Responses Of Global Hydrometeorological Extreme Events Under Climate Change

Global anthropogenic warming has altered the energy exchange channel and hydrological cycle between atmosphere and earth,which exacerbated the uneven spatiotemporal distribution of water resources and resulted in frequent extreme hydrometeorological hazards.Under global warming,the holding vapor capacity of atmospheric layer as well as atmospheric moisture are increasing,and then lead to intensification of extreme precipitations in most areas of the globe.As a link between the geosphere,biosphere and atmosphere in the earth system,the hydrological cycle is impacted by both climate change and human activity.With underlying surface condition change,the precipitation and runoff series show significant change and non-stationarity.In this context,this paper studies the spatial-temporal evolution of global extreme hydrometeorological events,and discusses the microscopic interpretation and underlying physical mechanisms of extreme event change under climate change.This paper also investigates the thermodynamic response of rainfall extremes to climate change under different time scales,and characterizes the responses of storm runoff extremes to climate and anthropogenic changes.Finally,we proposed a method for deriving the adaptive isolines and most likely flood quantiles by non-stationary copula-based models.The main research works and findings were summarized as follows:(1)The research goal and significance of investigating the responses of global hydrometeorological extreme events under climate change are clarified.The definition methods of extreme hydrometeorological events are briefly reviewed and discussed,and the up-to-date advancement and development about climate change impact research are summarized.(2)Based on the multi-source data of global satellite remote sensing,radar and gauged station networks,the historical evolution trend of the global extreme temperatures,rainfall and runoff are examined.The internal physical mechanisms of extreme events change are interpreted with meteorological process and rainfall generation principle,and the spatiotemporal inconsistency of such change is investigated.We find an overall positive trend of temperature at the global scale,except in the midwestern US and north-western Europe which exhibit distinct cooling trends.Most rainfall stations outside of Russia show positive trends for precipitation extremes,while a decreasing trend in precipitation dominates Russia during 1961-1980,because moisture advection from the ocean is limited.Global runoff extreme change is accompanied by large spatial variability in magnitude,and the American continent,southern Europe,Japan and northern Australia indicates an overall increasing trend,while runoff extremes over the Sahel areas in Africa and middle Asia have declined.(3)Based on high-resolution gridded rainfall dataset(MSWEP?V2),with consideration of all-event and wet-event extreme definition methods,we systematically compare the thermodynamic responses of daily and 3-houly precipitation extremes to global climate change with extended Clausius-Clapeyron(C-C)equation.The results indicate that both daily and 3-hourly precipitation extremes could be classified into three types: i)a monotonic increase with temperature,ii)a monotonic decrease with temperature or iii)a Hook-like structure.Precipitation extremes at different time scales usually exhibit sub C-C scaling in extratropical areas.The all-event and wet-event extreme definition methods have great impacts on scaling structure and rates,particularly in tropics.Under wet-event definition,the tropics usually exhibit negative scaling rates;while under all-event definition,middle America,northern and southern Africa,India Peninsula,Indochina Peninsula and coastal Australia mostly show positive scaling rates.(4)The responses of storm runoff extremes to naturally and anthropogenically driven changes in surface temperature and atmospheric moisture content are characterized,and the theoretical formula of comparing the scaling rate of precipitation and storm runoff extremes in terms of runoff coefficient is derived.We assess the influence of agriculture and irrigation,land use and land cover change on the scaling of runoff extremes and temperature,and systematically compare this with changes in precipitation extremes.The results indicate that,governing by both atmospheric thermodynamic and dynamics,storm runoff extremes usually exhibit Hook-like structure with temperature.The Hook-like structure is not constant,and peak point temperature will increase with warming,shifting the Hook curve to warmer temperatures in the future.Our results reveal that storm runoff extremes largely exhibit a super C-C scaling over most measured regions of the globe,which are larger than that of precipitation extremes with spatial and decadal variability.(5)The 438 catchments over the U.S.under the MOPEX dataset and Ganjiang River basin in China were selected as case studies to investigate the responses of precipitation and runoff extremes to climate change,and the results are similar with those of gauged station analysis.In Ganjiang River basin,two Global Climate Models(GCMs)are employed to simulate climate scenarios,and the statistical downscaling methods and Xinanjiang model are incorporated to simulate the hydrometeorological series under climate change.To solve the non-stationarity of runoff series,the annual precipitation amount is selected as explanatory variable to establish the time-varying copula model.Considering the most likely realization of flood events,the adaptive bivariate flood quantile derivation approach is proposed.The results show that the precipitation and runoff series under changing environment is non-stationary,and that the quantile isoline and the joint design floods both have large uncertainty,implying that impacts of climate change should be taken into consideration for future engineering design and water resources management...