摘要
水文响应单元(Hydrological Response Units,HRU)是SWAT模型模拟的基本单元,传统方法划分的水文响应单元在空间分布上不连续且难以确定其明确的空间位置,不能反映HRU间的相互作用和进行精确空间分析。利用GIS工具对土地利用和土壤类型数据进行概化处理,提出了HRU空间离散化的方法,实现了水文响应单元在空间上的准确定位。在此基础上,针对SWAT模型中同一子流域所有HRU采取相同延迟的弱点进行改进,并选择太湖地区西苕溪流域对改进的SWAT模型进行水文模拟验证。改进后,校正期港口站Nash效率系数ENS(Nash-Sut-cliffe Efficiency)从0.64提高到0.67,验证期ENS系数从0.70提高到0.76。研究表明:修正后的SWAT模型更能反映流域的水文特征,可以达到非常好的效果,考虑到HRU距离因素的径流延迟更为准确地刻画径流过程。实现HRU空间离散化将为模型改进和更小尺度的空间分析提供数据基础。
The SWAT（Soil and Water Assessment Tools） is a physically based semi-distributed hydrological model.The basic computational unit in the SWAT is the hydrological response unit（HRU） that is usually determined by land uses,soil types,and slope and aspect of an area.Each HRU is assumed to have a homogeneous hydrological response.HRUs obtained with the traditional approaches for delineation of hydrologic response units are often spatially discontinuous and it is difficult to locate them in a catchment.The interaction among HRUs cannot be precisely described and analyzed.In this study,the land use and soil type data from a typical watershed in the Taihu Lake region are processed using the geographic information system（GIS） tools.A new approach for spatial discretization of hydrologic response units is proposed and applied to the processed data.As the result,the location of each HRU in the watershed can be accurately identified.Accordingly,different values of the surface runoff lag time in the SWAT can be applied.The improved version of SWAT is tested on the Xitiaoxi watershed of the Taihu Lake basin.The result shows that a notable improvement is achieved in the hydrological simulation compared to the original SWAT simulation for both model calibration and validation periods.The Nash-Sutcliffe model efficiency coefficients have been improved from 0.64 to 0.67 for calibration and from 0.70 to 0.76 for validation,respectively.The improved SWAT model can better reflect the runoff lag process and achieve a better simulation result.The new approach for spatial discretization of hydrologic response units is an effective solution for the model optimization,and can provide better estimates to the spatial data analysis at finer scales.
出处
《水科学进展》
EI
CAS
CSCD
北大核心
2012年第1期14-20,共7页
Advances in Water Science
基金
国家水体污染控制与治理科技重大专项资助项目(2008ZX07526-007)
国家自然科学基金资助项目(41101079)~~
作者简介
宁吉才（1974-），男，山东莘县人，博士，主要从事水文模型与GIS应用研究。E-mail：jcning2005@163．com
