Development of a linear mixed-effects individual-tree basal area increment model for masson pine in Hunan Province,South-central China

ABSTRACT

An individual-tree basal area increment model was developed for masson pine based on 26276 observations of 13,138 trees in 987 sample plots from the 7th (2004), 8th (2009), and 9th (2014) Chinese National Forest Inventory in Hunan Province, South-central China. The model was built using a linear mixed-effects approach with sample plots included as random effects since the data have a hierarchical stochastic structure and biased estimates of the standard error of parameter estimates could be a consequence of applying ordinary least square (OLS) for regression. In addition, within-plot heteroscedasticity and autocorrelation were also considered. The final mixed-effects model was determined according to the Akaike information criterion (AIC), Bayesian information criterion (BIC), log-likelihood (Loglik), and the likelihoodratio test (LRT). The results revealed that initial diameter (DBH), the sum of the basal area (m²/ha) in trees with DBHs larger than the DBH of the subject tree (BAL), number of trees per hectare (NT), and elevation (EL) had a significant impact on individual-tree basal area increment. The mixed-effects model performed much better than the basic model produced using OLS. Additionally, the variance structure of the model errors was successfully modeled using the power function. However, the autocorrelation structures were not defined because there was no autocorrelation amongst the data. It is believed that the final model will contribute to the scientific management of the masson pine.     

Interaction analysis between slenderness ratio and resin content on mechanical properties of particleboard

The interaction between particle size and resin content is one of the most important structural parameters that can influence the accuracy of predictions about wood-composite properties. We developed three kinds of equation (linear, quadratic, and exponential) for each mechanical property of particleboard based on slenderness ratio and resin content at a constant density (0.7g cm -3 ). Results from SHAZAM software (version 9) suggested that the quadratic function was not significant, but the linear and exponential functions were significant. The interaction between particle size and resin content was analyzed by Maple 9 software. The results indicated that an exponential function can better describe the simultaneous effect of slenderness and resin content than a linear equation. Under constant resin content, particles with higher slenderness ratios increased more in modulus of rupture (MOR) and modulus of elasticity (MOE) than did particles with lower slenderness ratios. Edge withdrawal resistance (SWRe) values did not increase with increasing slenderness ratio.     

基于SVM的CFRW导电性建模预测研究

针对碳纤维增强木质复合材料（CFRW）导电性具有非线性的特点,采用基于支持向量回归机（SVM）的机器学习方法建立碳纤维增强木质复合材料导电性预测模型。结果表明,所建模型的预测精度高、泛化能力强.     

不同密度沙枣能源林生物量估测模型研究

沙枣是我国新兴的能源树种,为了简单、方便、准确地估算其生物量,笔者以地径、冠幅、树高作为自变量,利用数理统计方法筛选出不同密度下地上生物量的最佳估测模型。结果表明：不同密度的沙枣能源林生物量估测中需要各易测因子适当组合才能获得最高的相关性,拟合出最为理想的回归模型。     

利用MODIS数据反演森林生物量的方法研究

以东北三省为试验区,利用MOD IS时间序列上的植被指数NDVI,结合森林资源清查样地数据测算的林分生物量,以及年均降水量、年均蒸散量、年均湿润度、年均积温、经度、纬度等地理环境因子,采用多元线性回归的方法构建森林生物量遥感估测模型。通过样本检验,该模型对东北三省的森林生物量密度估测精度较高,与检验样本的平均每公顷生物量平均相差1.30%;通过样本定位检验,平均相差21.87%。该模型适宜在宏观尺度上估测森林生物量。     

