| 英文摘要: | In the twentieth and twenty-first centuries, flooding has taken a devastating societal and economic toll on the central United States, contributing to dozens of fatalities and causing billions of dollars in damage1, 2. As a warmer atmosphere can hold more moisture (the Clausius–Clapeyron relation), a pronounced increase in intense rainfall events is included in models of future climate3. Therefore, it is crucial to examine whether the magnitude and/or frequency of flood events is remaining constant or has been changing over recent decades. If either or both of these attributes have changed over time, it is imperative that we understand the underlying mechanisms that are responsible. Here, we show that while observational records (774 stream gauge stations) from the central United States present limited evidence of significant changes in the magnitude of floodpeaks, strong evidence points to an increasing frequency of flooding. These changes in flood hydrology result from changes in both seasonal rainfall and temperature across this region. Over the past century, the central United States (CUS) has been plagued by a series of large floods such as those that occurred in 1993, 2008, 2011, 2013 and 2014. These events had adverse societal consequences including decreased food production and displacement of communities/people, led to economic losses reaching billions of dollars1, 2, and portend future increases in flood activity. However, the question remains: is the character of recent flooding truly distinct from the long-term averages, or is it simply an artefact of our relatively short collective memory? Use of historical records to ascertain change over time globally has thus far proved inconclusive. The Intergovernmental Panel on Climate Change4 concluded that ‘there continues to be a lack of evidence and thus low confidence regarding the sign of trend in the magnitude and/or frequency of floods at a global scale.’ A number of observational studies that examined changes in the magnitude of annual maximum peak discharge over the CUS (refs 5, 6, 7, 8) reached similar conclusions. The lack of evidence for an increase in peak discharge becomes even clearer when examining trends in the magnitude of the annual maximum daily discharge data for 774 US Geological Survey (USGS) stream gauge stations across the CUS over the common 1962–2011 time period (Fig. 1a; consult Methods for more information on how the analyses are performed). Over most of the study area, no statistically significant trends are identified; annual peak discharge magnitude has apparently not been increasing over most of the twentieth and early twenty-first centuries. Overall, 158 (20%) of these stations exhibit statistically significant changes in the magnitude of flood peaks, and of these, 101 (13% of the total number) are characterized by a trend towards increasing flood magnitude, with many of them concentrated in the greater Chicago area. These results are consistent with previous studies5, 7, 8, 9, 10, which also failed to detect widespread evidence of changing flood magnitude over the CUS.
| http://www.nature.com/nclimate/journal/v5/n3/full/nclimate2516.html
|
