无损方法实现重庆地区马尾松材积模型研建

doi:10.13304/j.nykjdb.2022.0102

摘要/Abstract

摘要：

重庆地区立木材积表一直沿用的是四川马尾松材积模型，并不完全适用于实际情况，因此需要建立新的材积模型。因伐木进行数据采集较为困难，所以使用无损方法采集数据。通过摄影测量的方式测量林木材积，将样本分为80%的建模样本和20%的验证样本，使用建模样本建立多种一元、二元材积模型，使用验证样本应用《二元立木材积表编制技术规程》中相关指标对多种模型进行对比分析。结果表明，采用摄影测量方法获取数据，单木胸径及单木树高的偏差、均方根误差、相对偏差和相对均方根误差的值均偏低，结果完全满足调查需要，表明可以通过摄影测量的方式获取数据来进行材积模型的研建；通过多种指标相结合确定多种材积模型中效果最好的模型，R²达到了0.966，预估精度97.69%；新研建的二元材积模型相较于传统的四川马尾松材积模型在总相对误差、相对误差平均值、预估精度的评价指标效果更好，可以作为重庆地区马尾松立木材积表建立依据。

关键词: 摄影测量, 材积模型, 立木材积表

Abstract:

The volume model of Sichuan masson pine has been used in Chongqing， so it is not completely applicable to the actual situation. It is necessary to establish a new volume model. It is difficult to collect data by logging， so lossless method is used to collect data. The volume of trees was measured by photogrammetry，the samples were divided into modeling and validation samples， with 80% for modeling and 20% for validation. Various one-dimensional and two-dimensional timber volume models were established using the modeling samples， while the validation samples were used to compare and analyze the models using relevant indexes from the Technical Regulations for the Preparation of Binary Standing Timber Volume Tables. The results showed that when data was obtained by photogrammetry， the values of deviation， root mean square error， relative deviation， and relative root mean square error of single wood diameter at breast height and single wood height were low，which indicated that the research and development of timber volume models could be carried out by photogrammetry， and the results fully satisfied the investigation needs. Among the multiple timber volume models determined by the combination of multiple indicators， the model with the best results had an R² of 0.966 and a prediction accuracy of 97.69%. Therefore， the newly developed binary wood volume model in this study was more effective than the traditional Sichuan masson pine wood volume model in terms of total relative error， relative error average， and prediction accuracy. As a result， it could be used as the basis for establishing the wood volume table of Mawei pine stumpage in Chongqing.

Key words: photogrammetry, volume model, tree volume table

中图分类号:

S791.248

郁壮, 冯仲科, 张标, 马天天. 无损方法实现重庆地区马尾松材积模型研建[J]. 中国农业科技导报, 2023, 25(7): 97-104.

Zhuang YU, Zhongke FENG, Biao ZHANG, Tiantian MA. Research and Construction of Pinus massoniana Lamb. Volume Model in Chongqing Area by Non-destructive Method[J]. Journal of Agricultural Science and Technology, 2023, 25(7): 97-104.

图/表 7

图1 坐标系

Fig. 1 Coordinate system

表1 摄影测量精度分析

Table 1 Analysis of photogrammetry accuracy

单木因子

Individual tree factor

偏差

Bias

相对偏差

Relative bias/%

均方根误差

RMSE

相对均方根误差

Relative RMSE/%

表2 模型拟合结果

Table 2 Model fitting results

模型编号 Model number	材积模型 Volume model	参数a Parameter a	参数b Parameter b	参数c Parameter c
1	$V = a D b$	0.000 292 3	2.255 753 4	—
2	$V = a + b D c$	-0.014 308 5	0.000 371 2	2.194 731 6
3	$V = a + b D + c D 2$	0.031 762 8	-0.007 816 5	0.000 930 7
4	$V = a D b H c$	0.000 092 1	1.931 195 6	0.782 622 1
5	$V = a + b D 2 H$	0.024 016 8	0.000 035 9	—
6	$V = a D 2 + b D 2 H$	0.000 172 2	0.000 028 6	—

