This study highlights drought characteristics and the many responses to drought stresses employed by Turkana pastoralists of northwestern Kenya. Multiple data sources, including socioeconomic interviews with 302 households, focus group discussions, and informal interviews with pastoralists were used to capture various aspects of drought and drought adaptation and coping practices. Standardized precipitation index derived from long-term rainfall data obtained from the Kenya Meteorological Service was used to quantify different degrees of drought intensity between 1950 and 2012. Results revealed that extreme drought events were increasingly frequent, and have impacted negatively on pastoral livelihoods. In order to adapt to or cope with climatic anomalies, households are using a variety of strategies. In addition to the traditional short-term coping mechanisms, the long-term adaptation strategies used include diversification of livelihood sources; livestock mobility to track forage and water resources; diversification of herd composition to benefit from the varied drought and disease tolerance, as well as fecundity of diverse livestock species; and sending children to school for formal education as a long term investment expected to pay back through income from employment. Policies and development interventions that reduce risks, diminish livelihood constraints, and expand opportunities for increased household resilience to drought are critical complements to the existing pastoral strategies.

Conflicts related to livestock raiding are not new phenomena in many pastoral societies in the Horn of Africa. Traditionally, various pastoral communities use raiding as a cultural practice for restocking of herds, especially after periods of drought or outbreaks of diseases. However, in recent years, livestock raiding has become more frequent, violent and destructive. This paper elucidates, first, the motives behind the current livestock raiding, and second, it analyses how conflict affects livelihoods of pastoral communities. Between 2008 and 2011, focus group discussions and interviews were carried out with 376 members of the Turkana and Pokot communities and key informants. The study findings suggest that hunger and drought impacting on availability and access of resources are critical raiding motives among the Turkana, while increasing wealth and payment of dowry are the most important motives for the Pokot community. Violent conflict poses a significant threat to pastoral livelihoods which are already under pressure from recurrent drought, diseases and political marginalisation. The direct impact of violent raiding is felt in terms of loss of human life and property, reduction in livestock numbers, limited access to water and pasture resources and forced migration. Indirectly, violent conflicts create a strong and omnipresent perception of insecurity which results in ineffective resource utilisation, reduced mobility, food insecurity and closure of markets and schools. These factors combined undermine adaptation strategies and pastoralism altogether. Hence, a framework of conflict mitigation is needed which addresses the specific raiding motives of each group.

This study uses statistical and econometric tools to measure households' vulnerability in pastoral rangelands of Kenya. It considered 27 socio-economic and biophysical indicators obtained from 302 households' in-depth interviews to reflect climate vulnerability components: adaptive capacity, exposure and sensitivity. The theoretical framework used combines exposure and sensitivity to produce potential impact, which was then compared with adaptive capacity in order to generate an overall measure of vulnerability. Principal component analysis (PCA) was used to develop weights for different indicators and produce a household vulnerability index (HVI) so as to classify households according to their level of vulnerability. In order to understand the determinants of vulnerability to climate-induced stresses, an ordered probit model was employed with predictor variables. The results show that 27% of households were highly vulnerable, 44% were moderately vulnerable and 29% of households were less vulnerable to climate-induced stresses. Factor estimates of the probit model further revealed that the main determinants of pastoral vulnerability are sex of household head, age of household head, number of dependents, marital status, social linkages, access to extension services and early warning information, complementary source of income, herd size and diversity, herd structure, herd mobility, distance to markets, employment status, coping strategies and access to credit. Therefore, policies that address these determinants of vulnerability with emphasis on women's empowerment, education and income diversifications are likely to enhance resilience of pastoral households.