Darío Gómez

Artificial neural network for the identification of unknown air pollution sources

Artificial neural networks (ANN), whose performances to deal with pattern recognition problems is well known, are proposed to identify air pollution sources. The problem that is addressed is the apportionment of a small number of sources from a data set of ambient concentrations of a given pollutant. Three layers feed-forward ANN trained with a back-propagation algorithm are selected. A test case is built, based on a Gaussian dispersion model. A subset of hourly meteorological conditions and measured concentrations constitute the input patterns to the network that is wired to recover relevant emission parameters of unknown sources as outputs. The rest of the model data are corrupted adding noise to some meteorological parameters and we test the effectiveness of the method to recover the correct answer. The ANN is applied to a realistic case where 24 h SO2 concentrations were previously measured. Some of the limitations of the ANN approach, together with its capabilities, are discussed in this paper.

Monitoring trace metals in urban aerosols from Buenos Aires city. Determination by plasma-based techniques

A study was undertaken, within the framework of a 3 years national project, to assess the content of 13 elements in airborne particulate matter collected in representative zones of the metropolitan area of Buenos Aires. The sampling strategy followed consisted in collecting simultaneously 67 samples of PM10 particulate matter in 9 sampling sites covering an area of about 30 km2 during one week. The collection was performed on ash-free fibre-glass filters using high volume samplers. A combination of aqua regia and perchloric acid was used for leaching metals from filters. Key elements, namely Al, Ca, Cu, Fe, Mn, Mo, Ni, Pb, S, Sb, Sn, Zn and Zr, were determined by inductively coupled plasma-optical emission spectrometry (ICP-OES) and inductively coupled plasma-mass spectrometry (ICP-MS) at micro g g(-1) and ng g(-1) levels. Analyte concentration varied from 130 ng g(-1)(Mo) to over 30%(Ca). Multivariate statistical analysis was performed on the data set including the measured elemental compositions for the monitored period. The atmospheric concentration found for Pb confirms the decreasing levels of this element since the introduction of unleaded gasoline in 1995: 88 ng m(-3)(2001) < 220 ng m(-3)(1997) < 3900 ng m(-3)(1994). The average S concentration above 3 microg m(-3) is somehow unexpectedly high for Buenos Aires since the relatively low S content of liquid fuels and the massive usage of natural gas imply low emissions of this element from combustion activities. To the best of our knowledge, S concentrations are reported for the first time for this city.

Traffic‐Related Elements in Airborne Particulate Matter

Abstract This review covers the investigations done over the last few years to quantify key traffic‐related elements, namely, Pb, Pd, Pt, Rh, and Sb in atmospheric aerosols. From this standpoint, the role played by analytical techniques as diverse as electrothermal atomization atomic absorption spectrometry (ET‐AAS), total reflection X‐ray fluorescence (TXRF) spectrometry, and inductively coupled plasma‐mass spectrometry (ICP‐MS) is also surveyed as a substantial amount of research is presently being performed to further improve the instrumental approach. In their turn, neutron activation analysis (NAA) and particle‐induced X-ray emission (PIXE) are nuclear techniques that were and are employed in this context. The application of ICP‐MS for fractionation studies based on the separation of particles according to their aerodynamic diameter and the use of chemical extraction for metal partitioning in airborne particulate matter and their speciation analysis are also discussed. This survey contains 91 references and covers primarily the literature published over the last decade.

Fractionation of eleven elements by chemical bonding from airborne particulate matter collected in an industrial city in Argentina

