Climate extremes: predictability, impacts, and consequences at regional scales.

• Online Access: http://hdl.handle.net/2047/D20203893

Climate and weather extremes such as tropical cyclones, floods, and heat waves can have potentially devastating societal and economic impacts. Decision-makers responsible for managing critical infrastructures and emergency preparedness are looking for credible projections of climate change over near-term (0-30 year) at regional scales. Private sectors especially insurance industry which are in the business of insuring risk resulting from hurricanes, floods, and tornado hails do not consider climate change in their risk assessment models even though frequency and severity of these extremes have increased due to climate change. There is an immediate need not only for reliable and credible projections of climate change but also for an increased understanding of uncertainty in projected climate change. Further, there is a need to develop framework to do multi-sector impacts assessment in which climate uncertainty from all sources has been incorporated. In this dissertation, I started with evaluation of the performance of latest generation of global climate models, Coupled Model Intercomparison Project Phase 5 (CMIP5) in simulating current climatology and multi-model agreement in projected climate change. Subsequently I studied the performance of CMIP5 models in simulating and projecting wind extremes at regional scales. Both of these studies were focused on long-term climatology, the end of the century time horizon. Multi-sector stakeholders are looking for reliable projections of climate change at near-term planning horizons as most of the decisions are made at time scales of one-to-two decades. Consideration of climate uncertainty especially climate internal variability and model response variability becomes more important as they dominate signal of climate change. One of the important contribution of my research has been on enhancing the present understanding of the role of different sources of uncertainty with projection time horizons at multiple spatial scales for precipitation and temperature. Finally, the framework has been applied to study the how much thermoelectric power production will be at risk due to warmer and scarcer water under nonstationary climate change.