Assessment of shallow landslide susceptibility by means of multivariate statistical techniques

Several multivariate statistical analyses have been performed to identify the most influential geological and geomorphological parameters on shallow landsliding and to quantify their relative contribution. A data set was first prepared including more than 30 attributes of 230 failed and unfailed slopes. The performance of principal component analysis, t‐test and one‐way test, allowed a preliminary selection of the most significant variables, which were used as input variables for the discriminant analysis. The function obtained has classified successfully 88·5 per cent of the overall slope population and 95·6 per cent of the failed slopes. Slope gradient, watershed area and land‐use appeared as the most powerful discriminant factors. A landslide susceptibility map, based on the scores of the discriminant function, has been prepared for Ensija range in the Eastern Pyrenees. An index of relative landslide density shows that the results of the map are consistent. Copyright © 2001 John Wiley & Sons, Ltd.     

State of the Practice Worldwide: HCA-PCA-EWM Discrimination Model of Water Inrush Source in Mines

The hydrogeological conditions in the North China type coal mine area are complex and diverse, and water gushing accidents are frequent. Quickly and accurately identifying the source of water in these incidents is necessary to reduce the harmful effects of mine floods. Taking the Pingdingshan mining area as an example, what chemical composition data from each of the main aquifers was collected and a mine gushing water source model, based on Hierarchical Cluster Analysis—Principle Component Analysis—Entropy Weighted Membership (HCA-PCA-EWM), was established. First, typical water sample data from an aquifer were selected by systematic cluster analysis. Then, the remaining water samples were randomly divided into training and prediction samples. Principal component analysis was used to reduce the dimension of water chemical components in training samples. Finally, the discriminant model of the water gushing source was established on the basis of the entropy weight-membership degree principle. The accuracy of the HCA-PCA-EWM discriminant model approached 100%, while those of traditional PCA-Fisher and PCA-EWM discriminant models were 62.5 and 87.5%, respectively. From the above, the HCA-PCA-EWM discriminant model can effectively select typical aquifer water samples, significantly improve the accuracy of the model, and provide a theoretical basis for the effective identification of mine water gushing sources.     

Multiple Discriminant Analysis of Marine Sediment Data

A multivariate statistical approach was used in the analysis of inter-relationships among marine benthic infaunal structure, physical and chemical characteristics of sediment samples and toxicity data derived from laboratory tests at 16 locations in shallow, inshore waters of Hong Kong. The method involved classification (cluster analysis using the Bray–Curtis similarity index) and ordination (multi-dimensional scaling) of infaunal patterns and the use of multiple discriminant analysis to relate groupings of the locations to the selected environmental and toxicity data. Analysis of the combined sediment physical, biological, chemical and toxicity dataset by stepwise multiple discriminant analysis allowed identification of those variables most sensitive for discriminating location groups. The use of multiple discriminant analysis in sediment quality characterization was evaluated against the Sediment Quality Triad approach and other statistical techniques.     

An alternative approach to predict human response to blast induced ground vibration

The aim of this study is to create alternative approaches to predict human response to ground vibration.Site measurements were conducted in an urban quarry.The multiple discriminant analysis technique was used to develop classification models.The measured ground vibration values were grouped based on the limit values appearing in the literature.Two classification models were created to predict human response to ground vibration.In addition,classical predictor equations were developed to compare to the classification models.The best equations were chosen considering several error measures.All the models were tested on the independent data.The prediction accuracy of the discriminant analysis models is quite promising.Nearly eighty-nine percent of the test data were correctly classified.The classification models have a combined structure.The results of the models can be directly presented to the neighbors of the studied quarry.The models are not complex.They do not require additional software or information to practice and can be easily used by site engineers.     

