Empirical Forecasting of Slow-Onset Disasters for Improved Emergency Response: An Application to Kenya’s Arid Lands

Introduction The ability to forecast the onset, duration and severity of droughts, floods, and disease outbreaks with reasonable accuracy, especially in terms of their prospective human welfare effects, is critical to the design of timely and cost-effective early warnings and emergency response systems that can minimize the suffering of populations adversely affected by such relatively slow-onset events. As the consensus on climate change and its consequences grows, there is an increasing worry that the frequency of climate shocks will rise, with more frequent and serious humanitarian crises and ensuing demand for emergency response.