摘要
利用千烟洲亚热带人工针叶林气象和碳水通量观测数据,采用OAT局部敏感性分析方法对陆地生态过程(CEVSA)模型的参数进行了敏感性分析,识别了模型的关键参数,并利用差分进化马尔科夫链算法结合净生态系统碳交换量NEE观测数据优化了关键参数,对比分析了2003-2005年参数优化前后千烟洲人工针叶林净生态系统生产力NEP模拟的效果.结果表明:共有40个参数为敏感性参数,其中光合作用参数植物氮吸收Ns和植物氮吸收Nc1敏感性最高,异养呼吸参数地表微生物碳库分解速率K3和土壤微生物碳库分解速率K4敏感性最低,影响NEE的光合作用参数同时会影响蒸散发ET.16个关键参数中,有12个能够被NEE观测数据有效约束,但是对于光合参数rubisco对CO2浓度特定反应τ2、异养呼吸参数缓性土壤碳库分解速率K7和植被残体氮NITG及土壤容重kw,仅使用NEE数据无法有效约束.利用优化后的一套参数集模拟了2003-2005年千烟洲人工针叶林NEP的变化,发现较参数优化前有大幅度改善,尤其在每年植被生长季节最为显著,且优化后模拟的R2=0.32,优化前为0.19,均方根误差优化后为7.56,优化前为13.42,纳什效率系数优化后为0.22,优化前为-1.47.本研究为CEVSA模型中关键参数的筛选提供初步的指导,为模型在千烟洲亚热带人工针叶林的应用提供一套参数优化方案,对进一步加深理解模型参数和改进模型结构有借鉴意义.
Meteorological and carbon-water flux observation data of subtropical coniferous plantation in Qiangyanzhou and the OAT local sensitivity analysis method were used to conduct a sensitivity analysis of parameters based on the terrestrial ecological process model:CEVSA model,the key parameters of the model were identified.The differential evolution of Markov chain algorithm combined with net ecosystem carbon exchange(NEE)observation data were utilized to optimize the key parameters and to analyze the net ecosystem productivity(NEP)simulation of subtropical coniferous plantation in Qiangyanzhou before and after the parameter optimization from 2003 to 2005.The results showed that a total of 40 parameters were sensitive.The photosynthetic parameters:plant nitrogen absorption Nsand plant nitrogen absorption Nc1 were the two most sensitive parameters,while heterotrophic respiration parameters:potential decay rate of soil microbe K3 and potential decay rate of surface microb K4-were the two lowest sensitive parameters.Photosynthetic parameters which can affect NEE also affected evapotranspiration(ET).Of the 16 key parameters,12 could be effective constraint NEE observation data,but for photosynthetic parameters:rubisco reacted specifically to CO2 concentrationsτ2,heterotrophic respiration parameters:the potential decay rate of slow soil organic K7,Nitrogen content of plant residue NITG,and soil bulk density kw,the NEE data could not effectively constrain them.The optimized set of parameters was used to simulate the NEP changes in the subtropical coniferous plantation in Qiangyanzhou from2003 to 2005,it was found that there was a great improvement after parameter optimization,especially in the biggest vegetation-growing season in a year.The R2 of the optimized simulation was 0.32,while it was only 0.19 before optimization;the root mean square error was 7.56 after optimization,while it was13.42 before.The Nash Sutcliffe coefficient was 0.22 after optimization,while it was 1.47 before.This study can provide some preliminary guidance on the selection of key parameters based on the CEVSA model,a set of parameter optimization schemes for the application of the model in the subtropical coniferous plantation in Qiangyanzhou,thus being useful to people in further understanding the model parameters and improving the model structure.
作者
刘晓文
韩拓
陈惠玲
许洁
牛忠恩
朱高峰
Liu Xiao-wen;Han Tuo;Chen Hui-ling;Xu Jie;Niu Zhong-en;Zhu Gao-feng(School of Earth and Environmental Sciences,Lanzhou University,Lanzhou 730000,China;Institute of Geographic Sciences and Natural Resources Research,Chinese Academy of Sciences,Beijing 100101,China)
出处
《兰州大学学报（自然科学版）》
CAS
CSCD
北大核心
2020年第5期700-710,共11页
Journal of Lanzhou University(Natural Sciences)
基金
国家重点研究发展计划项目(2016YFC0500203)
关键词
CEVSA模型
参数敏感性分析
参数优化
优化评价
CEVSA model
parameter sensitivity analysis
parameter optimization
optimization evaluation
作者简介
通讯联系人:朱高峰(1978-),男,山东栖霞人,教授,博士,e-mail:zhugf@lzu.edu.cn,研究方向为生态水文.