A four-step chemical sequential extraction procedure was used to evaluate the distribution of Al, As, Cr, Cu, Fe, Mn, Ni, Pb, Ti, V and Zn in airborne particulate matter collected on glass fibre filters using a high-volume sampler. Two sets of samples were collected in 2001 (winter and summer campaigns) in representative zones of an industrial city of Argentina. The leaching scheme was applied to PM-10 particles and consisted in extracting the elements in four fractions, namely soluble and exchangeable elements; carbonates, oxides and reducible elements; bound to organic matter and sulfidic metals; and residual elements. Metals and metalloids at microg g(-1) level were determined in each fraction by inductively coupled plasma optical emission spectrometry (ICP OES). Analyte concentration varied from 14 microg g(-1) (equivalent to 1.0 ng m(-3)) for As to 11.8 mg g(-1) (equivalent to 2,089 ng m(-3)) for Al. Seven elements, namely Al, Cr, Fe, Mn, Pb, Ti and Zn showed similar distributions in both seasons while As was distributed in a significantly different manner in each season. The results exhibited low As contents in the first and second fractions that could be associated with routine coal combustion and a high content in the third and fourth fractions of the summer samples that could be linked to the use of pesticides. Aluminium, As, Cu, Mn, Ni, Ti, V and Zn were found in different percentages in the more bioavailable aqueous fraction with As, Mn, V and Zn exhibiting solubilities greater than 1% while Cr and Pb being insoluble. The content of Al, Cr, Cu, Fe, Ni, Pb, and Zn in the residual fraction was, in average, higher than 50%. A comparative assessment of the use of the underlying information available from fractionation studies compared to that obtained from total element content was done for Fe and Mn. It showed that the results obtained using chemical sequential extraction procedures allowed further discrimination of the potential air pollution sources.

Trace elements content in size-classified volcanic ashes as determined by inductively coupled plasma-mass spectrometry

Abstract At present there is an increasing concern as regards the release of potentially toxic metals into the environment. Volcanic eruptions are a natural source of metals and metalloids in the atmosphere. Toxic trace elements ejected during an eruptive episode may produce hazardous effects for people and the environment in areas close to the volcano. In this context, a study was undertaken to investigate the concentration of metal and metalloids in ashes ejected from Copahue volcano, Neuquen, Argentina. Two samples (A and B) of deposited particles were collected one day after the first eruption and size-fractionated in four sub-samples (S1, S2, S3 and S4). Analysis was performed by inductively coupled plasma-mass spectrometry (ICP-MS) and the accuracy for the entire analytical procedure was performed by means of the certified reference material CRM GBW 07105 Rocks (NRCCRM, China). The elements considered were: As, Cd, Cr, Cu, Hg, Ni, Pb, Sb, U, V and Zn. The adverse effect of potentially interfering species on the mass-spectrometric determination of these elements was also taken into account. The concentration intervals found in the four fractions are as follows (in μg g −1 ): As, 6.0–2.6; Cd, 0.71–0.36; Cr, 29.5–54.0; Cu, 132–49.0; Hg, 0.020–0.007; Ni, 36.0–26.0; Pb, 15.5–2.55; Sb, 1.07–0.30; U, 2.57–1.94; V, 152–106; Zn, 85.5–55.0. The elements with the highest concentrations were: Cu, V and Zn. All fractions, in both samples, were found to be enriched in some toxic trace elements in the following order Sb>Cd>As. On the contrary, samples were depleted in Ni, Cr and Hg. Lead was the element that exhibits a noticeable difference in concentration between the finer and coarser fractions.

A study of uniformity of elements deposition on glass fiber filters after collection of airborne particulate matter (PM-10), using a high-volume sampler

A study was conducted to evaluate the homogeneity of the distribution of metals and metalloids deposited on glass fiber filters collected using a high-volume sampler equipped with a PM-10 sampling head. The airborne particulate matter (APM)-loaded glass fiber filters (with an active surface of about 500cm(2)) were weighed and then each filter was cut in five small discs of 6.5cm of diameter. Each disk was mineralized by acid-assisted microwave (MW) digestion using a mixture of nitric, perchloric and hydrofluoric acids. Analysis was performed by axial view inductively coupled plasma optical emission spectrometry (ICP OES) and the elements considered were: Al, As, Cd, Cr, Cu, Fe, Mn, Ni, Pb, Sb, Ti and V. The validation of the procedure was performed by the analysis of the standard reference material NIST 1648, urban particulate matter. As a way of comparing the possible variability in trace elements distribution in a particular filter, the mean concentration for each element over the five positions (discs) was calculated and each element concentration was normalized to this mean value. Scatter plots of the normalized concentrations were examined for all elements and all sub-samples. We considered that an element was homogeneously distributed if its normalized concentrations in the 45 sub-samples were within +/-15% of the mean value ranging between 0.85 and 1.15. The study demonstrated that the 12 elements tested showed different distribution pattern. Aluminium, Cu and V showed the most homogeneous pattern while Cd and Ni exhibited the largest departures from the mean value in 13 out of the 45 discs analyzed. No preferential deposition was noticed in any sub-sample.

