基于随机森林的大兴安岭中部天然林更新数量影响因素识别

doi:10.11707/j.1001-7488.LYKX20250344

摘要/Abstract

摘要：

目的: 构建随机森林模型，识别大兴安岭中部地区天然林更新数量的关键影响因素，为该地区森林的可持续经营提供理论依据。方法: 基于新林林业局翠岗林场、新林林场和壮志林场共96块标准样地调查数据，从林分特征、立地条件、土壤条件、林木大小多样性、物种多样性和林分空间结构6方面选取29个基础指标，采用Poisson模型、负二项模型和随机森林算法分别构建兴安落叶松和白桦更新数量模型；经模型选优后，应用OOB置换法确定候选变量对兴安落叶松和白桦更新数量的贡献。结果: 十折交叉检验表明，随机森林更新数量预测模型精度显著高于Poisson模型和负二项模型，其中兴安落叶松和白桦随机森林更新数量模型的均方根误差（RMSE）分别为482和682 tree·hm^?2，平均绝对误差（MAE）分别为377和460 tree·hm^?2。经OOB置换法得出各变量的相对重要性，其中对兴安落叶松更新数量重要的变量依次为胸径Shannon指数（17.57%）、胸径Pielou均匀度指数（16.88%）、单位蓄积量（13.29%）、胸径Simpson指数（12.92%）和林分平均胸径（12.91%）；对白桦更新数量重要的变量依次为胸径Shannon指数（18.53%）、胸径Pielou均匀度指数（16.13%）、草本盖度（12.62%）、林分平均胸径（12.34%）和灌木盖度（11.31%）。结论: 林木大小多样性和林分密度是大兴安岭中部地区天然林更新数量的重要影响因子，可通过抚育间伐或补植方式确保科学合理的林分密度，进而促进天然更新和森林生态系统的自然演替。

关键词: 特征变量筛选, 随机森林, 更新, 数量预测模型

Abstract:

Objective: This study aims to explore and identify key factors influencing the quantity of natural forest regeneration by developing a random forest model, so as to provide a theoretical basis for the sustainable management of natural forests in the central of Daxing’anling region. Method: Based on the survey data from 96 standard sample plots in Cuigang forest farm, Xinlin forest farm, and Zhuangzhi forest farm under the Xinlin Forestry Bureau, 29 basic indicators were selected across six categories: stand characteristics, site conditions, soil conditions, tree size diversity, species diversity, and stand spatial structure. Poisson model, negative binomial model, and random forest regression model were used to construct regeneration quantity models for Larix gmelinii and Betula platyphylla. After model optimization, the OOB permutation method was used to evaluate the contribution of each predictor variable to the regeneration quantities of L. gmelinii and B. platyphylla. Result: The ten-fold cross-validation test showed that the accuracy of the random forest regeneration prediction model was significantly higher than that of the Poisson model and the negative binomial model. The root mean square error ( RMSE) was 482 and 682 tree·hm^?2, and mean absolute error (MAE） was 377 and 460 tree·hm^?2 for L. gmelinii and B. platyphylla, respectively. The OOB permutation method identified the most important variables for L. gmelinii regeneration as: Shannon index of diameter at breast height (DBH) (17.57%), Pielou’s evenness index of DBH (16.88%), stand volume per unit area (13.29%), Simpson index of DBH (12.92%), and mean DBH (12.91%). For B. platyphylla, the key variables were: Shannon index of DBH (18.53%), Pielou’s evenness index of DBH (16.13%), herbaceous cover (12.62%), mean DBH (12.34%), and shrub cover (11.31%). Conclusion: Tree size diversity and stand density are key factors influencing regeneration. Nurturing harvesting or replanting strategies can ensure scientific and reasonable stand density, thereby promoting natural regeneration and supporting the ecological succession of forest ecosystems.

Key words: screening of characteristic variables, random forests, regeneration, quantity prediction model

中图分类号:

S754.1

田栋元,刘兆刚,姜立春,董灵波. 基于随机森林的大兴安岭中部天然林更新数量影响因素识别[J]. 林业科学, 2026, 62(5): 69-79.

