子計畫：人為活動對區域天氣與氣候極端事件改變可能貢獻的定量分析評估(II-III)

It is still quite controversial whether we can attribute the change of occurrence frequency of local extreme weather and climate events to anthropogenic forcing. An increase in average temperatures will lead to an increase in the frequency or magnitude of some extreme events. Through the probabilistic extreme event attribution research framework and statistical analysis, climate scientists are looking for different weather and climate extremes for the attribution study. Prominent climate centers established operational attribution analysis to explain whether the extremes occurred can be link to past global warming induced by anthropogenic greenhouse gas emissions, as part of their climate services. There are many uncertainties in attribution studies. Particular challenges are the availability of long-term meteorological observations and the reliability of climate model simulations of the climate conditions generating an extreme weather event. Uncertainties are present in all probabilistic event attribution studies, but there is generally higher confidence in studies focusing on heatwaves than those focusing on extreme precipitation. Investigation of hurricanes and typhoons is currently limited by the ability of global climate models to simulate these events. Our proposed project, continued from the current first two year project, aims on those regional extreme weather and climate events that are not quantitatively estimate the potential human contribution previously, namely the heat wave over lower latitude and tropical cyclone, including their occurrence, intensity, and associated rainfall. Due to the huge impact of tropical cyclone and heat wave on the loss and damages by weather-related hazards, such quantitative measure of anthropogenic impact cam provide scientific basis for calculating the compensation to the climate-impacted countries discussed in Paris COP21 meeting. In our research framework, the change in the frequency and severity of a heat wave event under current conditions is calculated and compared with the probability and magnitude of the event if the effects of particular external forcing, such as due to human influence, had been absent. In our research, we use the CAM5-1, HadGEM3-A-N216, HadAM3P-N96 large ensemble simulation from the CLIVAR C20C+ Detection and Attribution project (Folland et al. 2014) to detect the heat wave events occurred in both historical all forcing run and natural forcing only run. The heat wave events are identified by partial duration series method (Huth et al., 2000). We test the sensitivity of heat wave thresholds from daily maximum temperature (Tmax) in warm season (from May to September) between 1960 and 2010. We consider the anthropogenic effect on the later period (2007-2012) when the warming due to human impact is more evident. Using Taiwan and surrounding area as our preliminary research target, We found the anthropogenic effect will increase the heat wave day per year more than double and make the mean starting(ending) day for heat waves events about 15-30 days earlier(later). Using the Fraction of Attribution Risk analysis to estimate the risk of frequency of heat wave day, our results show the anthropogenic forcing very likely increase the heat wave days over Taiwan by more than 104% ~ 376%.