基于随机森林模型的东北三省乔木林生物质碳储量预测

doi:10.11707/j.1001-7488.20220405

Abstract

Abstract:

Objective: Based on the data of permanent monitoring plots of the successive national forest inventories, a multi-factor forest growth model was constructed through the machine learning algorithm, which improves the simulation accuracy of forest growth and carbon sequestration, predicts the future carbon sink potential of arbor forests of the three provinces in northeast China, and explore the potential distribution of arbor forest carbon sink. It provides a scientific guidance for accurately locating the role of forests in northeast China in increasing sinks and reducing emissions, and scientifically formulating the national "Carbon Neutral" action path and target management. Method: Based on the data of permanent monitoring plots of the successive national forest inventories from 1999 to 2018, and combined with multiple influence factors such as regional climate, soil, forest stand and topography, etc., we used the random forest algorithm to construct a growth-loss model of the main dominant tree species (groups) in the region. Combining the future climate and arbor forest area expansion scenario, the biomass carbon stock (BCS) and BCS changes of arbor forests of the three provinces in northeast China from 2015 to 2060 were predicted. Result: The BCS of arbor forests of three provinces in northeast China reached 3 393.15 TgC in 2060, an increase of 1 895.23 TgC compared with that in 2015, and the annual mean BCS changes was 42.12 TgC ·a^-1 during 2015-2060. Natural forests played the major role in the BCS and BCS changes. The BCS of arbor forests in Liaoning, Jilin, and Heilongjiang will be increased from 139.19, 463.58, and 895.15 TgC in 2015 to 328.95, 915.83, and 2 148.37 TgC in 2060, respectively. The average biomass Carbon Density (BCD) of arbor forests will be increased from 32.71, 59.75, and 45.11 t ·hm^-2 in 2015 to 75.20, 109.32, and 85.24 t ·hm^-2 in 2060, respectively, and the average annual biomass carbon sinks of arbor forests in Liaoning, Jilin, and Heilongjiang from 2015 to 2060 are 4.22, 10.05, and 27.85 TgC ·a^-1, respectively. Conclusion: The random forest model constructed in this study performs well and can be used to predict the future BCS of arbor forests of three provinces in northeast China. From 2015 to 2060, the BCS of arbor forests of three provinces in northeast China will be increased by 1 895.23 TgC, which will still have a large carbon sink potential in the future. Heilongjiang Province has the largest carbon sink potential, with an annual increase of carbon storage up to 27.85 TgC ·a^-1, which is an important carbon sink area in the future; while Liaoning Province is weaker, with an annual increase of carbon storage by only 4.22 TgC ·a^-1. Strengthening the management of the young and middle-aged forests, moderately regenerating the mature and over-mature forests, and enhancing the carbon sink function of arbor forests of the three provinces in northeast China will help to improve the role of forests in achieving the goal of "Carbon Neutral".

Key words: growth-loss model, random forest algorithm, national forest resources inventory, multivariate, biomass carbon stock

CLC Number:

S718.55

Huiling Tian,Jianhua Zhu,Xiao He,Xinyun Chen,Zunji Jian,Chenyu Li,Xueyuan Guo,Guosheng Huang,Wenfa Xiao. Projected Biomass Carbon Stock of Arbor Forest of Three Provinces in Northeastern China Based on Random Forest Model[J]. Scientia Silvae Sinicae, 2022, 58(4): 40-50.