Dongyuan Tian,Zhaogang Liu,Lichun Jiang,Lingbo Dong. Identification of the Influencing Factors on the Regeneration Quantity of Natural Forests in the Central Part of Daxing’anling Mountains Based on Random Forest[J]. Scientia Silvae Sinicae, 2026, 62(5): 69-79.

图/表 7

表1

图1

表2

图2

表3

图3

图4

参考文献 0

	陈　莹, 董灵波, 刘兆刚. 帽儿山天然次生林主要林分类型最优树种组成. 北京林业大学学报, 2019, 41 (5): 118- 126.
	Chen Y, Dong L B, Liu Z G. Optimal species composition for the main forest types of secondary forest in Mao’ershan Mountain, northeastern China. Journal of Beijing Forestry University, 2019, 41 (5): 118- 126.
	党宏忠, 陈　帅, 钟　鹏, 等. 樟子松人工林自然更新过程中断的机制及可能调控途径. 林业科学, 2024, 60 (12): 158- 167.
	Dang H Z, Chen S, Zhong P, et al. Mechanism and possible regulatory approaches of interruption in the natural regeneration process of Pinus sylvestris var. mongolica plantations in China. Scientia Silvae Sinicae, 2024, 60 (12): 158- 167.
	丁家祺, 黄文丽, 刘迎春, 等. 基于机器学习和多源数据的湘西北森林地上生物量估测. 林业科学, 2021, 57 (10): 36- 48. doi: 10.11707/j.1001-7488.20211004
	Ding J Q, Huang W L, Liu Y C, et al. Estimation of forest aboveground biomass in northwest Hunan province based on machine learning and multi-source data. Scientia Silvae Sinicae, 2021, 57 (10): 36- 48. doi: 10.11707/j.1001-7488.20211004
	董灵波, 田栋元, 陈　莹, 等. 基于结构方程模型的兴安落叶松天然林更新影响因素. 应用生态学报, 2021, 32 (8): 2763- 2772. doi: 10.13287/j.1001-9332.202108.010
	Dong L B, Tian D Y, Chen Y, et al. Clarifying the factors affecting Larix gmelinii forest regeneration based on structural equation model. Chinese Journal of Applied Ecology, 2021, 32 (8): 2763- 2772. doi: 10.13287/j.1001-9332.202108.010
	冯晓燕, 刘　宁, 郭晋平, 等. 控制光照条件下华北山地4个混交树种幼苗幼树的形态响应和可塑性. 林业科学, 2013, 49 (11): 42- 50.
	Feng X Y, Liu N, Guo J P, et al. Morphological responses and morphological plasticity indices of seedlings and saplings of four woody species in montane forest of northern China to experimental light regimes. Scientia Silvae Sinicae, 2013, 49 (11): 42- 50.
	韩广轩, 王光美, 毛培利, 等. 山东半岛北部黑松海防林幼龄植株更新动态及其影响因素. 林业科学, 2010, 46 (12): 158- 164.
	Han G X, Wang G M, Mao P L, et al. Regeneration dynamics of young pinus thunbergii and its influencing factors in the coastal protection forests in northern Shandong peninsula. Scientia Silvae Sinicae, 2010, 46 (12): 158- 164.
	惠刚盈. 角尺度: 一个描述林木个体分布格局的结构参数. 林业科学, 1999, 35 (1): 39- 44. doi: 10.3321/j.issn:1001-7488.1999.01.006
	Hui G Y. The neighbourhood pattern: a new structure parameter for describing distribution of forest tree position. Scientia Silvae Sinicae, 1999, 35 (1): 39- 44. doi: 10.3321/j.issn:1001-7488.1999.01.006
	惠刚盈, 胡艳波, 赵中华. 结构化森林经营研究进展. 林业科学研究, 2018, 31 (1): 85- 93.
	Hui G Y, Hu Y B, Zhao Z H. Research progress of structure-based forest management. Forest Research, 2018, 31 (1): 85- 93.
	惠刚盈, 胡艳波. 混交林树种空间隔离程度表达方式的研究. 林业科学研究, 2001, 14 (1): 23- 27. doi: 10.3321/j.issn:1001-1498.2001.01.004
	Hui G Y, Hu Y B. Measuring species spatial isolation in mixed forests. Forest Research, 2001, 14 (1): 23- 27. doi: 10.3321/j.issn:1001-1498.2001.01.004
	惠刚盈, Klaus von Gadow, Matthias Albert. 一个新的林分空间结构参数: 大小比数. 林业科学研究, 1999, 12 (1): 4- 9.
	Hui G Y, Gadow K V, Albert M. A new parameter for stand spatial structure neighbourhood comparison. Forest Research, 1999, 12 (1): 4- 9.
	贾黎明, 翟明普, 尹伟伦. 油松、辽东栎混交林中生化他感作用的研究. 林业科学, 1995, 31 (6): 491- 498.
	Jia L M, Zhai M P, Yi W L. The study of allelopathy of Pinus tabulaeformis and Quercus liaotungensis. Scientia Silvae Sinicae, 1995, 31 (6): 491- 498.
	焦强英, 韩宗甫, 王炜烨, 等. 基于多源数据和机器学习方法的大兴安岭地区雷击火驱动因子及火险预测模型. 林业科学, 2023, 59 (6): 74- 87.
	Jiao Q Y, Han Z F, Wang W H, et al. Driving factors and forecasting model of lightning-caused forest fires in Daxing’anling mountains based on multi-sources data and machine learning method. Scientia Silvae Sinicae, 2023, 59 (6): 74- 87.
	雷相东. 机器学习算法在森林生长收获预估中的应用. 北京林业大学学报, 2019, 41 (12): 23- 36.
	Lei X D. Applications of machine learning algorithms in forest growth and yield prediction. Journal of Beijing Forestry University, 2019, 41 (12): 23- 36.
	李春明, 付　卓. 基于广义混合效应模型的云冷杉林天然更新计数方法研究. 西南林业大学学报(自然科学), 2020, 40 (5): 108- 114. doi: 10.11929/j.swfu.202002026