基于激光雷达数据的东北林区航空林分材积表编制

利用东北林区云冷杉林、落叶松林、樟子松林、红松林、栎树林、桦树林、杨树林、榆树林、椴树林和水胡黄林10种森林类型的1947个样地的激光雷达数据和地面实测蓄积量数据,首先通过多元线性回归和非线性回归方法,分别建立基于机载激光雷达数据的森林蓄积量回归估计模型,并通过对比分析,确定统一形式的基础回归模型;然后利用哑变量建模方法,建立基于不同森林类型参数和相同激光雷达变量的蓄积量模型。结果表明,研究建立的10种森林类型的线性蓄积量回归模型的解释变量个数在2~7之间,确定系数在0.460~0.858之间;非线性蓄积量回归模型的解释变量个数在2~4之间,确定系数在0.461~0.846之间。基于点云平均高度和平均强度建立的10种森林类型的二元蓄积量模型(研究称之为标准模型),其确定系数在0.440~0.815之间,平均预估误差在2.88%~4.42%之间,平均百分标准误差在16.76%~25.52%之间,预估精度基本达到森林资源规划设计调查技术规定要求。依据研究建立的10种森林类型的蓄积量模型,可以编制基于激光雷达数据的航空林分材积表,在森林资源调查实践中推广应用。     

玉溪市林分植被覆盖变化及驱动力分析

林分植被覆盖在很大程度上能够反映林分生态系统的健康与否,其时空变化是林分生态系统发展方向的判断依据。对玉溪市2010—2019年林分植被覆盖的时空格局变化进行了分析,并利用单因素和多元回归分析方法分别建立回归模型探索了林分植被覆盖改善与退化的驱动因素。结果表明:林分植被覆盖整体呈向好趋势,各驱动因素与斜率均不是简单线性关系;道路和居民区与林分植被覆盖改善速度呈负相关、与退化速度呈正相关;DEM、坡度、大型水体、公益林管护与林分植被覆盖改善速度呈正相关、与退化速度呈负相关;坡向与林分植被覆盖改善速度呈正相关,地形起伏度与林分植被覆盖退化速度呈负相关。林分植被覆盖改善与退化的驱动力不尽相同,影响机制也存在差异,这为更好制定生态保护相关政策提供了有力的科学依据。     

景谷县森林蓄积量遥感估测及其动态变化分析

以云南省景谷县为研究区,基于Landsat5和Landsat8遥感数据,利用2012年和2017年两期全国森林资源连续清查实地调查数据,建立多元逐步回归和随机森林模型对景谷县森林蓄积量进行遥感估测对比研究.结果显示:综合建模精度来看,随机森林法在相同样地数量条件下具有更好的估测效果;从估测结果与林地保护利用规划和林地变...     

Predicting moisture content and density distribution of Scots pine by microwave scanning of sawn timber

This study was carried out to investigate the possibility of calibrating a prediction model for the moisture content and density distribution of Scots pine (Pinus sylvestris) using microwave sensors. The material was initially of green moisture content and was thereafter dried in several steps to zero moisture content. At each step, all the pieces were weighed, scanned with a microwave sensor (Satimo 9,4GHz), and computed tomography (CT)-scanned with a medical CT scanner (Siemens Somatom AR.T.). The output variables from the microwave sensor were used as predictors, and CT images that correlated with known moisture content were used as response variables. Multivariate models to predict average moisture content and density were calibrated using the partial least squares (PLS) regression. The models for average moisture content and density were applied at the pixel level, and the distribution was visualized. The results show that it is possible to predict both moisture content distribution and density distribution with high accuracy using microwave sensors.     

Assessment of wood surface roughness: comparison of tactile roughness and three-dimensional parameters derived using a robust Gaussian regression filter

Japanese oak and Japanese beech were sanded by hand with abrasive papers of varying grit number. Two three-dimensional parameters selected to characterize their surface roughness – one parameter for the distribution of roughness-profile peaks and the other for the relative area of the roughness-profile peaks above the threshold height – were compared against tactile roughness. The parameters were obtained from roughness profiles as determined by a robust Gaussian regression filter (RGRF) using seven cutoffs. The RGRF filtering process was adjusted specifically for the evaluation of wood surface roughness. Except for a cutoff wavelength of 0.25mm, the RGRF lent itself well to the determination of roughness profiles. No distortion of roughness profiles occurred around deep valleys, and there was a good correlation between the parameters and tactile roughness.Part of this study was presented at the 2002 Kansai Branch Office lecture meeting of the Japan Society for Precision Engineering, Kyotanabe, August 2002