Figures/Tables 8

Table 1

Table 2

Table 3

Table 4

Fig.1

Table 5

Fig.2

Fig.3

References 0

	范春楠, 韩士杰, 郭忠玲, 等. 吉林省森林植被固碳现状与速率. 植物生态学报, 2016, 40 (4): 341- 353.
	Fan C N , Han S J , Guo Z L , et al. Present status and rate of carbon sequestration of forest vegetation in Jilin Province, Northeast China. Chinese Journal of Plant Ecology, 2016, 40 (4): 341- 353.
	国家林业和草原局. 中国森林资源报告(2014—2018). 北京: 中国林业出版社, 2019.
	State Forestry and Grassland Administration of China . Report of forest resources in China (2014—2018). Beijing: China Forestry Publishing House, 2019.
	郭颖婕, 刘晓燕, 郭茂祖, 等. 植物抗性基因识别中的随机森林分类方法. 计算机科学与探索, 2012, 6 (1): 67- 77. doi: 10.3778/j.issn.1673-9418.2012.01.005
	Guo Y J , Liu X Y , Guo M Z , et al. Identification of plant resistance gene with random forest. Journal of Frontiers of Computer Science & Technology, 2012, 6 (1): 67- 77. doi: 10.3778/j.issn.1673-9418.2012.01.005
	焦燕, 胡海清. 黑龙江省森林植被碳储量及其动态变化. 应用生态学报, 2005, 16 (12): 2248- 2252. doi: 10.3321/j.issn:1001-9332.2005.12.005
	Jiao Y , Hu H Q . Carbon storage and its dynamics of forest vegetations in Heilongjiang Province. Chinese Journal of Applied Ecology, 2005, 16 (12): 2248- 2252. doi: 10.3321/j.issn:1001-9332.2005.12.005
	李奇, 朱建华, 冯源, 等. 中国森林乔木林碳储量及其固碳潜力预测. 气候变化研究进展, 2018, 14 (3): 287- 294.
	Li Q , Zhu J H , Feng Y , et al. Carbon storage and carbon sequestration potential of the forest in China. Climate Change Research, 2018, 14 (3): 287- 294.
	李婉华, 陈宏, 郭坤, 等. 基于随机森林算法的用电负荷预测研究. 计算机工程与应用, 2016, 52 (23): 236- 243. doi: 10.3778/j.issn.1002-8331.1606-0203
	Li W H , Chen H , Guo K , et al. Research on electrical load prediction based on random forest algorithm. Computer Engineering and Applications, 2016, 52 (23): 236- 243. doi: 10.3778/j.issn.1002-8331.1606-0203
	罗云建, 王效科, 张小全, 等. 中国森林生态系统生物量及其分配研究. 北京: 中国林业出版社, 2013.
	Luo Y J , Wang X K , Zhang X Q , et al. Biomass and its allocation of forest ecosystems in China. Beijing: China Forestry Publishing House, 2013.
	马晓哲, 王铮. 中国分省区森林碳汇量的一个估计. 科学通报, 2011, 56 (6): 433- 439.
	Ma X Z , Wang Z . An estimation of forest carbon sinks by provinces and regions in China. Chinese Science Bulletin, 2011, 56 (6): 433- 439.
	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.
	王春梅, 邵彬, 王汝南. 东北地区两种主要造林树种生态系统固碳潜力. 生态学报, 2010, 30 (7): 1764- 1772.
	Wang C M , Shao B , Wang R N . Carbon sequestration potential of ecosystem of two main tree species in Northeast China. Acta Ecologica Sinica, 2010, 30 (7): 1764- 1772.
	郗婷婷, 李顺龙. 黑龙江省森林碳汇潜力分析. 林业经济问题, 2006, 26 (6): 519- 522.519-522, 526 doi: 10.3969/j.issn.1005-9709.2006.06.008
	Xi T T , Li S L . Analysis of forestry carbon mitigation potential in Heilongjiang Province. Issues of Forestry Economics, 2006, 26 (6): 519- 522.519-522, 526 doi: 10.3969/j.issn.1005-9709.2006.06.008
	许恩银, 王维枫, 聂影, 等. 中国林业碳贡献区域分布及潜力预测. 中国人口·资源与环境, 2020, 30 (5): 36- 45.
	Xu E Y , Wang W F , Nie Y , et al. Regional distribution and potential forecast of China's forestry carbon contributions. China Population, Resources and Environment, 2020, 30 (5): 36- 45.
	张煜星, 王雪军. 全国森林蓄积生物量模型建立和碳变化研究. 中国科学(生命科学), 2021, 51 (2): 199- 214.
	Zhang Y X , Wang X J . Study on forest volume-to-biomass modeling and carbon storage dynamics in China. Scientia Sinica (Vitae), 2021, 51 (2): 199- 214.
	中共中央国务院. 关于完整准确全面贯彻新发展理念做好碳达峰碳中和工作的意见. 中华人民共和国国务院公报, 2021, (31): 33- 38.
	Opinions of the Central Committee of the CPC and the State Council . Carbon dioxide peaking and carbon neutrality in full and faithful implementation of the new development. Philosophy Chinese Full Text, 2021, (31): 33- 38.
	周志华. 机器学习. 北京: 清华大学出版社, 2016.
	Zhou Z H . Machine learning. Beijing: Tsinghua University Press, 2016.
	Ashraf M I , Zhao Z Y , Bourque C P A , et al. Integrating biophysical controls in forest growth and yield predictions with artificial intelligence technology. Canadian Journal of Forest Research, 2013, 43 (12): 1162- 1171. doi: 10.1139/cjfr-2013-0090
	De'ath G . Boosted trees for ecological modeling and prediction. Ecology, 2007, 88 (1): 243- 251. doi: 10.1890/0012-9658(2007)88[243:BTFEMA]2.0.CO;2
	Doelman J C , Stehfest E , van Vuuren D P , et al. Afforestation for climate change mitigation: potentials, risks and trade-offs. Global Change Biology, 2020, 26 (3): 1576- 1591. doi: 10.1111/gcb.14887
