汉江上游郧县五峰段史前大洪水水文学恢复研究

通过深入的野外考察，在汉江上游郧县五峰段多个地点发现全新世古洪水事件的沉积学记录，为恢复史前特大洪水事件的洪峰水位和洪峰流量提供了物质基础。在系统获取河道水文参数的基础上，采用ArcGIS 耦合HEC-RAS模型，推算出其洪峰流量介于42 220~63 400m³/s 之间。对于河槽糙率的灵敏度检验表明，给定糙率变化±25%，模型模拟得到的洪峰流量误差在-10.6%~6.3%之间。这个误差范围远小于采用比降-面积法获得的结果。同时，采用该模型和同样的参数，依据该河段现代大洪水洪痕高程，恢复推算其洪峰流量，其结果与实测数据误差在-3.9%~1.0%之间。与其它方法相比较，这种新方法有效地提高了水文参数选取准确性，使得古洪水水文恢复计算结果更为可靠。本研究为超长尺度大洪水水文学恢复提供了新的途径，也获得了汉江上游万年尺度特大洪水的水文学数据资料，对汉江上游水利水电枢纽工程建设、水资源调度和防洪减灾具有重要的现实意义。

Hydrological reconstructions of the pre-historical great floods in the Wufeng Reach of Yunxian County in the upper Hanjiang River

Palaeoflood hydrological studies were carried out in the upper reaches of the Hanjiang River. Typical palaeoflood slackwater deposits of the Holocene period were identified at several sites along the banks of the Wufeng reach in Yunxian County. They have recorded four episodes of extraordinary palaeoflood events (12,600-12,400 a BP, 4200-4000 a BP, 3200-2800 a BP and 1900-1700 a BP). The palaeoflood peak stages were estimated with the elevations and the deposition depths of the SWD. The palaeoflood peak discharges were estimated by using the HEC-RAS one dimensional model running within an ArcGIS environment. The results indicated that these palaeoflood discharges were between 42,220 and 63,400 m³/s. A sensitivity test performed on the model indicated that for a 25% variation in roughness values, an error between -10.6% and 6.3% was introduced into the results of peak discharge. The error is much smaller than that obtained by using the slope-area methods. Based on the flood stage indicators, the peak discharges of modern floods were also reconstructed in the same reach using the HEC-RAS model and the same hydraulic parameters. The error between the reconstructed and gauged peak flood discharges is between -3.9% and 1.0%. These showed that the palaeoflood peak discharges reconstructed by using the HEC-RAS model are reliable. The reconstructed peak discharges of the palaeoflood are much larger than those of the gauged largest floods in the upper reaches of the Hanjiang River. The flood data series of the river are therefore extended to a 10,000-year time-scale. This result provides reference in hydrological engineering, water resource management and flood mitigation on the river. And it is also very helpful in understanding the relationships between extreme flood events and global climate change.