	Li C M, Fu Z. The natural regeneration count model of spruce-fir mixed stand based on generalized mixed effect models. Journal of Southwest Forestry University (Natural Sciences), 2020, 40 (5): 108- 114. doi: 10.11929/j.swfu.202002026
	李欣海. 随机森林模型在分类与回归分析中的应用. 应用昆虫学报, 2013, 50 (4): 1190- 1197.
	Li X H. Using “random forest” for classification and regression. Chinese Journal of Applied Entomology, 2013, 50 (4): 1190- 1197.
	李雪云, 潘　萍, 臧　颢, 等. 闽楠天然次生林自然更新的影响因子研究. 林业科学研究, 2017, 30 (5): 701- 708. doi: 10.13275/j.cnki.lykxyj.2017.05.001
	Li X Y, Pan P, Zang H, et al. Study on factors affecting natural regeneration of natural secondary Phoebe bournei forest. Forest Research, 2017, 30 (5): 701- 708. doi: 10.13275/j.cnki.lykxyj.2017.05.001
	刘　斌, 陈　维, 陈伏生, 等. 九连山次生阔叶林幼苗生长对氮磷添加的响应. 应用生态学报, 2020, 31 (8): 2533- 2540. doi: 10.13287/j.1001-9332.202008.020
	Liu B, Chen W, Chen F S, et al. Responses of seedling growth in subtropical secondary broad-leaved forest to nitrogen and phosphorus addition in Jiulian Mountain, China. Chinese Journal of Applied Ecology, 2020, 31 (8): 2533- 2540. doi: 10.13287/j.1001-9332.202008.020
	娄淑兰, 刘目兴, 易　军, 等. 三峡山地不同类型植被和坡位对土壤水文功能的影响. 生态学报, 2019, 39 (13): 4844- 4854.
	Lou S L, Liu M X, Yi J, et al. Influence of vegetation coverage and topographic position on soil hydrological function in the hillslope of the Three Gorges area. Acta Ecologica Sinica, 2019, 39 (13): 4844- 4854.
	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. 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. Journal of Beijing Forestry University, 2019, 41 (9): 9- 19.
	任学敏, 朱　雅, 陈兆进, 等. 太白山锐齿槲栎林乔木更新特征及其影响因子. 林业科学, 2019, 55 (1): 11- 21. doi: 10.11707/j.1001-7488.20190102
	Ren X M, Zhu Y, Chen Z J, et al. Regeneration of arbor trees and its contributing factors in an oak forest in Taibai Mountain, China. Scientia Silvae Sinicae, 2019, 55 (1): 11- 21. doi: 10.11707/j.1001-7488.20190102
	索奥丽, 陈　锋, 龚俊伟, 等. 华北落叶松林和油松林地表可燃物特征、土壤理化性质及林下植物多样性对林间草地放牧的响应. 林业科学, 2025, 61 (10): 60- 73. doi: 10.11707/j.1001-7488.LYKX20250161
	Suo A L, Chen F, Gong J W, et al. Response of surface fuel characteristics, soil physicochemical properties, and understory plant diversity to forest grassland grazing in Larix gmelinii var. principis-rupprechtii and Pinus tabuliformis forests. Scientia Silvae Sinicae, 2025, 61 (10): 60- 73. doi: 10.11707/j.1001-7488.LYKX20250161
	田惠玲, 朱建华, 何　潇, 等. 基于随机森林模型的东北三省乔木林生物质碳储量预测. 林业科学, 2022, 58 (4): 40- 50. doi: 10.11707/j.1001-7488.20220405
	Tian S L, Zhu J H, He X, et al. Projected biomass carbon stock of arbor forest of three provinces in northeastern China based on random forest model. Scientia Silvae Sinicae, 2022, 58 (4): 40- 50. doi: 10.11707/j.1001-7488.20220405
	王君杰, 姜立春. 利用分位数组合预测兴安落叶松枝下高. 北京林业大学学报, 2021, 43 (3): 9- 17.
	Wang J J, Jiang L C. Predicting height to crown base for Larix gmelinii using quantile groups. Journal of Beijing Forestry University, 2021, 43 (3): 9- 17.
	肖　晨, 田栋元, 马　榕, 等. 兴安落叶松天然林更新数量相容性预测模型. 应用生态学报, 2023, 34 (9): 2345- 2354.
	Xiao C, Tian D Y, Ma R, et al. Compatibility predictive model for regeneration quantities of Larix gmelinii natural forest in Daxing’anling Mountains, China. Chinese Journal of Applied Ecology, 2023, 34 (9): 2345- 2354.
	易　晨, 李德成, 张甘霖, 等. 土壤厚度的划分标准与案例研究. 土壤学报, 2015, 52 (1): 220- 227. doi: 10.11766/trxb201402180069
	Yi C, Li D C, Zhang G L, et al. Criteria for partition of soil thickness and case studies. Acta Pedologica Sinica, 2015, 52 (1): 220- 227. doi: 10.11766/trxb201402180069
	由珈齐, 李明泽, 范文义, 等. 基于高光谱和激光雷达数据的林分类型识别. 林业科学, 2021, 57 (5): 119- 129. doi: 10.11707/j.1001-7488.20210511
	You J Q, Li M Z, Fan W Y, et al. Stand type identification based on hyperspectral and LiDAR data. Scientia Silvae Sinicae, 2021, 57 (5): 119- 129. doi: 10.11707/j.1001-7488.20210511
	张凡兵, 龙　婷, 金文斌, 等. 红松林天然更新对连续小梯度海拔变化的响应. 东北林业大学学报, 2017, 45 (5): 1- 10. doi: 10.3969/j.issn.1000-5382.2017.05.001
