Bayesian parameter estimation of dominant height growth model for Changbai larch (Larix olgensis Henry) plantations

YAO Dan-dan; LEI Xiang-dong; ZHANG Ze-lu

doi:10.13332/j.1000-1522.20140221

Journal of Beijing Forestry University > 2015 > 37(3): 94-100. > DOI: 10.13332/j.1000-1522.20140221

YAO Dan-dan, LEI Xiang-dong, ZHANG Ze-lu. Bayesian parameter estimation of dominant height growth model for Changbai larch (Larix olgensis Henry) plantations[J]. Journal of Beijing Forestry University, 2015, 37(3): 94-100. DOI: 10.13332/j.1000-1522.20140221

Citation:

PDF (1509 KB)

Bayesian parameter estimation of dominant height growth model for Changbai larch (Larix olgensis Henry) plantations

1.
1 Institute of Forest Resource Information Techniques, Chinese Academy of Forestry, Beijing, 100091, P. R. China
2.
2 Helong Forestry Bureau, Helong, Jilin, 133500, P. R. China.

More Information

Received Date: June 30, 2014
Revised Date: September 11, 2014
Published Date: March 30, 2015

Graphical Abstract

Abstract

Abstract

Bayesian inference is an alternative method of statistical inference based on prior and data information. It has become an important statistical method for forest growth modeling. With 1 687 pairs of the dominant height and age data of Changbai larch (Larix olgensis Henry) plantations, we developed dominant height-age model based on Richards equation by using classical and Bayesian methods, and discussed model reliability with small sample size and Bayesian method. To test model performance for small sample size, four sample size options were employed including all data,10%, 5% and 2% of all sample randomly selected. We examined model performance and the distribution of parameters among methods for parameter estimation covering classical statistics (nonlinear least squares method), Bayesian statistics with uninformative prior and informative prior. Models were evaluated by root mean square error (RMSE), DIC and the confidence intervals of parameters. Results showed that the results of Bayesian statistics with small sample size was very close to those of classical statistics with large sample size, while the model reliability using Bayesian method was better than classical method, and RMSE with 5% sample was the smallest. Bayesian method with informative priors has the best performance for 5% sample. Compared with non-informative priors, the fitting results by Bayesian with informative priors were better, distribution was more concentrated and it had less uncertainty. The distribution of parameters estimated from Bayesian method with informative prior was largely overlapped with that from classical method. In addition, the results of Bayesian statistics with informative priors based on three different sample sizes showed that both the standard deviations of parameters and the RMSE of model were the smallest with 5% sample. It indicated that the fitting precision of Bayesian statistics and parameters uncertainty also had a certain relationship with sample size. The study confirmed the advantages of Bayesian method in model parameter estimation of forest growth for small sample size and using informative priors.
- dominant height,
- Richards growth equation,
- Bayesian method,
- classical statistics,
- Larix olgensis Henry

FullText(HTML)

References (42)