表2 模型拟合结果

Table 2 Model fitting results

模型编号 Model number	材积模型 Volume model	参数a Parameter a	参数b Parameter b	参数c Parameter c
1	$V = a D b$	0.000 292 3	2.255 753 4	—
2	$V = a + b D c$	-0.014 308 5	0.000 371 2	2.194 731 6
3	$V = a + b D + c D 2$	0.031 762 8	-0.007 816 5	0.000 930 7
4	$V = a D b H c$	0.000 092 1	1.931 195 6	0.782 622 1
5	$V = a + b D 2 H$	0.024 016 8	0.000 035 9	—
6	$V = a D 2 + b D 2 H$	0.000 172 2	0.000 028 6	—

表3 模型评价指标结果

Table 3 The model evaluates the index results

模型类型 Model type	材积模型 Volume model	相关指数 R²	总相对误差 RS/%	相对误差平均值 REA/%	预估精度 P/%
一元材积模型 One-variable volume model	$V = a D b$	0.925	-1.73	-3.44	95.05
	$V = a + b D c$	0.925	-0.54	11.29	95.10
	$V = a + b D + c D 2$	0.926	-0.63	1.38	95.14
二元材积模型 Two-variable volume model	$V = a D b H c$	0.966	0.39	1.86	97.69
	$V = a + b D 2 H$	0.963	0.05	-7.29	97.22
	$V = a D 2 + b D 2 H$	0.965	1.12	3.99	97.48

表3 模型评价指标结果

Table 3 The model evaluates the index results

模型类型 Model type	材积模型 Volume model	相关指数 R²	总相对误差 RS/%	相对误差平均值 REA/%	预估精度 P/%
一元材积模型 One-variable volume model	$V = a D b$	0.925	-1.73	-3.44	95.05
	$V = a + b D c$	0.925	-0.54	11.29	95.10
	$V = a + b D + c D 2$	0.926	-0.63	1.38	95.14
二元材积模型 Two-variable volume model	$V = a D b H c$	0.966	0.39	1.86	97.69
	$V = a + b D 2 H$	0.963	0.05	-7.29	97.22
	$V = a D 2 + b D 2 H$	0.965	1.12	3.99	97.48