	Fang J , Chen A , Peng C , et al. Changes in forest biomass carbon storage in China between 1949 and 1998. Science, 2001, 292 (5525): 2320- 2322. doi: 10.1126/science.1058629
	Fang J Y , Guo Z D , Hu H F , et al. Forest biomass carbon sinks in East Asia, with special reference to the relative contributions of forest expansion and forest growth. Global Change Biology, 2014, 20 (6): 2019- 2030. doi: 10.1111/gcb.12512
	Fernández-Martínez M , Vicca S , Janssens I A , et al. Addendum: nutrient availability as the key regulator of global forest carbon balance. Nature Climate Change, 2014, 4 (7): 643. doi: 10.1038/nclimate2282
	Gong P , Liu H , Zhang M N , et al. Stable classification with limited sample: transferring a 30-m resolution sample set collected in 2015 to mapping 10-m resolution global land cover in 2017. Science Bulletin, 2019, 64 (6): 370- 373. doi: 10.1016/j.scib.2019.03.002
	Hu H F , Wang S P , Guo Z D , et al. The stage-classified matrix models project a significant increase in biomass carbon stocks in China's forests between 2005 and 2050. Scientific Reports, 2015, 5, 11203. doi: 10.1038/srep11203
	Jevšenak J , Skudnik M . A random forest model for basal area increment predictions from national forest inventory data. Forest Ecology and Management, 2021, 479, 118601. doi: 10.1016/j.foreco.2020.118601
	Kuhn M , Johnson K . Applied predictive modeling. New York: Springer, 2013.
	Lin B Q , Ge J M . Valued forest carbon sinks: how much emissions abatement costs could be reduced in China. Journal of Cleaner Production, 2019, 224, 455- 464. doi: 10.1016/j.jclepro.2019.03.221
	Lun F , Liu Y , He L , et al. Life cycle research on the carbon budget of the Larix principis-rupprechtii plantation forest ecosystem in North China. Journal of Cleaner Production, 2018, 177, 178- 186. doi: 10.1016/j.jclepro.2017.12.126
	Mina M , Huber M O , Forrester D I , et al. Multiple factors modulate tree growth complementarity in Central European mixed forests. Journal of Ecology, 2018, 106 (3): 1106- 1119. doi: 10.1111/1365-2745.12846
	Ni J . Carbon storage in Chinese terrestrial ecosystems: approaching a more accurate estimate. Climatic Change, 2013, 119 (3/4): 905- 917.
	Ou Q X , Lei X D , Shen C C . Individual tree diameter growth models of larch-spruce-fir mixed forests based on machine learning algorithms. Forests, 2019, 10 (2): 187. doi: 10.3390/f10020187
	Prasad A M , Iverson L R , Liaw A . Newer classification and regression tree techniques: bagging and random forests for ecological prediction. Ecosystems, 2006, 9 (2): 181- 199. doi: 10.1007/s10021-005-0054-1
	Qi G , Chen H , Zhou L , et al. Carbon stock of larch plantations and its comparison with an old-growth forest in northeast China. Chinese Geographical Science, 2016, 26 (1): 10- 21. doi: 10.1007/s11769-015-0772-z
	Qiu Z X , Feng Z K , Song Y N , et al. Carbon sequestration potential of forest vegetation in China from 2003 to 2050: predicting forest vegetation growth based on climate and the environment. Journal of Cleaner Production, 2020, 252, 119715. doi: 10.1016/j.jclepro.2019.119715
	Richards K R , Stokes C . A review of forest carbon sequestration cost studies: a dozen years of research. Climatic Change, 2004, 63 (1/2): 1- 48. doi: 10.1023/B:CLIM.0000018503.10080.89
	Sharma T , Kurz W A , Stinson G , et al. A 100-year conservation experiment: impacts on forest carbon stocks and fluxes. Forest Ecology and Management, 2013, 310, 242- 255. doi: 10.1016/j.foreco.2013.06.048
	Stinson G , Kurz W A , Smyth C E , et al. An inventory-based analysis of Canada's managed forest carbon dynamics, 1990 to 2008. Global Change Biology, 2011, 17 (6): 2227- 2244. doi: 10.1111/j.1365-2486.2010.02369.x
	Tang X L , Zhao X , Bai Y F , et al. Carbon pools in China's terrestrial ecosystems: new estimates based on an intensive field survey. Proceedings of the National Academy of Sciences of the United States of America, 2018, 115 (16): 4021- 4026. doi: 10.1073/pnas.1700291115
	Wei S G , Dai Y J , Liu B Y , et al. A China data set of soil properties for land surface modeling. Journal of Advances in Modeling Earth Systems, 2013, 5 (2): 212- 224. doi: 10.1002/jame.20026