基于SPT-APD-DDA的砂土液化评价方法研究

传统的规范法在建立经验判别准则过程中存在大量不确定性因素,工程中亟需针对复杂场地液化判别方法的综合性研究。通过对上海市包头路段沉管抢修项目49组砂土土样的研究,在标准贯入试验法(SPT)定性评价的基础上采用提出的绝对差值百分比法(APD),划分出35组判别结果精确及14组存在判别误差的土样。借鉴判别分析理论思想,将平均粒径D₅₀、不均匀系数C_u、比贯入阻力P_s、标准贯入点深度d_s、地下水埋深d_w和标准贯入锤击数N_63.5等物理力学参数加入液化评价,利用精确土样建立适宜研究区工程地质特性的距离判别分析模型(DDA),对14组存在判别误差或失误的土层进行定量评价。研究结果表明：建立的砂土液化综合评价体系对研究区具有较强的针对性与适应性,从多角度全面、客观地评价了研究区砂土液化情况。最终通过合理的抗液化沉陷措施,保障了后续工程的稳步进行,为类似地质条件场区的砂土液化稳定性评价提供参考和借鉴。     

Characterization of spatial patterns in river water quality using chemometric pattern recognition techniques

This study employed three chemometric data mining techniques (factor analysis (FA), cluster analysis (CA), and discriminant analysis (DA)) to identify the latent structure of a water quality (WQ) dataset pertaining to Kinta River (Malaysia) and to classify eight WQ monitoring stations along the river into groups of similar WQ characteristics. FA identified the WQ parameters responsible for variations in Kinta River's WQ and accentuated the roles of weathering and surface runoff in determining the river's WQ. CA grouped the monitoring locations into a cluster of low levels of water pollution (the two uppermost monitoring stations) and another of relatively high levels of river pollution (the mid-, and down-stream stations). DA confirmed these clusters and produced a discriminant function which can predict the cluster membership of new and/or unknown samples. These chemometric techniques highlight the potential for reasonably reducing the number of WQVs and monitoring stations for long-term monitoring purposes.     

Supervised neural network recognition of habitat zones of rice invertebrates

This study aimed to evaluate effectiveness and performance of several supervised neural network models and make pattern recognition on invertebrate habitat zones. Probabilistic, general regression, and linear neural networks, and discriminant analysis were used to recognize both known and unknown invertebrate habitat zones. The results showed that neural network models were better than traditional discriminant analysis in the recognition of known habitat zones. There was not distinctive variation in recognition from different neural network models. Sensitivity analysis indicated that the learning rate of the neural network would influence recognized results. An unknown invertebrate species from Lepidoptera was recognized to be soil-dweller (dryland) by both neural network models and discriminant analysis. In sensitivity analysis it was additionally recognized to be the type of plant canopy (terrestrial). Overall the species was estimated to be a soil-dweller (dryland) or live on plant canopy (terrestrial). It was concluded that neural network models can perform better than conventional statistic models in pattern recognition, but a comprehensive comparison among various models is necessary in order to achieve a high reliable recognition and prediction. Furthermore, sensitivity analysis can lead to an in-depth grasp on the mechanism in the recognition and is thus needed.     

Pattern-Recognition-Verfahren in der Umweltanalytik und Ökotoxikologie: Methodische Grundlagen

Modern methods of analyses of environment yield multivariate data sets of measured objects which are used as the basis of the evaluation and interpretation of substances, differentiated effects, but also of samples or processes. The integrated evaluation and assessment of such multivariate data requires the application of mathematical-statistical methods, the method of pattern recognition being particularly suitable. Pattern recognition program systems constitute a compilation of different methods, as e.g. the multivariate linear regression, classification methods (discriminant analysis, multivariate variance analysis), cluster algorithms, multidimensional scaling, factorial analysis. A few possibilities of such pattern comparisons and identifications are represented by the example of gas chromatography and of the structure/effect analysis for the mutagenity to be expected of 117 substances.     

