基于贝叶斯法的针阔混交林树高与胸径混合效应模型

doi:10.11707/j.1001-7488.20191110

摘要/Abstract

摘要： [目的] 在多树种多层次针阔混交林中，基于贝叶斯混合效应模型法构建树高与胸径关系的混合效应模型，以提高预测模型参数稳定性，为揭示混交林多树种生长规律、资源分配差异及森林质量精准提升提供科学依据。[方法] 以河北省塞罕坝机械林场华北落叶松和白桦针阔混交林为研究对象，基于112块标准地（30 m×30 m）调查数据，选取6个包含不同林分因子的理论方程作为构建混交林不同树种树高与胸径关系的基础模型，选择出拟合精度较高的模型，分别采用两水平非线性混合效应模型法和贝叶斯混合效应模型法构建立包含哑变量的多树种树高与胸径关系模型。[结果] 包含林分优势高和林分断面积组合变量的Richards方程拟合效果最好，模型确定系数（R²）、均方根误差（RMSE）和绝对误差（Bias）分别为0.849 5、2.378 6和0.365 4；贝叶斯混合效应模型法拟合精度略高于传统非线性混合效应模型法：基于传统非线性混合效应模型法的华北落叶松树高与胸径关系模型的RMSE和Bias分别为0.930 4和0.103 4，白桦树高与胸径关系模型的RMSE和Bias分别为0.982 7和0.112 6；基于贝叶斯混合效应模型法的华北落叶松树高与胸径关系模型的RMSE和Bias分别为0.910 5和0.096 8，白桦树高与胸径关系模型的RMSE和Bias分别为0.963 3和0.100 2。[结论] 基于贝叶斯混合效应模型法构建的非线性混合效应模型，充分考虑混交林多树种树高与胸径关系模型参数的不确定性，模型预测效果更具可靠性和稳定性。

关键词: 贝叶斯统计, 非线性混合效应模型, 哑变量, 树高-胸径, 混交林

Abstract: [Objective] This paper established the nonlinear mixed effects model for height-diameter relationship based on Bayesian statistics in multi-storied and multi-species mixed forests. The purpose of this study was to provide some references for growth regularity of multiple tree species, differences in resource allocation and precision improvement of forest quality.[Method] A total of 112 temporary plots were established in Larix principis-rupprechtii and Betula platyphylla mixed forest of Saihanba national forest park, Hebei Province, China. Plot size was 30 m×30 m. We selected 6 typical models including different stand factors to fit height-diameter relationship. And the best-fit model was chose as the basis for building mixed-effects models by the method of Bayesian and nonlinear mixed models. We also added dummy variables to the mixed-effects models in order to solve intra-plot variability resulting from species difference. The goodness-of-fit criteria used were the coefficient of determination(R²), the absolute error of estimate(Bias)and the root mean square error(RMSE).[Result] Richards equation including dominant height and basal area of stand provided the most accurate prediction of height with the highest R²(0.849 5), the lowest Bias(2.378 6)and RMSE(0.365 4).The fitting accuracy of Bayesian non-linear mixed effect method was slightly higher than that of traditional non-linear mixed effects model method. Parameter estimation method of traditional non-linear mixed effect model had the best fits with the fit statistics values(RMSE=0.930 4; Bias=0.103 4)for L. principis-rupprechtii and values(RMSE=0.982 7; Bias=0.112 6)for B. platyphylla. Parameter estimation method of Bayesian nonlinear mixed effect model had the best fits with the fit statistics values(RMSE=0.910 5; Bias=0.096 8)for L. principis-rupprechtii and values(RMSE=0.963 3; Bias=0.100 2)for B. platyphylla.[Conclusion] The non-linear mixed effect model based on Bayesian theory considered the uncertainties of parameters in the model of tree height-diameter relationship of multi-species. The prediction results have better reliability and stability.

Key words: Bayesian statistics, nonlinear mixed-effects model, dummy variable, height-diameter, mixed forest

中图分类号:

S758.5⁺7

王冬至, 张冬燕, 李永宁, 张志东, 李大勇, 黄选瑞. 基于贝叶斯法的针阔混交林树高与胸径混合效应模型[J]. 林业科学, 2019, 55(11): 85-94.

Wang Dongzhi, Zhang Dongyan, Li Yongning, Zhang Zhidong, Li Dayong, Huang Xuanrui. Height-Diameter Relationship for Conifer Mixed Forest Based on Bayesian Nonlinear Mixed-Effects Model[J]. Scientia Silvae Sinicae, 2019, 55(11): 85-94.

