基于Lg波识别判据的特性分析

Lg波是检测、识别地下核爆炸的重要震相.通常情况下,地下核爆炸的低频段Lg波能量相对较强,基于Lg波的识别判据在高频段表现较好.然而在低频范围内却表现出难以用球对称爆炸源模式解释的现象,如Lg/Pg判据在1 Hz附近失效等现象.本文利用理论地震图方法,分析了伴随层裂过程产生的重要辅助源——CLVD源对Lg波低频成分的调...     

A Study of 2MASS Low Surface Brightness Galaxies

A study is being made of a sample of about 4,000 Low Surface Brightness (LSB) galaxies detected in the 2MASS all-sky near-infrared survey. These objects have similar central K-band surface brighnesses as optically selected LSBs. About 1,000 of them were observed in HI, and one third of these were detected. BVRI CCD photometry was obtained of 65 objects. These multi-wavelength data provide us with important information on a large, homogenous sample of LSB galaxies (e.g., colors, gas content). They will be used to construct models of the star formation histories and evolution of LSBs, and for a study of their Tully-Fisher (TF) relation, particularly to search for deviations from the ‘classical’ TF relation. This revised version was published online in September 2006 with corrections to the Cover Date.     

Discriminant analysis of a Cretaceous foraminifer using shrunken estimators

Canonical variate analysis of living and fossil organisms, based on morphological characters, can sometimes greatly distort the biological interpretation (i.e., reification) of the coefficients of the eigenvectors forming the canonical variates through the inclusion of redundant within-group directions. Instability is associated with the smallest eigenvalues, particularly if these do not greatly differ from zero. In the present study of 46 borehole samples of the Late Cretaceous foraminifer Afrobolivina afraReyment, stability of the canonical variate coefficients is attained by removal of a near redundant direction of within-group variation. This leads to improved interpretability of the morphometric relationships in this species. on leave from Division of Mathematics and Statistics, CSIRO, Wembley, 6014, Western Australia.     

Detecting Holocene movements of the woodland–steppe ecotone in northern China using discriminant analysis

Discriminant functions link vegetation zones and surface pollen along the woodland–steppe ecotone in northern China. These links were utilised to reconstruct palaeovegetation using fossil pollen data from four lake sediment sites within the ecotone. Holocene movements of the vegetation zones were derived from the above results and were interpreted as the results of climatic, topographic and edaphic factors. This study shows that climatic change was the driving factor for ecotonal movement, but that the positions of vegetational zones were nevertheless controlled by topographical and edaphic conditions, and consequently boundaries between different vegetational zones responded individualistically. Copyright © 2001 John Wiley & Sons, Ltd.     

基于PCA-SVM算法对稀土元素与稀土判别指标耦合数据集的铀矿床分类

不同类型铀矿床的沥青铀矿/晶质铀矿具有不同的稀土元素组成,其组成可作为判别铀矿床类型的重要指标。采用基于Python语言的主成分分析（principal component analysis,PCA）与支持向量机（support vector machines,SVM）结合的分类模型,对收集到的全球已知6种类型铀矿床的216组沥青铀矿/晶质铀矿稀土元素数据进行研究。以216组数据为训练集,通过数据清洗、特征缩放、PCA特征提取、网格搜索和交叉验证参数寻优构建SVM分类模型,对24组同变质型胡家峪晶质铀矿进行智能识别。研究结果显示：仅使用稀土元素的14维训练集最优模型判定胡家峪晶质铀矿类型的测试准确率为0.4%;由稀土元素、稀土总量、轻重稀土比、铕异常组成的17维训练集最优模型的测试准确率为75.0%,较14维训练集提高74.6%,模型泛化能力强;而通过传统稀土元素配分曲线、w（ΣREE）-（LREE/HREE）_N图解不能有效判定胡家峪晶质铀矿类型。本次研究表明,PCA-SVM算法对增有传统稀土判别指标数据集进行挖掘可有效厘定铀氧化物成因类型,效果明显优于单纯的稀土元素数据集以及传统的稀土配分曲线、w（ΣREE）-（LREE/HREE）_N图解。     