Rapid increase in copper concentrations in a new marina, San Diego Bay

Concentrations of copper in water rose rapidly following the introduction of boats to a new marina in San Diego Bay. Two months after the marina reached half its capacity, a majority of water samples exceeded chronic and acute criteria for dissolved copper, and copper concentrations in several samples exceeded the highest concentrations observed in another marina that has been listed as an impaired water body. A box model suggested that a small fraction of the leached copper was sequestered in sediment. Copper concentrations in water entering the marina from the bay was more than half the chronic concentration limit, so only 50% of marina boat capacity could be accommodated without exceeding the chronic criterion more than 50% of the time. Copper concentrations in water may increase rapidly following boat introduction in small marinas, but could return to pre-introduction levels by controlling boat numbers or reducing use of copper-based paints.     

Spectrophotometric discrimination of river dissolved organic matter

There is a need to be able to differentiate the dissolved organic matter (DOM) fraction in river waters. Research in the 1970s and 1980s has attempted to utilize both absorbance and fluorescence to distinguish between DOM fractions in river waters, but both were limited by the available technology. Total organic carbon content has, therefore, been widely used as a standard method of measuring DOM concentration, although it has little power to differentiate DOM fractions. Recent advances in fluorescence spectrophotometry enable rapid and optically precise analysis of DOM. Here, we show how a combination of both fluorescence and absorbance can be used to discriminate statistically between spatial variations of DOM in tributaries in a small catchment of the Ouseburn, NE England. The results of the discriminant analysis suggest that about 70% of the samples can be correctly classified to its tributary. Discriminant function 1 explains 60·8% of the variance in the data and the fulvic‐like fluorescence intensity has the largest absolute correlation within this function; discriminant function 2 explains a further 21·5% of the variance and the fulvic‐like fluorescence emission wavelength has the largest absolute correlation within this function. The discriminant analysis does not correctly classify all tributaries every time, and successfully discriminates between the different tributaries 70% of the time. Occasions when the tributary waters are less well discriminated are due to either episodic pollution events (at two sites) or due to tributaries that have strong seasonal trends in spectrophotometric parameters, which allows the sites to be misclassified. Results suggest that spectrophotometric techniques have considerable potential in the discrimination of DOM in rivers. Copyright © 2002 John Wiley & Sons, Ltd.     

Vergleichende Untersuchung der Diäten vonLymnaea auricularia undL. peregra (Gastropoda: Basommatophora) im Zürichsee

Two very similar species of freshwater snails,Lymnaea auricularia andL. peregra coexist in Lake Zürich. We examined their diets to determine possible differences. Microscopic analysis of stomach and hind gut contents revealed no marked differences between the diets of the two snail species collected from the same substrate. However, in both species we found significant differences between the composition of the substrate and that of the stomach contents, using discriminant analysis. We conclude that both species of snails feed preferentially on patches of favourable food supply and that their diets overlap almost totally. The coexistence of the two species under these conditions suggests that food does not limit their density in the area studied.     

AMS analysis of I in Japanese soil samples collected from background areas far from nuclear facilities

