用于数值分类的聚类分析   总被引：24，自引：0，他引：24  

么枕生 《海洋湖沼通报》1994,(2):1-12

这里所谓聚类分析包括传统的聚类分析和作者提出的聚类统计检验分析。传统的聚类分析是常用的重要数值分类方法。但是，这个方法由于缺乏显著性检验，在国际上已长期公认是一个在多元统计中并不完整的方法。本文把聚类统计检验分析在其具有显著性检验的优越性上，和传统聚类分析作了对比分析。聚类统计检验分析区分有均匀聚类与中心聚类。前者用（１６）作为聚类判据，后者则用（２７），并且各用相应的公式（１７）－（２１）与（２  相似文献   

Toward a generic method for studying water renewal, with application to the epilimnion of Lake Tanganyika   总被引：3，自引：0，他引：3  

Olivier Gourgue  Eric Deleersnijder  Laurent White   《Estuarine, Coastal and Shelf Science》2007,74(4):628

We present a method, based on the concept of age and residence time, to study the water renewal in a semi-enclosed domain. We split the water of this domain into different water types. The initial water is the water initially present in the semi-enclosed domain. The renewing water is defined as the water entering the domain of interest. Several renewing water types may be considered depending on their origin. We present the equations for computing the age and the residence time of a certain water type. These timescales are of use to understand the rate at which the water renewal takes place. Computing these timescales can be achieved at an acceptable extra computer cost.The above-mentioned method is applied to study the renewal of epilimnion (i.e. the surface layer) water in Lake Tanganyika. We have built a finite element reduced-gravity model modified to take into account the water exchange between the epilimnion and the hypolimnion (i.e. the bottom layer), the water supply from precipitation and incoming rivers, and the water loss from evaporation and the only outgoing river. With our water renewal diagnoses, we show that the only significant process in the renewal of epilimnion water in Lake Tanganyika is the water exchange between the epilimnion and the hypolimnion, other phenomena being negligible.  相似文献   

东海区小黄鱼渔业生物学现状的分析   总被引：15，自引：1，他引：15  

林龙山  程家骅 《中国海洋大学学报(自然科学版)》2004,34(4):565-570

以2001年东海区渔业资源监测调查资料为依据，对小黄鱼的体长、体重、生长、死亡等渔业生物学特征及临界年龄等进行了分析研究。结果表明，目前东海区小黄鱼优势体长组为110～130mm，占41．34％；优势体重组为20～40g，占50．91％；总死亡系数为2．05，自然死亡系数为0．58，捕捞死亡系数为1．47，临界年龄为2．18龄。对照东海区小黄鱼生物学特征的历史状况分析，目前东海区小黄鱼个体小型化、低龄化、性成熟提早现象明显，生物学参数发生较大变化。建议在东海区现行资源状态下，小黄鱼的可捕标准应充分考虑其生物学现状加以科学调整。  相似文献   

两参量的海面阻力系数模式的探讨   总被引：4，自引：0，他引：4  

汪炳祥 《海洋与湖沼》1997,28(1):96-103

从风浪的能量平衡方程出发，引进若干风要素与波要素以及波要素之间的定性关系，经演算可导出海面阻力系数（Ｃｐ）或是风速（Ｕ）和波龄（β）或是Ｕ和波高（Ｈ）的函数，然后沿用最小二乘法，终将得出４组１２个回归方程。当β（或β）或Ｈ为某一给定值，惟有Ｕ为唯一参量时，所提各式均可简化为非线性方程：ＣＤ＝ａ＋ｂ，Ｕ＋ｃ．Ｕ＾２；式中ａ，ｂ和ｃ为三个经验系数，就所检验的例子而言，本文的结果与实际的符合前人的为好。  相似文献   

Population Structure and Secondary Production of the Mediterranean Octocoral Lophogorgia ceratophyta (L. 1758)     

