论文部分内容阅读
以2003—2008年常规气象梯度观测数据为驱动数据,应用多层模型模拟了不同冠层分层数下长白山温带针阔混交林净生态系统生产力(NEP),对比涡度相关通量观测数据,以确定系数(R~2)最大及均方根误差(RMSE)最小作为评判标准,分析评价了不同冠层分层数对NEP模拟精度的影响,得出如下结论:(1)模型在不同分层情况下均具有较好的NEP日变化再现能力,当冠层分层数较少时,表现出NEP低估的倾向,而分层数较多时有高估倾向;(2)当分层数较少时,随着分层数的增加,模型模拟NEP日变化的能力增强,模拟误差变小;当分层数增加到一定数值,随着分层数的继续增加,模型模拟NEP日变化的能力趋于平稳,而模拟误差呈增大趋势;(3)采用多层模型对长白山温带针阔混交林生态系统NEP进行模拟时,以将冠层分为4~7层为宜。研究结果可为陆地生态系统NEP的模型研究提供参考。
Using the conventional meteorological gradient observation data from 2003 to 2008 as the driving data, the net ecosystem productivity (NEP) of temperate coniferous and broad-leaved mixed forests in Changbai Mountain under different canopy stratification numbers was simulated by using multi-layer model, and the eddy covariance fluxes were compared. In order to determine the maximum of coefficient (R ~ 2) and the minimum of root mean square error (RMSE), the influence of different canopy layers on NEP simulation accuracy was analyzed and evaluated. The conclusions are as follows: (1) The results showed that NEP had a tendency of underestimation when the number of canopy stratification was small, and the tendency of overestimation when there were many stratification; (2) When the number of stratification was less , With the increase of number of stratification, the ability of model to simulate diurnal variation of NEP is enhanced and the simulation error is reduced. When the number of stratification increases to a certain value, the model simulates diurnal variation of NEP as the number of stratification continues to increase (3) When applying multi-layer model to simulate the NEP in the temperate coniferous and broad-leaved mixed forest in Changbai Mountain, it is advisable to divide the canopy into 4 ~ 7 layers. The results can provide a reference for the model research of terrestrial ecosystem NEP.