Long-term effectiveness of indigenous and introduced soil and water conservation measures in soil loss and slope gradient reductions in the semi-arid Ethiopian lowlands

Abstract

Across the world, on-site and off-site effects of soil erosion accelerated by anthropogenic land-use/cover changes have been partly offset by the implementation of soil and water conservation (SWC) measures. However, studies evaluating the effectiveness of SWC over longer periods are currently rare. The dimensions of the terraced fields and SWC were measured at 242 sample fields that had slope gradients between 2.0 and 6.5% in three villages in the semi-arid Ethiopian Rift Valley (Ethiopian lowlands). In the study area, both indigenous contour weed hedgerows (denber) and introduced bund techniques, such as soil and stone bunds, are observed as major SWC measures. The establishment of the oldest SWC dated back to 65 years ago. Long-term effectiveness of the SWC was assessed in terms of the reduction in soil loss rates and slope gradient change of the terraced fields. Long-term quantitative assessment of SWC across the world has been conducted only in highlands; however, no case studies were conducted in lowlands. It is partly because the model that assessed the long-term effectiveness of SWC measures for the highlands appears to be inapplicable to lowlands. In the lowland model used in this study, the mid-point of the field was connected with that of the upper and lower fields by straight lines, which were assumed to represent the original slopes before the SWC establishment. The difference in slope gradient between the present slope and the original slope of the field without SWC was used for the estimates of slope reduction. The measured mass of the soil loss zone of the field was used to determine the long-term mean intra-field soil loss rate, while that of the sediment deposition zone was used to determine the long-term mean intra-field sediment accumulation rate behind the lower SWC. The mean net soil loss from the terraced fields was 19.6 Mg ha−1 yr−1, which was higher than that in the semi-arid northwestern highlands (Tigray); however, that from the stone bund fields was 13.7 Mg ha−1 yr−1, which was lower than that in Tigray. The soil and stone bunds retained 52–72% of the intra-field soil loss, compared to 34% retained by the denber. Slope gradients were reduced annually by 0.03% grade for the denber fields and 0.17–0.28% grades for the soil and stone bunds fields. Uncertainty analysis proved these estimates had relative uncertainty of <10% in general. All the SWC measures investigated in this study were effective in reducing soil loss from cropland and in reducing slope gradients of cropland in Ethiopian lowlands. Farmers’ efforts to place eliminated weeds on the lower hedgerows and bunds and a subsequent regeneration mechanism of weed hedgerows sustain the long-term effectiveness of not only denbers but also other bunds even without any extra maintenance efforts by farmers in the Rift Valley.