Prediction of bark thickness for <i>Larix gmelinii</i> based on machine learning

Zhang Mengku; Jiang Lichun

doi:10.12171/j.1000-1522.20210097

Journal of Beijing Forestry University > 2022 > 44(6): 54-62. > DOI: 10.12171/j.1000-1522.20210097

Zhang Mengku, Jiang Lichun. Prediction of bark thickness for Larix gmelinii based on machine learning[J]. Journal of Beijing Forestry University, 2022, 44(6): 54-62. DOI: 10.12171/j.1000-1522.20210097

Citation:

PDF (816 KB)

Prediction of bark thickness for Larix gmelinii based on machine learning

Zhang Mengku,
Jiang Lichun^,

School of Forestry, Key Laboratory of Sustainable Forest Ecosystem Management of Ministry of Education, Northeast Forestry University, Harbin 150040, Heilongjiang, China

More Information

Received Date: March 14, 2021
Revised Date: May 24, 2021
Available Online: May 22, 2022
Published Date: June 24, 2022

Graphical Abstract

Abstract

Abstract

Objective This paper aims to study the application of multiple machine learning algorithms in the prediction of bark thickness, to compare and analyze the influence of different individual tree factors on the prediction of bark thickness, and to provide new methods for the prediction of bark thickness.
Method Four machine learning algorithms (neural network, support vector regression, decision tree, random forest) were constructed based on the bark thickness data of Dahurian larch (Larix gmelinii) in Daxing’anling Mountains of northeastern China. Their performance in predicting bark thickness was compared with six traditional bark thickness models. The determination coefficient (R²), root mean square error (RMSE), mean absolute error (MAE) and Akaike information criterion (AIC) were used to evaluate different models and algorithms.
Result (1) Among the six basic models, Model 5 showed better prediction results. In the comparison between the basic model and the machine learning models, all the machine learning models fitted better accuracy than the traditional model Model 5 except for the CART4 model. (2) Among the machine learning models, the fitting and prediction accuracy of ANN4 and SVR3 were similar, and RF4 was the best. (3) The input variables of RF4 were diameter at breast height (DBH), tree height (H), and relative tree height (H_r). Based on the training samples, the R² of random forest increased from 0.675 2 to 0.723 4, RMSE decreased from 0.575 5 to 0.531 0 compared with Model5. Based on the testing samples, the R² of random forest increased from 0.666 9 to 0.710 5, RMSE decreased from 0.616 9 to 0.544 6 compared with Model 5.
Conclusion Compared with the basic bark thickness model, random forest, support vector machine regression and artificial neural network in machine learning algorithm can improve the prediction accuracy of bark thickness. The prediction effect of random forest is slightly better and suitable for the prediction of bark thickness of Dahurian larch in this region.
- bark thickness,
- support vector regression,
- neural network,
- random forest,
- decision tree

FullText(HTML)

References (29)

