无人机激光雷达人工林参数估测试验

doi:10.13466/j.cnki.lyzygl.2023.03.012

摘要/Abstract

摘要：

为探讨小区域森林资源调查监测中先进、可靠和可行的技术路径,对无人机激光雷达(UAV-LiDAR)森林参数估测和制图进行试验。采用13个刻画森林冠层三维结构、具有明确森林计测学和生态学解释意义的UAV-LiDAR变量,通过有规则的穷举法进行变量组合,得到86个森林参数估测模型式,每个模型式含2~5个变量;采用样地数据对全部模型式进行拟合和检验,得到6个森林参数估测优选模型。结果表明:松树、桉树人工林平均高、断面积和蓄积量估测模型的决定系数(R²)为0.616~0.853,相对均方根误差(rRMSE)为10.85%~18.79%,平均预报误差(MPE)为3.80%~9.72%。无人机激光雷达可实现森林参数的精确估测和制图,为小区域森林资源调查提供了全新技术手段,并且有效克服了传统地面调查存在的诸多问题。但是,在无人机激光雷达森林资源调查应用中,为进一步提高精度、降低调查成本,仍有很多技术问题需要加强研究。

关键词: 森林资源, 林分调查因子, 估测, 模型, 遥感

Abstract:

To explore advanced,reliable,and feasible technical schemes for small-scale forest resource inventory and monitoring,unmanned aerial vehicle-based LiDAR (UAV-LiDAR) was tested for estimating and mapping forest inventory attributes.Thirteen UAV-LiDAR-derived metrics,which depict the three-dimensional structural aspects of the forest canopy and have clear forest mensuration and ecology significance,were used to construct 86 multiplicative power formulations consisting of 2~5 predictors for forest inventory attribute estimation by using a rule-based exhaustive combination.All the formulations were calibrated and validated using the sample plot data,and six optimal models were achieved.The results indicated that the coefficients of determination (R²) of the mean stand height,basal area,and volume estimation for the pine and eucalyptus planted forests were 0.616~0.853,the relative root mean squared errors (rRMSE) were 10.85%~18.79%,and the mean predictive errors (MPE) were 3.80%~9.72%.With its ability to accurately estimate and map forest attributes,UAV-LiDAR provides an innovative technological tool for small-scale forest resource inventory,and effectively overcomes many of the problems of conventional field measurements.However,there are still numerous technical issues that need to be further investigated in the application of UAV-LiDAR to forest resource inventory to improve accuracy and reduce inventory costs.

Key words: forest resources, stand factor, estimation, model, remote sensing

中图分类号:

S771.8

周梅, 李春干, 杨承伶, 李振. 无人机激光雷达人工林参数估测试验[J]. 林业资源管理, 2023, 0(3): 90-97.

ZHOU Mei, LI Chungan, YANG Chengling, LI Zhen. Experiments on Estimating Planted Forest Inventory Attributes Based on UAV-LiDAR Data[J]. FOREST RESOURCES WANAGEMENT, 2023, 0(3): 90-97.

