Detect And Attribute Anthropogenic Influence On Extreme Precipitation Changes At Different Spatial Scales

Under the background of global warming,global extreme weather and climate events occur frequently,and social production and ecosystem are seriously damaged,among which extreme precipitation,floods and secondary disasters have caused huge economic losses.Therefore,it is an urgent need for the country to conduct in-depth study of extreme precipitation in order to formulate disaster prevention and mitigation measures.Currently,a lot of research work has been carried out on the change of global extreme weather and climate events,although the conclusion that human activities cause the increase of global extreme precipitation has high credibility,but it is still unclear about the extend to human activities influence the change of extreme precipitation.Therefore,we analyze the possible impact of human activities on extreme precipitation changes at different spatial scales and different return periods using HadGEM3 model dataset,combined with quantile regression and generalized extreme value distribution method.Based on this,the response of extreme precipitation in the Asian Australian monsoon region to global warming under the background of continuous intensification of human activities is further explored.The research results mainly include the following two aspects:Detection and attribution is to detect and quantify the changes caused by external forcing and identify the relative contribution of man-made and natural forcing to climate.In this paper,the changes of global land extreme precipitation intensity and frequency at different spatial scales and different return periods(50th,90th,98th,99th,99.5th,99.8th,99.9th)are detected and attributed.Based on the analysis of the generalized extreme value distribution,the annual maximum single day precipitation and the annual maximum five consecutive days precipitation show that the contribution of human activities to the intensity and frequency of extreme precipitation in different return periods can be significantly detected on the global scale,hemispheric scale,continental scale and regional scale(95%confidence level),but the detection attribution results on the grid scale have great uncertainty.The contribution of human activities to extreme precipitation is easier to detect in frequency than in intensity.The choice of spatial scale and return period will greatly affect the detectability of extreme precipitation events.The detectability of attribution is constrained by the spatial scale,and there is great uncertainty in detecting extreme precipitation changes on a smaller spatial scale.As the regional scale is suitable for extreme precipitation detection attribution,the Asian Australian monsoon region with dense population,abundant monsoon precipitation and frequent extreme precipitation is selected to further study the impact of human activities on extreme precipitation.The HadGEM3-A model can better simulate the temporal and spatial variation characteristics of extreme precipitation in the Asian Australian monsoon region in the historical period(1960-2013).The linear correlation coefficient between the simulated and observed annual maximum daily precipitation exceeds+0.64(99%confidence level).Through quantile regression,it is found that with the increase of extreme degree,the extreme precipitation under natural forcing test shows a decreasing trend,with a minimum of 3.2%/?,while the extreme precipitation under total forcing test including human activities shows an increasing trend,with a maximum of+4.5%/?.The regions significantly affected by human activities are located in the south of the East Asian monsoon region,the north of the East Asian monsoon region and the Indochina Peninsula.Therefore,human activities lead to the increase of extreme precipitation in the Asian Australian monsoon region.The more extreme the intensity,the faster the growth rate of precipitation.