VPM模型最大光能利用效率参数优化对不同森林生态系统GPP模拟的影响  

Impacts of Optimizing Maximum Light Use Efficiency Parameter in VPM on GPP Simulation in Different Forest Ecosystems

作　　者：杨甫禹 张弥[1] 肖薇[1] 石婕 Yang Fuyu;Zhang Mi;Xiao Wei;Shi Jie(Center on Atmospheric Environment,Nanjing University of Information Science and Technology School of Ecology and Applied Meteorology,Nanjing University of Information Science and Technology,Nanjing 210044)

机构地区：[1]南京信息工程大学大气环境中心、南京信息工程大学生态与应用气象学院,南京210044

出　　处：《林业科学》2025年第3期86-99,共14页Scientia Silvae Sinicae

基　　金：国家自然科学基金国际(地区)合作与交流资助项目(42061144004)。

摘　　要：【目的】分析植被光合作用模型(VPM)中关键参数最大光能利用率(ε_(0))在不同气候区森林生态系统是否存在差异及其主要原因,并选择出具有普遍适用性的最大光能利用率参数化方案,以期深化对森林生态系统植被生产力估算过程中不确定性的认知,为提高模型模拟精度与降低模型参数不确定性提供参考。【方法】利用4种参数化方案——BPLUT查表法、Michealis-Menten光响应曲线方程拟合、生长季增强型植被指数最大值(EVI_(max))指数拟合以及Mointeith方程推导法对VPM中的最大光能利用率(ε_(0))进行估算,基于4种参数化方案的估算结果对中国地区4个森林生态系统总初级生产力(GPP)进行模拟,与涡度相关观测到的总初级生产力(GPP)进行比较并结合各项模型评价指标(决定系数R^(2)、均方根误差RMSE、一致性系数d及平均相对误差MRE)对VPM模拟结果进行评估。【结果】长白山、千烟洲、鼎湖山与西双版纳站点的ε_(0)值分别为:(0.65±0.14)、(0.47±0.10)、(0.44±0.09)和(0.69±0.12)g·mol^(−1)。在季节尺度上,利用各站点最优参数化方案模拟的GPP与观测GPP相比,其均方根误差(RMSE)较最不适用的参数化方案在长白山、千烟洲、鼎湖山和西双版纳分别降低了55.1%、38.1%、48.6%和34.3%;在年际尺度上,各站点最优参数化方案模拟GPP的平均相对误差(MRE)在长白山、千烟洲、鼎湖山和西双版纳分别为−7.9%、−24.3%、−7.4%和−3.0%,小于最不适用的参数化方案模拟的结果(长白山:35.8%;千烟洲:−53.4%;鼎湖山:29.8%;西双版纳:25.4%)。【结论】不同参数化方案在同一个站点的ε_(0)值差异较大,且相同参数化方案下,不同站点间的ε_(0)值存在差异,造成这种差异的主要原因与各参数化方案本身的结构属性及不同区域水热条件差异有关。Mointeith方程推导法为长白山、千烟洲与鼎湖山地区ε_(0)最优的参数化方案;生长�【Objective】To analyze whether there are differences in the maximum light use efficiency(ε_(0)),a key parameter in the vegetation photosynthesis model(VPM),in forest ecosystems of different climatic zones and the main reasons for such differences,and to select a parameterization scheme for the maximum light use efficiency that has general applicability.In order to deepen the knowledge of uncertainty in the process of estimating vegetation productivity in forest ecosystems and to provide a reference for improving model simulation accuracy and reducing the uncertainty of model parameters.【Method】The maximum light use efficiency(ε_(0))was estimated using the four selected parameterization schemes,BPLUT look-up table method,Michealis-Menten light response curve equation fitting,maxmiun enhanced vegetation index(EVI_(max))of the growing season exponential fitting,and Mointeith equation estimation method.Simulation of gross primary productivity(GPP)of four forest ecosystems in China based on the estimation results of the four parameterization schemes,comparison with eddy covariance observed GPP,and evaluation of VPM simulation results by combining various model evaluation indexes(coefficient of determination R^(2),root mean square error RMSE,the Willmott’s index of agreement d and mean relative error MRE).【Result】Theε_(0) values were(0.65±0.14g),(0.47±0.10),(0.44±0.09),and(0.69±0.12)g·mol,respectively for the Changbaishan,Qianyanzhou,Dinghushan,and Xishuangbanna.On the seasonal scale,compared with the GPP based on observation,root mean square error(RMSE)of GPP simulated by the optimal parameterization scheme forε_(0) is reduced by 55.1%(Changbaishan),38.1%(Qianyanzhou),48.6%(Dinghushan),and 34.3%(Xishuangbanna),respectively,than that simulated by the least applicable parameterization scheme.On the inter-annual scale,the mean relative errors(MRE)of the GPP simulated by the optimal parameterization scheme were-7.9%,-24.3%,-7.4%,and-3.0%respectively at Changbaishan,Qianyanzhou,Dinghushan,and Xishuang

关 键 词：最大光能利用率 参数化方案 VPM模型 总初级生产力 

分 类 号：S718.55[农业科学—林学]