References

[1]	Martin R E, Crist J B. Elements of bark structure and terminology[J]. Wood and Fiber Science, 1970, 2(3): 269−279.
[2]	成俊卿, 杨家驹, 刘鹏. 中国木材志[M]. 北京: 中国林业出版社, 1992. Cheng J Q, Yang J J, Liu P. Woods of China[M]. Beijing: China Forestry Publishing House, 1992.
[3]	Franceschi V R, Krokene P, Christiansen E, et al. Anatomical and chemical defenses of conifer bark against bark beetles and other pests[J]. New Phytologist, 2005, 167(2): 353−376. doi: 10.1111/j.1469-8137.2005.01436.x
[4]	Wilson B G, Witkowski E T F. Seed banks, bark thickness and change in age and size structure (1978–1999) of the African savanna tree, Burkea africana[J]. Plant Ecology, 2003, 167(1): 151−162. doi: 10.1023/A:1023999806577
[5]	Wesolowski A, Adams M A, Pfautsch S. Insulation capacity of three bark types of temperate Eucalyptus species[J]. Forest Ecology and Management, 2014, 313: 224−232. doi: 10.1016/j.foreco.2013.11.015
[6]	Marshall H D, Murphy G E, Lachenbruch B. Effects of bark thickness estimates on optimal log merchandising[J]. Forest Products Journal, 2006, 56(11): 87−92.
[7]	Heath L S, Hansen M, Smith J E, et al. Investigation into calculating tree biomass and carbon in the FIADB using a biomass expansion factor approach[J]. Forest Service, 2009, 37(1): 167−173.
[8]	Laasasenaho J, Melkas T, Aldén S. Modelling bark thickness of Picea abies with taper curves[J]. Forest Ecology and Management, 2005, 206(1−3): 35−47. doi: 10.1016/j.foreco.2004.10.058
[9]	Malone T, Liang J. A bark thickness model for white spruce in Alaska northern forests[J]. International Journal of Forestry Research, 2014, 2009: 1−5.
[10]	张兴龙, 姜立春. 兴安落叶松树干去皮直径预测模型[J]. 林业科学研究, 2015, 28(1): 67−73. Zhang X L, Jiang L C. Inside bark diameter prediction models for Dahurian larch[J]. Forestry Research, 2015, 28(1): 67−73.
[11]	Stängle S M, Weiskittel A R, Dormann C F, et al. Measurement and prediction of bark thickness in Picea abies: assessment of accuracy, precision, and sample size requirements[J]. Canadian Journal of Forest Research, 2016, 46(1): 39−47. doi: 10.1139/cjfr-2015-0263
[12]	唐诚, 王春胜, 庞圣江, 等. 西南桦人工林树皮厚度模型模拟[J]. 林业科学, 2017, 53(7): 85−93. doi: 10.11707/j.1001-7488.20170709 Tang C, Wang C S, Pang S J, et al. Simulating bark thickness for Betula alnoides plantations[J]. Scientia Silvae Sinicae, 2017, 53(7): 85−93. doi: 10.11707/j.1001-7488.20170709
[13]	欧强新, 雷相东, 沈琛琛, 等. 基于随机森林算法的落叶松−云冷杉混交林单木胸径生长预测[J]. 北京林业大学学报, 2019, 41(9): 9−19. Ou Q X, Lei X D, Shen C C, et al. Individual tree DBH growth prediction of larch-spruce-fir mixed forests based on random forest algorithm[J]. Journal of Beijing Forestry University, 2019, 41(9): 9−19.
[14]	雷相东. 机器学习算法在森林生长收获预估中的应用[J]. 北京林业大学学报, 2019, 41(12): 23−36. doi: 10.12171/j.1000-1522.20190356 Lei X D. Applications of machine learning algorithms in forest growth and yield prediction[J]. Journal of Beijing Forestry University, 2019, 41(12): 23−36. doi: 10.12171/j.1000-1522.20190356
[15]	车少辉, 张建国, 段爱国, 等. 杉木人工林胸径生长神经网络建模研究[J]. 西北农林科技大学学报(自然科学版), 2012, 4(3): 84−92. Che S H, Zhang J G, Duan A G, et al. Modelling tree diameter growth for Chinese fir plantations with neural networks[J]. Journal of Northwest A&F University (Natatural Science Edition), 2012, 4(3): 84−92.
[16]	沈剑波, 雷相东, 李玉堂, 等. 基于bp神经网络的长白落叶松人工林林分平均高预测[J]. 南京林业大学学报(自然科学版), 2018, 42(2): 147−154. Shen J B, Lei X D, Li Y T, et al. Prediction mean height for Larix olgensis plantation based on Bayesian-regularization BP neural network[J]. Journal of Nanjing Forestry University (Natural Sciences Edition), 2018, 42(2): 147−154.