Analytical procedures in the determination of iodine-129 (half-life: 1.6×10⁷ y) have been studied using accelerator mass spectrometry (AMS), with special references to the separation procedures of iodine from soil samples for the AMS measurement. Iodine was successfully volatilized from soil samples by pyrohydrolysis at 1000 °C and collected in a trap solution. Iodine was purified from the matrix by solvent extraction. Finally, it was precipitated as silver iodide to make a target for AMS. In order to obtain information on the ¹²⁹I/¹²⁷I ratio in a chemical blank (or iodine carrier), we have determined the ratios in several iodine reagents and found that the ratios fell in a narrow range around 1.7×10⁻¹³. The detection limit for soil sample (1 g material) by the present method was about 0.01 mBq/kg or 4×10⁻¹¹ as the ratio of stable iodine (¹²⁹I/¹²⁷I ratio), i.e. these values were much better than that by neutron activation analysis (NAA) used in our previous studies. We have applied this method in the analysis of soil samples collected from different places in Japan. We could successfully determine ¹²⁹I in soil samples with low ¹²⁹I concentrations, which could not be detected by NAA. Sample size necessary for the soil analysis by AMS was only about 0.5 g or less, whereas about 100 g of the sample were required for NAA [Muramatsu, Y., Ohmomo, Y., 1986. Iodine-129 and iodine-127 in environmental samples collected from Tokaimura/ Ibaraki, Japan. Sci. Total Environ. 48, 33-43]. Using this method, new data were obtained for the ¹²⁹I levels in 20 soil samples collected from background areas far from nuclear facilities, and the ranges were 1.4×10⁻⁵−4.5×10⁻³ Bq/kg as ¹²⁹I concentrations and 3.9×10⁻¹¹−2.2×10⁻⁸ as ¹²⁹I/¹²⁷I ratios. These values are useful in understanding the ¹²⁹I levels in Japanese environments. Higher ¹²⁹I concentrations were observed in forest soils than those in field and rice paddy soils should be related to the interception effect of atmospheric ¹²⁹I due to tree canopies. Relatively high ¹²⁹I/¹²⁷I ratios found in rice paddy soils could be explained by their low stable iodine concentrations which were caused by the desorption of stable iodine from the rice paddies during the cultivation.     

Assessment of temporal and spatial variation of coastal water quality and source identification along Macau peninsula

Cluster analysis (CA), discriminant analysis (DA) and principal component analysis (PCA) were comprehensively coupled to explore and identify the spatial and temporal variation and potential pollution sources in coastal water quality along Macau peninsula. The results show that the 12 months could be grouped into two periods, June–September and the remaining months, and the entire area divided into two clusters, one located at the western sides, and the other on the southeast and southern sides of the Macau peninsula. Through backward stepwise DA, pH, Cl⁻, TSS, Color and TP, Chloride, Color, NH₄ ⁺, DO, COD were discriminant variables of spatial and temporal variation, with 84.82 and 76.54% correct assignments, respectively. Fecal pollution, organic pollution and soil weathering are among the major sources for coastal water quality deterioration along Macau peninsula. This study illustrates that application of multivariate statistical techniques was beneficial to gain knowledge for further optimizing the monitoring network and controlling coastal water quality along Macau peninsula.     

An Approach for Classification of Groundwaters by Synchronous Fluorescence Spectroscopy and Discriminant Analysis

Monitoring and evaluation of groundwater quality in drinking water protectorates is of particular interest if the uncontrolled entry of pollutants, for example by infiltrating river water, cannot be excluded. A fluorescence spectroscopic method is presented as possible alternative to the conventional expensive hydrochemical investigations. This method uses the information yielded by synchronous fluorescence spectra. After a Fourier transformation of the original spectra and a following discriminant analysis, the samples can be classified in different groups corresponding to the different types of groundwater. The method was tested in the drinking water protectorate of a waterworks of a German metropolis (Halle, Sachsen-Anhalt). We found a prediction rate of about 90% in the investigated case.     

Multivariate discriminant analysis distinguishes metal- from non metal-related biomarker responses in the clam Chamaelea gallina

Molecular biomarkers are among the most sensitive and earliest responses to pollutants. However, lack of detailed knowledge on variability of responses and their possible seasonal variation limit their use. In addition, the seasonality of biological processes modulates the response of organisms to pollutant stressors. Using multivariate statistics, we have studied the influence of environmental and biological factors on the response of a battery of molecular biomarkers in the clam Chamaelea gallina collected along the South-Spanish littoral. Multivariate discriminant analysis clearly distinguished biomarker response between clean and polluted areas, using heavy metals as indicator of pollution. Such differences disappeared when the dataset was normalised for metal content, thus indicating that pollution was the main significant cause of the changes observed between clean and polluted sites. In conclusion, this work shows that, when applying a complete biomarker panel, multivariate statistical tools can be used to discern pollutant- from non pollutant-related responses.     