	Zhang F B, Long T, Jin W B, et al. Natural regeneration of primitive Korean pine forests in response to the variation in continuous small elevation gradient. Journal of Northeast Forestry University, 2017, 45 (5): 1- 10. doi: 10.3969/j.issn.1000-5382.2017.05.001
	张　雷, 刘世荣, 孙鹏森, 等. 气候变化对马尾松潜在分布影响预估的多模型比较. 植物生态学报, 2011, 35 (11): 1091- 1105. doi: 10.3724/SP.J.1258.2011.01091
	Zhang L, Liu S R, Sun P S, et al. Comparative evaluation of multiple models of the effects of climate change on the potential distribution of Pinus massoniana. Chinese Journal of Plant Ecology, 2011, 35 (11): 1091- 1105. doi: 10.3724/SP.J.1258.2011.01091
	张　雷, 王琳琳, 张旭东, 等. 随机森林算法基本思想及其在生态学中的应用: 以云南松分布模拟为例. 生态学报, 2014, 34 (3): 650- 659. doi: 10.5846/stxb201306031292
	Zhang L, Wang L L, Zhang X D, et al. The basic principle of random forest and its applications in ecology: a case study of Pinus yunnanensis. Acta Ecologica Sinica, 2014, 34 (3): 650- 659. doi: 10.5846/stxb201306031292
	张凌宇, 刘兆刚. 2019. 基于地理加权回归模型的大兴安岭中部天然次生林更新分布. 林业科学, 55(11): 105−116.
	Zhang L Y, Liu Z G, 2019. Regeneration and distribution of natural secondary forests in the central part of Daxing’an mountains based on geographically weighted regression model. Scientia Silvae Sinicae, 55(11): 105−116. ［in Chinese］
	张雄清, 雷渊才, 雷相东, 等. 基于计数模型方法的林分枯损研究. 林业科学, 2012, 48 (8): 54- 61.
	Zhang X Q, Lei Y C, Lei X D, et al. Predicting stand-level mortality with count data models. Scientia Silvae Sinicae, 2012, 48 (8): 54- 61.
	张宇娇, 赵恒谦, 付含聪, 等. 基于无人机遥感多特征的单木地上生物量反演模型. 林业科学, 2025, 61 (8): 129- 141. doi: 10.11707/j.1001-7488.LYKX20240390
	Zhang Y J, Zhao H Q, Fu H C, et al. Inversion model of aboveground biomass at individual tree scale based on the multiple features of UAV remote sensing. Scientia Silvae Sinicae, 2025, 61 (8): 129- 141. doi: 10.11707/j.1001-7488.LYKX20240390
	张兹鹏, 王君杰, 刘索名, 等. 形率对白桦单木材积和生物量预测精度的影响. 林业科学, 2022, 58 (5): 31- 39.
	Zhang Z P, Wang J J, Liu S M, et al. Effect of form quotient on prediction accuracy of individual tree volume and biomass of Betula platyphylla. Scientia Silvae Sinicae, 2022, 58 (5): 31- 39.
	竹万宽, 王志超, 许宇星, 等. 基于机器学习算法的雷州半岛桉树复层混交林土壤呼吸模拟. 林业科学, 2026, 62 (1): 67- 82.
	Zhu W K, Wang Z C, Xu Y X, et al. Simulation of Soil respiration in a stratified mixed stand of Eucalyptus spp. and Manglietia glauca in Leizhou Peninsula based on machine learning algorithms. Scientia Silvae Sinicae, 2026, 62 (1): 67- 82.
	祝子枭, 刘兆刚, 董灵波, 等. 环境因子对大兴安岭天然落叶松次生林主要树种更新的影响. 东北林业大学学报, 2020, 48 (6): 135- 141.
	Zhu Z X, Liu Z G, Dong L B, et al. Effects of environmental factors on regeneration of major species in larch secondary forest of Daxing’an Mountains. Journal of Northeast Forestry University, 2020, 48 (6): 135- 141.
	Boulesteix A L, Bender A, Lorenzo Bermejo J, et al. Random forest Gini importance favours SNPs with large minor allele frequency: impact, sources and recommendations. Briefings in Bioinformatics, 2012, 13 (3): 292- 304.
	Breiman L. Random forests. Machine Learning, 2001, 45 (1): 5- 32.
	Cutler D R, Edwards T C Jr, Beard K H, et al. Random forests for classification in ecology. Ecology, 2007, 88 (11): 2783- 2792.
	Diao J J, Feng T, Li M S, et al. Use of vegetation change tracker, spatial analysis, and random forest regression to assess the evolution of plantation stand age in southeast China. Annals of Forest Science, 2020, 77 (2): 27.
	Forbes C, Evans M, Hastings N, et al. 2010. Statistical Distributions. John Wiley & Sons, Inc.
	Degenhardt F, Seifert S, Szymczak S. Evaluation of variable selection methods for random forests and omics data sets. Briefings in Bioinformatics, 2019, 20 (2): 492- 503.
	Iverson L R, Prasad A M, Matthews S N, et al. Estimating potential habitat for 134 eastern US tree species under six climate scenarios. Forest Ecology and Management, 2008, 254 (3): 390- 406.
	Khaine I, Woo S Y, Kwak M, et al. Factors affecting natural regeneration of tropical forests across a precipitation gradient in Myanmar. Forests, 2018, 9 (3): 143.
	Nicodemus K K. Letter to the editor: on the stability and ranking of predictors from random forest variable importance measures. Briefings in Bioinformatics, 2011, 12 (4): 369- 373.

