论文部分内容阅读
叶面积指数(LAI)是森林生态系统碳循环研究的重要观测数据,也是驱动森林生态系统模型模拟碳循环的重要参数.本文以毛竹林和雷竹林为研究对象,首先利用双集合卡尔曼滤波,同化两种竹林生态系统观测站点2014—2015年MODIS LAI时间序列数据,然后将同化的高质量毛竹LAI和雷竹LAI作为输入数据驱动BEPS模型,模拟两种竹林生态系统总初级生产力(GPP)、净生态系统碳交换量(NEE)和总生态系统呼吸(TER)等碳循环数据,并用通量站实际观测值评价模拟结果;另外,还对比不同质量LAI对碳循环模拟的影响.结果表明:双集合卡尔曼滤波同化得到的毛竹林和雷竹林LAI与实测LAI之间的相关关系极为显著,R~2分别为0.81和0.91,且均方根误差和绝对偏差均较小,极大地提高了MODIS LAI的产品精度;在同化得到的LAI驱动下,BEPS模型模拟的毛竹林GPP、NEE和TER与实际观测值之间的R~2分别为0.66、0.47和0.64,雷竹林分别为0.66、0.45和0.73,模拟结果均好于三次样条帽盖算法平滑LAI模拟得到的GPP、NEE和TER,其中,毛竹林、雷竹林NEE的模拟精度提高幅度最大,分别为11.2%和11.8%.
Leaf area index (LAI) is an important observation data of carbon cycle research in forest ecosystem and also an important parameter for driving forest ecosystem model to simulate carbon cycle.In this paper, bamboo and Phyllostachys pubescens forest were taken as research objects, Assimilating the MODIS LAI time series data of two kinds of bamboo forest ecosystems observation stations from 2014 to 2015, and then using the assimilated high quality bamboo LAI and bamboo LAI as input data to drive the BEPS model to simulate the total primary productivity (GPP) of two bamboo forest ecosystems, Carbon cycle data such as net ecosystem carbon exchange (NEE) and total ecosystem respiration (TER) were calculated, and the simulation results were evaluated by actual observations of flux station. In addition, the effects of different quality LAI on carbon cycle simulation were compared.The results showed that: The correlation between LAI and measured LAI of Phyllostachys pubescens plantations and Phyllostachys pubescens plantations assimilated by the double-set Kalman filter is extremely significant with R ~ 2 values of 0.81 and 0.91, respectively, and the rms and absolute deviations are both small and greatly improved MODIS LAI product accuracy; Under the assimilation of LAI driven by the BEPS model of bamboo forest GPP, NEE and TER and the actual observations of R ~ 2 were 0.66,0.47 and 0. 64 and Ray bamboo forest were 0.66, 0.45 and 0.73, respectively. The simulation results were better than the GPP, NEE and TER obtained by the smooth LAI simulation of the cubic spline cap algorithm. Among them, the simulation accuracy of NEE in Moso bamboo forest and Lei bamboo forest increased the most respectively 11.2% and 11.8%.