References

[1]	ZHANG X Q, ZHANG J G, DUAN A G. Tree-height growth model for Chinese fir plantation based on Bayesian method[J]. Scientia Silvae Sinicae, 2014, 50(3): 69-75.
[1]	ELLISON A M. Bayesian inference in ecology [J].Ecology Letters, 2004, 7(6): 509-520.
[2]	JIA N G. The estimation problem of local tree height-diameter at breast height [J]. Journal of Beijing Forestry College, 1984,6(3):1-9.
[2]	MCCARTHY M A. Bayesian methods for ecology [M]. Cambridge: Cambridge University Press, 2007.
[3]	CHENG X P, MIZUSAKI D, WANG S H, et al. Construction of tree growth model based on spatial autocorrelation [J]. Journal of Beijing Forestry University, 2012, 34 (5):113-119.
[3]	LI R, STEWART B, WEISKTTEL A. A Bayesian approach for modelling non-linear longitudinal/hierarchical data with random effects in forestry [J]. Forestry, 2012, 85(1): 17-25.
[4]	DAVIDIAN M, GILTINAN D M. Nonlinear models for repeated measurement data [M]. New York: Chapman and Hall, 1995.
[4]	JIA N G. Bayesian statistics and its application prospect in the forestry [J]. Journal of Beijing Forestry University, 1988, 10 (4): 91-96.
[5]	GREEN E J, STRAWDERMAN W E. A comparison of hierarchical Bayes and empirical Bayes methods [J]. Forest Science, 1992, 38(2):350-366.
[5]	LIANG W J, WANG X C, LIU F J. The preparation of main forestation tree species site index table in Jilin Province[J]. Jilin Forestry Science and Technology, 1999, 143(6): 1-5.
[6]	JIANG G J, ZHENG X X. Structural characteristics of natural spruce-fir forest in Wangqing Forestry Bureau of Jilin Province, northeastern China [J]. Journal of Beijing Forestry University, 2012, 34(4): 36-41.
[6]	GREEN E J, ROESH A F, SMITH A F M, et al. Bayesian estimation for the three-parameter Weibull distribution with tree diameter data [J]. International Biometric Society, 1994, 50(1):254-260.
[7]	ZHANG J G, DUAN A G. Approach to theoretical growth equations for modelling stands diameter structure of Chinese fir plantations [J]. Scientia Silvae Sinicae, 2003, 39(6): 55-61.
[7]	BULLOCK B P, BOONE E L. Deriving tree diameter distributions using Bayesian model averaging[J]. Forest Ecology and Management, 2007, 242(2-3): 127-132.
[8]	GREEN E J, STRAWDERMAN W E. Predictive posterior distributions from a Bayesian version of a slash pine yield model [J].Forest Science, 1996, 42(4): 456-464.
[8]	WEI X H, SUN Y J, MA W. A height growth model for Cunninghamia lanceolata based on Richards’ equation[J]. Journal of Zhejiang Agriculture and Forestry University, 2012, 29(5): 661-666.
[9]	LI F R, WU J M, LU S L. The comparison between Richards growth function and the Schnute model [J]. Journal of Northeast Forestry University, 1993, 21(4): 15-24.
[9]	WYCKOFF P H, CLARK J S. Predicting tree mortality from diameter growth: a comparison of maximum likelihood and Bayesian approaches [J].Canadian Journal of Forest Research, 2000, 30(1): 156-167.
[10]	METCALF C J E, MMAHO S M, CLARK J S. Overcoming data sparseness and parametric constraints in modeling of tree mortality: a new nonparametric Bayesian model [J].Canadian Journal of Forest Research, 2009, 39(9): 1677-1687.
[10]	XING L F, FA Y L, CHEN W Z, et al. Richards growth model’s utility of standing timber [J]. Journal of Shandong Forestry Science and Technology, 1997(5): 16-19.
[11]	NYSTROM K, STAHL G. Forecasting probability distributions of forest yield allowing for a Bayesian approach to management planning [J]. Silva Fennica, 2001, 35(2): 185-201.
[11]	MAO S S, TANG Y C. Bayesian statistics[M]. 2nd ed. Beijing: China Statistics Press, 2012.
[12]	XUE Y, CHEN L P. Statistical modeling and R software[M]. Beijing: Tsinghua University Press, 2007.
[12]	张雄清, 张建国, 段爱国. 基于贝叶斯法估计杉木人工林树高生长模型[J]. 林业科学, 2014, 50(3): 69-75.
[13]	ZHANG X Q, DUAN A G, ZHANG J G, et al. Estimating height-diameter models with Bayesian method[J]. The Scientific World Journal, 2014,1:1-27.
[14]	CLARK J S, WOLOSIN M, DIETZE M, et al. Tree growth inference and prediction from diameter censuses and ring widths[J]. Ecological Applications, 2007, 17(7): 1942-1953.
[15]	TATSUMI S, OWARI T. Bayesian modeling of neighborhood competition in uneven-aged mixed-species stands [J]. Formath, 2013, 12(1):191-209.
[16]	贾乃光. 局部树高-胸径的估计问题[J].北京林学院学报, 1984,6(3):1-9.
[17]	程希平, 水崎大二郎, 王四海,等. 基于空间自相关构建树木生长模型[J]. 北京林业大学学报, 2012,34 (5):113-119.
[18]	BOX G E P, TIAO G C. Bayesian inference in statistical analysis [M]. Massachusetts: Wiley Press, 1992.
[19]	SEBER G A F, WILD C J. Nonlinear regression [M]. Massachusetts: Wiley Press, 1989.
[20]	贾乃光. 贝叶斯统计学及其应用于林业的前景 [J].北京林业大学学报, 1988,10(4):91-96.
[21]	梁万军, 王宪成, 刘福金. 吉林省主要造林树种立地指数表的编制[J].吉林林业科技, 1999, 143(6): 1-5.
[22]	蒋桂娟, 郑小贤. 吉林省汪清林业局云冷杉天然林结构特征研究[J]. 北京林业大学学报, 2012, 34(4): 36-41.
[23]	张建国, 段爱国. 理论生长方程对杉木人工林林分胸径结构的模拟研究[J]. 林业科学, 2003, 39(6): 55-61.
[24]	魏晓慧, 孙玉君, 马炜. 基于Richards方程的杉木树高生长模型[J].浙江农林大学学报, 2012, 29(5): 661-666.
[25]	李凤日, 吴俊民, 鲁胜利. Richards函数与Schnute生长模型的比较 [J].东北林业大学学报, 1993, 21(4): 15-24.
[26]	邢黎峰, 法永乐, 陈文周,等. Richards林木生长模型及其实用性 [J].山东林业科技, 1997(5): 516-19.
[27]	茆诗松, 汤银才. 贝叶斯统计[M]. 2版.北京:中国统计出版社, 2012.
[28]	GELMAN A, CARLIN J B, STERN H S, et al. Bayesian data analysis[M]. 2nd ed. New York: Chapman and Hall, 2003.
[29]	SPIEGELHALTER D J, BESTN G, CARLIN B P, et al. Bayesian measures of model complexity and fit[J]. Journal of Royal Statistical Society: Series B, 2002,64(4):583-639.
[30]	薛毅, 陈丽萍. 统计建模与R软件[M]. 北京:清华大学出版社, 2007.

