In the drought-prone Upper Blue Nile River (UBNR) basin of Ethiopia, soil erosion by water results in significant consequences that also affect downstream countries. However, there have been limited comprehensive studies of this and other basins with diverse agroecologies. We analyzed the variability of gross soil loss and sediment yield rates under present and expected future conditions using a newly devised methodological framework. The results showed that the basin generates an average soil loss rate of 27.5tha-1yr-1 and a gross soil loss of ca. 473Mtyr-1, of which, at least 10% comes from gully erosion and 26.7% leaves Ethiopia. In a factor analysis, variation in agroecology (average factor score=1.32) and slope (1.28) were the two factors most responsible for this high spatial variability. About 39% of the basin area is experiencing severe to very severe (>30tha-1yr-1) soil erosion risk, which is strongly linked to population density. Severe or very severe soil erosion affects the largest proportion of land in three subbasins of the UBNR basin: Blue Nile 4 (53.9%), Blue Nile 3 (45.1%), and Jema Shet (42.5%). If appropriate soil and water conservation practices targeted ca. 77.3% of the area with moderate to severe erosion (>15tha-1yr-1), the total soil loss from the basin could be reduced by ca. 52%. Our methodological framework identified the potential risk for soil erosion in large-scale zones, and with a more sophisticated model and input data of higher spatial and temporal resolution, results could be specified locally within these risk zones. Accurate assessment of soil erosion in the UBNR basin would support sustainable use of the basin's land resources and possibly open up prospects for cooperation in the Eastern Nile region.

This paper reviews Ethiopia’s experience and research progress in past soil and water conservation (SWC) efforts and suggests possible solutions for improvement. Although indigenous SWC techniques date back to 400 BC, institutionalized SWC activity in Ethiopia became significant only after the 1970s. At least six national SWC-related programs have been initiated since the 1970s and their focus over time has shifted from food relief to land conservation and then to livelihoods. The overall current soil erosion rates are highly variable and large by international standards, and sheet, rill, and gully erosion are the dominant processes. The influence of human activities on the landscape has traditionally been deleterious, but this trend seems to have recently reversed in some parts of the country following the engagement of the communities in land management. The efficiency of SWC measures show mixed results that are influenced by the type of measures and the agro-ecology under which they were implemented; in general, the relative performance of the interventions is better in the drylands as compared with humid areas. Methodological limitations also occur when addressing the economic aspects related to benefits of ecosystem services and other externalities. Although farmers have shown an increased understanding of the soil erosion problem, SWC efforts face a host of barriers related to limited access to capital, limited benefits, land tenure insecurity, limited technology choices and technical support, and poor community participation. In general SWC research in Ethiopia is fragmented and not comprehensive, mainly because of a lack of participatory research, field observations, and adoptable methods to evaluate impacts. A potentially feasible approach to expand and sustain SWC programs is to attract benefits from global carbon markets. Moreover, a dedicated institution responsible for overseeing the research–extension linkage of SWC interventions of the country should be established.

Understanding the effect of land use and sustainable land management (SLM) practices on runoff and soil loss (SL) is essential for adopting suitable strategies to control soil erosion. The purpose of this study was to analyze runoff and SL from different land use types and evaluate the effectiveness of different SLM practices through monitoring runoff and sediment from 42 runoff plots (30 m × 6 m) in different agro-ecologies of the Upper Blue Nile basin of Ethiopia. Four treatments for croplands (control, soil bund, Fanya juu, and soil bund reinforced with grass) and three treatments for non-croplands (control, exclosure, and exclosure with trenches) were investigated during the rainy seasons. The results showed that runoff and SL varied greatly depending on agro-ecology, land use type, and SLM practice. Seasonal runoff ranged from 52 to 810 mm in 2015 and 37 to 898 mm in 2016, whereas SL ranged from 0.07 to 39.67 t ha-1 and 0.01 to 24.70 t ha-1. The highest rates were observed from untreated grazing land in the midland agro-ecology, largely because of heavy grazing and the occurrence of intense rain events. Runoff and SL were both significantly lower (P < 0.05) in SLM plots than in control plots. On average, seasonal runoff was reduced by 11% to 68%, and SL by 38% to 94% in SLM plots. Soil bund reinforced with grass in croplands and exclosure with trenches in non-croplands were found to be the most effective SLM practices for reducing both runoff and SL. Integrating structural and vegetative measures was therefore found to be the best way to control soil erosion and its consequences. Additional investigation is needed in consideration of ecological succession and other possible effects of these types of integrated measures, for example, the effects on soil properties, biomass, and biodiversity.