项 Item	变量 Variables	最小值 Min.	平均值 Mean	最大值 Max.	标准差 SD	变异系数 CV （%）
林分特征 Stand characteristics	平均胸径Mean DBH/cm	9.50	13.37	21.10	2.63	19.67
	平均树高Mean height/m	8.80	11.75	18.20	2.13	18.13
	林木密度Tree density/（ tree·hm^?2）	567.00	1 466.00	2 680.00	543.33	37.06
	单位蓄积量Volume/（m³·hm^?2）	53.47	127.58	223.11	39.40	30.88
	郁闭度Canopy density	0.25	0.63	0.90	0.12	19.05
	灌木盖度Shrub coverage （%）	0.00	15.48	84.00	20.07	129.65
	草本盖度Herb coverage （%）	0.03	19.99	84.25	25.82	129.16
立地条件 Site conditions	海拔Elevation/m	404.00	514.45	676.30	67.66	13.15
	坡度Slope	0.00	7.00	25.00	6.41	0.92
	坡向Aspect	1.00	4.96	8.00	2.52	50.75
	坡位Site	1.00	2.03	3.00	0.62	30.52
土壤条件 Soil conditions	酸碱度pH	4.03	5.28	6.41	0.40	7.63
	OM/（g·kg^?1）	1.01	4.50	6.81	18.10	30.06
	TN/（g·kg^?1）	0.67	3.21	9.11	1.68	52.36
	TP/（g·kg^?1）	1.21	1.53	2.46	0.24	15.46
	TK/（g·kg^?1）	6.14	8.28	10.85	1.28	15.41
林木大小多样性 Tree size diversity	DSW	1.54	2.11	2.89	0.27	13.03
	DS	0.76	0.86	0.96	0.04	4.66
	DP	0.72	0.88	1.00	0.07	7.45
	HSW	1.23	1.76	2.12	0.18	10.52
	HS	0.67	0.80	0.87	0.04	5.07
	HP	0.72	0.87	0.99	0.06	6.45
物种多样性 Species diversity	TSSW	0.05	0.66	1.19	0.25	37.96
	TSS	0.02	0.39	0.64	0.14	37.19
	TSP	0.07	0.63	1.00	0.20	32.18
林分空间结构 Stand spatial structure	W	0.41	0.50	0.75	0.05	9.12
	U	0.39	0.50	0.70	0.04	8.21
	M	0.03	0.34	0.61	0.14	40.45
	SDP	0.00	0.63	1.00	0.39	61.93
更新数量 Regeneration number	兴安落叶松 Larix gmelinii /（ tree·hm^?2）	90.00	899.00	4 550.00	999.39	111.17
更新数量 Regeneration number	白桦Betula platyphylla /（ tree·hm^?2）	40.00	1 035.00	7 800.00	1 406.53	135.90

