文献类型：期刊文献

中文题名：WASSI-C生态水文模型响应单元空间尺度的确定——以杂古脑流域为例

英文题名：Determination of spatial scale of response unit for the WASSI-C eco-hydrological model—a case study on the upper Zagunao River watershed of China

作者：刘宁[1] 孙鹏森[1] 刘世荣[1] 孙阁[2]

第一作者：刘宁

机构：[1]中国林业科学研究院森林生态环境与保护研究所,国家林业局森林生态环境重点实验室;[2]Eastern Forest Environmental Threat Assessment Center,USDA Forest Service

年份：2013

卷号：37

期号：2

起止页码：132-141

中文期刊名：植物生态学报

外文期刊名：Chinese Journal of Plant Ecology

收录：CSTPCD;;Scopus;北大核心:【北大核心2011】；CSCD:【CSCD2013_2014】；

基金：国家科技支撑项目(2012BAD22B01);中国林业科学研究院院所基金海外人才专项(CAFYBB2008007);林业公益性行业科研专项(200804001)

语种：中文

中文关键词：生态水文模型;水文响应单元;空间尺度

外文关键词：eco-hydrological model; hydrological response unit （HRU）; spatial scale

分类号：P343

摘要：模型基本单元空间尺度的确定是大尺度生态水文模型应用的前提,也是提高模型模拟精度的关键。该文以长江流域岷江上游的杂谷脑河上游流域为例,通过设置最小流域面积阈值,构建不同的水文响应单元划分方案,探讨生态水文模型WASSI-C响应单元的最佳空间响应尺度。结果表明:模型响应单元空间尺度的变化对模型精度有显著影响,模拟效果存在随响应单元划分面积阈值增加先提高再稳定的趋势,面积阈值小于85km2时,模型的模拟效果较好。此外,面积阈值小于85km2时,模型模拟的水、碳循环变量验证的拟合相关性系数和效率系数均趋于稳定,因此可以将模型水文响应单元流域划分的面积阈值确定为85km2。基于这一尺度的模拟与验证研究,分析了WASSI-C模型中关键变量设置对模拟结果的影响。
Aims Optimal spatial scale of hydrological response unit （HRU） is a precondition for eco-hydrological modeling as it is essential to improve accuracy. Our objective was to evaluate the spatial scale of HRU for application of the WASSI-C model. Methods We determined the best HRU scale for the eco-hydrological model （WASSI-C） through examining the modeling accuracies at different HRU thresholds. This study focused on a large watershed, the upper Zagunao River watershed, situated in the upper reach of the Minjiang River, Yangtze River Basin, China. Important findings Variation of spatial scales in HRU significantly affected the modeling accuracy. With the increase of the spatial scale of HRU, the accuracy of simulated results first increased then remained relatively un-changed and then decreased, suggesting existence of a threshold around 85 km2 in HRU for this model for this watershed. We validated the model using this optimum spatial scale and discussed the potential to improve model output by addressing input parameters such as temperature.