参考文献

崔令军, 张雄清, 段爱国, 等. 2016. 基于分层贝叶斯法的杉木人工林最大密度线. 林业科学, 52(9):95-102.
(Cui L J, Zhang X Q, Duan A G, et al. 2016. A hierarchical Bayesian model to predict maximum-size density line for Chinese fir plantation in southern China. Scientia Silvae Sinicae, 52(9):95-102.[in Chinese])
刘春云, 方文静, 蔡琼, 等. 2017. 中国落叶松林胸径-树高相关关系的探讨. 北京大学学报:自然科学版, 53(6):1081-1088.
(Liu C Y, Fang W J, Cai Q, et al. 2017. Allometric relationship between tree height and diameter of larch forests in China. Acta Scientiarum Naturalium Universitatis Pekinensis, 53(6):1081-1088.[in Chinese])
娄明华, 张会儒, 雷相东, 等. 2016. 天然云冷杉针阔混交林单木胸径树高空间自回归模型研究. 北京林业大学学报, 38(8):1-9.
(Lou M H, Zhang H R, Lei X D, et al. 2016. An individual height-diameter model constructed using spatial autoregressive models within natural spruce-fir and broadleaf mixed stands. Journal of Beijing Forestry University, 38(8):1-9.[in Chinese])
卢军, 张会儒, 雷相东, 等. 2015. 长白山云冷杉针阔混交林幼树树高-胸径模型. 北京林业大学学报, 37(11):10-25.
(Lu J, Zhang H R, Lei X D, et al. 2015. Height-diameter models for saplings in a spruce-fir mixed forest in Changbai Mountains. Journal of Beijing Forestry University, 37(11):10-25.[in Chinese])
王冬至, 张冬燕, 张志东, 等. 2016. 基于非线性混合模型的针阔混交林树高与胸径关系. 林业科学, 52(1):30-36.
(Wang D Z, Zhang D Y, Zhang Z D, et al. 2016. Height-diameter relationship for conifer mixed forest based on nonlinear mixed-effects model. Scientia Silvae Sinicae, 52(1):30-36.[in Chinese])
肖锐, 陈东升, 李凤日, 等. 2015. 基于两水平混合模型的杂种落叶松胸径和树高生长模拟. 东北林业大学学报, 43(5):33-37.
(Xiao R, Chen D S, Li F R, et al. 2015. Simulating DBH and height growth of trees for hybrid larch plantation with two-level mixed effect model. Journal of Northeast Forestry University, 43(5):33-37.[in Chinese])
姚丹丹, 雷相东, 张则路. 2015. 基于贝叶斯法的长白落叶松林分优势高生长模型研究. 北京林业大学学报, 37(3):94-100.
(Yao D D, Lei X D, Zhang Z L. 2015. Bayesian parameter estimation of dominant height growth model for Changbai larch(Larix olgensis Henry)plantations. Journal of Beijing Forestry University, 37(3):94-100.[in Chinese])
臧颢, 雷相东, 张会儒, 等. 2016. 红松树高-胸径的非线性混合效应模型研究. 北京林业大学学报, 38(6):8-16.
(Zang H, Lei X D, Zhang H R, et al. 2016. Nonlinear mixed-effects height-diameter model of Pinus koraiensis. Journal of Beijing Forestry University, 38(6):8-16.[in Chinese])
张雄清, 张建国, 段爱国. 2014. 基于贝叶斯法估计杉木人工林树高生长模型. 林业科学, 50(3):69-75.
(Zhang X Q, Zhang J G, Duan A G. 2014. Tree-height growth model for Chinese fir plantation based on Bayesian method. Scientia Silvae Sinicae, 50(3):69-75.[in Chinese])
张雄清, 张建国, 段爱国. 2015. 杉木人工林林分断面积生长模型的贝叶斯法估计. 林业科学研究, 28(4):538-542.
(Zhang X Q, Zhang J G, Duan A G. 2015. Application of Bayesian method in stand basal area prediction of Chinese fir plantation. Forest Research, 28(4):538-542.[in Chinese])
Aertsen W, Kint V, Van Orshoven J, et al. 2010. Comparison and ranking of different modelling techniques for prediction of site index in Mediterranean mountain forests. Ecological Modelling, 221(8):1119-1130.
Ahmadi K, Alavi S J. 2016. Generalized height-diameter models for Fagus orientalis Lipsky in Hyrcanian forest, Iran. Journal of Forest Science, 62(9):413-421.
Ahmadi K, JalilAlavi S, Kouchaksaraei M T, et al. 2013. Non-linear height-diameter models for oriental beech(Fagus orientalis Lipsky)in the Hyrcanian forests, Iran. Biotechnology, Agronomy, Society and Environment, 17(3):431-440.
Botkin D B, Janak J F, Wallis J R. 1972. Some ecological consequences of a computer model of forest growth. The Journal of Ecology,60(3):849-872.
Bullock B P, Boone E L. 2007. Deriving tree diameter distributions using Bayesian model averaging. Forest Ecology and Management, 242:127-132.
Calama R, Montero G. 2004. Interregional nonlinear height diameter model with random coefficients for stone pine in Spain. Canadian Journal of Forest Research, 34(1):150-163.
Colbert K C, Larsen D R, Lootens J R. 2002. Height-diameter equations for thirteen midwestern bottomland hardwood species. Northern Journal of Applied Forestry, 19(4):171-176.
Cripps E, Pecht M. 2017. A Bayesian nonlinear random effects model for identification of defective batteries from lot samples. Journal of Power Sources, 342:342-350.
Dorado F C, Diéguez-Aranda U, Anta M B, et al. 2006. A generalized height-diameter model including random components for radiata pine plantations in northwestern Spain. Forest Ecology and Management, 229(3):202-213.
Duan G S, Gao Z, Wang Q, et al. 2018. Comparison of different height-diameter modelling techniques for prediction of site productivity in natural uneven-aged pure stands. Forests, 9(2):63.
