摘要
根据2008和2009年野外试验资料，应用CoupModel模型，对位于长江三峡库区末端的重庆市四面山阔叶林（木荷×石栎）和针叶林（杉木×马尾松）0-80cm土层的土壤水分和温度进行了逐日模拟验证，并在此基础上采用OAT方法（即每次只改变其中1个参数的方法），对模型参数进行了敏感性分析。验证结果表明，CoupModel能较好地模拟土壤水分和温度的动态变化，2种林分土壤水分决定系数（R。）、平均误差（ME）、均方根误差（RMSE）、Nash效率系数（NSE）分别为0．81-0．94，-0．01％-0．26％，0．24％-1．83％和0．80～O．87；土壤温度的R2，ME，RMSE，NSE分别为0．92～0．99，-0．14～0．06℃，0．18-0．34℃，0．90-0．98。说明该模型在该地区表现出良好的适用性。参数敏感性分析表明，对土壤水分模拟结果影响较大的参数有孔隙分布参数、进气吸力、残留含水量、饱和含水量、饱和导水率和蒸发阻力系数（PsiRs—lp），对土壤温度模拟结果影响较大的参数有土壤热传导系数、有机质层厚度、植被反照率、消光系数、水汽压亏缺和蒸发阻力系数。
A physical process-based model（CoupModel） was applied to simulate daily soil moisture and tem- perature of broadleaved forest （Schima superba ×Lithocarpus glaber） and coniferous forest （Cunninghamia lanceolata × Pinus massoniana） in Simian Mountain, which located in the Three Gorges Reservoir area. Then, the OAT method（one factor at a time） was adopted to analyze sensitivities of the model parameters. The validation results showed that CoupModel could fairly simulate the dynamic changes of soil moisture and temperature. Determination coefficient（Re）, mean error（ME）, root mean square error（RMSE）, and Nash- sutcliffe efficiency coefficient（NSE） of soil moisture were 0.81-0.94, --0.01%-0.26%, 0. 24%-1. 83% and 0.80-0.87 respectively, while they were 0.92-0.99, --0.14-0.06 ℃, 0.18-0.34 ℃, and 0.90- 0.98 for soil temperature. It meant that the model had a good applicability in this region. The results of the sensitivity analysis indicated that, many parameters had great influence on the simulation of soil moisture, such as lambda, air entry, residual water, saturation, matrix conductivity and PsiRs-lp. While scaling coef- ficient, organic layer thick, plant albedo, light extinction coefficient, cond VPD and PsiRs-lp impacted on simulation of soil temperature.
出处
《水土保持通报》
CSCD
北大核心
2013年第6期295-301,共7页
Bulletin of Soil and Water Conservation
基金
国家林业局林业科技推广项目"三峡库区低山丘陵区水土保持型植被群落建设技术"(2012-42)
国家自然科学基金项目(41271300
30900866)
中央高校基本科研业务费专项(TD2011-2)
