时间步长和产汇流方法对小流域次洪模拟的影响

为研究半干旱半湿润流域复杂产流机制下不同时间步长和产汇流方法对次洪模拟的影响,选取北张店流域为研究对象,对1993～2005年具有代表性的12场次洪降雨径流资料,利用HEC-HMS模型分别以10,15,20 min为时间步长,采用2种方案进行模拟计算并对比分析结果。方案1为初损稳渗法+Clark单位线法;方案2为格林-安普特法+SCS单位线法;河道汇流演算均为马斯京根法,基流计算均为指数消退法,结果表明:时间步长的选取对次洪模拟结果影响较大,两种方案均在15 min时间步长下模拟精度最高,表明时间步长的选取并非是越短越精确。时间步长为15 min时方案1的确定性系数(DC)均值为0.813,方案2的DC均值为0.764,且方案1在洪峰、洪量和峰现时差模拟方面均优于方案2。

Influence of time-steps and runoff generation and confluence calculation methods on flood simulation in small watersheds

In order to study the influence of different time steps and runoff generation and confluence calculation methods on flood simulation in semi-humid and semi-arid regions, we simulated 12 representational flood events in Beizhangdian Basin from 1993 to 2005. The HEC-HMS model was applied to simulate the hydrological process under different time steps (10,15,20 min) and two schemes. The first scheme was the initial-loss and stable seepage method＆Clark unit line method, the second one was the Green and Ampt method＆SCS unit line method, and both schemes used the Muskingum method for river channel routing and the exponential regression method for base flow calculation. The results show that the time step has a great influence on the simulation results and both schemes had the highest simulation accuracy under 15min time step, indicating that the shorting time step does not always lead to accuracy improvement. When the time step was 15min, the average deterministic coefficient of the first scheme was 0.813, and the second scheme was 0.764. The first scheme is superior to the second one in terms of flood peak, flood volume and flood peak time.