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.     

Basalt geochemistry and tectonic discrimination within continental flood basalt provinces

Continental flood basalts are usually regarded as a single tectonomagmatic entity but frequently quoted examples exhibit a variety of tectonic settings. In one well-studied, classic, flood basalt province, the Mesozoic Karoo province of southern Africa, magmatism occurred in the following tectonic settings: (a) continental rifting leading to ocean-floor spreading in the South Atlantic Ocean (Etendeka suite of Namibia); (b) stretched continental lithosphere and rifting not leading directly to ocean-floor formation (Lebombo suite of southeastern Africa); and (c) an a-tectonic, within-plate, continental setting characterized by an absence of faulting or warping (Lesotho highlands and Karoo dolerites of South Africa). By means of spidergrams of the elements Rb, Ba, Th, Nb, K, La, Ce, Sr, Nd, P, Hf, Zr, Sm, Ti, Tb, Y, V, Ni and Cr, uncontaminated tholeiites from (c) above [i.e. the Lesotho-type continental flood basalts (LTCFB)] are compared with mid-ocean ridge basalts (MORB), ocean-island tholeiites (OIT), and tholeiites and calc-alkali basalts from subduction environments. The comparison reveals the LTCFBs are geochemically distinct. The differences are reflected in relative enrichments or depletions of the more incompatible elements (Rb-Ce) to less incompatible elements (Ce-Y), i.e. the overall slope of the spidergrams, and in anomalous enrichments or depletions of one or more of the elements Th, K, Nb, Sr, Ti, Hf, and Zr. The distinctive geochemical character of the Lesotho LTCFBs is interpreted in terms of a lithospheric mantle source for the basalts. This is supported by isotopic data. There are no major geochemical differences between Lesotho CFBs and basalts of the rift-related Etendeka and Lebombo suites, although the latter are somewhat enriched in Rb, Ba and K. However, unlike the Lesotho basalts, the Lebombo and Etendeka basalts are associated with voluminous silicic volcanics or intrusive centres and late-stage dolerites having MORB/OIT (i.e. asthenospheric) geochemical characteristics. The flood basalt/silicic magmatism/late-stage dyke swarm association is characteristic of several rift or thinned lithosphere environments (e.g., Ethiopia, Skye, eastern Greenland) but in many of these the flood basalts have ocean-island basalt (OIT) geochemical characteristics. The Lesotho-type CFB geochemistry is exhibited by the Grande Ronde Basalt of the Columbia River Group (a possible subduction-related flood basalt province) and the basic rocks associated with Mesozoic rifting in the North and South Atlantic. Basalt geochemistry alone is unhelpful in determining the tectonic setting of CFBs although the rift-related environments may be identified by the petrology and geochemistry of the whole igneous suite. A two-source model is proposed for the mantle-derived basic rocks in rift-related CFB provinces. Early enriched basalts are derived from the lithosphere and, following pronounced lithospheric attenuation or rifting, later MORB-like melts are emplaced from the rising asthenosphere. The presence of both Lesotho- and OIT-type geochemical patterns in rift-related CFBs suggests that the lithosphere exhibits different styles of enrichment.     

Molekular Connectivity Indizes als strukturchemische Stoffparameter ökotoxikologischer Relevanz

The molecular connectivity index developed by Randic as a numerical expression characterizes the position and bond of atoms in the molecular structure. Since significant correlations between the molar refraction as well as electron polarizability and ecotoxicologically relevant parameters of substances were determined in the preceding investigations, such connections are investigated also for this index and sorption coefficients, bioconcentration factors and toxicity data. By the example of 56 differentiated organic substances, significant regression relations are demonstrated between bioconcentration factors, sorption coefficients and toxicity data as well as the valency molecular connectivity indices. By these regressions it is possible to determine essential distribution and action parameters of chemical substances from the structure for assesing an ecotoxicological risk. The possibilities resulting from this for predicting ecotoxicologically relevant parameters of substances are discussed.     

Identification of regions enriched or depleted in radioelements through nondistributional analysis of aerial radiometric data

Throughout the aerial radiometric reconnaissance survey portion of the U.S. Department of Energy's National Uranium Resource Evaluation (NURE) program, the identification of outliers (anomalies) was an important approach to locating regions with radio-element concentrations that are either higher or lower than expected. The method introduced herein to locate such regions involves three steps: selection of a high (or low) threshold for the variate of interest; use of the sample percentile to identify all points of interest; and movement of a window over the selected data to locate significant clusters of observations. These steps, applied to aerial radiometric ²¹⁴Bi (equivalent uranium) data collected over the Copper Mountain area in Wyoming, resulted in the identification of areas enriched in that variate.     

Certainty, uncertainty, and the spatiality of disease: a West Nile Virus example

The problem is not uncertainty—proposed here as an inevitable condition—but the chimera of certainty asserted by most contemporary researchers. Problems of data definition, collection, and their use are reviewed in terms of spatial epidemiology and health data with examples drawn from several areas of contemporary health research. The argument is that preconceptions limit data modeled in a manner assuming its completeness. The result, as the West Nile Virus example seeks to demonstrate, may obscure other patterns and limit avenues of research.