Stochastic gradient boosting classification trees for forest fuel types mapping through airborne laser scanning and IRS LISS-III imagery

This paper presents an application of Airborne Laser Scanning (ALS) data in conjunction with an IRS LISS-III image for mapping forest fuel types. For two study areas of 165 km² and 487 km² in Sicily (Italy), 16,761 plots of size 30-m × 30-m were distributed using a tessellation-based stratified sampling scheme. ALS metrics and spectral signatures from IRS extracted for each plot were used as predictors to classify forest fuel types observed and identified by photointerpretation and fieldwork. Following use of traditional parametric methods that produced unsatisfactory results, three non-parametric classification approaches were tested: (i) classification and regression tree (CART), (ii) the CART bagging method called Random Forests, and (iii) the CART bagging/boosting stochastic gradient boosting (SGB) approach. This contribution summarizes previous experiences using ALS data for estimating forest variables useful for fire management in general and for fuel type mapping, in particular. It summarizes characteristics of classification and regression trees, presents the pre-processing operation, the classification algorithms, and the achieved results. The results demonstrated superiority of the SGB method with overall accuracy of 84%. The most relevant ALS metric was canopy cover, defined as the percent of non-ground returns. Other relevant metrics included the spectral information from IRS and several other ALS metrics such as percentiles of the height distribution, the mean height of all returns, and the number of returns.     

Texture augmented detection of macrophyte species using decision trees

Image classification using multispectral sensors has shown good performance in detecting macrophytes at the species level. However, species level classification often does not utilize the texture information provided by high resolution images. This study investigated whether image texture provides useful vector(s) for the discrimination of monospecific stands of three floating macrophyte species in Quickbird imagery of the South Nation River. Semivariograms indicated that window sizes of 5 × 5 and 13 × 13 pixels were the most appropriate spatial scales for calculation of the grey level co-occurrence matrix and subsequent texture attributes from the multispectral and panchromatic bands. Of the 214 investigated vectors (13 Haralick texture attributes * 15 bands + 9 spectral bands + 10 transformations/indices), feature selection determined which combination of spectral and textural vectors had the greatest class separability based on the Mann–Whitney U-test and Jefferies–Matusita distance. While multispectral red and near infrared (NIR) performed satisfactorily, the addition of panchromatic-dissimilarity slightly improved class separability and the accuracy of a decision tree classifier (Kappa: red/NIR/panchromatic-dissimilarity – 93.2% versus red/NIR – 90.4%). Class separability improved by incorporating a second texture attribute, but resulted in a decrease in classification accuracy. The results suggest that incorporating image texture may be beneficial for separating stands with high spatial heterogeneity. However, the benefits may be limited and must be weighed against the increased complexity of the classifier.     

Hyperspectral remote sensing analysis of short rotation woody crops grown with controlled nutrient and irrigation treatments

Hyperspectral remote sensing research was conducted to document the biophysical and biochemical characteristics of controlled forest plots subjected to various nutrient and irrigation treatments. The experimental plots were located on the Savannah River Site near Aiken, SC. AISA hyperspectral imagery were analysed using three approaches, including: (1) normalized difference vegetation index based simple linear regression (NSLR), (2) partial least squares regression (PLSR) and (3) machine-learning regression trees (MLRT) to predict the biophysical and biochemical characteristics of the crops (leaf area index, stem biomass and five leaf nutrients concentrations). The calibration and cross-validation results were compared between the three techniques. The PLSR approach generally resulted in good predictive performance. The MLRT approach appeared to be a useful method to predict characteristics in a complex environment (i.e. many tree species and numerous fertilization and/or irrigation treatments) due to its powerful adaptability.     

黄土的湿陷性与击实试验指标关系研究

进行了黄土的击实试验,提出了击实率的概念,拓展了CART(classification and regression trees)决策树算法的功能。将其用于相关性挖掘,进行了原状黄土湿陷性与击实率等指标的相关性挖掘,结果表明,击实率与湿陷系数具有相关性,击实率与最优含水率状态的击实率呈显著负相关性。根据试验及分析结果提出黄土在击实过程中的变形特性可反映黄土的湿陷性,黄土的击实效果同其湿陷性具相关性。这一研究为黄土湿陷性的评价提出了一条新的思路,同时成果具有实用性。     

古火灾历史重建的研究进展

火是地球系统的重要组成部分,与气候、植被、生物地球化学循环和人类活动密切相关。火对全球气候和生态系统的影响已成为目前全球变化研究的一个热点。火灾发生后会在周围的环境中留下许多燃烧产物,如黑碳、木炭屑、多环芳烃、左旋葡萄糖等,它们广泛存在于海洋、湖泊、河流、土壤和陆地风成沉积物中;还会留下一些火灾痕迹,如树木火疤、土壤磁...     

A continental analysis of correlations between tree patterns in African savannas and human and environmental variables

This study analyses possible relationships between natural processes taking place in savannas and the tree patterns found in savannas. This can lead to new hypotheses about which processes are driving savanna physiognomy. To do so tree patterns were quantified for African savannas from historical aerial photographs applying frequently used landscape metrics. Also, additional data for these areas were collected to quantify the processes taking place at these locations. Correlations between tree pattern indices and explaining factors were analysed. We found a negative trend between tree cover and density of sheep and goats, but no relationship between tree cover and density of cattle, suggesting that small livestock have an effect on tree cover, but that larger livestock (or obligate grazers) do not. Also, a positive correlation between human population density and tree cover was found. Possible explanations for the found relations are discussed. Subsequent ways to analyse the latter correlation are discussed, and the potential of the presented historical database of aerial photographs is highlighted.     

