Like many other lakes in the world, the interconnected Abaya and Chamo lakes in the Southern Main Ethiopian Rift are affected by rapid sediment accumulation. Although land degradation is a well-known issue in this part of the African continent, the main sediment sources, their spatial distribution and interaction in the Abaya–Chamo lakes’ basin have not yet been documented. Here, we present a systematic inventory, characterization, and spatial analysis of landslides and gullies as concentrated sediment sources, for four representative river catchments impacted by landscape rejuvenation. Using Google Earth imagery and field surveys, we mapped with high accuracy a total of 7336 gullies and 430 landslides. Recent landslides observed during the last decade were few, small and shallow, and appear to have played a minor role in the current sediment dynamics. Large landslides are old and inactive. Although they do not contribute to the current sediment budget, they contribute indirectly to landscape dynamics by favoring the occurrence of gullies. Overall, large percentages of severe to extremely degraded areas of gully erosion are located in rejuvenated landscapes, especially at the level of the old landslides. Many active gullies are connected to the river network, as such acting as the source of sediment. Our analysis is a step towards understanding the nature and control of anthropic activities on sediment production in the region. We also highlight the importance of considering the interactions between sediment sources and the connectivity of the geomorphological system.

Quantification of suspended sediment dynamics is a proxy for the intensity of soil erosion processes in a catchment. However, direct measurement of suspended sediments is difficult. Long-term-based discharge-suspended sediment relationships often fail to provide sufficiently accurate results due to the temporal variations in sediment supply. This study evaluates temporal variations in suspended sedimentdischarge relations for six stations in two selected catchments within the Gamo Highlands in the southern Ethiopian Rift Valley. During a three-year period (2018-2020), discharge was monitored at a 10-min interval using a TD-diver, while 1938 samples of suspended sediment concentration were taken at specific events. Observed suspended sediment concentration varied between 0.04 and 130 kg m À3 for discharge ranging between quasi-zeros to 339 m 3 s À1 . The results show that pooled annual datasets lead to unsatisfying sediment rating curves and large errors in estimated sediment load. However, stratification of the data into three rainy season periods strongly improved the performance of the sediment rating curves. The development of a vegetative cover throughout the rainy season largely controls sediment supply from hillslopes to the river channel and hence the shape of the sediment rating curves. Furthermore, localized bank erosion and gully head dynamics also lead to important interannual changes in sediment rating curves within one river system, as well as between different rivers. Thus, attention should be given to possible errors due to high temporal and spatial variability while using discharge-suspended sediment relations to quantify sediment load.

No abstract

No abstract