图/表 6

表1

表2

表3

表4

图1

图2

参考文献 27

[1]	Næsset E T. Gobakken J, Holmgren H, et al. Laser scanning of forest resources:The Nordic experience[J]. Scandinavian Journal of Forest Research, 2004, 19(6):482-499. doi: 10.1080/02827580410019553
[2]	White J C, Tompalski P, Vastaranta M, et al. A model development and application guide for generating an enhanced forest inventory using airborne laser scanning data and an area-based approach[R]. Victoria: Canadian Wood Fibre Centre, 2017.
[3]	李春干, 李振. 机载激光雷达大区域亚热带森林参数估测的普适性模型式[J]. 林业科学, 2021, 57(10):23-35.
[4]	代华兵, 李春干, 庞勇, 等. 基于天空地一体化森林资源调查的小班因子设置与信息获取方法[J]. 林业资源管理, 2021(2):180-188.
[5]	李春干, 代华兵. 中国森林资源调查:历史、现状与趋势[J]. 世界林业研究, 2021, 34(6):72-80.
[6]	Sankey T, Donager J, McVay J, et al. UAV lidar and hyperspectral fusion for forest monitoring in the southwestern USA[J]. Remote Sensing of Environment, 2017, 195:30-43. doi: 10.1016/j.rse.2017.04.007
[7]	Liu Kun, Shen Xin, Cao Lin, et al. Estimating forest structural attri-butes using UAV-LiDAR data in Ginkgo plantations[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2018, 146:465-482. doi: 10.1016/j.isprsjprs.2018.11.001
[8]	Cao Lin, Liu Kun, Shen Xin, et al. Estimation of forest structural parameters using UAV-LiDAR data and a process-based model in Ginkgo planted forests[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2019, 12(11):4175-4190. doi: 10.1109/JSTARS.2019.2918572
[9]	D'Oliveira M V N, Broadbent E N, Oliveira L C, et al. Aboveground biomass estimation in Amazonian Tropical Forests:A comparison of aircraft- and gatorEye UAV-borne LiDAR data in the chico mendes extractive reserve in Acre,Brazil[J]. Remote Sensing, 2020, 12:1754. doi: 10.3390/rs12111754
[10]	Corte A P D, de Vasconcello B N, Rex F E, et al. Applying high-resolution UAV-LiDAR and quantitative structure modelling for estimating tree attributes in a crop-livestock-forest system[J]. Land, 2022, 11:507. doi: 10.3390/land11040507
[11]	Xu Dandan, Wang Haobin, Xu Weixin, et al. LiDAR applications to estimate forest biomass at individual tree scale:Opportunities,challenges and future perspectives[J]. Forests, 2021, 12(5):550. doi: 10.3390/f12050550
[12]	Corte A P D, Souza D V, Rex F E, et al. Forest inventory with high-density UAV-Lidar:Machine learning approaches for predicting individual tree attributes[J]. Computers and Electronics in Agriculture, 2020, 179:105815. doi: 10.1016/j.compag.2020.105815
[13]	Cao Lin, Liu Kai, Shen Xin, et al. Estimation of forest structural parameters using UAV-LiDAR data and a process-based model in Ginkgo planted forests[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2019, 12(11):4175-4189. doi: 10.1109/JSTARS.2019.2918572
[14]	Peng Xi, Zhao Anjiu, Chen Yongfu, et al. Comparison of modeling algorithms for forest canopy structures based on UAV-LiDAR:A case study in tropical china[J]. Forests, 2020, 11:1324. doi: 10.3390/f11121324
[15]	Neuville R Bates J S, Jonard F. Estimating forest structure from UAV-mounted LiDAR point cloud using machine learning[J]. Remote Sensing, 2021, 13:352. doi: 10.3390/rs13030352
[16]	Næset E. Predicting forest stand characteristics with airborne scanning laser using a practical two-stage procedure and field data[J]. Remote Sensing of Environment, 2002, 80(1):88-99. doi: 10.1016/S0034-4257(01)00290-5
[17]	周梅, 王新华, 李春干, 等. 不同样地面积对人工林林分参数的影响[J]. 西部林业科学, 2018, 47(1):110-116.
[18]	Li Chungan, Lin Xin, Dai Huabing, et al. Effects of plot size on airborne LiDAR-derived metrics and predicted model performances of subtropical planted forest attributes[J]. Forests, 2022, 13:2124. doi: 10.3390/f13122124
[19]	Li Chungan, Chen Zhongchao, Zhou Xiangbei, et al. Generalized models for subtropical forest inventory attribute estimations using a rule-based exhaustive combination approach with airborne LiDAR-derived metrics[J]. Giscience & Rremote Sensing, 2023, 60(1):2194601.
[20]	Ferster C J, Coops N C, Trofymow J A. Aboveground large tree mass estimation in a coastal forest in British Columbia using plot-level metrics and individual tree detection from lidar[J]. Canadian Journal of Remote Sensing, 2009, 35(3):270-275. doi: 10.5589/m09-014
[21]	Hall S A, Burke I C, Box D O, et al. Estimating stand structure using discrete-return lidar:An example from low density,fire prone ponderosa pine forests[J]. Forest Ecology and Management, 2005, 208(1):189-209. doi: 10.1016/j.foreco.2004.12.001
[22]	曾伟生, 唐守正. 立木生物量方程的优度评价和精度分析[J]. 林业科学, 2011, 47(11):106-113.
[23]	Coops N C, Tompalski P, Goodbody T R H, et al. Modelling lidar-derived estimates of forest attributes over space and time:A review of approaches and future trends[J]. Remote Sensing of Environment, 2021, 260:112477. doi: 10.1016/j.rse.2021.112477
[24]	Gobakken T, Næsset E. Assessing effects of laser point density,ground sampling intensity,and field sample plot size on biophysical stand properties derived from airborne laser scanner data[J]. Canadian Journal of Forest Research, 2008, 38:1095-1109. doi: 10.1139/X07-219
[25]	余铸, 李春干, 苏凯. 等. 基于垂直结构分类的机载激光雷达森林参数估测[J/OL]. 桂林理工大学学报.(2022-05-06)[2023-05-16]. http://kns.cnki.net/kcms/detail/45.1375.n.20220429.1213.002.html.
[26]	曾伟生, 孙乡楠, 王六如, 等. 基于机载激光雷达数据的森林蓄积量模型研建[J]. 林业科学, 2021, 57(2):31-38.
[27]	Liu Hao, Cao Lin, She Guanghui, et al. Extrapolation assessment for forest structural parameters in planted Forests of southern China by UAV-LiDAR samples and multispectral satellite imagery[J]. Remote Sensing, 2022, 14(11):2677. doi: 10.3390/rs14112677

