Chemometric analysis of soil pollution data using the Tucker N-way method.

N-way methods, particularly the Tucker method, are often the methods of choice when analyzing data sets arranged in three- (or higher) way arrays, which is the case for most environmental data sets. In the future, applying N-way methods will become an increasingly popular way to uncover hidden information in complex data sets. The reason for this is that classical two-way approaches such as principal component analysis are not as good at revealing the complex relationships present in data sets. This study describes in detail the application of a chemometric N-way approach, namely the Tucker method, in order to evaluate the level of pollution in soil from a contaminated site. The analyzed soil data set was five-way in nature. The samples were collected at different depths (way 1) from two locations (way 2) and the levels of thirteen metals (way 3) were analyzed using a four-step-sequential extraction procedure (way 4), allowing detailed information to be obtained about the bioavailability and activity of the different binding forms of the metals. Furthermore, the measurements were performed under two conditions (way 5), inert and non-inert. The preferred Tucker model of definite complexity showed that there was no significant difference in measurements analyzed under inert or non-inert conditions. It also allowed two depth horizons, characterized by different accumulation pathways, to be distinguished, and it allowed the relationships between chemical elements and their biological activities and mobilities in the soil to be described in detail.

Environmetric modeling and interpretation of river water monitoring data.

This environmetric study deals with modeling and interpretation of river water monitoring data from the basin of the Saale river and its tributaries the Ilm and the Unstrut. For a period of one year of observation between September 1993 and August 1994 a data set from twelve campaigns at twenty-nine sampling sites from the Saale river and six campaigns from the river Ilm at seven sampling sites and from river Unstrut at ten sampling sites was collected. Twenty-seven chemical and physicochemical properties were measured to estimate the water quality. The application of cluster analysis, principal components analysis, and apportioning modeling on absolute principal components scores revealed important information about the ecological status of the region of interest:identification of two separate patterns of pollution (upper and lower stream of the rivers);identification of six latent factors responsible for the data structure with different content for the two identified pollution patterns; anddetermination of the contribution of each latent factor (source of emission) to the formation of the total concentration of the chemical burden of the river water. As a result more objective ecological policy and decision making is possible.

Exploratory analysis of data sets with missing elements and outliers.

The main goal of the presented paper was to develop a general strategy allowing exploration of contaminated data sets with missing elements, based on application of robust PLS for initial estimation of missing elements. Using robust distance, the outlying elements were identified. After their identification and replacing by missing elements, the expectation-maximization algorithm (which can be built in into different computational procedures, such as principal component analysis and its generalisation to the N-way data-the TUCKER3 model) was used for construction of the final model.

Interaction of Porphyromonas gingivalis with KB cells: comparison of different clinical isolates.

The ability of different Porphyromonas gingivalis strains (15 clinical isolates and ATCC 33277) to attach to and invade KB cells, in relation to other properties such as release of interleukin (IL)-6 and IL-8, cytotoxicity, proteolytic activity and types of fimbriae genes present, was examined. A hierarchical cluster analysis based on adherence and internalization resulted in four groups. Eight of the 15 clinical isolates belonged to a cluster group whose adherence and internalization were about 10% those of the ATCC strain. A negative correlation between lysine-specific protease activity and adherence was found. In all cases the released concentrations of IL-6 and IL-8 were very low. Only one strain was found to be cytotoxic to KB cells. Principal components analysis demonstrated correlations between adherence, internalization and autoaggregation. Most strains had fimA type I and II, type I being associated with elastase-like activity. The ability of P. gingivalis to invade epithelial cells may be a key factor for maintaining periodontal disease.

Resolution of uncertainty of a four-stage sequential extraction procedure using analysis of variance.

Sequential extraction procedures are mostly applied to soils and sediments. Investigations were made for testing the applicability of a selected extraction scheme on spoil pile material. Therefore, different spoil pile samples from uranium mining were extracted into four steps (exchangeable, reducible, oxidizable, and aqua regia soluble phase) and the contents of the elements Cr, Cu, Fe, K, Ni, and Zn were analyzed. To substitute the sequential extraction by a sequence of single extractions and thus keep uncertainties of extraction on a low level, both procedures were compared regarding conformity of their results. The effects of sample handling as well as those of analysis on the measured metal contents were determined by means of analysis of variance and sources of uncertainties were discussed.

