Comparison of artificial neural network with compatible biomass model for predicting aboveground biomass of individual tree

doi:10.13287/j.1001-9332.202201.001

Abstract

Abstract: Forest biomass is an important index in forest development planning and forest resource monitoring. In order to provide a more efficient and low-biased method for estimating individual tree biomass, we introduced artificial neural network here. We used the data of aboveground biomass of 101 Larix olgensis trees harvested from the Dongzhelenghe Forest Farm in Heilongjiang Province to develop four aggregation model systems (AMS), based on different combination of the variables (diameter at breast height, tree height, crown width). The weighted functions were used to eliminate heteroscedasticity. Then, we trained artificial neural network (ANN) biomass model based on the optimal combination. The models were tested by the leave-one-out cross-validation method to compare the accuracy of the two biomass estimation methods. The results showed that biomass model based on only one variable, diameter at breast height, could accurately estimate the biomass of L. olgensis. Adding two indices, tree height and crown width, could improve the fitting performance of models, with AMS4 performing the best among the four addictive model systems. The biomass models developed by the two methods both could estimate biomass at tree level accurately, with the coefficient of determination (R²) of each component was higher than 0.87. Compared with the AMS4, R² of leaf biomass model was about 0.05 higher, and that of other organs were also about 0.01 higher in artificial neural network model system. In addition, the root mean square error (RMSE) and other indicators were also significantly smaller. For example, the RMSE of tree stem and aboveground biomass were smaller by 2.135 kg and 3.908 kg, respectively. The model's validation statistics mean relative error (MRE) performed better. In general, ANN was a flexible and reliable biomass estimation method, which was worthy consideration when predicting tree component biomass or aboveground biomass.

Key words: artificial neural network, compatible model, seemingly unrelated regression, aboveground biomass

LIANG Rui-ting, WANG Yi-fu, QIU Si-yu, SUN Yu-jun, XIE Yun-hong. Comparison of artificial neural network with compatible biomass model for predicting aboveground biomass of individual tree[J]. Chinese Journal of Applied Ecology, 2022, 33(1): 9-16.

References 25

[1]	董利虎, 李凤日. 大兴安岭东部主要林分类型乔木层生物量估算模型. 应用生态学报, 2018, 29(9): 2825-2834
[2]	Asrat Z, Eid T, Gobakken T, et al. Aboveground tree biomass prediction options for the dry Afromontane fore-sts in south-central Ethiopia. Forest Ecology and Mana-gement, 2020, 473: 118335
[3]	Dong L, Zhang L, Li F. Developing two additive biomass equations for three coniferous plantation species in Northeast China. Forests, 2016, 7: 136
[4]	罗云建, 张小全, 王效科, 等. 森林生物量的估算方法及其研究进展. 林业科学, 2009, 45(8): 129-134
[5]	曹磊, 刘晓彤, 李海奎, 等. 广东省常绿阔叶林生物量生长模型. 林业科学研究, 2020, 33(5): 61-67
[6]	王佳慧, 李凤日, 董利虎. 基于不同预测变量的天然椴树可加性地上生物量模型构建. 应用生态学报, 2018, 29(11): 3685-3695
[7]	王金池, 邓华锋, 黄国胜, 等. 天然云杉相容性生物量估算模型. 应用生态学报, 2017, 28(10): 3189-3196
[8]	曾伟生, 唐守正. 利用度量误差模型方法建立相容性立木生物量方程系统. 林业科学研究, 2010, 23(6): 797-803
[9]	符利勇, 雷渊才, 曾伟生. 几种相容性生物量模型及估计方法的比较. 林业科学, 2014, 50(6): 42-54
[10]	Fu L, Lei Y, Wang G, et al. Comparison of seemingly unrelated regressions with error-in-variable models for developing a system of nonlinear additive biomass equations. Trees, 2015, 30: 839-857
[11]	Dong L, Zhang Y, Zhang Z, et al. Comparison of tree biomass modeling approaches for larch (Larix olgensis Henry) trees in Northeast China. Forests, 2020, 11: 202
[12]	Lei Y, Fu L, Affleck DLR, et al. Additivity of nonlinear tree crown width models: Aggregated and disaggregated model structures using nonlinear simultaneous equations. Forest Ecology and Management, 2018, 427: 372-382
[13]	曹磊, 李海奎. 两种相容性生物量模型的比较——以广东省3个阔叶树种为例. 生态学杂志, 2019, 38(6): 1916-1925
[14]	Özçelık R, Diamantopoulou MJ, Eker M, et al. Artificial neural network models: An alternative approach for reliable aboveground pine tree biomass prediction. Forest Science, 2017, 63: 291-302
[15]	Silva JPM, Silva MLMD, Silva EFD, et al. Computational techniques applied to volume and biomass estimation of trees in Brazilian savanna. Journal of Environmental Management, 2019, 249: 3-12
[16]	Vahedi A. Artificial neural network application in comparison with modeling allometric equations for predicting above-ground biomass in the Hyrcanian mixed-beech forests of Iran. Biomass and Bioenergy, 2016, 88: 66-76
[17]	徐奇刚, 雷相东, 国红, 等. 基于多层感知机的长白落叶松人工林林分生物量模型. 北京林业大学学报, 2019, 41(5): 97-107
[18]	Widagdo F, Li F, Zhang L, et al. Aggregated biomass model systems and carbon concentration variations for tree carbon quantification of natural Mongolian oak in Northeast China. Forests, 2020, 11: 397
[19]	Dong L, Zhang L, Li F. A three-step proportional weighting system of nonlinear biomass equations. Forest Science, 2015, 61: 35-45
[20]	Dong L, Zhang L, Li F. Additive biomass equations based on different dendrometric variables for two dominant species (Larix gmelini Rupr. and Betula platyphylla Suk.) in natural forests in the Eastern Daxing'an Mountains, Northeast China. Forests, 2018, 9: 261
[21]	Dong L, Widagdo F, Xie L, et al. Biomass and volume modeling along with carbon concentration variations of short-rotation poplar plantations. Forests, 2020, 11: 780
[22]	刘秀红, 姜春前, 徐睿, 等. 相容性单木生物量模型估计方法的比较——以青冈栎为例. 林业科学, 2020, 56(9): 164-173
[23]	欧强新, 李海奎, 雷相东, 等. 基于清查数据的福建省马尾松生物量转换和扩展因子估算差异解析——3种集成学习决策树模型的比较. 应用生态学报, 2018, 29(6): 2007-2016
[24]	辛士冬, 严云仙, 姜立春. 基于不同可加性方法的黑龙江省红松人工林林分生物量模型. 应用生态学报, 2020, 31(10): 3322-3330
[25]	Widagdo F, Li F, Zhang L, et al. Aggregated biomass model systems and carbon concentration variations for tree carbon quantification of natural Mongolian oak in Northeast China. Forests, 2020, 11: 397