图2 预测材积与摄影测量材积对比

Fig. 2 Comparison of predicted volume and photogrammetric volume

表4 模型评价指标结果

Table 4 Model evaluates the index results

模型类型

Model type

材积模型

Volume model

参考文献 34

1	吴伊. 鲁中地区栾树材积表与碳测算表的编制[D].泰安:山东农业大学, 2022.
	WU Y. Compilation of Luan tree volume table and carbon calculation table in central Shandong [D]. Tai'an: Shandong Agricultural University, 2022.
2	骆期邦,宁辉. 二元立木材积动态模型研究[J]. 林业科学研究, 1992, 5(3):263-270.
	LUO Q B, NING H. A study on standard volume dynamic model [J]. For. Res., 1992, 5(3): 263-270.
3	王希群,王安琪. 德国林业科学教育的先驱——柯塔[N]. 中国绿色时报,2015-04-10(4).
4	孟宪宇. 测树学[M].北京:中国林业出版社, 2006:1-350.
	MENG X Y. Forest Measuration [M]. Beijing:China Forestry Publishing House, 2006:1-350.
5	曾伟生,骆期邦.论林业数表模型的研建方法[J].中南林业调查规划,2001(2):1-4.
	ZENG W S, LUO Q B. On method of mathematical modeling for forestry tables [J]. Central South For. Inventory Plan., 2001(2):1-4
6	贺东北,曾伟生,骆期邦.通用性二元立木材积模型的建模样本研究[J].中南林业调查规划,2001(1):1-9.
	HE D B, ZENG W S, LUO Q B. Research on collection of sample for modeling of commonly used two-way tree volume model [J]. Central South For. Inventory Plan., 2001(1):1-9.
7	VIBRANS A C, MOSER P, OLIVEIRA L Z, et al.. Generic and specific stem volume models for three subtropical forest types in southern Brazil [J]. Ann. For. Sci., 2015,72: 865-874.
8	程文生,冯仲科,于景鑫.中国主要树种通用二元材积模型与推导形数模型研究[J].农业机械学报,2017,48(3):245-252.
	CHENG W S, FENG Z K, YU J X. Development of generic standard volume model and derived form factor model for major tree species in China [J]. Trans. Chin. Soc. Agric. Mach., 2017,48(3):245-252.
9	李曙我,贺传峰,姚闻,等.安福县杉木人工林二元立木材积动态模型的研究[J].江西林业科技,2014,42(1):22-29.
	LI S W, HE C F, YAO W, et al.. Study on the two-way tree volume model of Cunninghamia lanceolata plantation in Anfu county, Jiangxi province [J]. Jiangxi For. Sci. Technol., 2014,42(1):22-29.
10	何游云,张玉波,李俊清,等.利用无人机遥感测定岷江冷杉单木树干生物量[J].北京林业大学学报, 2016, 38(5): 42-49.
	HE Y Y, ZHANG Y B, LI J Q, et al.. Estimation of stem biomass of individual Abies faxoniana through unmanned aerial vehicle remote sensing [J]. J. Beijing For. Univ., 2016, 38(5):42-49.
11	冯仲科,罗旭,马钦彦,等. 基于三维激光扫描成像系统的树冠生物量研究[J]. 北京林业大学学报, 2007,29(): 52-56.
	FENG Z K, LUO X, MA Q Y, et al.. An estimation of tree canopy biomass based on 3D laser scanning imaging system [J]. J. Beijing For. Univ., 2007,29(S2): 52-56.
12	陈周娟,程光,卜元坤,等.基于无人机倾斜影像的阔叶林单木参数提取[J].林业资源管理, 2022(1):132-141.
	CHEN Z J, CHENG G, BU Y K, et al.. Single tree parameters extraction of broad-leaved forest based on UAV tilting photography [J]. For. Resour. Manage., 2022(1):132-141.
13	徐诗宇.基于地面激光雷达的中国南方城市绿化树种碳储量测算及研究[D]. 杭州:浙江农林大学,2018.
	XU S Y. Measurement and study of carbon storage for urban greening tree species in southern China based on terrestrial laser scanning [D]. Hangzhou: Zhejiang A&F University, 2018.
14	MAGNUSSEN S, NORD-LARSEN T, RIIS-NIELSEN T. Lidar supported estimators of wood volume and aboveground biomass from the Danish national forest inventory (2012-2016) [J]. Remote Sens. Environ., 2018, 211(15) : 146-153.
15	CHEN Y, ZHU X. An integrated GIS tool for automatic forest inventory estimates of Pinus radiata from LiDAR data [J]. GISci. Remote Sens., 2013, 50(6) : 667-689.
16	孙晨,程志刚,毛晓亮,等.近44a四川地区极端气候变化趋势及特征分析[J].兰州大学学报(自然科学版), 2017,53(1):119-126.
	SUN C, CHENG Z G, MAO X L, et al.. Extreme climatic change trend and features in Sichuan for the latest 44 years [J]. J. Lanzhou Univ. (Nat. Sci.), 2017,53(1):119-126.
17	刘英明,刘贤鹏.马尾松速丰林立地条件选择研究[J].四川林业科技, 1990(3):45-48.
	LIU Y M, LIU X P. Study on site condition selection of Pinus massoniana Sufeng forest [J]. J. Sichuan For. Sci. Technol., 1990(3):45-48.