Use of Observations below Detection Limit for Model Calibration

Censored (nondetect) values occur when chemical concentrations in water samples are near or below the level that can be measured by an analysis method. It is common to either delete or substitute values for nondetect observations for use in model calibration, but this practice can bias the estimated parameter values and the model predictions. A more realistic representation of the system is obtained from the calibration if we include such observations in a manner reflecting that we know only the value is below the detection limit. Consequently, we propose use of the censored-residual approach to including nondetect values as observations for calibration. In this approach, residuals are calculated as the detection limit minus the simulated value when the simulated value exceeds the detection limit, and the residual is assigned a value of zero when the simulated value is below the detection limit. The new censored-residual approach is particularly advantageous when calibrating transport models to low concentration data.     

利用磷灰石裂变径迹约束脆性断裂活动的时限

本文通过}西隆起区、沂沭断裂带不同性质断裂构造解析,并对其构造带及构造带两侧未变形地层进行野外样品采集、测试、分析的实例总结,认为利用磷灰石裂变径迹低温测年技术,辅助以样品磷灰石裂变径迹冷却史模拟分析,可以直接对地质体不同性质断裂活动的时间进行约束.鉴于磷灰石裂变径迹对低温特殊的敏感性与反映温度区间的局限性,利用该方法约束断裂时限时,需注意包括野外采样等多个环节,需要结合大量的区域地质资料综合分析,必要时要辅助以其他同位素测年手段进行综合约束,以获得断裂活动的可靠信息.     

中国大陆余震序列类型的综合判定

在对中国大陆294次5级以上地震序列主震后不同时段、11个单项序列参数计算结果的基础上,分别建立不同时段、以不同序列参数为自变量的Fisher判别函数,用于序列类型的综合判定。在Fisher判别函数建立过程中,除以常规序列参数作为函数自变量输入外,还引入主震滑动角参数,以使判别函数在一定程度上包含主震破裂形式的信息。统计结果表明,序列类型识别正确率随资料长度(震后时段)增加而提高。不同类型序列的识别正确率不同,孤立型序列的识别正确率大于0.93,主余型序列的识别正确率为0.73~0.93,多震型序列的正确识别率为0.57~0.86,均明显高于随机投票0.33的自然对应率,因而其序列分类能力应得到正的评分。     

Silica standardization: a discriminant technique applied to a volcanic arc system

A classic tholeiitic, calc-alkaline, and shoshonitic island-arc sequence has developed in western Luzon, the Philippines, as a result of subduction defined by an eastward-dipping Benioff zone. A simple technique of standardizing the data allows chemical concentrations to be recast to “andesite equivalent” compositions. Based on the excellent linear correlations that exist between SiO₂ and other elements, individual analyses were standardized to 55% SiO₂ by projecting parallel to least-squares regression lines. Raw and SiO₂-standardized data were then treated by an univariate one-factor analysis of variance and a multivariate discriminant function analysis.Results of the analysis of variance indicate that at the 99% confidence level; 20 of the 27 chemical variates discriminate between at least two of the three geographically distinct magma series; SiO₂-standardized data generally discriminate better than do raw data; and K₂O, Ba, Rb, Sr, and Zr provide the best discrimination. Better classification of samples into the appropriate series is obtained from results of the discriminant function analysis using the SiO₂-standardized data rather than results using the raw data. With this standardizing technique it was possible to detect offsets of SiO₂-standardized K₂O isopleths along the Manila Bay fracture zone that correspond to offsets in contours of depths for earthquake foci.     

花岗岩非稳态流变实验

脆塑性转化带对于研究岩石圈变形、断层强度和变形机制以及强震的孕育和发生具有重要意义.文中采用汶川地震震源区彭灌杂岩中具有代表性的细粒花岗岩样品,在固体压力介质三轴实验系统上开展了高温高压非稳态流变实验研究.实验设计模拟了汶川地震区地壳10～30km深度的实际温度和压力,温度为190～490℃,压力为250～750MPa...