loss for sustainable land management planning at the Gelana sub-watershed, northern highlands of Ethiopia. International Journal of River Basin Management, 16(1): 41–50. CSA (Central Statistical Agency). 2007. Summary and Statistical Report of the 2007 Population and Housing Census Results. Federal Democratic Republic of Ethiopia Population Census Commission, Addis Ababa, Ethiopia. Das, A., Milan, M., Bhaskar, D. and Asim. R.G. 2012. Analysis of drainage morphometry and watershed prioritization in Bandu watershed Purulia, West Bengal through remote sensing and GIS. International Journal of Geomatics and Geosciences, 2(4): 995–1013. Ermias Teferi, Dagnachew Legese, Belay Semane and Woldeamlak Bewket. 2009. Prioritization of micro-watersheds on the basis of soil erosion risk in the source region of the Blue Nile River Using RUSLE Model, Remote Sensing and GIS: Case study in the Muga watershed. Pp.50– 65. In: Fisum Hagos, Minale Kassie, TsehayWoldegiorgis, Yesuf Mohammednur, Zenebe Gebreegziabher (eds.). Proceedings of Collaborative National Workshop on Sustainable Land Management Research and Institutionalization of Future Collaborative Research, 8–9 May, 2009, Addis Ababa. Florinsky, I.V. 2012. Digital terrain analysis in soil science and geology. Academic Press, Kidlington, Oxford OX5 1GB, UK, 379. Gezahegn Weldu, Anteneh Derribew, Solomon Tekalign and Reddy, U. 2018. Spatial modeling of soil erosion risk and its implication for conservation planning: The case of the Gobele watershed, East Hararghe Zone, Ethiopia. Land, 7: 25, doi:10.3390/land7010025. Hadley, R.F. and Schumm, S.A. 1961. Sediment sources and drainage basin characteristics in upper Cheyenne River basin. Water Supply Paper 1531- B, U.S. Geological Survey. Hlaing, K.T., Haru, Y.S. and Aye, M.M. 2008. Using GIS-based distributed soil loss modeling and morphometric analysis to prioritize watershed for soil conservation in Bago river basin of Tesfaye Multi-criteria Watershed Prioritization for Soil and Water Conservation 127 Lower Myanmar. Front. Earth Science, 2(4): 465– 478. Horton, R.E. 1932. Drainage basin characteristics. Transactions. America Geophysical Union, 13: 350–361. Horton, R.E. 1945. Erosional development of streams and their drainage basins; hydrological approach to quantitative morphology. Bulletin of Geological Society of America, 56 (3): 275–370. Melton, M.A. 1957. An analysis of the relation among elements of climate, surface properties and geomorphology. Technical Office of National Research, project NR Columbia University. Miller, V.C. 1953. A quantitative geomorphic study of drainage basin characteristics in the Clinch mountain area. Technical Report 3. Department of Geology, Columbia University, New York. Mohr, P.A. 1971. The Geology of Ethiopia. Haile Sellassie I University Press, Addis Ababa, Ethiopia. Moore, I.D, Grayson, R.B. and Ladson, A.R. 1991. Digital terrain modelling: A review of hydrological, geomorphological, & biological applications. Hydrological Processes, 5(1): 3–30. Mulatie Mekonen and Assefa Melesse. 2011. Soil erosion mapping and hotspot area identification using GIS and Remote Sensing in Northwest Ethiopian Highlands, near Lake Tana. Pp. 207–224. In: Assefa Melesse (ed.). Nile River Basin. Springer, The Netherlands. Rahman, A.S, Ajeez, S.A., Aruchamy, S. and Jegankumar, R. 2015. Prioritization of sub watershed based on morphometric characteristics using fuzzy analytical hierarchy process and geographical information system– A study of Kallar watershed, Tamil Nadu. International Conference on Water Resources, Coastal and Ocean Engineering, Aquatic Procedia, 4: 1322– 1330. Rouse, J.W. 1973. Monitoring the vernal advancement and retrogradation (Green wave effect) of natural vegetation. Type II report for Period September 1973 - March 1973. Goddard Space Flight Center Green Belt, Maryland. Russo, R.C., Rashleigh, B. and Ambrose, R.B. 2008. Watershed management in the United States. Pp. 173–198. In: Gönenç, I.E., Vadineanu, A., Wolflin, J.P. and Russo, R.C. (eds.). Sustainable Use and Development of Watersheds. Springer, The Netherlands. Schumn, S.A. 1956. Evolution of drainage systems and slopes in badlands at Perth Amboy, New Jersey. Geological Society of America Bulletin, 67: 597–646. Sharma, D.D. and Thakur, B.R. 2016. Prioritization of micro watersheds in giri catchment for conservation and planning. Transactions, 38(2): 267–280. Smith, K.G. 1950. Standards for grading textures of erosional topography. American Journal of Science, 248(9): 655–668. Sonneveld, B.G. 2002. Land under pressure: The impact of water erosion on food production in Ethiopia. PhD Dissertation, Vrije Universiteit, The Netherlands. Strahler, A.N. 1957. Quantitative analysis of watershed geomorphology. Transactions American Geophysical Union, 38(6): 913–920. Strahler, A.N. 1964. Quantitative geomorphology of drainage basins and channel networks. Pp. 439– 476: In: Chow, V. (Ed.). Handbook of applied hydrology. McGraw Hill Book Company, New York. Temesgen Gashaw, Taffa Tulu and Mekuria Argaw. 2017. Erosion risk assessment for prioritization of conservation measures in Geleda watershed, Blue Nile basin, Ethiopia. Environmental System Research, 6(1) https://doi.org/10.1186/s40068- 016-0078-x. Vittala, S.S, Govindaiah, S. and Gowda, H.H. 2008. Prioritization of sub-watersheds for sustainable development and management of natural resources: An integrated approach using remote sensing, GIS and socio-economic data. Current Science, 95(3): 345–354. Wischmeier, W.H. and Smith, D.D. 1978. Predicting rainfall erosion losses–A guide to conservation planning. Agricultural Handbook, Number 537 USDA. Woldeamlak Bewket. 2003. Towards integrated watershed management for resource conservation in Chemoga watershed, Northwestern highlands of Ethiopia. Tropical Resource Management Papers, No. 44. ISBN 90- 6754-708-5. Woldeamlak Bewket and Ermias Teferi. 2009. Assessment of soil erosion hazard and prioritization for treatment at the watershed level: case study in the Chemoga watershed, Blue Nile basin, Ethiopia. Land Degradation and Development, 20(6): 609–622. https://doi.org/10.1186/s40068-016-0078-x https://doi.org/10.1186/s40068-016-0078-x Tesfaye East African Journal of Sciences Volume 15 (2): 115-128 128