The use of diagnostic ratios, biomarkers and 3-way Kohonen neural networks to monitor the temporal evolution of oil spills.

Oil spill identification relies usually on a wealth of chromatographic data which requires advanced data treatment (chemometrics). A simple approach based on Kohonen neural networks to handle three-dimensional arrays is presented. A suite of 28 diagnostic ratios was considered to monitor six oils along four months. It was found that some traditional diagnostic ratios were not stable enough. In particular, alkylated PAHs (e.g. 1-methyldibenzothiophene, 4-methylpyrene, 27bbSTER and the TA21 and TA26 triaromatic steroids) seemed less resistant to medium-weathering than biomarkers. One (or two) ratios were found to differentiate each product: 30O, 28ab (and 25nor30ab), C3-dbt/C3-phe, 27Ts, TA26 and 29Ts characterized Ashtart, Brent, Maya, Sahara, IFO and Prestige oils, respectively.

Impact of medium-distance pollution sources in a Galician suburban site (NW Iberian peninsula).

This work studies airborne quality in a geographical area that has not been investigated broadly: a suburban site nearby A Coruña (Galicia, NW Iberian Peninsula). In contrast to major Spanish cities, the site has Atlantic characteristics: rainy, scarce calm weather and infrequent prolonged sunny periods. The relationships between several gaseous pollutants (NO, NO2, NOx, CO, O3, PM10, PM2.5 and PM1) and their temporal trends (daily, monthly and seasonal) were evaluated. The aim was to unravel whether medium- and long-distance sources were impacting upon the site. Univariate studies focused on factorizing the pollutants according to a codifying factor (wind direction, hour of the day, season and month). Multivariate studies (Varimax-rotated factorial analysis) were done separately on both weekdays and weekends. The intensity of the daily maxima for NO, NO2, NOx and CO was lower during the weekends, with O3 behaving opposite. PM average values agreed with previous historical reports for a rural background station relatively close to the site and they decreased daily between 11:00 and 19:00 h, likely because of the marine breeze. With moderate wind speeds the pollutants were associated to medium-distance pollution sources, mainly the city of A Coruña and a combination of industrial pollution sources (a power plant, a solid waste incinerator and a regional airport).

Objective chemical fingerprinting of oil spills by partial least-squares discriminant analysis.

An objective method based on partial least-squares discriminant analysis (PLS-DA) was used to assign an oil lump collected on the coastline to a suspected source. The approach is an add-on to current US and European oil fingerprinting standard procedures that are based on lengthy and rather subjective visual comparison of chromatograms. The procedure required an initial variable selection step using the selectivity ratio index (SRI) followed by a PLS-DA model. From the model, a "matching decision diagram" was established that yielded the four possible decisions that may arise from standard procedures (i.e., match, non-match, probable match, and inconclusive). The decision diagram included two limits, one derived from the Q-residuals of the samples of the target class and the other derived from the predicted y of the PLS model. The method was used classify 45 oil lumps collected on the Galician coast after the Prestige wreckage. The results compared satisfactorily with those from the standard methods.

Comparing roadsoils pollution patterns extracted by MOLMAP and classical three-way decomposition methods.

A recent approach based on self-organizing maps (SOMs) to extract patterns from three-way data, named MOLMAP, was applied in a four-seasons study on soil pollution and its results compared with three different conventional approaches: Parallel factor analysis (PARAFAC), matrix augmented principal components analysis (MA-PCA) and Procrustes rotation. Each sampling season comprised 92 roadsoil samples and 12 analytical variables (Cd, Co, Cu, Cr, Fe, Mn, Ni, Pb, Zn, loss on ignition, pH and humidity). It was found that all techniques yielded highly similar results as the samples became organized in two major groups, each with a differentiated pollution pattern. This confirmed MOLMAP as a reliable option to handle environmental three-way datasets and to extract accurate pollution patterns.

Self organizing maps for analysis of polycyclic aromatic hydrocarbons 3-way data from spilled oils.

