Abstract:

Increase of the intensity and frequency of extreme flood events has been a big challenge of regional water resources security. Based on daily discharge data during 1957-2011 in the Kaidu watershed, annual maximum daily discharge, maximum daily discharge in spring and maximum daily discharge in summer at Bayblk and Dashankou hydrological stations are extracted, respectively. Linear trend method, Mann-Kendall test and Pettitt test are used to analysis the variation of the 6 series. Then, the series are selected to describe the extreme distributions of GEV, and frequency of extreme floods is analyzed. The results indicated that, there are no significant trend and no change point in the 6 series in which the annual maximum daily discharge at Bayblk station, the annual maximum daily discharge at Dashankou station and the maximum daily discharge in spring at Bayblk station fitted the Frechet distribution, and the maximum daily discharge in spring at Dashankou station, the maximum daily discharge in summer at Bayblk station and the maximum daily discharge in summer at Dashankou station fitted the Gumbel distribution. Since 1980s, frequency of extreme floods in the Kaidu River obviously increased; the number of floods in summer continuous increase at the Bayblk station, and the number of floods both in spring and summer increase at Dashankou station; occurrence time of spring floods shift to an earlier date at both two hydrological stations. The increase of precipitation in winter, and the rise of temperature in spring have important contribution to spring flood changes; the increase of extreme rainfall in summer would be dominant factors for changes of summer floods.

Key words: climate change, extreme flood, GEV distribution, inland river