[17]	Diamantopoulou M J, Özcelik R, Yavuz H. Tree-bark volume prediction via machine learning: a case study based on black alder’s tree-bark production[J]. Computers and Electronics in Agriculture, 2018, 151: 431−440. doi: 10.1016/j.compag.2018.06.039
[18]	Che S, Tan X, Xiang C, et al. Stand basal area modelling for Chinese fir plantations using an artificial neural network model[J]. Journal of Forestry Research, 2019, 30(5): 1641−1649. doi: 10.1007/s11676-018-0711-9
[19]	Zhao Q, Yu S, Zhao F, et al. Comparison of machine learning algorithms for forest parameter estimations and application for forest quality assessments[J]. Forest Ecology and Management, 2019, 434: 224−234. doi: 10.1016/j.foreco.2018.12.019
[20]	Tian Y, Wu B, Su X, et al. A crown contour envelope model of Chinese fir based on random forest and mathematical modeling[J]. Forests, 2021, 12(1): 48.
[21]	Li R, Weiskittel A R. Estimating and predicting bark thickness for seven conifer species in the Acadian Region of North America using a mixed-effects modeling approach: comparison of model forms and subsampling strategies[J]. European Journal of Forest Research, 2011, 130(2): 219−233. doi: 10.1007/s10342-010-0423-y
[22]	Jang W, Keyes C R, Graham J M. Evaluation of predictive models for Douglas-fir bark thickness at breast height following 12 biomass harvest treatments[J]. Biomass and Bioenergy, 2016, 84: 118−123. doi: 10.1016/j.biombioe.2015.11.015
[23]	Williams V L, Witkowski E T F, Balkwill K. Relationship between bark thickness and diameter at breast height for six tree species used medicinally in South Africa[J]. South African Journal of Botany, 2007, 73(3): 449−465. doi: 10.1016/j.sajb.2007.04.001
[24]	Yuan H, Yang G, Li C, et al. Retrieving soybean leaf area index from unmanned aerial vehicle hyperspectral remote sensing: Analysis of RF, ANN, and SVM regression models[J]. Remote Sensing, 2017, 9(4): 309. doi: 10.3390/rs9040309
[25]	Ding Y, Zhang H, Wang Z, et al. A comparison of estimating crop residue cover from sentinel-2 data using empirical regressions and machine learning methods[J]. Remote Sensing, 2020, 12(9): 1470. doi: 10.3390/rs12091470
[26]	Breiman L. Random forest[J]. Machine Learning, 2001, 45: 5−32. doi: 10.1023/A:1010933404324
[27]	王晓林, 蔡可旺, 姜立春. 落叶松树皮厚度变化规律的研究[J]. 森林工程, 2011, 27(2): 8−11. doi: 10.3969/j.issn.1001-005X.2011.02.003 Wang X L, Cai K W, Jiang L C. Study on bark thickness of dahurian larch[J]. Forest Engineering, 2011, 27(2): 8−11. doi: 10.3969/j.issn.1001-005X.2011.02.003
[28]	Diamantopoulou M J, Milios E. Modelling total volume of dominant pine trees in reforestations via multivariate analysis and artificial neural network models[J]. Biosystems Engineering, 2010, 105(3): 306−315. doi: 10.1016/j.biosystemseng.2009.11.010
[29]	Binoti D H B, Binoti M L M S, Leite H G, et al. Support vector machine to estimate volume of eucalypt trees[J]. Revista Árvore, 2016, 40(4): 689−693.

Cited By

Cited by

Periodical cited type(1)

徐媛，陈锦玲，陈玉梅，李璐璐，李惠敏，秦新民. 干旱胁迫下花生转录组与miRNA测序及相关基因的表达. 贵州农业科学. 2021(01): 1-9 .

Other cited types(3)

Get Citation

PDF

XML

Article views (752) PDF downloads (93) Cited by(4)

Turn off MathJax

Article Contents

Abstract

References

Prediction of bark thickness for Larix gmelinii based on machine learning

Abstract

References

Cited by

Periodical cited type(1)

Other cited types(3)

Catalog

Export File

Citation

Format

Content