Date Published: May 31, 2018
Author(s): Frédéric Kosmowski.
•Following a strong El-Niño, some regions of Ethiopia experienced major droughts during the 2015/16 agricultural season.•This study investigates the effect of terraces and contour bunds on yields during the 2015 drought.•At the national level, terraced plots have slightly lower yields than non-terraced plots.•Terraced plots acted as a buffer against the 2015 Ethiopian drought, while contour bunds did not.•This study provides evidence that terraces have the potential to help farmer deal with current climate risks.
Research suggests that conventional agriculture creates unsustainable erosion rates that can result in decreased agricultural potential (Montgomery, 2007). In Ethiopia, soil erosion is considered a severe constraint for land resource productivity. Since the 1980s, massive soil conservation programs have been carried out by governmental and non-governmental organizations. Following a top-down approach, soil water management practices have been promoted according to the land’s physical limitations and erosion risks. These interventions, often implemented at the water catchment level, are designed to foster community labor mobilization (Desta et al., 2005).
At the national level, slightly less than one third of sampled plots are terraced – this figure is similar for contour bunds. Tigray and Amhara are the two regions where terraces have seen the highest adoption rates, with 55% and 46% respectively. Contour bunds are more important in Tigray (41%) and SNNP regions (40%). No significant differences exist in terms of adoption rates between agro-ecological areas.
In this article, the treatment effect of terracing and contour bunds on yields was estimated using propensity score matching on a dataset representative of Ethiopia. Three key insights have emerged from this study. First, results showed that terraced plots have significantly lower yields than unterraced plots, confirming and extending previous findings (Shiferaw and Holden, 2001; Adimassu et al., 2012). This slight yield penalty (−9.5%) is consistent with the surface reduction occupied by benches or stones. Second, we found little evidence to suggest that treatment effects differ by agro-ecological areas, a finding that contradicts the results of Kato et al. (2011). This conflicting finding could be attributed to the difference in proxies used for rainfall amount. Kato et al. (2011) utilized historical rainfall data. Here, rainfall amount is matched with planting and harvesting dates of each surveyed plots. Another result, consistent with Kato et al. (2011) is that contour bunds have the potential to increase yields in highlands. However, contour bunds should be used in combination (Gebreegziabher et al., 2009; Kato et al., 2011), while terracing stands out as a potentially independent practice. When terraces are not possible, a combination of soil water management practices should thus be recommended when designing climate change adaptation plans. Third, the results support our main hypothesis that terraces, carried out primarily to control water runoff and erosion, acted as a buffer during the 2015 Ethiopian drought. Contour bunds, a less intensive soil water management practice, did not minimize drought impacts. These findings demonstrate the high relevance of terraces for climate-smart agriculture. While available evidence suggests that terracing offers benefits for sustainability (Vancampenhout et al., 2006; Raes et al., 2007; Adimassu et al., 2012; Adgo et al., 2013), this study brings further evidence that terraces can help farmers to build resilience in the face of extreme events. Indeed, trade-offs exist between the risk mitigation benefits of terracing with the yield reduction in years with adequate rainfall. In drought-prone areas, the practice should be promoted as a way to increase farmer’s adaptive capacity.