M. Mistri 《Marine Ecology》1995,16(3):181-188

Abstract. The gorgonian Lophogorgia ceratophyta thrives in turbid water at Tinetto Rock, La Spezia Gulf, Ligurian Sea. Its age structure suggests that the population is in a steady state; nevertheless, it is difficult to discern whether the main factors controlling this population are biological or event-related. Secondary production was estimated by means of an increment-summation method and yielded 0.54 g ash-free dry weight (AFDW)*m^-1 a^-1. This estimate does not take into account biomass losses due to reproductive output and non-lethal predation. Standing stock biomass was 5.44 g AFDW e m^-2. The annual P/B ratio was 0.10 and turnover time about 10a  相似文献   

Extremely Shallow-water Morphobathymetric Surveys: The Valle Fattibello (Comacchio, Italy) Test Case     

Luca Gasperini 《Marine Geophysical Researches》2005,26(2-4):97-107

  相似文献   

盐田水体遥感分类方法研究   总被引：3，自引：1，他引：3  

王晶晶  张鹰  陶菲 《海洋技术学报》2005,24(1):67-71

以连云港台北盐场为研究区，介绍了监督分类法和神经网络分类法及其在盐田水体遥感分类中的具体应用。研究结果表明，用神经网络分类法进行遥感影像自动分类，其分类精度高，显示了其在遥感领域较为广阔的应用前景。  相似文献   

Determining the age of water and long-term transport timescale of the Chesapeake Bay   总被引：5，自引：0，他引：5  

Jian Shen  Harry V. Wang 《Estuarine, Coastal and Shelf Science》2007,74(4):585-598

The concept of age of water (AW) is applied to the Chesapeake Bay to investigate the long-term transport properties for dissolved substances. A real-time calibrated hydrodynamic Chesapeake Bay model in 3 Dimensions (CH3D), employing a boundary-fitted curvilinear grid, is used for the study. The long-term transport properties, represented by AW, are investigated under the conditions of low river inflow of 1995 and high river inflow of 1996, as well as for constant mean inflows. The influences of freshwater, density-induced circulation, and wind-induced transport on age distribution have been investigated. Model results show that river inflows, wind stress, and density-induced circulation play important roles in controlling the long-term transport in the Bay. The model results shows that it requires 120–300 days for a marked change in the characteristics of the pollutant source discharged into the Bay from the Susquehanna River to affect significantly the conditions near the mouth under different hydrodynamic conditions. An increase of river discharge results in increases of downstream residual current and gravitational circulation, and thus reduces AW. The density-induced circulation contributes to the transport substantially. The dissolved substances discharged into the Bay are transported out of the Bay more rapidly when the estuary becomes more stratified. Southeasterly and southwesterly winds have strong impacts on the transport compared to the northeasterly and northwesterly winds. The former increases lateral and vertical mixing significantly. Consequently, the gravitational circulation is reduced and the transport time is increased by 50%. The model results provide useful information for understanding the long-term transport processes in the Bay.  相似文献   

Age, length and reproductive parameters of harbour porpoises Phocoena phocoena (L.) from West Greenland     

Lockyer  C.; Heide-Jorgensen  M. P.; Jensen  J.; Kinze  C. C.; Sorensen  T. Buus 《ICES Journal of Marine Science》2001,58(1):154-162

  相似文献   

利用极化SAR数据进行土地覆盖分类研究     

余海坤  张永红  汪云甲 《海洋测绘》2006,26(3):34-38

基于相关矩阵特征向量的目标分解将地物回波复杂的散射过程分解成相互独立的三种单一散射分量:单向散射、双向散射和交叉散射,分别对应各自的目标相关矩阵。目标分解技术降低了散射回波之间的相关性,有利于分析地物散射机理,有助于提高分类精度。对荷兰F levoland地区全极化数据进行分解,经过试验和相关性分析,选用7种数据形成多参数数据组合,对其进行最大似然监督分类,同时进行常规三种极化加相位差的分类和基于复W ishart分布的最大似然分类,逐像元计算混淆矩阵,分析对比三种分类结果的精度,试验表明:相对于常规数据组合分类,基于复W ishart分布的监督分类可以小幅度提高分类精度,而利用目标分解得到多参数组合数据进行分类则有大幅度的提高。  相似文献