In this paper, the application of a new method based on self-organizing maps (SOM; termed MOLMAP, molecular map of atom-level properties) to handle 3-way data generated in a monitoring environmental study is presented. The study comprised 50 polycyclic aromatic hydrocarbons (PAHs) analyzed in samples derived from the weathering of six oil products (four crude oils and two fuel oils) spilled under controlled conditions for about 4 months. MOLMAP yielded useful information on each mode of the data cube: weathering samples, spilled oil products, and PAHs. Thus, the different behaviors of the six oils were ascertained, along with their particular evolution on time, and their weathering patterns were studied in terms of the original PAHs. Thus, the two heaviest products (two fuel oils) were characterized by two neurons whose more relevant weights were associated to heavy PAHs, as C(1)-fluoranthene, C(2)-fluoranthene, benzo(ghi)perylene, and dibenz(ah)anthracene. The six spilled products were projected on different regions on both the MOLMAP-SOM and a subsequent principal components analysis (PCA) scatter plot, developed using the so-called MOLMAP-scores. Besides, it was possible to further differentiate between unweathered, or slightly weathered, samples and the most weathered ones. The more relevant PAHs characterizing those samples were assessed studying the weights of the neurons in which the samples got projected.

Chemically driven variable selection by focused multimodal genetic algorithms in mid-IR spectra.

Four genetic-algorithm-based approaches to variable selection in spectral data sets are presented. They range from a pure black-box approach to a chemically driven one. The latter uses a fitness function that takes into account not only typical parameters like the number of errors when classifying a training set but also the chemical interpretability of the selected variables. In order to cope with the fact that multiple solutions may be acceptable, a multimodal genetic algorithm (GA) is employed and the most satisfactory solution selected. The multimodal GA uses two populations (denominated "hybrid two populations" GA or HTP-GA): a classical population, from which potential solutions emerge, and a new population, which maintains diversity in the search space (as required by multimodal problems). Results show that the HTP-GA approach improves the chemical understanding of the selected solution (compared to other GA approaches) and that the classification capabilities of the approach are still good. All of the GA strategies for variable selection were compared with a classical parametric technique, Procrustes rotation, which does not consider interpretability.

Selecting the optimum number of partial least squares components for the calibration of attenuated total reflectance-mid-infrared spectra of undesigned kerosene samples.

Selecting the correct dimensionality is critical for obtaining partial least squares (PLS) regression models with good predictive ability. Although calibration and validation sets are best established using experimental designs, industrial laboratories cannot afford such an approach. Typically, samples are collected in an (formally) undesigned way, spread over time and their measurements are included in routine measurement processes. This makes it hard to evaluate PLS model dimensionality. In this paper, classical criteria (leave-one-out cross-validation and adjusted Wold's criterion) are compared to recently proposed alternatives (smoothed PLS-PoLiSh and a randomization test) to seek out the optimum dimensionality of PLS models. Kerosene (jet fuel) samples were measured by attenuated total reflectance-mid-IR spectrometry and their spectra where used to predict eight important properties determined using reference methods that are time-consuming and prone to analytical errors. The alternative methods were shown to give reliable dimensionality predictions when compared to external validation. By contrast, the simpler methods seemed to be largely affected by the largest changes in the modeling capabilities of the first components.

Development of a fast analytical tool to identify oil spillages employing infrared spectral indexes and pattern recognition techniques.

A fast analytical tool based on attenuated total reflectance mid-IR spectrometry is presented to evaluate the origin of spilled hydrocarbons and to monitor their fate on the environment. Ten spectral band ratios are employed in univariate and multivariate studies (principal components analysis, cluster analysis, density functions - potential curves - and Kohonen self organizing maps). Two indexes monitor typical photooxidation processes, five are related to aromatic characteristics and three study aliphatic and branched chains. The case study considered here comprises 45 samples taken on beaches (from 2002 to 2005) after the Prestige carrier accident off the Galician coast and 104 samples corresponding to weathering studies deployed for the Prestige's fuel, four typical crude oils and a fuel oil. The univariate studies yield insightful views on the gross chemical evolution whereas the multivariate studies allow for simple and straightforward elucidations on whether the unknown samples match the Prestige's fuel. Besides, a good differentiation on the weathering patterns of light and heavy products is obtained.

Handling complex effects in slurry-sampling-electrothermal atomic absorption spectrometry by multivariate calibration.

