基于无人机多源数据的梭梭(Holoxylon ammodendron)地上生物量估算

Estimation of aboveground biomass of Haloxylon ammodendron based on UAV multi-source data

梭梭(Holoxylon ammodendron)是荒漠生态系统的重要组成,其地上生物量(aboveground biomass,AGB)在一定程度上反映了群落及生态系统的健康程度,对荒漠生态系统碳储量的估算具有重要意义。以乌兰布和沙漠梭梭为研究对象,基于无人机激光雷达提取的株高(H)、冠幅(C)参数和可见光数据构建的植被指数(VI)以及二者结合的3类特征指标,采用多元逐步回归、主成分回归和偏最小二乘回归,对梭梭地上生物量进行估算研究,利用决定系数R^(2)和均方根误差RMSE进行评价。结果表明:(1)无人机LiDAR点云数据提取的株高(R^(2)=0.85,RMSE=0.32 m)和冠幅(R^(2)=0.80,RMSE=0.73 m)精度较高。(2)多元逐步回归、主成分回归和偏最小二乘回归模型对梭梭地上生物量拟合效果均较好,R^(2)均大于0.8,而偏最小二乘回归(PLSR)模型的均方根误差最小(2.86 kg·株^(-1)),可以作为梭梭AGB估算的最优模型。利用无人机激光雷达数据可以有效地提取梭梭单株参数因子(H、C),并开展荒漠低矮植被地上生物量的估算研究。

As one of the important components of desert ecosystem,the aboveground biomass(AGB)of Haloxylon ammodendron reflects the health of community and ecosystem to a certain extent,which is of great significance to the carbon cycle of desert ecosystem.In this paper,H.ammodendron in Ulan Buh Desert was taken as the research object.Based on the plant height(H),crown width(C),visible light vegetation index(VI)extracted by UAV laser radar and the three characteristic indexes of the combination of the two,multiple stepwise regression(MSR),principal component regression(PCR)and partial least squares regression(PLSR)were used to estimate the above-ground biomass of H.ammodendron.The determination coefficient R^(2) and root mean square error RMSE were used for evaluation.The results showed that:(1)The accuracy of plant height(R^(2)=0.85,RMSE=0.32 m)and crown width(R^(2)=0.80,RMSE=0.73 m)extracted from UAV Li DAR point cloud data was high.(2)MSR,PCR and PLSR models had good fitting effects,and R^(2) was greater than 0.8,while the root mean square error of PLSR model was the smallest(2.86 kg·plant^(-1)),which could be used as the optimal model for AGB estimation of H.ammodendron.The results show that the use of UAV Li DAR data can effectively extract the single tree parameter factors(H,C)of H.ammodendron,and carry out the estimation of aboveground biomass of low vegetation in desert.