[1]	Wen Haoyu, Zhang Jie, Li Huiyu, Gao Caiqiu, Wang Chao, Zhang Qingzhu, Jiang Jing, Liu Guifeng. Analysis of fast-growing and stability characteristics of Betula platyphylla progeny at multiple locations based on BLUP-GGE biplot[J]. Journal of Beijing Forestry University, 2024, 46(10): 53-62. DOI: 10.12171/j.1000-1522.20240010
[2]	Fang Fan, Jia Liming, Jiang Xiaohui, Bai Qian, Su Shuchai. Selection of superior individual Pistacia chinensis based on 24 phenotypic traits[J]. Journal of Beijing Forestry University, 2024, 46(2): 40-50. DOI: 10.12171/j.1000-1522.20220447
[3]	LIU Yu, XU Huan-wen, ZHANG Guang-bo, WANG You-ju, TENG Wen-hua, JIANG Jing. Multipoint growth trait test of half-sibling offspring and excellent family selection of Betula platyphylla[J]. Journal of Beijing Forestry University, 2017, 39(3): 7-15. DOI: 10.13332/j.1000-1522.20160154
[4]	LIU Chao-yi, LIU Gui-feng, FANG Gong-gui, JIANG Chuan-ming, JIANG Jing. Comparison of tetraploid Betula platyphylla wood fiber traits and selection of superior seed trees[J]. Journal of Beijing Forestry University, 2017, 39(2): 9-15. DOI: 10.13332/j.1000-1522.20160091
[5]	LIU Yu, XU Huan-wen, TENG Wen-hua, JIANG Jing, LIU Gui-feng. Full-sib progeny test and early selection in superior families of Betula platyphylla[J]. Journal of Beijing Forestry University, 2017, 39(2): 1-8. DOI: 10.13332/j.1000-1522.20160138
[6]	ZHANG Zi-jie, YANG Shan-xun, ZENG Yan-jiang, WANG Rong-gang, WANG Li-ming, PANG Xiao-ming, LI Yue. Variation within clones and families and superior individual selection in different cultivars of Camellia oleifera‘Ruanzhi’.[J]. Journal of Beijing Forestry University, 2016, 38(10): 59-68. DOI: 10.13332/j.1000-1522.20160104
[7]	JIA Qing-bin, ZHANG Han-guo, ZHANG Lei, HOU Dan. Selection of superior hybrid larch families and growth rhythm analysis[J]. Journal of Beijing Forestry University, 2016, 38(2): 52-60. DOI: 10.13332/j.1000-1522.20150343
[8]	XU Huan-wen, LIU Yu, LI Zhi-xin, PENG Ru-sheng, SHANG Fu-qiang, ZOU Jian-jun, LIU Gui-feng, JIANG Jing. Analysis of the stability and superiority of five-year-old birch crossbreed families based on a multi-site test.[J]. Journal of Beijing Forestry University, 2015, 37(12): 24-31. DOI: 10.13332/j.1000-1522.20140466
[9]	LI Yan-xia, ZHANG Han-guo, DENG Ji-feng, ZHANG Lei. Correlations among wood density, wood physical mechanics index and growth trait, and selection of elite families for production of building products in Larix olgensis[J]. Journal of Beijing Forestry University, 2012, 34(5): 6-14.
[10]	LI Qing-jian, LI Yue, CHEN Zhi-qiang, SHI Rui-ting, ZHANG Jian-qiu, WANG Guo-zhu, ZHANG Yu-ling. Genetic variation and selection of fine families of Ulmus pumila L. in seedling stage under salinealkali stress[J]. Journal of Beijing Forestry University, 2012, 34(3): 53-57.

Cited By

Cited by

Periodical cited type(1)

党英侨，殷晶晶，陈传佳，孙丽丽，刘鹏，曹传旺. 转舞毒蛾LdCYP6AN15v1基因果蝇品系对氯虫苯甲酰胺胁迫响应. 林业科学. 2017(06): 94-104 .

Other cited types(2)

Get Citation

PDF

XML

Article views (1961) PDF downloads (46) Cited by(3)

Turn off MathJax

Article Contents

Abstract

References

Bayesian parameter estimation of dominant height growth model for Changbai larch (Larix olgensis Henry) plantations

Abstract

References

Related Articles

Cited by

Periodical cited type(1)

Other cited types(2)

Catalog

Bayesian parameter estimation of dominant height growth model for Changbai larch (Larix olgensis Henry) plantations

Abstract

References

Related Articles

Cited by

Periodical cited type(1)

Other cited types(2)

Catalog

Export File

Citation

Format

Content