Regular and irregular periodic orbits

We distinguish between regular orbits, that bifurcate from the main families of periodic orbits (those that exist also in the unperturbed case) and irregular periodic orbits, that are independent of the above. The genuine irregular families cannot be made to join the regular families by changing some parameters. We present evidence that all irregular families appear inside lobes formed by the asymptotic curves of the unstable periodic orbits. We study in particular a dynamical system of two degrees of freedom, that is symmetric with respect to the x-axis, and has also a triple resonance in its unperturbed form. The distribution of the periodic orbits (points on a Poincaré surface of section) shows some conspicuous lines composed of points of different multiplicities. The regular periodic orbits along these lines belong to Farey trees. But there are also lines composed mainly of irregular orbits. These are images of the x-axis in the map defined on the Poincaré surface of section. Higher order iterations of this map , close to the unstable triple periodic orbit, produce lines that are close to the asymptotic curves of this unstable orbit. The homoclinic tangle, formed by these asymptotic curves, contains many regular orbits, that were generated by bifurcation from the central orbit, but were trapped inside the tangle as the perturbation increased. We found some stable periodic orbits inside the homoclinic tangle, both regular and irregular. This proves that the homoclinic tangle is not completely chaotic, but contains gaps (islands of stability) filled with KAM curves.     

中国矮曲林的分布特征及生态意义

矮曲林是山地森林带在严酷的自然环境影响下发展演化而成的特殊植物群落,对中国山地植被和垂直带的研究具有重要的地理和生态意义。本文选择20世纪30年代以来在国内外文献中发表的52个矮曲林数据,结合Worldclim气候数据等,对中国矮曲林的类型和分布及其相应的气候条件进行了研究。结果表明：①中国的矮曲林主要包括山顶苔藓矮曲林、寒温性偃松矮曲林和岳桦矮曲林3类;其中,山顶苔藓矮曲林主要分布于秦巴山区以南的山地,而偃松矮曲林和岳桦矮曲林主要分布于东北地区。②山顶苔藓矮曲林主要由亚热带、热带地区的常绿阔叶林在山顶或山脊受山顶效应等的影响演化而来;而寒温性矮曲林主要由寒温性针叶林或针阔混交林在山顶或山脊条件下长期演化而来。山顶苔藓矮曲林自东向西随地势升高,其分布的海拔高度也逐渐升高;寒温性偃松矮曲林和岳桦矮曲林分布高度自南向北逐渐降低。③矮曲林分布在气候林线以下的山地,其气候条件足以支撑山地森林的发育,但由于山顶效应的存在(如强风、气候干冷或湿冷及土壤贫瘠等),只能发育矮曲林而非垂直地带性的山地森林。④中国东南部的很多山地高度小于3000 m,山顶效应明显,矮曲林广泛发育,造成了假林线的普遍存在。     

基于NDVI时间序列影像的张掖市农作物种植结构提取

HJ卫星(环境与灾害监测预报小卫星)数据具有较高的时间和空间分辨率,可以有效地进行农作物种类识别,利用其时间序列数据进行作物信息提取具有很大优势。以甘肃省张掖市为研究区,利用2012年4月~2012年10月15个时相的HJ卫星CCD(电荷耦合元件,charge-coupled device)数据,通过分析玉米、油菜、大麦和小麦农作物在整个生长阶段的时间序列NDVI(归一化植被指数,normalized difference vegetation index)变化特点,提取四类农作物NDVI随时间变化的特征曲线,利用决策树分类方法,对2012年张掖市农作物进行分类,并以外业调查得到的实际地面验证点对分类结果进行验证。验证结果表明,可以有效地识别玉米、油菜、大麦和小麦4类农作物,制图精度分别为94.74%、82.35%、78.95%和90%;总体分类精度达到86.43%。而后提取了张掖市2014年农作物种植面积和分布格局,通过对比分析发现,2012~2014年,张掖市玉米、油菜的种植面积有不同程度的增加,大麦和小麦的种植面积各有一定程度的较少。可以为张掖市主要作物的种植结构调整提供科学依据,还可以为黑河流域农作物信息提取研究提供参考。     

A Model Test on the Behavior of a Static Drill Rooted Nodular Pile Under Compression

A static drill rooted nodular pile is a new type of composite pile foundation with high bearing capacity, and mud emissions can be largely reduced using the static drill rooted method. This report presents a model test on the behavior of this composite pile in a test box. The load-displacement response, axial force, skin friction, and mobilized base load are discussed in the report; in particular, the force in the cemented soil was investigated based on the measured data. Moreover, the finite element software ABAQUS was used to help investigate this behavior more thoroughly. It was determined that the function of the cemented soil around the pile shaft was different from that at the enlarged pile base; the stress in the cemented soil around the shaft increased suddenly when nearing the pile base; the ultimate skin friction obtained in the model test was larger than that estimated in the field test; and the relative displacement between the precast nodular pile and the cemented soil could be ignored during the loading process, which corresponded to the result of the field test and demonstrated that the nodular pile and cemented soil act as one entity during the loading process.