RESUMEN
The main purpose of this study was to characterize the livelihood resilience of smallholder farmers in the face of recurring droughts in northeast Ethiopia. The data was collected using a cross-sectional survey of 274 households and five focus group discussions. Principal component analysis and multiple linear regression models were employed to analyze the data. The Livelihood Resilience Index (LRI), framed on absorptive, adaptive, and transformative capacities, was used to quantify the households' livelihood resilience. The results indicated that about 57% of the surveyed respondents were non-resilient, while 43% were resilient to different degrees. Abay Tekeze watershed (ATW) livelihood zone exhibits the highest proportion of resilient households (57.4%), while North Wollo highland belg has the lowest proportion (22.7%). The high resilience in ATW was attributed to the relatively lower persistence of droughts, better accessibility in enabling institutions, more access to agricultural inputs, and the training and support given to farmers. The better-off were more resilient (90.9%) than the medium (52.1%) and the poor (34.6%) households. Among the latent dimensions, sensitivity with ß value -0.372, climate change and variability (-0.33), and enabling institutions and environments (0.288) showed a significant (p < 0.0001) influence on LRI. This was followed by adaptive capacity and food access (0.249), agricultural practice and technology (0.213), and asset possession (0.19), in respective order. It implies that the absorptive capacity of households showed the leading influence in determining LRI, while adaptive and transformative capacities had nearly similar low effects. Thus, it is recommended that future planning for building livelihood resilience and drought risk interventions in the area should address the levels of resilience identified and the relative importance of each latent dimension indicated.
RESUMEN
Agriculture is the most sensitive sector which has largely been affected by the impacts of drought. The study aims to detect and characterize agricultural droughts using MODIS-based multiple indices in North Wollo, Ethiopia. Two Moderate Resolution Imaging Spectroradiometer (MODIS) datasets (MOD13Q1 and MOD11A2) for the period 2000 to 2019 were used to generate Normalized Difference Vegetation Index (NDVI) and Land Surface Temperature (LST). Accordingly, NDVI anomaly, Vegetation Condition Index (VCI), Temperature Condition Index (TCI), and Vegetation Health Index (VHI) were computed to characterize agricultural droughts during the crop growing season. Both the NDVI anomaly and VCI confirmed that there was no single drought-free year in the area throughout the study period. TCI showed relatively exaggerated drought stress than the other indices. However, VHI indicated lower area coverage and a lower level of stress than its aggregates (VCI and TCI). Specifically, 2002, 2004, 2009, 2010, and 2015 were all identified as severe drought years, where over 60% of the area was affected by droughts. Results of the regression analysis indicated that VCI, TCI, and VHI were having significant positive trends with precipitation in the majority of the districts. Using the aggregated drought frequency of each index, 13.5, 73.7, and 12.8% of the area were under moderate, high, and extremely high levels of agricultural drought occurrence, respectively, and the likelihood of implied risks. Therefore, all the districts of North Wollo were affected by persistent drought stress. Such drought recurrences have the potential to impose significant impacts on the agro-based livelihoods of the local community demanding ongoing drought monitoring and the application of effective early warning systems.