基于植被分布的河道糙率分区及率定方法

天然河流的河道综合糙率呈现出空间上的差异性和随水位（或流量）变化的动态性，但目前缺乏相关参数化方法来定量描述河道糙率的动态变化规律。尝试通过参数化方法开展受河道植被影响显著河流的糙率反演研究，用以提升模型精度。基于植被分布将河道断面划分为若干糙率不同的子区，通过率定河道断面各分区的糙率，从而反演糙率—水位曲线。在此基础上通过分析河道植被覆盖情况与河道断面特点对糙率曲线变化的影响，推求了基于分区糙率的河道断面综合糙率计算公式，从而定量描述分区糙率与综合糙率的关系。以漓江干流为例，采用该方法率定漓江干流（大溶江至阳朔段）水动力模型。结果表明:漓江干流综合糙率随水位在0.022~0.180间变化；在1.5 m的临界水深下，断面可划分为底床植被区（n=0.210）与非植被区（n=0.006），能较好地反演糙率—水位曲线并获得理想的水位模拟效果。漓江底床植被繁茂是糙率随水位变化的根本原因，断面边滩的坡度变化是糙率与水位曲线梯度变化的主要驱动因素，两者的共同作用使得糙率随着水位呈现两段式的非线性变化。

Roughness zoning and calibration of channel cross-section based on vegetation distribution

The synthesis roughness of natural rivers varies with cross-sections, as well as water level and discharge. However, there are a few methods to parameterize the roughness curves (e. g., a roughness-water level curve). In order to improve the model accuracy, a significant effect of vegetation on channel roughness is considered in this research to parameterize the roughness. The channel cross-section is divided into several zones with different partition roughness based on vegetation distribution. After calibrating partition roughness values, the roughness-water level curve is obtained by using a water level-discharge relationship method. By analyzing the impact of channel cross-section characteristics and vegetation cover on the roughness curve, a formula is given for calculating the synthesis roughness of channel cross-section based on partition roughness values. This formula quantitatively describes the relationship between partition roughness and comprehensive roughness. The proposed method was used to calibrate the roughness of the hydrodynamic model of the Lijiang River. The results show that the synthesis roughness of the main channel varies in relation to water level ranging from 0.022 to 0.180. The critical depth was set at 1.5 m to divide the cross-section of the channel into the bottom-bed vegetation zone (n=0.210) and non-vegetation zone (n=0.006). Thus, the roughness-water level curve can be inverted, causing reasonable water level simulation results to be obtained. Dense vegetation in the Lijiang riverbed is the main factor causing the synthesis roughness changing with water level. The slope change of the channel cross-section is the main driving factor inducing the gradient change of the roughness-water level curve. These two factors change the cross-section synthesis roughness non-linearly in relation to water level. The results can provide a reliable reference for the hydrological forecasting of rivers in mountainous areas.