冰川冻土 ›› 2019, Vol. 41 ›› Issue (1): 147-157.doi: 10.7522/j.issn.1000-0240.2019.0003

• 寒旱区水文水资源 • 上一篇    


赵奕1, 南卓铜1, 李祥飞1, 徐毅1, 张凌2   

  1. 1. 南京师范大学 虚拟地理环境教育部重点实验室, 江苏 南京 210023;
    2. 中国科学院 西北生态环境资源研究院, 甘肃 兰州 730000
  • 收稿日期:2017-10-09 修回日期:2018-08-19 发布日期:2019-03-16
  • 通讯作者: 南卓铜。
  • 作者简介:赵奕(1994-),男,江苏常州人,2016年在南京师范大学获学士学位,现为南京师范大学硕士研究生,从事地表过程建模及信息系统
  • 基金资助:

On applicability of a fully distributed hydrological model in the cold and alpine watershed of Northwest China

ZHAO Yi1, NAN Zhuotong1, LI Xiangfei1, XU Yi1, ZHANG Ling2   

  1. 1. Key Laboratory of Virtual Geographic Environment of Ministry of Education, Nanjing Normal University, Nanjing 210023, China;
    2. Northwest Institute of Eco-Environment and Resources, Chinese Academy of Science, Lanzhou 730000, China
  • Received:2017-10-09 Revised:2018-08-19 Published:2019-03-16

摘要: 分布式水文-土壤-植被模型(Distributed Hydrology Soil Vegetation Model,DHSVM)是基于栅格离散的分布式水文模型,对地表水热循环的各个过程能进行很精细地刻画,被广泛应用于世界各地很多类型的流域的高时空分辨率的水文模拟,然而它在高寒山区的适用性并不清楚。基于300 m数字高程模型,应用DHSVM模型对典型的高寒山区流域八宝河流域2001-2009年的水文过程展开模拟,并采用流域出口祁连站的水文实测数据对模型进行了精度评价。参数敏感性分析表明,土壤横向导水率、田间持水量和植被反照率等是该区域主要的敏感性参数。模型默认参数会高估高寒山区流域的潜在蒸散发量,导致夏季径流量远小于观测值。通过参数率定,模型校准期(2001-2004)的模拟日径流和月径流Nash效率系数分别达到0.72和0.87;而模型验证期(2005-2009)分别为0.60和0.74。结果表明,DHSVM模型基本具备了模拟高寒山区流域降水-径流过程的能力。然而,由于DHSVM模型缺少对高寒山区流域土壤的冻融过程的刻画,春季径流的模拟精度明显受到影响,需要在将来重点改进。

关键词: DHSVM模型, 高寒山区, 八宝河流域, 参数敏感性分析, 径流模拟

Abstract: Distributed-Hydrology-Soil-Vegetation-Model (DHSVM) is a grid-based distributed hydrological model and has been widely used to simulate hydrological processes at high spatiotemporal resolution across the world owing to its particular calculation of surface water and heat balance. However, its applicability in cold and alpine regions remains unclear. This paper employed DHSVM to simulate hydrological processes during the period of 2001 to 2009 at a 300 meters and 3 hours modeling resolutions in the Babao River basin, a representative mountainous river basin located within the Qilian Mountains in the cold region of Northwest China. The applicability was thus validated with observations at the basin outlet. Parametric sensitivity analysis shows that lateral conductivity, field capacity, leaf area index and albedo are some most sensitive parameters. The default model parameters lead to overestimation of potential evaporation, and consequent underestimation of streamflow simulation in summers. Using calibrated parameters, the model can achieve good simulation with Nash-Sutcliffe efficiency coefficients of 0.72 and 0.87 in the calibration period (2001-2004), and 0.60 and 0.74 in the validation period (2005-2009) for daily and monthly simulations, respectively. This study concludes that DHSVM is generally capable for simulating hydrological processes at high spatial and temporal resolutions in cold and alpine regions, although it is insufficient in representing freezing and thawing processes occurred in soil, resulting in lower accuracy of streamflow simulation in springs, which should be addressed in future when modelling in those areas.

Key words: DHSVM, cold and alpine region, the Babao River basin, sensitivity analysis, streamflow simulation


  • P334+.92