高级检索+
曾琪, 余坤勇, 姚雄, 郑文英, 张今朝, 艾婧文, 刘健. 基于PROSAIL辐射传输模型的毛竹林分冠层反射率模拟研究[J]. 植物科学学报, 2017, 35(5): 699-707. DOI: 10.11913/PSJ.2095-0837.2017.50699
引用本文: 曾琪, 余坤勇, 姚雄, 郑文英, 张今朝, 艾婧文, 刘健. 基于PROSAIL辐射传输模型的毛竹林分冠层反射率模拟研究[J]. 植物科学学报, 2017, 35(5): 699-707. DOI: 10.11913/PSJ.2095-0837.2017.50699
Zeng Qi, Yu Kun-Yong, Yao Xiong, Zheng Wen-Ying, Zhang Jin-Zhao, Ai Jing-Wen, Liu Jian. Research on canopy reflectance modeling of Phyllostachys pubescens forest based on the PROSAIL canopy radiative transfer model[J]. Plant Science Journal, 2017, 35(5): 699-707. DOI: 10.11913/PSJ.2095-0837.2017.50699
Citation: Zeng Qi, Yu Kun-Yong, Yao Xiong, Zheng Wen-Ying, Zhang Jin-Zhao, Ai Jing-Wen, Liu Jian. Research on canopy reflectance modeling of Phyllostachys pubescens forest based on the PROSAIL canopy radiative transfer model[J]. Plant Science Journal, 2017, 35(5): 699-707. DOI: 10.11913/PSJ.2095-0837.2017.50699

基于PROSAIL辐射传输模型的毛竹林分冠层反射率模拟研究

Research on canopy reflectance modeling of Phyllostachys pubescens forest based on the PROSAIL canopy radiative transfer model

  • 摘要: 冠层光谱反射率直接关系到毛竹(Phyllostachys pubescens Mazel)林冠层参数的反演,对毛竹林地土壤肥力间接估测具有重要意义。以PROSPECT5、PROSAIL模型为基础,从叶片尺度和冠层尺度分析模型参数对叶片和冠层反射率的影响,构建毛竹冠层叶面积指数(LAI)-冠层反射率查找表并通过代价函数选取最优冠层反射率,从而实现毛竹林分冠层反射率的准确模拟。结果表明,在叶片尺度,PROSPECT模型参数敏感性从高到低依次为叶肉结构参数(N) > 叶绿素含量(Cab) > 等效水厚度(EWT) > 干物质含量(Cm) > 类胡萝卜素含量(Car);在冠层尺度,PROSAIL模型参数敏感性从高到低依次为LAI > Cab > EWT > Cm > N > Car > ALA(平均叶倾角);叶片尺度反射率整体大于冠层尺度反射率;在400~900 nm波长范围内,PROSAIL模型模拟的冠层光谱反射率与实测光谱反射率拟合效果较好,相对误差为6.71%。

     

    Abstract: Canopy spectral reflectance is directly related to inversion of Phyllostachys pubescens Mazel canopy parameters, which is of great significance in the indirect estimation of P. pubescens forest soil fertility. The effects of model parameters on leaf and canopy reflectance at the leaf and canopy scales were analyzed based on PROSPECT and PROSAIL models. PROSAIL canopy radiative transfer was used to establish the leaf area index (LAI) and a canopy reflectance lookup table for P. pubescens forest, with optimum canopy reflectance then determined by the cost function to achieve accurate simulation of P. pubescens forest canopy reflectance. Results showed that at the leaf scale, the sensitivity of the input parameters in the PROSPECT model decreased in the order of leaf structure parameters (N) > chlorophyll content (Cab) > equivalent water thickness (EWT) > dry matter content (Cm) > carotenoid content (Car). At the canopy scale, the sensitivity of the input parameters in the PROSAIL model decreased in the order of LAI > Cab > EWT > Cm > N > Car > ALA (average leaf angle). Leaf reflectance was greater than canopy reflectance. In the 400 to 900 wavelength range, canopy spectral reflectance of the PROSAIL model exhibited better fitting than the measured reflectance, with a relative error of 6.71%.

     

/

返回文章
返回