Ducey M J. 2009. The ratio of additive and traditional stand density indices. Western Journal of Applied Forestry, 24(1):5-10.
Eby W M, Oyamakin S O, Chukwu A U. 2017. A new nonlinear model applied to the height-DBH relationship in Gmelina arborea. Forest Ecology and Management, 397:139-149.
Eerikäinen K. 2009. A multivariate linear mixed-effects model for the generalization of sample tree heights and crown ratios in the finish national forest inventory. Forest Science, 55(6):480-493.
Ellison A M. 2004. Bayesian inference in ecology. Ecology Letters, 7(6):509-520.
Gómez-García E, Diéguez-Aranda U, Castedo-Dorado F, et al. 2013. A comparison of model forms for the development of height-diameter relationships in even-aged stands. Forest Science, 60(3):560-568.
Green E J, Strawderman W E. 1996. Predictive posterior distributions from a Bayesian version of a slash pine yield model. Forest Science, 42:456-464.
Grégoire T G, Schabenberger O, Barrett J P. 1995. Linear modelling of irregularly spaced, unbalanced, longitudinal data from permanent-plot measurements. Canadian Journal of Forest Research, 25(1):137-156.
Huang S, Wiens D P, Yang Y, et al. 2009. Assessing the impacts of species composition, top height and density on individual tree height prediction of quaking aspen in boreal mixed woods. Forest Ecology and Management, 258(7):1235-1247.
Jayaraman K, Zakrzewski W T. 2001. Practical approaches to calibrating height-diameter relationships for natural sugar maple stands in Ontario. Forest Ecology and Management, 148(3):169-177.
Larjavaara M, Muller-Landau H C. 2013. Measuring tree height:a quantitative comparison of two common field methods in a moist tropical forest. Methods in Ecology and Evolution, 4(9):793-801.
Li R, Stewart B, Weiskittel A. 2011. A Bayesian approach for modelling non-linear longitudinal/hierarchical data with random effects in forestry. Forestry, 85(1):17-25.
Mensah S, Pienaar O L, Kunneke A, et al. 2018. Height-diameter allometry in South Africa's indigenous high forests:assessing generic models performance and function forms. Forest Ecology and Management, 410:1-11.
Metcalf C J E, McMahon S M, Clark J S. 2009. Overcoming data sparseness and parametric constraints in modeling of tree mortality:a new nonparametric Bayesian model. Canadian Journal of Forest Research, 39:1677-1687.
Newton P F, Amponsah I G. 2007. Comparative evaluation of five height-diameter models developed for black spruce and jack pine stand-types in terms of goodness-of-fit, lack-of-fit and predictive ability. Forest Ecology and Management, 247(3):149-166.
Nigh G. 2002. Site index conversion equations for mixed trembling aspen and white spruce stands in northern British Columbia. Silva Fennica, 36(4):789-797.
Özçelik R, Cao Q V, Trincado G, et al. 2018. Predicting tree height from tree diameter and dominant height using mixed-effects and quantile regression models for two species in Turkey. Forest Ecology and Management, 419:240-248.
Peng C, Zhang L, Zhou X, et al. 2004. Developing and evaluating tree height-diameter models at three geographic scales for black spruce in Ontario. Northern Journal of Applied Forestry, 21(2):83-92.
Richards F J. 1959. A flexible growth function for empirical use. Journal of Experimental Botany,10(2):290-301.
Saud P, Lynch T B, KC A, et al. 2016. Using quadratic mean diameter and relative spacing index to enhance height-diameter and crown ratio models fitted to longitudinal data. Forestry, 89(2):215-229.
Temesgen H, Gadow K. 2004. Generalized height-diameter modelsan application for major tree species in complex stands of interior British Columbia. European Journal of Forest Research, 123(1):45-51.
Temesgen H, Hann D W, Monleon V J. 2007. Regional height-diameter equations for major tree species of southwest Oregon. Western Journal of Applied Forestry, 22(3):213-219.
Temesgen H, Zhang C H, Zhao X H. 2014. Modelling tree height-diameter relationships in multi-species and multi-layered forests:a large observational study from Northeast China. Forest Ecology and Management, 316:78-89.
Yang R C, Kozak A, Smith J H G. 1978. The potential of Weibull-type functions as flexible growth curves.Canadian Journal of Forest Research,8(4):424-431.
Zhang X Q, Duan A G, Zhang J G, et al. 2014. Estimating tree height-diameter models with the Bayesian method. The Scientific World Journal,(1):1-9.

