Machine learning methods for basal area prediction of Fagus orientalis Lipsky stands based on national forest inventory

IF 2.1 3区农林科学 Q2 FORESTRY

Trees Pub Date : 2025-03-19 DOI:10.1007/s00468-025-02616-y

Seyedeh Fatemeh Hosseini, Hamid Jalilvand, Asghar Fallah, Hamed Asadi, Mahya Tafazoli

{"title":"Machine learning methods for basal area prediction of Fagus orientalis Lipsky stands based on national forest inventory","authors":"Seyedeh Fatemeh Hosseini, Hamid Jalilvand, Asghar Fallah, Hamed Asadi, Mahya Tafazoli","doi":"10.1007/s00468-025-02616-y","DOIUrl":null,"url":null,"abstract":"<div><h3>Key Message</h3><p>Machine learning models accurately predict <i>F. orientalis</i> stand basal area in the Hyrcanian forest using environmental variables, with the RF model performing best. Elevation is the most important predictor.</p><h3>Abstract</h3><p>Accurate prediction of tree basal area (BA) as an important forest stand structural characteristic is essential for sustainable forest management. The aim of this study was to use four machine learning methods, including generalized linear model (GLM), k-nearest neighbors (KNN), support vector machine (SVM), and random forest (RF), to predict and assess the stand BA of <i>Fagus orientalis</i> Lipsky using national forest inventory data and a comprehensive set environmental variables. Modeling was performed using a 10-fold spatial cross-validation technique to counteract the effect of spatial auto-correlation in predictor and response data, as well as to reduce the dependency between training and test data. The RF model outperformed the others by having the best match between measured and predicted stand BA values, with the highest squared correlation coefficient (<span>\\({R}_{\\text{Train}}^{2}\\)</span> = 0.77; <span>\\({R}_{\\text{Test}}^{2}\\)</span>= 0.76) and the lowest root mean square error (<span>\\({\\text{RMSE}}_{\\text{Train}}\\)</span>= 2.70; <span>\\({\\text{RMSE}}_{\\text{Test}}\\)</span>= 2.90) and mean absolute error (<span>\\({\\text{MAE}}_{\\text{Train}}\\)</span>= 1.74; <span>\\({\\text{MAE}}_{\\text{Test}}\\)</span>=1.76). Among all investigated variables, elevation showed the highest correlation with stand BA of <i>F. orientalis</i> in the Hyrcanian forest. The relation was positive and restricted to the range of approximately 700 to 1200 m. The RF and GLM models indicated the bulk density as the second-most important variable, whereas the SVM and kNN models indicated the air temperature as the second important variable. In general, this research identifies key variables influencing the stand BA of <i>F. orientalis</i>, providing valuable insights for forest management and conservation efforts. These findings contribute to a better understanding of forest dynamics in the Hyrcanian region and can inform targeted management strategies.</p></div>","PeriodicalId":805,"journal":{"name":"Trees","volume":"39 2","pages":""},"PeriodicalIF":2.1000,"publicationDate":"2025-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Trees","FirstCategoryId":"2","ListUrlMain":"https://link.springer.com/article/10.1007/s00468-025-02616-y","RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"FORESTRY","Score":null,"Total":0}

引用次数: 0

Key Message

Machine learning models accurately predict F. orientalis stand basal area in the Hyrcanian forest using environmental variables, with the RF model performing best. Elevation is the most important predictor.

Abstract

Accurate prediction of tree basal area (BA) as an important forest stand structural characteristic is essential for sustainable forest management. The aim of this study was to use four machine learning methods, including generalized linear model (GLM), k-nearest neighbors (KNN), support vector machine (SVM), and random forest (RF), to predict and assess the stand BA of Fagus orientalis Lipsky using national forest inventory data and a comprehensive set environmental variables. Modeling was performed using a 10-fold spatial cross-validation technique to counteract the effect of spatial auto-correlation in predictor and response data, as well as to reduce the dependency between training and test data. The RF model outperformed the others by having the best match between measured and predicted stand BA values, with the highest squared correlation coefficient (\({R}_{\text{Train}}^{2}\) = 0.77; \({R}_{\text{Test}}^{2}\)= 0.76) and the lowest root mean square error (\({\text{RMSE}}_{\text{Train}}\)= 2.70; \({\text{RMSE}}_{\text{Test}}\)= 2.90) and mean absolute error (\({\text{MAE}}_{\text{Train}}\)= 1.74; \({\text{MAE}}_{\text{Test}}\)=1.76). Among all investigated variables, elevation showed the highest correlation with stand BA of F. orientalis in the Hyrcanian forest. The relation was positive and restricted to the range of approximately 700 to 1200 m. The RF and GLM models indicated the bulk density as the second-most important variable, whereas the SVM and kNN models indicated the air temperature as the second important variable. In general, this research identifies key variables influencing the stand BA of F. orientalis, providing valuable insights for forest management and conservation efforts. These findings contribute to a better understanding of forest dynamics in the Hyrcanian region and can inform targeted management strategies.

查看原文本刊更多论文

基于国家森林清查的东方山毛榉林分基底面积预测的机器学习方法

关键信息：利用环境变量，机器学习模型能够准确预测海卡尼亚森林东方林林分基面积，其中RF模型表现最好。海拔是最重要的预测因子。摘要树基面积（BA）作为林分结构的重要特征，其准确预测对森林可持续经营具有重要意义。本研究旨在利用广义线性模型（GLM）、k近邻模型（KNN）、支持向量机（SVM）和随机森林（RF） 4种机器学习方法，在国家森林清查数据和综合环境变量的基础上，预测和评估东方柴（Fagus orientalis Lipsky）林分BA。使用10倍空间交叉验证技术进行建模，以抵消预测和响应数据中空间自相关的影响，并减少训练和测试数据之间的依赖性。RF模型在林分BA测量值与预测值之间的拟合度最高，相关系数平方最高(\({R}_{\text{Train}}^{2}\) = 0.77；\({R}_{\text{Test}}^{2}\) = 0.76)，最小均方根误差(\({\text{RMSE}}_{\text{Train}}\) = 2.70；\({\text{RMSE}}_{\text{Test}}\) = 2.90)，平均绝对误差(\({\text{MAE}}_{\text{Train}}\) = 1.74；\({\text{MAE}}_{\text{Test}}\) =1.76)。在所有被调查的变量中，海拔与海卡尼亚森林东方木林分BA的相关性最高。这种关系是正的，并且限制在大约700 ～ 1200 m的范围内。RF和GLM模型表明体积密度是第二重要的变量，而SVM和kNN模型表明空气温度是第二重要的变量。总的来说，本研究确定了影响东方木林分BA的关键变量，为森林管理和保护工作提供了有价值的见解。这些发现有助于更好地了解海卡尼亚地区的森林动态，并为有针对性的管理战略提供信息。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Trees 农林科学-林学

CiteScore

4.50

自引率

4.30%

发文量

113

审稿时长

3.8 months

期刊介绍： Trees - Structure and Function publishes original articles on the physiology, biochemistry, functional anatomy, structure and ecology of trees and other woody plants. Also presented are articles concerned with pathology and technological problems, when they contribute to the basic understanding of structure and function of trees. In addition to original articles and short communications, the journal publishes reviews on selected topics concerning the structure and function of trees.