年份Year	辽宁Liaoning		吉林Jilin		黑龙江Heilongjiang		合计Total
年份Year	人工 Planted	天然 Natural	人工 Planted	天然 Natural	人工 Planted	天然 Natural	人工 Planted	天然 Natural
2016—2020	1.42	0.71	7.41	3.70	64.20	32.10	73.03	36.52
2021—2025	1.32	0.66	6.89	3.45	59.74	29.87	67.95	33.98
2026—2030	1.25	0.63	6.52	3.26	56.50	28.25	64.27	32.14
2031—2035	1.20	0.60	6.24	3.12	54.05	27.03	61.48	30.74
2036—2040	1.16	0.58	6.02	3.01	52.19	26.10	59.37	29.69
2041—2045	1.13	0.56	5.86	2.93	50.79	25.39	57.77	28.89
2046—2050	0.44	0.22	2.29	1.15	19.89	9.94	22.62	11.31
2051—2055	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00
2056—2060	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00
2016—2060	7.92	3.96	41.23	20.62	357.36	178.68	406.51	203.25

起源 Origin	优势树种(组) Dominant tree species (group)	现有林面积 Existing forest area/ 10⁴ hm²	生物量-蓄积模型B_{j, p}=a_j×V_{j, p}^b_j×λ_j Biomass-volume model					生物量含碳率 Biomass carbon fraction
起源 Origin	优势树种(组) Dominant tree species (group)	现有林面积 Existing forest area/ 10⁴ hm²	样本数 Sample number	a	b	相关系数 Correlation coefficient (R²)	校正系数 Correction coefficient (λ)	生物量含碳率 Biomass carbon fraction
天然林 Natural forest	云杉Picea asperata	13.02	25	5.413	0.633	0.966	1.012	0.493
	落叶松Larix gmelinii	227.79	38	2.985	0.746	0.902	1.022	0.489
	油松Pinus tabulaeformis	6.35	72	2.475	0.752	0.951	1.013	0.517
	栎类Quercus	348.65	18	1.682	0.918	0.978	1.007	0.480
	白桦Betula platyphylla	319.48	12	8.329	0.467	0.627	1.050	0.487
	榆树Ulmus pumila	39.64	59	3.300	0.741	0.884	1.035	0.476
	杨树Populus	64.01	19	1.703	0.803	0.884	1.027	0.491
	针叶混交林Mixed coniferous forest	45.99	11	6.699	0.538	0.808	1.012	0.502
	阔叶混交林Mixed broadleaved forest	1 179.05	20	1.526	0.908	0.898	1.028	0.479
	针阔混交林Mixed coniferous and broadleaved forest	248.10	54	3.088	0.734	0.832	1.033	0.494
	小计Subtotal	2 568.28

