摘要
基于云南省昭通市大关县天星镇林分二调数据,以林分蓄积量作为因变量,将郁闭度、平均胸径、平均海拔和坡度作为自变量,采用最小二乘法模型(OLS)和地理加权回归模型(GWR),对林分蓄积量的空间效应进行分析。结果表明:1)距离在6 075 m以内,林分蓄积量在总体上呈现出显著的空间自相关性,且随距离的增大自相关性逐渐减小最后趋于稳定;2)随着间隔距离的增加,基台值逐渐减小,块金值逐渐增加,在距离达到6 075 m时趋于稳定;3)平均胸径、郁闭度和平均海拔与林分蓄积量呈现出正相关性,而坡度与林分蓄积量具有负相关性;4)GWR模型的R2(0.73)和预测精度P(0.60)均大于OLS模型R2(0.34)和预测精度P(0.54)。林分蓄积量在空间尺度上具有异质性,且GWR模型能很好地描述林分蓄积量的空间异质性,采用合理的经营方式以促进林木胸径、郁闭度的生长,同时考虑海拔和坡度的影响,从而促进林分蓄积量的增长。
Based on the forest management inventory in Tianxing Township,Daguan County,Zhaotong City,Yunnan Province,with stand volume as the dependent variable and canopy closure,average DBH,average altitude and slope as independent variables,the least squares model(OLS)and geographically weighted regression models(GWR)were used to analyze the spatial effects of stand stock.The results showed that 1)within the distance of 6075 m,the stand volume presented a significant spatial autocorrelation on the whole,and the autocorrelation gradually decreased with the increase of the distance and finally stabilized.2)With the increase of separation distance,the abutment value gradually decreased,the nugget value gradually increased,and tended to be stable when the distance reached 6075 m.3)The average DBH,canopy closure and average altitude showed positive correlation with the stand volume,while the slope was negatively correlated with the stand volume.The stand volume had a negative correlation with slope degree.4)The R 2(0.73)and prediction accuracy P(0.60)of the GWR model were both greater than those of the OLS model R 2(0.34)and prediction accuracy P(0.54).The stand volume has heterogeneity on the spatial scale,and the GWR model can well describe the spatial heterogeneity of the stand volume.Reasonable management methods should be adopted to promote the growth of tree diameter at breast height and canopy density,and at the same time,the effects of altitude slope degree should be taken into consideration to promote the growth of stand volume.
作者
和云润
陈朝情
殷唐燕
刘畅
张加龙
魏晓燕
HE Yun-run;CHEN Chao-qing;YIN Tang-yan;LIU Chang;ZHANG Jia-long;WEI Xiao-yan(Forestry College,Southwest Forestry University,Kunming 650224,Yunnan,China;Yunnan Provincial Surveying and Mapping/Yunnan Provincial Geomatics Centre,Kunming 650003,Yunnan,China)
出处
《西北林学院学报》
CSCD
北大核心
2023年第3期169-175,共7页
Journal of Northwest Forestry University
基金
国家自然科学基金项目(42061074,41701470)
云南省国有自然资源资产权益管理试点项目(632171)
云南省高层次人才培养支持计划“青年拔尖人才”专项(YNWR-QNBJ-2020-164)。
关键词
空间异质性
空间非平稳性
林分蓄积量
最小二乘法模型
地理加权回归模型
spatial heterogeneity
spatial nonstationarity
forest stand volume
ordinary least square
geographical weighted regression model
