Modeling drought vulnerability and coping traps among pastoral households in Borena Ethiopia

Pastoral livelihoods remain highly vulnerable to recurrent drought, yet empirical understanding of how drought exposure, coping responses, and adaptive capacity interact remains limited. This study aims to develop and empirically validate an integrated structural model of drought-induced livelihood vulnerability. The model examines the pathways through which drought exposure, socio-ecological vulnerability, coping responses, and adaptive capacity influence livelihood outcomes among pastoral households. Partial Least Squares Structural Equation Modeling (PLS-SEM) was applied to survey data collected from 384. Results show that drought exposure and resource degradation explain a substantial proportion of socio-ecological vulnerability (R2 = 0.511). Higher vulnerability is associated with greater coping intensity (β = 0.361, p < 0.001), while coping intensity negatively influences livelihood outcomes (β = −0.553, p < 0.001). Conversely, adaptive capacity positively influences livelihood outcomes (β = 0.510, p < 0.001). Among the adaptive capacity components, indigenous knowledge has a comparatively higher standardized path coefficient (β = 0.328), followed by institutional support (β = 0.301) and technology adoption (β = 0.261). Coping intensity is positively associated with livelihood diversification (β = 0.167, p < 0.05), although diversification alone has limited influence on livelihood outcomes. The findings reveal a vulnerability–coping pattern in which drought stress increases reliance on reactive coping strategies, while adaptive capacity provides a pathway to improved livelihood outcomes. By empirically distinguishing coping responses from adaptive capacity, the study extends the Sustainable Livelihoods Framework and highlights the importance of integrating indigenous knowledge with institutional and technological interventions to reduce drought-induced livelihood vulnerability. Drought and resource degradation jointly explain 51.1% of socio-ecological vulnerability to drought (R2 = 0.511). Coping shows negative association with livelihoods (β = − 0.553), while adaptive capacity enhances them (β = 0.510). Indigenous knowledge (β = 0.328) has a slightly stronger association with livelihoods than technology (β = 0.261) as a predictor of adaptive capacity, challenging techno-centric narrative. Livelihood diversification is distress-driven (β = 0.167), not proactive, revealing that diversification alone fails to guarantee resilience.