基于RF与BPNN的海南清澜港红树植物叶片生态化学计量高光谱反演  

Hyperspectral retrieval of leaf ecological stoichiometry of mangrove species with RF and BPNN models in Qinglangang Mangrove Nature Reserve,Hainan

作　　者：李化哲 窦志国 聂磊超 王俊杰 高常军[4] 唐希颖 翟夏杰 李伟[1,2] LI Huazhe;DOU Zhiguo;NIE Leichao;WANG Jun-jie;GAO Changjun;TANG Xiying;ZHAI Xiajie;LI Wei(Institute of Wetland Research,Chinese Academy of Forestry,Beijing Key Laboratory of Wetland Ecological Frunction and Restoration,Beijing 100091,China;Institute of Ecological Conservation and Restoration,Chinese Academy of Forestry,Beijing 100091,China;College of Life Sciences and Oceanography,Shenzhen University,Shenzhen 518060,Guangdong,China;Guang-dong Acadermy of Forestry,Guangdong Provincial Key Laboratory of Siluiculture,Protection and Utilization,Guang-zhou 510520,China)

机构地区：[1]中国林业科学研究院湿地研究所,湿地生态功能与恢复北京市重点实验室,北京100091 [2]中国林业科学研究院生态保护与修复研究所,北京100091 [3]深圳大学生命与海洋科学学院,广东深圳518060 [4]广东省林业科学研究院,广东省森林培育与保护利用重点实验室,广州510520

出　　处：《生态学杂志》2024年第8期2523-2530,共8页Chinese Journal of Ecology

基　　金：国家重点研发计划项目(2017YFC0506200)资助。

摘　　要：红树林是天然的海岸防御屏障,在沿海防灾减灾中具有不可替代的作用,了解红树林生长状况颇为重要。植物的生态化学计量能够反映其养分贮存和供应能力,应用高光谱数据量化红树植物的生态化学计量,探讨红树植物叶片高光谱反演的精确性和稳定性,可以为红树林生长状况的快速遥感监测提供技术参考。本研究在海南清澜港红树林自然保护区采集海莲(Bruguiera sexangula)、角果木(Ceriops tagal)、正红树(Rhizophora apiculata)3种优势植物的高光谱数据,对其C、N、P含量及其生态化学计量特征进行反演。结果表明:3种红树植物叶片的C、N、P含量及其生态化学计量存在显著性差异,说明3种红树植物对养分的利用存在差异;利用R^(2)、RMSE以及RPD进行评价,随机森林模型(Random Forest,RF)优于反向传播神经网络模型(Back Propagation Neural Network,BPNN);利用高光谱数据可以实现对红树林C、N、P含量及其生态化学计量特征的准确反演,2种模型对每种红树植物的反演效果也存在差异,整体上RF模型的反演准确性与稳定性好,是反演红树林生态化学计量的较优选择。Mangrove forest is a natural coastal defense barrier,which plays an irreplaceable role in coastal disaster prevention and mitigation.Therefore,it is important to understand the growth status of mangroves.The ecological stoichiometry of plants can reflect their nutrient storage and supply capacity.Applying hyperspectral data to quantify the ecological stoichiometry of mangrove plants and exploring the accuracy and stability of hyperspectral retrieval of leaves can provide a technical reference for rapid remote sensing monitoring of mangrove growth conditions.In this study,we collected hyperspectral data of leaves of three dominant mangrove species(Bruguiera sexangula,Ceriops tagal,and Rhizophora apiculata)in Qinglangang Mangrove Nature Reserve,Hainan,and retrieved the contents and stoichiometry of C,N,and P.The results showed that there were significant differences in the contents and stoichiometry of C,N,and P among the three species,indicating differences in nutrient utilization of the three mangrove species.The Random Forest(RF)model outperformed Back Propagation Neural Network(BPNN)mod-el in retrieving C,N,P contents and their ecological stoichiometry considering R^(2),RMSE and RPD.This study demonstrated that the contents and stoichiometry of C,N,and P in mangrove leaves could be accurately estimated by leaf hyperspectral data.RF model is recommended for the hyperspectral retrieval of mangrove ecological stoichiometry when considering model accuracy and robustness.

关 键 词：机器学习模型 高光谱 生态化学计量 红树林 

分 类 号：S718.5[农业科学—林学]