高级检索+

亚热带常绿针叶林光能利用率模型优化与总初级生产力估算

Model Optimization and GPP Estimation of Light Energy Utilization in Subtropical Evergreen Coniferous Forest

    摘要
  • 摘要: 为使光能利用率模型在亚热带常绿针叶林区得到更好的应用,以江西省太和县千烟洲为例,利用2003—2005年卫星遥感数据(GLASS LAI叶面积指数产品数据集、MODIS地表反射率产品数据集（MOD09A1）和MODIS GPP产品（MOD17A2）)及中国陆地生态系统通量研究网络（ChinaFLUX）千烟洲常绿针叶林的通量观测数据,对13种不同光能利用率模型中的光合有效辐射吸收比率（Fraction of absorbed photosynthetically active radiation, FPAR）、7种水分限制因子（fθ）和3种温度限制因子（f_t）进行模型重组,通过对比通量站观测值与273种模型组合的估算值的决定系数及均方根误差（RMSE）得到更优的模型组合。同时使用该优化模型进行总初级生产力（Gross Primary Productivity, GPP）估算,并进行敏感性分析。结果表明,优化所得到的优选模型组合:用EVI表示的光合有效辐射吸收比率FPAREVI、3PG模型中的水分影响因子fθ-3PG、TEM模型中的温度影响因子ft-TEM(R2=0.86,RMSE=0.47μmol/（m2·s）)的模拟效果最好。优化模型的GPP模拟值均优于MODIS陆地四级标准数据产品（MOD17A2）。敏感性分析结果表明,最大光量子效率αmax、增强型植被指数（Enhanced vegetation index, EVI）和光合有效辐射（Photosynthetically active radiation, PAR）为模型的直接线性变量,对模型输出结果影响最大,其他参数敏感性由大到小排序依次为:光合最低温度Tmin、光合最适温度Topt、月均温度T、光合最高温度Tmax。因此,本文的优化模型具有较强的实用意义,对进一步提高GPP估算精度具有重要意义。

     

    Abstract: In order to better apply the light energy utilization model in subtropical evergreen coniferous forest areas, take Qianyanzhou, Taihe County, Jiangxi Province as an example, using satellite remote sensing data（GLASS LAI leaf area index product data set） from 2003 to 2005, MODIS surface reflectance product data set（MOD09 A1） and MODIS GPP product（MOD17 A2）) and China Terrestrial Ecosystem Flux Research Network（ChinaFLUX） the flux observation data of evergreen coniferous forests in Qianyanzhou, fraction of absorbed photosynthetically active radiation（FPAR）, 7 kinds of moisture limitation factors（fθ） and 3 kinds of temperature limitation factors（ft） in the 13 different energy efficiency models were reconstructed. A better model combination was obtained by comparing the coefficient of determination and the root mean square error（RMSE） of the flux station observations with the estimated values of 273 model combinations. At the same time, the optimization model was used to estimate gross primary productivity（GPP） and conduct sensitivity analysis. The results showed that the optimal model combination obtained by optimizing: the photosynthetically active radiation absorption ratio expressed in EVI, the moisture influence factor fθ-3 PG in the 3 PG model, and the temperature influence factor ft-TEM in the TEM model（R~2=0.86, RMSE=0.47 μmol/m~2·s） had the best simulation effect. The GPP simulation values of the optimized model were better than MODIS terrestrial level four standard data products（MOD17 A2）. Sensitivity analysis resultsshowed that the maximum light quantum efficiency αmax,enhanced vegetation index（ EVI） and photosynthetically active radiation（ PAR） were the direct linear variables of the model and had the greatest impact. Other parameters were determined by the order from big to small was light and minimum temperature Tmin,photosynthetic optimum temperature Topt,monthly average temperature T,and photosynthetic maximum temperature Tmax. Therefore,the optimization model in this paper had strong practical significance and was of great significance to further improve the accuracy of GPP estimation.

     

    • HTML全文
    • 参考文献(0)
    • 相关文章
    • 施引文献
  • 资源附件(0)

/

返回文章
返回