[1]	万盼,刘文桢,刘瑞红,王鹏,王宏翔,惠刚盈. 结构化经营对栎松混交林林分空间结构及稳定性的影响[J]. 林业科学, 2020, 56(4): 35-45.
[2]	何怀江, 张忠辉, 张春雨, 郝珉辉, 姚杰, 解蛰, 高海涛, 赵秀海. 采伐强度对东北针阔混交林林分生长和物种多样性的短期影响[J]. 林业科学, 2019, 55(2): 1-12.
[3]	曾伟生, 贺东北, 蒲莹, 肖前辉. 含地域和起源因子的马尾松立木生物量与材积方程系统[J]. 林业科学, 2019, 55(2): 75-86.
[4]	薛春泉, 徐期瑚, 林丽平, 何潇, 罗勇, 赵菡, 曹磊, 雷渊才. 广东主要乡土阔叶树种含年龄和胸径的单木生物量模型[J]. 林业科学, 2019, 55(2): 97-108.
[5]	陈金磊, 方晰, 辜翔, 李雷达, 刘兆丹, 王留芳, 张仕吉. 中亚热带2种森林群落组成、结构及区系特征[J]. 林业科学, 2019, 55(2): 159-172.
[6]	曾伟生,陈新云,杨学云. 我国人工杨树生物量建模和生产力分析[J]. 林业科学, 2019, 55(11): 1-8.
[7]	王金池,黄清麟,马志波,黄如楚,郑群瑞. 永安市半天然马尾松阔叶混交林的树种组成与多样性[J]. 林业科学, 2019, 55(11): 19-26.
[8]	孟令君,张春雨,姚杰,赵秀海. 吉林蛟河针阔混交林乔木幼苗存活对密度和生境的响应[J]. 林业科学, 2019, 55(11): 172-180.
[9]	刘丹, 李玉堂, 洪玲霞, 国红, 谢阳生, 张卓立, 雷相东, 唐守正. 基于最大熵模型的吉林省主要天然林潜在分布适宜性[J]. 林业科学, 2018, 54(7): 1-15.
[10]	雷相东, 符利勇, 李海奎, 李玉堂, 唐守正. 基于林分潜在生长量的立地质量评价方法与应用[J]. 林业科学, 2018, 54(12): 116-126.
[11]	姚兰, 艾训儒, 易咏梅, 黄永涛, 冯广, 刘峻城. 鄂西南亚热带常绿落叶阔叶混交林优势种群的结构与动态特征[J]. 林业科学, 2017, 53(2): 10-18.
[12]	冯广, 艾训儒, 姚兰, 刘峻城, 黄永涛, 林勇. 鄂西南亚热带常绿落叶阔叶混交林的自然恢复动态及其影响因素[J]. 林业科学, 2016, 52(8): 1-9.
[13]	李晖, 曾伟生. 不同区域落叶松二元立木材积表的检验及差异分析[J]. 林业科学, 2016, 52(6): 157-162.
[14]	惠刚盈, 张弓乔, 赵中华, 胡艳波, 刘文桢, 张宋智, 白超. 天然混交林最优林分状态的π值法则[J]. 林业科学, 2016, 52(5): 1-8.
[15]	林文树, 穆丹, 王丽平, 邵立郡, 吴金卓. 针阔混交林不同演替阶段表层土壤理化性质与优势林木生长的相关性[J]. 林业科学, 2016, 52(5): 17-25.