参数Parameter	兴安落叶松Larix gmelinii						白桦Betula platyphylla
参数Parameter	Poisson	P	VIF	NB	P	VIF	Poisson	P	VIF	NB	P	VIF
截距Intercept	4.228 0	***		7.114 1	***		2.034 0	***		2.534 4	***
ELE	?0.003 1	***	2.082 2	?0.004 7	***	1.499 2
OM	0.048 1	***	1.548 0
TP	?1.755 0	***	4.955 3	?0.887 5	*	1.532 0
TK	0.043 8	***	4.108 5
DSW				1.567 5	***	1.045 8	0.826 0	***	8.115 0
DS	4.685 0	***	3.719 1
DP							2.855 0	***	4.215 4
HSW							0.169 8	***	1.619 5	0.947 4	*	1.449 0
MDBH										0.230 6	***	2.112 5
MH	0.132 1	***	5.671 0				0.028 7	***	5.523 9
TD	?0.000 1	***	3.191 4				?0.000 5	***	3.210 8
V	0.001 1	***	3.570 9				0.001 3	***	3.851 6	?0.006 0	*	1.657 2
M	1.270 0	***	1.054 7
SC							0.002 9	***	1.822 5
HC							0.007 4	***	2.486 3	0.006 4	*	1.368 8