Characterization of elemental pollution and its variations in sediments and suspended particulate matter from the River Elbe via multi-element analysis combined with chemometric data evaluation.

Sediments and suspended particulate matter taken from about 100 sampling sites along the River Elbe from the source to the mouth during the period 1992 and 1998 have been analyzed for the concentrations of more than 60 elements in each sample. The analytical data were subsequently processed by means of multivariate statistics in order to characterize the charge of the River Elbe with inorganic pollutants to elucidate pollution trends. Using factor analysis 18 elements were found to be enriched by human activities. By means of cluster analysis--using the 18 anthropogenically influenced elements as variables--the sampling sites were aggregated into groups having similar element distributions. Thus, the entire stretch of the River Elbe was divided into three sections of characteristic elemental pollution. Finally, the trends in anthropogenic pollution within the characteristic regions since the early nineties were assessed. It turned out that most of the anthropogenically influenced elements showed a decrease in the upper and middle parts of the river, whereas no decrease could be verified in the estuary region.

Comparison of detection limits in environmental analysis--is it possible? An approach on quality assurance in the lower working range by verification.

Detection limit, reporting limit and limit of quantitation are analytical parameters which describe the power of analytical methods. These parameters are used for internal quality assurance and externally for competing, especially in the case of trace analysis in environmental compartments. The wide variety of possibilities for computing or obtaining these measures in literature and in legislative rules makes any comparison difficult. Additionally, a host of terms have been used within the analytical community to describe detection and quantitation capabilities. Without trying to create an order for the variety of terms, this paper is aimed at providing a practical proposal for answering the main questions for the analysts concerning quality measures above. These main questions and related parameters were explained and graphically demonstrated. Estimation and verification of these parameters are the two steps to get real measures. A rule for a practical verification is given in a table, where the analyst can read out what to measure, what to estimate and which criteria have to be fulfilled. In this manner verified parameters detection limit, reporting limit and limit of quantitation now are comparable and the analyst himself is responsible to the unambiguity and reliability of these measures.

Inert sampling and sample preparation--the influence of oxygen on heavy metal mobility in river sediments.

Two approaches have been used to investigate changes in the nature of metal binding in river sediments caused by atmospheric oxygen. Firstly, non-inert and inert sample preparation were applied, in combination with sequential extraction, to determine for which metals inert sample preparation is necessary for correct determination of metal mobility under environmental conditions. Secondly, the metal contents of sediments sampled before and after a river weir were fractionated by sequential extraction to study the effect of the oxygen impact at the weir on heavy metal mobility in the sediments. Different grain-size fractions from one sample were also extracted, to enable selection of the upper grain-size limit most suitable for answering this analytical question. The results showed the need for the inert sample preparation technique for Cd, Zn, Pb, Mn, and Fe, but not for Co, Ni, Cu, and Cr. No significant change of heavy metal mobility at the weir could be proved, although the mobilization behavior of some elements was different. The optimum upper grain-size limit was 63 microm.

Chemometric characterization of soil and plant pollution: Part 1: Multivariate data analysis and geostatistical determination of relationship and spatial structure of inorganic contaminants in soil.

Geostatistical and multivariate methods of data analysis are used to describe patterns of soil pollution with inorganic contaminants in Celje County, Slovenia. Groups of contaminants and polluted sites were identified using cluster analysis and confirmed with multidimensional variance and discriminant analysis. Factor analysis yields an identification of not directly observable relationships between the contaminants. The spatial structure and distribution of contaminants were assessed by applying semivariogram analysis and kriging interpolation method. Zinc, Cd and Cu were identified as a pollutant emitted from the zinc smelter, Pb also from other sources, and Cr and Ni mostly from geological parent material.