M变换在NOAA/AVHRR数据变化检测中的应用

将基于典型相关分析的正交变换的基础理论用于NOAA/AVHRR数据的多元变化检测 ,对具体实施步骤进行了深入的探讨 ,并对部分中间结果进行了分析。实验表明 ,该方法应用于NOAA/AVHRR数据的多元变化检测具有明显的优势 ,克服了传统方法存在的缺陷 ,具有良好的应用前景。     

敦煌绿洲地下水微量元素分布特征及其成因

采用多元统计和水文地球化学分析对中国西北地区敦煌绿洲地下水微量元素分布特征及其成因进行了研究.结果表明:地下水中大多数微量元素含量较低,水质总体良好.地下水微量元素具有明显的地带分布规律,从山前洪积扇扇顶补给地带,到中下游细土平原径流排泄区,多数金属元素含量增加,部分在中游出现极大值.蒸发浓缩作用和水岩相互作用控制了研...     

Analysis of multivariate point distributions and chemical grouping of rocks

Large sets of rock compositions can be grouped by the analysis of corresponding point distributions in multidimensional space, where the separate dimensions represent the chemical variables. The point density around each composition is estimated in a small, multivariate, rectangular interval. Establishment of the identity of specimens contained in the interval, and systematic comparison of the point densities surrounding each of them, leads to the recognition of density maxima, which represent the statistical modes of the rock types present in the set. The remaining specimens by the same operations can be assigned to the groups formed. Experimental results are given for two sets, one comprising a wide variety of metasedimentary and metavolcanic rocks from the ore-bearing Skellefte district and adjacent parts of V?sterbotten, Sweden, and the other representing a less variable population of basic rocks from Norrbotten, Sweden. Linear discriminant analysis and existing geological information indicate that the groups obtained are statistically and geologically valid.     

Regionalization of unit hydrograph parameters: 1. Comparison of regression analysis techniques

Hydrologic regionalization is a useful tool that allows for the transfer of hydrological information from gaged sites to ungaged sites. This study developed regional regression equations that relate the two parameters in Nash's IUH model to the basin characteristics for 42 major watersheds in Taiwan. In the process of developing the regional equations, different regression procedures including the conventional univariate regression, multivariate regression, and seemingly unrelated regression were used. Multivariate regression and seeming unrelated regression were applied because there exists a rather strong correlation between the Nash's IUH parameters. Furthermore, a validation study was conducted to examine the predictability of regional equations derived by different regression procedures. The study indicates that hydrologic regionalization involving several dependent variables should consider their correlations in the process of establishing the regional equations. The consideration of such correlation will enhance the predictability of resulting regional equations as compared with the ones from the conventional univariate regression procedure.     

Oligochaete distribution patterns in two German hardwater lakes of different trophic state

We investigated the effect of contrasting trophic conditions on the distribution and dominance patterns of Oligochaeta species assemblages in lake sediments and their relation to the environmental variables depth, season, and substrate. The study was performed on the highly eutrophic Lake Haussee and the oligotrophic Lake Stechlin, both hardwater lakes in the Baltic Lake District of Northern Germany. Quantitative monthly and seasonal sampling took place over one year at 14 representative sites, covering littoral and profundal sediments of both lakes. Between-lake differences in the profundal were clearcut with an absence of any zoobenthos in Lake Haussee and a peculiar meiobenthic species assemblage in Lake Stechlin (Collado et al. 1999). Between-lake differences in the littoral, however, were small and mainly attributable to a small number of species exclusive to Lake Stechlin and an overall higher abundance of oligochaetes, especially naidids, in Lake Haussee. Species-richest family in both lakes were Naididae; Tubificidae were dominant in Lake Stechlin; in Lake Haussee Tubificidae and Naididae were equally abundant. Most striking were diversity of habitat types and species distribution patterns in the littoral. Multivariate analysis (CA, CCA) showed that the distribution patterns of oligochaete species assemblages are significantly correlated with depth, season and substrate. Depth is the major factor when the whole water body is considered. When confining to the littoral, species abundance and distribution are strongly related to seasonality and substrate type. Naididae show, in general, maximal abundances in autumn and summer and a preference for plants, plant debris and soft sediments; Tubificidae are more abundant in spring and prefer mineral substrate. The relation between seasonality, substrate and food availability is discussed.     

湖泊水质预测模型的建立及其应用

根据东湖多年监测数据分析，建立了多元相关预测模型、基于灰色理论灰色系统的预测模型和质量平衡模型。为了提高灰色预测精度，利用残差对模型进行了相应的修正，并利用３种模型对影响东湖水质主要因子——经济、人口、ＢＯＤ、ＣＯＤ、ＴＮ、ＴＰ进行了２０００年的预测，提出了东湖污染防治对策     

Projective ordination by SIMCA: A dynamic strategy for cost-efficient environmental monitoring around offshore installations

A method for environmental monitoring using benthic species profiles as input is developed in this work. The method, referred to as projective ordination, utilises local principal component modelling (SIMCA) to obtain a cross-validated model which spans the natural variation in a region around offshore oil-producing installations. The borderline between regions with disturbed and non-disturbed species communities is subsequently decided from the residual distribution. This distribution is used to design an approximate F-test for assessing whether a community at a particular sampling location is disturbed or not. If so, the nature of the disturbance is determined by projecting the data on the PC model.Projective ordination utilises information from previous surveys to define the permissible variation in species communities, i.e. the limit of the natural variation. In addition, the method is dynamic, in a sense that the sampling locations may vary from survey to survey.Furthermore, our analysis shows that the customary number of replicate samples per station can be reduced. Modelling with only four randomly chosen replicates out of the complete set of five for each sampling location, does not affect the model significantly. On the other hand, taking only 3 replicates into account leads to significant divergences.A model based on the 1990 and 1993 surveys at the Statfjord field is presented as an example of the technique.