人工林 Planted forest	云杉Picea asperata	13.15	25	5.413	0.633	0.966	1.012	0.493
	落叶松Larix gmelinii	208.17	38	2.985	0.746	0.902	1.022	0.489
	樟子松Pinus sylvestris var. mongolica	28.78	39	5.117	0.602	0.889	1.017	0.511
	油松Pinus tabulaeformis	41.40	72	2.475	0.752	0.951	1.013	0.517
	栎类Quercus	12.56	18	1.682	0.918	0.978	1.007	0.480
	杨树Populus	146.56	19	1.703	0.803	0.884	1.027	0.491
	针叶混交林Mixed coniferous forest	76.32	11	6.699	0.538	0.808	1.012	0.502
	阔叶混交林Mixed broadleaved forest	90.66	20	1.526	0.908	0.898	1.028	0.479
	小计Subtotal	617.59

起源Origin	优势树种(组)Dominant tree species (group)	样地数Sample plot number	最优模型mtry mtry of the optimal model	10折交叉验证评价指标10-fold cross-validation evaluation index
起源Origin	优势树种(组)Dominant tree species (group)	样地数Sample plot number	最优模型mtry mtry of the optimal model	R²	MAE/m³	RMSE/m³
天然林 Natural forest	云杉Picea asperata	43	8	0.740	0.677	1.050
	落叶松Larix gmelinii	660	29	0.653	0.457	0.606
	油松Pinus tabulaeformis	31	28	0.481	1.191	1.364
	栎类Quercus	1 657	29	0.661	0.506	0.726
	白桦Betula platyphylla	861	13	0.392	0.866	1.158
	榆树Ulmus pumila	24	20	0.829	0.977	1.042
	杨树Populus	73	14	0.670	1.112	1.488
	针叶混交林Mixed coniferous forest	169	16	0.684	1.350	1.913
	阔叶混交林Mixed broadleaved forest	4 376	5	0.308	1.237	1.643
	针阔混交林Mixed coniferous and broadleaved forest	474	5	0.396	1.422	1.820

人工林 Planted forest	云杉Picea asperata	29	29	0.828	2.573	2.908
	落叶松Larix gmelinii	939	19	0.409	1.902	2.519
	樟子松Pinus sylvestris var. mongolica	134	1	0.350	1.990	2.586
	油松Pinus tabulaeformis	220	28	0.570	0.375	0.451
	栎类Quercus	33	26	0.416	1.877	2.137
	杨树Populus	713	28	0.466	1.979	2.585
	针叶混交林Mixed coniferous forest	30	19	0.485	1.923	2.501
	阔叶混交林Mixed broadleaved forest	69	13	0.443	1.445	1.865

起源 Origin	实际样地数 Actual sample plot number	预测样地数Predictive sample plot number		误差率 Error rate (%)	评价指标Evaluation index
起源 Origin	实际样地数 Actual sample plot number	发生枯损 Mortality	未发生枯损 Non-mortality	误差率 Error rate (%)	R²	MAE	RMSE
天然林 Natural forest	发生枯损 Mortality	2 596	4	0.15	0.671	0.399	0.806
天然林 Natural forest	未发生枯损 Non-mortality	138	5 630	2.39	0.671	0.399	0.806
人工林 Planted forest	发生枯损 Mortality	1 188	12	1.00	0.731	0.598	1.126
人工林 Planted forest	未发生枯损 Non-mortality	2	965	0.21	0.731	0.598	1.126

