尼罗罗非鱼(Oreochromis niloticus)不同月龄性状的主成分与判别分析

采用主成分与判别分析的方法,研究尼罗罗非鱼形态性状的增长规律,并判定其最佳生长季节体格与月龄的关系。选择2—5月龄尼罗罗非鱼各100尾,测量体长、头长、躯干长、体高、尾柄长、尾柄高、体宽和体重共8个性状,并对其进行主成分与判别分析。主成分分析结果表明:尼罗罗非鱼各月龄的体型特征参数间均有不同程度的正相关(P<0.05),体重与体长、体高的相关性最显著;不同月龄尼罗罗非鱼性状的主成分有所不同,主成分1:2—5月龄均相同为增重因子;主成分2:2—4月龄均为尾柄因子,而5月龄是躯干因子;主成分3:2月龄是头部因子,3—4月龄是躯干因子,5月龄是尾柄因子。通过建立判别式来判断错过最佳生长季节尼罗罗非鱼的体格与大小相符的月龄,判别结果表明,总的判别准确率为99.25%,其中2—4月龄尼罗罗非鱼的判别准确率为100%。     

日本囊对虾(Marsupenaeus japonicus)3个野生种群和1个养殖种群的形态差异与判别分析

采用3种多元分析方法,对3个野生种群和1个养殖种群的13个形态比例参数进行比较研究。聚类分析结果表明,广东湛江和海南临高的种群形态最接近,福建厦门种群的趋异程度最大。主成分分析构建了6个主成分,其贡献率分别为:第1主成分23.084%,第2主成分17.976%,第3主成分15.321%,第4主成分13.84%,第5主成分12.707%,第6主成分7.770%,累积贡献率为90.699%。逐步判别选入6个贡献率较大的性状进行判别分析,F检验结果显示,4个种群形态差异显著(P0.01);同时建立了4个种群的判别函数,其判别准确率P1、P2均为100%。综合判别率为100%。3种多元分析结果均认为,4个种群日本囊对虾在形态上已产生一定程度的差异,且集中表现在体重、头胸甲的性状上。     

初夏亚洲季风区环流低频振荡与长江下游持续暴雨

利用主成份分析和非整数波功率谱分析研究了1991年初夏亚洲季风区逐日500 hPa环流的时空分布特征。结果表明，主成份分析得到的前5个特征向量的空间分布与各个季风系统的活动有关，其时间系数存在显著的12—22天准周期振荡和28—31，43—65天的低频振荡周期，它们与长江下游暴雨形成有密切联系。当第一主成份从谷点上升且第二主成份稳定地增大（减小）时，长江下游出现持续暴雨。亚洲副热带海洋加热异常和海陆热力差异产生的不均匀加热分布激发的各种低频波及其相互作用导致向热带外能量频散的异常，是引起长江下游持续暴雨的     

对我国国土规划几个主要问题的重新认识

本文基于对我国国土规划近十年实践中所反映的问题的分析,阐述了应改变国土规划的基本性质、以空间结构布局为其核心内容,以及国土规划一般不宜在市域或地区及以下地域层次开展的观点,进而就国土规划特性新的内涵、国土规划的内容重点等问题进行了初步探讨。     

我国夏季气温、降水场的时空特征分析

本文用主分量及转动主分量分析方法对我国夏季气温及降水量场(1951—1985年期间)的时空特征进行了研究。从所提取的空间模式中发现气温的解释方差较降水的大,且具有较好的均匀性;从对应的时间分量分析发现气温与降水均具有2—3年的主要变化周期,与东亚大气环流的关系分析中以气温表现较为密切。 对气温和降水场时空特征的稳定性分析表明,无论在持续性、周期性及与东亚大气环流的关系上,进入70年代后均有较明显的变化。比较表明气温场的稳定性较降水为好。 气温场与降水场相互关系分析发现它们有显著的反相关,表现显著的地区为长江中下游、华南及华北等。     

主应变比异常理论在青海祁连地震预报中的应用

本文利用主应变比异常理论分析了祁连建站以来地应变观测资料，认为肃南──祁连断裂段上及附近发生的地震是由于构造应力场的变化，在引张应力状态下，正断层或走滑断层易于活动而导致的。     

多剩磁成分的主成分法分离及其应用

岩石的多剩磁成分的分离是古地磁研究的重要环节之一.Kirschvink（1980）提出的主成分分析法（PCA）能有效地进行多剩磁成分的方向分离.作者认为,最大角偏差（MAD）值的选择应视各磁成分的剩磁谱组合特征而定,剩磁相似度的提出,可以帮助确定MAD值选取的最佳值.在主成分法分离的剩磁方向的基础上,作者提出了一种多剩磁成分的强度分离方法,籍此可以进一步研究各分离磁组分的剩磁谱特征及其成因,并将以上方法应用于新疆柯坪地区晚古生代地层的剩磁分离.     

Statistical empirical index of chemical weathering in igneous rocks: A new tool for evaluating the degree of weathering

Chemical weathering indices are useful tools in characterizing weathering profiles and determining the extent of weathering. However, the predictive performance of the conventional indices is critically dependent on the composition of the unweathered parent rock. To overcome this limitation, the present paper introduces an alternative statistical empirical index of chemical weathering that is extracted by the principal component analysis (PCA) of a large dataset derived from unweathered igneous rocks and their weathering profiles. The PCA analysis yields two principal components (PC1 and PC2), which capture 39.23% and 35.17% of total variability, respectively. The extent of weathering is reflected by variation along PC1, primarily due to the loss of Na₂O and CaO during weathering. In contrast, PC2 is the direction along which the projections of unweathered felsic, intermediate and mafic igneous rocks appear to be best discriminated; therefore, PC1 and PC2 represent independent latent variables that correspond to the extent of weathering and the chemistry of the unweathered parent rock. Subsequently, PC1 and PC2 were then mapped onto a ternary diagram (MFW diagram). The M and F vertices characterize mafic and felsic rock source, respectively, while the W vertex identifies the degree of weathering of these sources, independent of the chemistry of the unweathered parent rock.

The W index has a number of significant properties that are not found in conventional weathering indices. First, the W index is sensitive to chemical changes that occur during weathering because it is based on eight major oxides, whereas most conventional indices are defined by between two and four oxides. Second, the W index provides robust results even for highly weathered sesquioxide-rich samples. Third, the W index is applicable to a wide range of felsic, intermediate and mafic igneous rock types. Finally, the MFW diagram is expected to facilitate provenance analysis of sedimentary rocks by identifying their weathering trends and thereby enabling a backward estimate of the composition of the unweathered source rock.     

自由网平差方法及统计特性新探

本文结合参数平差和主成分估计理论,导出了误差方程中含多重共线性时未知参数的求解公式,并以定理的形式,证明了主成分估计的解是极小范数解。由此,将主成分估计推广到秩亏网平差中,同时导出了未知参数估值之协因数阵的计算公式,同时,证明了自由网平差的传统特性。     

COMMENTS ON THE NIPALS ALGORITHM

The Non-linear lterative Partial Least Squares(NIPALS)algorithm is used in principal componentanalysis to decompose a data matrix into score vectors and eigenvectors(loading vectors)plus a residualmatrix.N1PALS starts with some guessed starting vector.The principal components obtained by NIPALSdepends on the starting vector;the first principal component could not always be computed.Wold hassuggested a starting vector for NIPALS,but we have found that even if this starting vector is used,thefirst principal component cannot be obtained in all cases.The reason why such a situation occurs isexplained by the power method.A simple modification of the original NIPALS procedure to avoid gettingsmaller eigenvalues is presented.