歷史與推估未來的極端降雨與熱浪事件改變：觀測、模擬、歸因

Natural disasters due to events of extreme precipitation, drought, and heatwaves have a huge influence on security of human life and properties, societal economy and development. Extreme rainfall leads to flooding and landslide including risk to human life, damage to buildings and infrastructure, and loss of crops and livestock. Extended heatwave not only affect human health, but also have large impact on crop yields and food security and increase strain on infrastructure (power, water and transport). From past IPCC reports, it is commonly expected that precipitation extremes and frequency of heat waves will increase as the climate warms and with more moisture. Although there are plenty of studies working on the topic, usually the projected change are expressed as long-term mean change of extremes or as changes in return level and return period based on statistical extreme value analysis. The uniqueness of the proposed research work is on search of events by spatial and temporal connected extremes occurrences. The extremes can be heavy rainfall, heat waves, and drought etc. The occurrence or severity of events can be used as threshold to test the sensitivity of methodology in determining the event duration, affected area and integrated magnitude of event. The outcome can be compared and ranked through record history and with different regions. The total magnitude of unsegregated extreme events should link more directly to the potential damages of natural disasters. The joint events from different extremes that lead to compound hazard can be further explored. The research will be start from applying the methodology to the growing historical records and observational analysis. The result can be evaluated from the well-known historical extreme events in the literatures and past media coverages. The same objective approach be applied to global and regional climate model simulation to obtain the characteristics of spatial and temporal evolution of extreme events in the model and to further assess the capability of model in capturing the observed statistics of extreme events in their historical runs. Another advantage of using event approach and compared to the mean changes is that the event associated environmental thermal and dynamical conditions in different climate regions can be constructed to further delineate the underlying physical processes separately. The compounding processes that lead to compound disasters can also jointly discussed. The proposed study not only analyzing the global and regional aspects of different type of precipitation and temperature extreme events change through time in the past and into the future, but also the associated environmental and meteorological conditions that cause the events to happen both individually or jointly. Additional effort to quantify the attributable risks of extreme events due to human impact or other individual forcing can be studied using IPCC CMIP designed numerical experiments with large ensemble members.