森林类型	样地数量	林木密度/ (株/hm²)	平均直径		平均高		断面积		蓄积量
森林类型	样地数量	林木密度/ (株/hm²)	均值/ cm	CV/ %	均值/ cm	CV/ %	均值/ (m²/hm²)	CV/ %	均值/ (m³/hm²)	CV/ %
松树林	33	717~2 233	26.2	24.6	17.5	26.2	32.9	23.4	242.7	30.5
桉树林	35	1 067~3 450	11.1	28	15.2	30.6	18.3	53.9	150.6	65.3

变量	含义	冠层三维结构的刻画角度	变量组
hp95	95%分位数高度	冠层高度	高度变量
Hmean	点云平均高	冠层高度	高度变量
Hstd	点云高度的标准差	冠层高度	高度变量
Hcv	点云高度的变动系数	冠层高度	高度变量
CC	郁闭度(修正)	冠层密度	密度变量
dp50	50%分位数密度	冠层密度	密度变量
dp75	75%分位数密度	冠层密度	密度变量
LADmean	叶面积密度均值	垂直结构异质性	垂直结构变量
LADstd	叶面积密度标准差	垂直结构异质性	垂直结构变量
LADcv	叶面积密度变动系数	垂直结构异质性	垂直结构变量
VFPmean	枝叶垂直剖面均值	垂直结构异质性	垂直结构变量
VFPstd	枝叶垂直剖面标准差	垂直结构异质性	垂直结构变量
VFPcv	枝叶垂直剖面变动系数	垂直结构异质性	垂直结构变量

森林类型	森林参数	模型式
松树林	蓄积量(VOL)/m³	VOL_Pine=a0Hmean^a¹CC^a²VFPstd^a³Hstd^a⁴dp75^a⁵
	断面积(BA)/m²	BA_Pine=a0Hmean^a¹CC^a²VFPstd^a³Hcv^a⁴dp50^a⁵
	平均高(H)/m	H_Pine=a0hp60^a¹hp70^a²hp80^a³CC^a⁴dp50^a⁵
桉树林	蓄积量(VOL)/m³	VOL_Eucalyptus=a0hp95^a¹CC^a²VFPstd^a³Hcv^a⁴dp75^a⁵
	断面积(BA)/m²	BA_Eucalyptus=a0hp95^a¹CC^a²VFPstd^a³Hstd^a⁴dp75^a⁵
	平均高(H)/m	H_Eucalyptus=a0hp60^a¹hp80^a²

森林类型	森林参数	样地数量	模型参数估计值							修正因子 (CF)		拟合指标						检验指标
森林类型	森林参数	样地数量	a0	a1	a2	a3	a4	a5		修正因子 (CF)		R²		rRMSE/ %		MPE/ %		R²	rRMSE/ %		MPE/ %
松树林	VOL	33	0.172 70	1.474 9	1.329 4	-0.163 6	-0.263 800	-1.381 2	1.009 8		0.783		12.44		4.69		0.708			13.86		5.36
	BA	33	0.694 70	0.701 1	0.887 0	-0.204 4	-0.365 000	-0.981 7	1.008 4		0.676		11.94		4.51		0.616			14.26		5.04
	H	33	-0.080 12	-5.177 1	8.299 5	-2.391 0	1.136 600	-1.012 3	1.007 0		0.846		10.30		3.88		0.794			11.06		4.15
桉树林	VOL	35	-0.288 50	1.837 9	0.661 7	0.211 3	-0.002 804	0.144 5	1.018 4		0.943		15.71		5.82		0.853			17.79		6.22
	BA	35	-0.453 30	1.252 4	0.592 9	0.132 7	-0.054 760	0.161 8	1.019 1		0.903		15.91		5.89		0.844			16.72		6.72
	H	35	0.055 13	-2.560 6	3.471 9				1.005 9		0.899		9.87		3.59		0.833			10.85		3.81