Potential of urban forest structural diversity to predict species diversity based on"air-ground"data:A case of Harbin
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 coeffi-cient of determination(R2)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 R2 of 0.58 and a △AIC of 0.Models that only include the coverage and openness indicators perform best in predicting Shan-non-wiener diversity,with an R2 of 0.40 and a △AIC of 0,whereas models that solely incorporate external het-erogeneity indicators excel in predicting Simpson diversity,with an R2 of 0.49 and a △AIC 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 R2 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.
urban forestsstructural diversityspecies diversityair-ground datapredictive potential of spe-cies diversity
万方数据