项目Item	年份 Year	天然林Natural forest			人工林Planted forest			乔木林Arbor forest
项目Item	年份 Year	碳储量 Carbon storage/ TgC	年碳汇量 Annual carbon sink/ (TgC·a^-1)	碳密度 Carbon density/ (MgC·hm^-2)	碳储量 Carbon storage/ TgC	年碳汇量 Annual carbon sink/ (TgC·a^-1)	碳密度 Carbon density/ (MgC·hm^-2)	碳储量 Carbon storage/ TgC	年碳汇量 Annual carbon sink/ (TgC·a^-1)	碳密度 Carbon density/ (MgC·hm^-2)
现有林地 Existing woodland	2015	1 321.48	—	51.45	176.43	—	28.57	1 497.92	—	47.02
	2020	1 487.56	33.22	57.92	219.44	8.60	35.53	1 707.00	41.82	53.58
	2025	1 646.33	31.75	64.10	253.74	6.86	41.08	1 900.06	38.61	59.64
	2030	1 800.86	30.91	70.12	283.45	5.94	45.90	2 084.31	36.85	65.42
	2035	1 951.94	30.22	76.00	309.89	5.29	50.18	2 261.83	35.50	71.00
	2040	2 100.40	29.69	81.78	334.47	4.92	54.16	2 434.87	34.61	76.43
	2045	2 243.45	28.61	87.35	356.34	4.37	57.70	2 599.79	32.98	81.60
	2050	2 384.38	28.19	92.84	377.24	4.18	61.08	2 761.62	32.37	86.68
	2055	2 523.51	27.83	98.26	397.52	4.06	64.37	2 921.03	31.88	91.69
	2060	2 660.88	27.47	103.61	417.31	3.96	67.57	3 078.19	31.43	96.62

新增林地 New woodland	2015	0	—	0	0	—	0	0	—	0
	2020	3.04	0.61	8.32	11.64	2.33	15.93	14.68	2.94	13.40
	2025	7.92	0.98	11.23	31.18	3.91	22.12	39.10	4.89	18.49
	2030	14.42	1.30	14.05	56.94	5.15	27.74	71.36	6.45	23.18
	2035	22.36	1.59	16.76	87.36	6.08	32.75	109.72	7.67	27.42
	2040	31.65	1.86	19.41	121.41	6.81	37.23	153.06	8.67	31.29
	2045	42.15	2.10	21.96	158.10	7.34	41.18	200.26	9.44	34.78
	2050	52.45	2.06	25.80	191.58	6.70	47.13	244.03	8.75	40.02
	2055	62.04	1.92	30.52	219.60	5.60	54.02	281.63	7.52	46.19
	2060	71.27	1.85	35.06	243.69	4.82	59.95	314.95	6.66	51.65

Projected Biomass Carbon Stock of Arbor Forest of Three Provinces in Northeastern China Based on Random Forest Model

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 8

References 0

Related Articles 6

Recommended Articles

Metrics

Comments

[1]	Zhongqiu Sun,Jinping Gao,Fayun Wu,Xianlian Gao,Yang Hu,Jianxin Gao. Estimating Forest Stock Volume via Small-Footprint LiDAR Point Cloud Data and Random Forest Algorithm [J]. Scientia Silvae Sinicae, 2021, 57(8): 68-81.
[2]	Qiu Shuai, Shen Baichun, Li Tingting, Guo Juan, Wang Ji, Sun Lina, Chen Xuping, Hu Shaoqing. A Method of Osmanthus fragrans Cultivars Identification Based on Random Forest Algorithm and SRAP Molecular Markers [J]. Scientia Silvae Sinicae, 2018, 54(1): 32-45.
[3]	Liang Huiling, Lin Yurui, Yang Guang, Su Zhangwen, Wang Wenhui, Guo Futao. Application of Random Forest Algorithm on the Forest Fire Prediction in Tahe Area Based on Meteorological Factors [J]. Scientia Silvae Sinicae, 2016, 52(1): 89-98.
[4]	Liu Enbin, Shi Yongjun, Li Yongfu, Zhou Guomo, Yang Dong. Non Spatial Structural Characteristic of Moso Bamboo Forest and Its Dynamics in Zhejiang Province [J]. Scientia Silvae Sinicae, 2013, 49(9): 1-7.
[5]	hi Gangrong;Xing Haitao. Eco-Anatomical Characteristics of Eight Tree Species in Xiangshan Mountain, Huaibei [J]. Scientia Silvae Sinicae, 2007, 43(3): 28-33.
[6]	Feng Zhiqiang. THE MARKING METHOD OF EQUAL PROBABILITY FOR EQUAL MARK AND ITS APPLICATION TO THE SELECTION OF AMELIORATED WALNUT VARIETIES [J]. , 1989, 25(4): 382-388.