协同空地数据的城市森林冠层结构预测物种多样性潜势研究——以哈尔滨市为例  

Potential of urban forest structural diversity to predict species diversity based on“air-ground”data:A case of Harbin

作　　者：王蕾[1,2] 贾佳 翟雅琳 荆忠伟 许大为 姚允龙[1] Wang Lei;Jia Jia;Zhai Yalin;Jing Zhongwei;Xu Dawei;Yao Yunlong(College of Landscape Architecture,Northeast Forestry University,Harbin 150040,Heilongjiang,China;Key Lab for Garden Plant Germplasm Development&Landscape Eco-restoration in Cold Regions of Heilongjiang Province,Harbin 150040,Heilongjiang,China)

机构地区：[1]东北林业大学园林学院,黑龙江哈尔滨150040 [2]黑龙江省寒区园林植物种质资源开发与景观生态修复重点实验室,黑龙江哈尔滨150040

出　　处：《地理科学》2024年第8期1481-1491,共11页Scientia Geographica Sinica

基　　金：国家自然科学基金项目(42171246);中央高校基本科研业务费专项资金(2572023AW37)资助。

摘　　要：本研究基于“空地”数据联岛的城市森林结构多样性监测技术,准确估算了林分尺度结构多样性参数,量化了森林高度、覆盖与开放度、外部与内部异质性特征,探讨了其对物种多样性的预测能力。结果表明:①覆盖与开放度、内部和外部异质性特征对物种多样性具有较高的预测能力(0.07<R^(2)<0.47);②结合所有结构多样性指标的模型对物种丰富度的预测能力更优(R^(2)=0.58,ΔAIC=0),仅包括覆盖与开放度指标的模型对香浓多样性指数(Shannonwiener)的预测能力更优(R^(2)=0.40,ΔAIC=0),仅包括外部异质性指标的模型对辛普森指数(Simpson)的预测能力更优(R^(2)=0.49,ΔAIC=0);③不同物种丰富度水平显著影响结构多样性参数与Shannon-wiener和Simpson的动态相关程度与相关关系,结构多样性特征联合调优的参数组合(GFP+VCI+Entropy)对预测物种多样性具有较好的潜力(0.48<R^(2)<0.56)。研究基于结构多样性空地监测推进森林生态系统正向演替的知识决策,为有效提升城市森林物种多样性提供科学依据。The diversity of urban forest structure directly quantifies the ecological niche occupancy of species distribution and canopy structure,and the advancement of drone technology offers a pivotal opportunity for multispectral coupling monitoring of urban forest species and structures.This article based on“air-ground”data,monitors the structural diversity of urban forests,accurately estimating forest height,cover and openness,external and internal heterogeneity characteristics at the stand scale,and explores the predictive capability of structural diversity on species diversity.The results indicate:1)The traits of coverage and openness,internal heterogeneity,and external heterogeneity have a strong predictive power for species diversity,with a coefficient of determination(R^(2))ranging from 0.07 to 0.47.2)A multiple linear regression model that incorporates all structural diversity indicators offers superior predictive capability for Richness,achieving an R^(2) of 0.58 and aΔAIC of 0.Models that only include the coverage and openness indicators perform best in predicting Shannon-wiener diversity,with an R^(2) of 0.40 and aΔAIC of 0,whereas models that solely incorporate external heterogeneity indicators excel in predicting Simpson diversity,with an R^(2) of 0.49 and aΔAIC of 0.3).Different levels of species richness significantly affect the relationship between structural diversity parameters and both Shannon-wiener and Simpson diversity indices.An optimized set of structural diversity parameters(GFP+VCI+Entropy)shows good potential for predicting species diversity,with an R^(2) between 0.48 and 0.56.This study aims to monitor forest structural diversity based on“air-ground”technological methods,advancing the knowledge decision-making leap in urban forest ecosystems and providing scientific support for effectively enhancing urban forest species diversity.

关 键 词：城市森林 结构多样性 物种多样性 空地数据 物种多样性预测能力 

分 类 号：S718.55[农业科学—林学]