Fractionation of elements by particle size of ashes ejected from Copahue Volcano, Argentina

The volcano Copahue, Neuquén province, Argentina has shown infrequent explosive eruptions since the 18th century. Recently, eruptive activity and seismicity were registered in the period July-October, 2000. As a consequence, ash clouds were dispersed by winds and affected Caviahue village located at about 9 km east of the volcano. Samples of deposited particles from this area were collected during this episode for their chemical analysis to determine elements of concern with respect to the health of the local population and its environment. Different techniques were used to evaluate the distribution of elements in four particle size ranges from 36 to 300 microm. X-ray powder diffraction (XRD) was selected to detect major components namely, minerals, silicate glass, fragments of rocks and sulfurs. Major and minor elements (Al, Ca, Cl, Fe, K, Mg, Mn, Na, S, Si and Ti), were detected by energy dispersive X ray analysis (EDAX). Trace element (As, Cd, Cr, Cu, Hg, Ni, Pb, Sb, U, V and Zn) content was quantified by inductively coupled plasma-mass spectrometry (ICP-MS). Nuclear activation analysis (NAA) was employed for the determination of Ce, Co, Cs, Eu, Hf, La, Lu, Rb, Sc, Sm, Ta and Yb. An enrichment was observed in the smallest size fraction of volcanic ashes for four elements (As, Cd, Cu and Sb) of particular interest from the environmental and human health point of view.

Air pollution sources of PM10 in Buenos Aires City

To elucidate the sources of PM10 air pollution from the experimental information collected in a local air quality monitoring campaign we have applied two methods, effective variance and genetic algorithms, in the solution of the chemical mass balance. The comparison of these two mathematical approaches show that the identification of the possible sources and the evaluation of its contributions are quite independent of them. The role of possible different sources for major and trace elements and the significance of standardizing available data is also addressed. We also present a simple method for identifying the number of candidate sources, a key element defining the dimension of the search space.

The mean angular distance among objects and its relationships with Kohonen artificial neural networks.

This job refers to classification of multidimensional objects and Kohonen artificial neural networks. A new concept is introduced, called the mean angular distance among objects (MADO). Its value can be calculated as the cosine of the mean centered vectors between objects. It can be expressed in matrix form for any number of objects. The MADO allows us to interpret the final organization of the objects in a Kohonen map. Simulated examples demonstrate the relationship between MADO and Kohonen maps and show a way to take advantage of the information present in both of them. Finally, a real analytical chemistry case is analyzed as an application on a big data set of an air quality monitoring campaign. It is possible to discover in it a subgroup of objects with different characteristics than those of the general trend. This subgroup is linked to the existence of an unidentified SO(2) source that, a priori, has not been taken into account.

2D mapping by Kohonen networks of the air quality data from a large city.

The 15-variable environmental data (7 concentrations: CO, SO2, O3, NOx, NO, NO2, particulate matter smaller than 10 micron (PM10), and 8 weather data: cloudiness, rainfall, insolation factor (Isfi), temperature, pressure at two locations, and wind intensity with direction) in a period of 45 days with 1-h intervals were extracted from a larger database of concentrations recorded in minute intervals for the same time period. The monitoring site was located in the City of Buenos Aires in a relatively heavy traffic crossroad of two avenues. The data required special pretreatment where the hourly content of rain, wind intensity, wind velocity, and cloudiness were concerned. The new variable named insolation factor (relative UV radiation) calculated on the basis of the general meteorological data, the geographic position of the monitoring site, cloudiness, date, and the time of the recording was composed. The relative intensity of UV radiation was modeled by a Gaussian function, multiplied by a cloudiness factor. Based on the 14-variable input and the 1-variable output (ozone) data, first, the clustering of all 980 data records was made. The top map clustering showing the ozone concentration was related to the maps of all 14 variables. The link between O3 clusters, NO2, and Isfi weight levels is shown and discussed. As a preliminary result of this study some of the most interesting correlations between the maps and remaining variables are given.