18	曹受金,潘百红.气候变化对不同纬度马尾松径向生长的影响[J].中南林业科技大学学报, 2015,35(12):10-19.
	CAO S J, PAN B H. Effects of climate change on Pinus massoniana growth in different latitudes in Nanling regions of south China [J]. J. Central South Univ. For. Technol., 2015,35(12):10-19.
19	史京京,雷渊才,赵天忠.森林资源抽样调查技术方法研究进展[J].林业科学研究,2009,22(1):101-108.
	SHI J J, LEI Y C, ZHAO T Z. Progress in sampling technology and methodology in forest inventory [J]. For. Res., 2009,22(1):101-108.
20	骆期邦,贺东北,宁辉,等.立木材积表编制方法的研究[J].中南林业调查规划,1992(1):28-38.
	LUO Q B, HE D B, NING H, et al.. Research on the compiling method of the volume table of standing trees [J]. Central South For. Inventory Plan., 1992(1):28-38.
21	李海奎,宁金魁.基于树木起源、立地分级和龄组的单木生物量模型[J].生态学报,2012,32(3):740-757.
	LI H K, NING J K. Individual tree biomass model by tree origin,site classes and age groups [J]. Acta Ecol. Sin., 2012,32(3):740-757.
22	于景鑫,刘金成,冯仲科.森林盗伐的数字摄影测量方法及应用[J].南京林业大学学报(自然科学版),2015,39(5):87-92.
	YU J X, LIU J C, FENG Z K. Statistical methods and application of forest logging digital photogrammetry [J]. J. Nanjing For. Univ.(Nat. Sci.), 2015,39(5):87-92.
23	MOKROŠ M, MIKITA T, SINGH A, et al.. Novel low-cost mobile mapping systems for forest inventories as terrestrial laser scanning alternatives [J/OL]. Int. J. Appl. Earth Observ. Geoinf., 2021, 104: 102512 [2022-01-12]..
24	DONG Y Q, FAN G P, ZHOU Z W, et al.. Low cost automatic reconstruction of tree structure by AdQSM with terrestrial close-range photogrammetry [J/OL]. Forests, 2021, 12(8) : 1020[2022-01-12]. .
25	MINCH C, DVORAK J, JACKSON J, et al.. Creating a field-wide forage canopy model using UAVs and photogrammetry processing [J/OL]. Remote Sens., 2021, 13(13): 2487 [2022-01-12 ]. .
26	刘怡青,田育红,宋含章,等. 胸径和林分密度决定内蒙古东部落叶松林种内竞争[J].生态学杂志,2018,37(3):847-853.
	LIU Y Q, TIAN Y H, SONG H Z, et al.. DBH and stand density regulate intraspecific competition of larix gmelinii in eastern Inner Mongolia [J]. Chin. J. Ecol., 2018,37(3):847-853.
27	卢昌泰,黄山如,王利蓉,等.云南松地径一元材积表的编制[J].林业资源管理,1992(3):49-51.
	LU C T, HUANG S R, WANG L R, et al.. Compilation of the univariate volume table of ground diameter in Pinus yunnanensis [J]. For. Resour. Manage., 1992(3):49-51.
28	曾伟生.论一元立木材积模型的研建方法[J].林业资源管理,2004(1):21-23.
	ZENG W S. On the methodology for establishing one-way tree volume models [J]. For. Resour. Manage., 2004(1):21-23.
29	曾伟生,杨学云,陈新云.单木和林分水平一元与二元材积模型的预估精度对比[J].中南林业调查规划,2017,36(4):1-6.
	ZENG W S, YANG X Y, CHEN X Y. Comparison on prediction precision of one-variable and two-variable volume modelson tree-level and stand-level [J]. Central South For. Inventory Plan., 2017,36(4):1-6.
30	翁国庆,闫宏伟,白卫国,等. 二元立木材积表编制技术规程: [S].北京:中国标准出版社,2013.
31	翁国庆,白卫国,王维芳,等. 一元立木材积表编制技术规程: [S].北京:中国标准出版社,2015.
32	MOREIRA B M, GOYANES G, PINA P, et al.. Assessment of the influence of survey design and processing choices on the accuracy of tree diameter at breast height (DBH) measurements using UAV-based photogrammetry [J/OL]. Drones, 2021, 5(2) : 43 [2022-01-12]. .
33	HYYPPÄ E, YU X, KAARTINEN H, et al.. Comparison of backpack, handheld, under-canopy UAV, and above-canopy UAV laser scanning for field reference data collection in boreal forests [J/OL]. Remote Sens., 2020, 12(20): 3327 [2022-01-12]. .
34	KUCUKKAYA A G. Photogrammetry and remote sensing in archeology [J]. J. Quant. Spectrosc. Radiat. Transf., 2004, 88(1-3): 83-88.

[1]	练一宁1，冯仲科1*，刘培斌2，陈世林1,2. 智能手机单片摄影测树技术研究与试验[J]. 中国农业科技导报, 2020, 22(3): 56-63.
[2]	成竺欣1，王秀兰1*，沈亚骏1，常晨1，冯仲科1，蒋君志伟1，孙素芬2，张浪3,4，马文苑1. 手持式超站仪摄影测量研建材积模型[J]. 中国农业科技导报, 2020, 22(3): 72-78.