Analysis of solid samples by slurry-sampling-electrothermal atomic absorption spectrometry (SS-ETAAS) can imply spectral and chemical interferences caused by the large amount of concomitants introduced into the graphite furnace. Sometimes they cannot be solved using stabilized temperature platform furnace (STPF) conditions or typical approaches (previous sample ashing, use of chemical modifiers, etc.), which are time consuming and quite expensive. A new approach to handle interferences using multivariate calibrations (partial least squares, PLS, and artificial neural networks, ANN) is presented and exemplified with a real problem consisting on determining Sb in several solid matrices (soils, sediments and coal fly ash) as slurries by ETAAS. Experimental designs were implemented at different levels of Sb to develop the calibration matrix and assess which concomitants (seven ions were considered) modified the atomic signal mostly. They were Na+ and Ca2+ and they induced simultaneous displacement, depletion (enhancement) and broadening of the atomic peak. Here it is shown that these complex effects can be handled in a reliable, fast and cost-effective way to predict the concentration of Sb in slurry samples of several solid matrices. The method was validated predicting the concentrations of five certified reference materials (CRMs) and studying its robustness to current ETAAS problems. It is also shown that linear PLS can handle eventual non-linearities and that its results are comparable to more complex (non-linear) models, as those from ANNs.

Monitoring photooxidation of the Prestige's oil spill by attenuated total reflectance infrared spectroscopy.

The recent release of ca. 70,000 tonnes of a heavy fuel oil from the Prestige-Nassau carrier along the Spanish northern coast, mainly along Galicia, was monitored using attenuated total reflectance-mid IR spectrometry. The fuel was characterized and differentiated from 10 products commonly transported along the Galician coast (and their series of weathered samples) using factor analysis. The Prestige's fuel was weathered under natural conditions and under infrared radiation to study its evolution on time. A correlation was established using the 1690-1700 cm(-1) carbonyl peak, where from it was deduced that IR radiation weathered the product two times faster than natural conditions. The use of 10 weathering indexes was carried out to confirm the main patterns given by factor analysis and to seek out which main functional groups and structures increased or decreased during weathering. It was found that the carbonyl and sulphoxide indexes varied greatly, as well as the total aromaticity and long chains ones. The substitution-related indexes pointed out that highly substituted aromatic structures increased although the total amount of isolated CH groups in aromatic structures reached a plateau.

Screening the origin and weathering of oil slicks by attenuated total reflectance mid-IR spectrometry.

The combination of attenuated total reflectance-fourier transform mid-infrared spectrometry (ATR-FTMIR) and multivariate pattern recognition is presented as a fast and convenient methodology to ascertain the source product an oil slick comes from and to evaluate the extent of its weathering. Different types of hydrocarbons (including crude oils, several heavy distillates and the Prestige's heavy fuel oil) were spilled on metallic containers designed ad hoc and their fate monitored by ATR-FTMIR. Not only environmental conditions were considered for weathering but artificial IR- and UV-irradiation. Pattern-recognition studies revealed that the different hydrocarbons clustered at different locations on the score plots and that the samples corresponding to each oil became ordered according to the extent of their weathering. Among them, fuel oil samples coming from the recent disaster of the Prestige tanker off the Galician shoreline showed a distinctive behaviour. Comparison of natural-, IR- and UV-weathering of a crude oil showed that IR solar radiation can be important in oil-weathering, in addition to broadly-reported UV degradation.

Classification of commercial apple beverages using a minimum set of mid-IR wavenumbers selected by Procrustes rotation.

When infrared spectral data are used in classification and/or multivariate regression methods there can be problems related to both chemical understanding and computation speed due to the large number of wavenumbers in each spectrum. Here, it is shown that the Procrustes rotation technique can be used to select a minimum set of spectral variables (wavenumbers) to perform classification and regression. Procrustes rotation was coupled to several multivariate methods as PLS, SIMCA and potential curves (a maximum likelihood classification method). The practical problem of implementing a screening methodology for classifying apple juice-based beverages according to their contents of "pure" apple juice was addressed using attenuated total reflectance, mid-IR spectroscopy. It is found that two of the original wavenumbers are almost as good predictors as all the 176 initial ones.

Generation and mid-IR measurement of a gas-phase to predict security parameters of aviation jet fuel.

The worldwide use of kerosene as aviation jet fuel makes its safety considerations of most importance not only for aircraft security but for the workers' health (chronic and/or acute exposure). As most kerosene risks come from its vapours, this work focuses on predicting seven characteristics (flash point, freezing point, % of aromatics and four distillation points) which assess its potential hazards. Two experimental devices were implemented in order to, first, generate a kerosene vapour phase and, then, to measure its mid-IR spectrum. All the working conditions required to generate the gas phase were optimised either in a univariate or a multivariate (SIMPLEX) approach. Next, multivariate prediction models were deployed using partial least squares regression and it was found that both the average prediction errors and precision parameters were satisfactory, almost always well below the reference figures.