树种 Tree species	模型 Model	检验指标Test indicators
树种 Tree species	模型 Model	RMSE/ （tree·hm^?2）	rRMSE （%）	MAE/ （tree·hm^?2）	MAPE （%）
兴安落叶松 Larix gmelinii	Poisson	1 312	142.63	899	98.38
	NB	1 310	142.62	898	98.38
	RF	482	55.72	377	80.92
白桦 Betula platyphylla	Poisson	1 588	150.32	1 012	98.07
	NB	1 586	150.31	1 011	98.05
	RF	682	68.00	460	84.85

[1]	竹万宽,王志超,许宇星,黄润霞,陶怡,钟源源,杜阿朋. 基于机器学习算法的雷州半岛桉树复层混交林土壤呼吸模拟[J]. 林业科学, 2026, 62(1): 67-82.
[2]	郝嘉珩,郭毅超,李浩,朱爱青,石雷. 基于时序遥感指数的中国竹林植被信息提取[J]. 林业科学, 2025, 61(9): 1-11.
[3]	熊晓燕,李彩霞,柴国奇,陈龙,贾翔,雷令婷,张晓丽. 联合UAV-LiDAR和GEDI数据的区域森林地上生物量估算[J]. 林业科学, 2025, 61(8): 142-153.
[4]	黄栋才,郭鑫,王得祥,王韵淑,张欣,霍雪莹. 不同经营方法对秦岭华北落叶松林分生长和林下植被的影响[J]. 林业科学, 2024, 60(8): 57-66.
[5]	朱诗豪,吴志伟,李政杰,李顺. 赣南马尾松林地表细小死可燃物含水率动态及模型[J]. 林业科学, 2024, 60(5): 158-168.
[6]	刘守佳,何拓,陆杨,焦立超,郭娟,Alex CWiedenhoeft,殷亚方. 桃花心木木材定量解剖特征变异性及其识别[J]. 林业科学, 2024, 60(5): 169-176.
[7]	党宏忠,陈帅,钟鹏,韩辉,张日升,张学利,石长春. 樟子松人工林自然更新过程中断的机制及可能调控途径[J]. 林业科学, 2024, 60(12): 158-167.
[8]	周小成, 黄婷婷, 李媛, 肖祥希, 朱洪如, 陈芸芝, 冯芝淸. 结合遥感林龄因子的亚热带森林蓄积量估算方法[J]. 林业科学, 2023, 59(4): 88-99.
[9]	田惠玲,朱建华,何潇,陈新云,简尊吉,李宸宇,郭学媛,黄国胜,肖文发. 基于随机森林模型的东北三省乔木林生物质碳储量预测[J]. 林业科学, 2022, 58(4): 40-50.
[10]	刘鑫源,杨光,宁吉彬,耿道通,于宏洲,邸雪颖. 红松人工林地表可燃物燃烧释放颗粒物质量及影响因素[J]. 林业科学, 2022, 58(3): 97-106.
[11]	郭旭展,陈巧,张晓芳,洪亮,尤媛媛,唐守正,符利勇. 基于无人机高分辨率影像的油松新造林健康树冠提取[J]. 林业科学, 2022, 58(10): 111-120.
[12]	王金池,黄清麟,严铭海,黄如楚,郑群瑞. 由巨桉人工林转型的13年生青冈栎天然林特征[J]. 林业科学, 2021, 57(9): 13-20.
[13]	孙忠秋,高金萍,吴发云,高显连,胡杨,高剑新. 基于机载激光雷达点云和随机森林算法的森林蓄积量估测[J]. 林业科学, 2021, 57(8): 68-81.
[14]	由珈齐,李明泽,范文义,全迎,王斌,莫祝坤,祝子枭. 基于高光谱和激光雷达数据的林分类型识别[J]. 林业科学, 2021, 57(5): 119-129.
[15]	高艳丽,杨智杰,张丽,熊德成. 不同更新方式对亚热带常绿阔叶林土壤氮矿化的影响[J]. 林业科学, 2021, 57(4): 24-31.