气候变暖条件下水文循环变化检测与归因研究的几点认识

Some Points of View on Detection and Attribution of Observed Changes in Hydrological Cycle Under Global Warming

气候变化检测与归因的实践指南(GPGP)综合了4种检测归因方法,它囊括了目前研究这一因果链采用的不同途径。自1990年政府间气候变化专门委员会(IPCC)第一次评估报告以来,气候变化的检测与归因从气温升高和其他系统或变量变化趋势的研究开始,经历了从全球、半球、海洋与陆地、七大洲乃至区域尺度的细化进程。人为气候变化的指纹检测与归因方法已普遍用于全球大尺度水文循环要素——纬度带平均降水、强降水、径流、比湿、水汽含量等时空型态变化的研究,个别地也应用到流域尺度,如以冰雪融水补给为主的河流的天然流量、积雪和最低温度变化的研究。对于以降水为主,且受人类活动干扰较大的流域,观测的水文循环要素变化的检测与归因,大多数聚焦于观测的气候要素及非气候变量对其的影响,而不回答变化是自然的还是人为强迫引起的。这种方法要求高质量长系列的观测数据、先进的统计检验技术以及物理基础好的水文模型。对于水文循环要素对气候变化敏感的流域,若观测数据满足要求,则采用正规的检测与归因方法或最优指纹法识别外强迫及驱动因子的影响研究,将具有重要科学意义和发展前景。

The Good Practice Guidance Paper （GPGP） presents four different methods,which encompasses existent different approaches to investigate the causation chain. Since 1990,the detection and attribution study of climate changes has undergone a refinement process in spatial scale of temperature rise （warming） from the global to half-hemispheric scale and from the continental to regional scale,and an expansion process in research topic from lands to oceans,and from the climate system to other related systems. The fingerprinting detection and attribution method of anthropogenic climate changes has been extensively applied in studies on spatiotemporal pattern changes of global/large scale hydrological cycle elements,such as latitudinal band average precipitation,extreme precipitation,specific humidity,lower tropospheric moisture content,and individually in the streamflow impacted by external forcings,natural runoff,snow pack and minimum temperature on a river basin scale. For the river basins recharged by precipitation and impacted by human activities,the detection and attribution studies of hydrological cycle elements mainly focus on effects of observed climatic elements and non-climatic variables on streamflow,without the emphasis on causation by the anthropogenic component of climate change. For this kind of approach,long term observed climatic and non-climatic data,advanced statistic technology and hydrological model with better physical basis are vital for the correct recognition of causation process. In the river basins with high qualified long term observed climate and non-climate data,where hydrological variables are sensitive to climate changes,the optimal fingerprinting method will be valuable in detection and attribution studies.