大气电场的FFT频谱分析及雷暴预报研究

采用傅里叶变换方法将时域数据转换到频域,针对首次地闪发生(简称“雷暴天气”)和首次地闪不会发生(简称“非雷暴天气”)的大气电场强度数据做了幅度谱分析.研究发现,首次地闪前0.5h的大气电场数据与非雷暴天气中任意时段0.5h的大气电场数据的幅度谱分布具有明显差异,两类天气的幅度谱分布有一定规律可循.在此基础上对样本整理分类,将两类天气的平均幅度谱视为判别标准,利用欧氏距离判别法,对63个未知天气样本进行判别分类和效果检验,结果表明,此方法可供短时临近的首次地闪预警工作中予以参考.     

Discriminant analysis on land grading

This paper proposes the discriminant analysis on land grading after analyzing the common methods and discussing the Fisher's discriminant in detail. Actually this method deduces the dimension from multi to single, thus it makes the feature vectors inn-dimension change to a scalar, and use this scalar to classify samples. This paper illustrates the result by giving an example of the residential land grading by the discriminant analysis.     

Using hyperspectral reflectance data to assess biocontrol damage of giant salvinia

Field hyperspectral reflectance data were studied at 50 wavebands (10-nm bandwidth) over the 400- to 900-nm spectral range to determine their potential for distinguishing among giant salvinia (Salvinia molesta Mitchell) plants subjected to four population levels of salvinia weevils (Cyrtobagous salviniae Calder and Sands) to develop feeding damage to the plants. The four populations included a control with no insects and those with low, medium and high insect populations. The plants were studied in two experiments on each of two dates: 14 October 2010 and 21 July 2011. Two procedures were used to determine the optimum bands for discriminating among treatments: least significant difference (LSD) and stepwise discriminant analysis. The LSD comparison test results for both October and July experiments showed that generally the best bands for separating among treatments occurred in the green (505–595 nm), red (605–635 nm), red-near-infrared (NIR; 695–745 nm) edge and NIR (755–895 nm) regions where three to four treatments could be distinguished. Stepwise discriminant analysis identified four bands in the green, red and red-NIR edge to be significant to discriminate among the four treatments in Experiment 1 in October. For Experiment 2 in October, discriminant analysis identified five bands in the blue, green, red and NIR regions to be significant for distinguishing among the treatments. In Experiment 1 in July, five bands in the blue, green, red-NIR edge and NIR regions were found to be significant to discriminate among the treatments. For Experiment 2 in July, discriminant analysis identified four bands in the blue, green and red-NIR edge regions to be significant to discriminate among the treatments.     

基于Fisher判别分析法的岩溶塌陷预测

为了更准确地预测岩溶塌陷,基于Fisher判别分析法,首先选择内聚力、内摩擦角、覆盖层厚度、覆盖层厚度减高水位埋深、覆盖层厚度减低水位埋深5个最主要的影响因素作为判别因子;接着以桂林市20组岩溶塌陷案例为学习样本进行计算,建立相应的Fisher线性判别函数;然后利用回代估计法对上述20纽样本的预测结果进行逐一检验,其正确率为100％;随后将上述预测结果与前人运用逐步判别、神经网络判别等方法得到的结果进行比较分析,认为Fisher判别分析法具有不需进行模型和参数的选择以及不受人为因素的影响等优点;最后将建立的Fisher线性判别函数用于另外的10组实际案例进行预测分析,预测结果与实际情况完全吻合。结果表明：用Fisher判别分析法进行岩溶塌陷预测,简易方便,正确率高,实用性好;该方法是岩溶塌陷预测的一种新方法。     

水库诱发地震震级预测的统计研究

于收集到的全球102例水库及已发地震的资料,应用隶属函数方法综合分析了水库基本属性、震中区岩性、库坝区基本烈度和震中区断层类型等与水库诱发地震之间的关系,从统计学角度给出了水库诱发最大地震震级的判定方法．通过回溯检验和费舍尔判别检验给出了预测震级的相对误差和正确识别率,总体预测效果较好,可为将来水库的设防和最大地震震级的判定提供统计学上的依据．