The problem of air pollution is a frequently recurring situation and its management has social and economic considerable effects. Given the interaction of the numerous factors involved in the raising of the atmospheric pollution rates, it should be considered that the relation between the intensity of emission produced by the polluting source and the resulting pollution is not immediate. The aim of this study was to realise and to compare two support decision system (neural networks and multivariate regression model) that, correlating the air quality data with the meteorological information, are able to predict the critical pollution events. The development of a back-propagation neural network is presented to predict the daily PM₁₀ concentration 1, 2 and 3 days early. The measurements obtained by the territorial monitoring stations are one of the primary data sources; the forecasting of the major weather parameters available on the website and the forecasting of the Saharan dust obtained by the “Centro Nacional de Supercomputaciòn” website, satellite images and back trajectories analysis are used for the weather input data. The results obtained with the neural network were compared with those obtained by a multivariate linear regression model for 1 and 2 days forecasting. The relative root mean square error for both methods shows that the artificial neural networks (ANN) gives more accurate results than the multivariate linear regression model mostly for 1 day forecasting; moreover, the regression model used, in spite of ANN, failed when it had to fit spiked high values of PM₁₀ concentration.

A yearlong (December 2003 to February 2005) monitoring program was undertaken for urban roadside measurement of benzene, toluene, ethyl benzene, m- and p-xylene, and o-xylene (BTEX) at three different sites of Kolkata, India. The concentrations of monoaromatic hydrocarbons were found to be sufficiently high. Chemical mass balance model was applied to identify the sources and estimate their percentage contribution. Vehicular exhaust emission was found to be the dominant source of the target compounds and contributed 38.8-44.8% toward total volatile organic compound (VOC) level. Assuming that the vehicular exhaust fraction of the ambient BTEX level was due to the vehicular activity in the adjacent road of the monitoring site, vehicular emission factors for individual VOCs were estimated by running CALINE4 dispersion model in an inverse way. The total emission factor, average for all vehicles, was found to be in the range of 9.1 to 43.1 mg vehicle⁻¹ km⁻¹ for BTEX. From the measured vehicular composition during sampling, the category-wise emission factors for light-duty vehicles (LDVs), medium-duty (MDVs), and heavy-duty vehicles (HDVs), were also estimated by constrained nonlinear regression analysis. The emission factor of benzene for heavy, medium, and light vehicles was found to be 13.4, 21.0, and 31.2 mg vehicle⁻¹ km⁻¹ respectively.

Comprehensive survey of major rivers in the Tokyo metropolitan area was conducted for clarifying the emission sources of perfluorooctane sulfonate (PFOS) in Tokyo. PFOS was found at all sampling sites at concentrations ranging from 0.5 to 58 ng L⁻¹; in addition to this, it was also indicated that unknown PFOS emission sources are present in the midstream of the Tama River basin. The relationship between PFOS and perfluorooctanoic acid (PFOA) was constant at a ratio of 10:3 (PFOS/PFOA) throughout the Tama River basin. The sum of daily load amounts of PFOS from Tokyo's major rivers to Tokyo Bay reached 215 g day⁻¹. This value corresponds to 12.8 μg day⁻¹ per person using the sum of wastewater treatment district populations. In contrast, an estimation of PFOS contribution of domestic wastewater was also attempted, and the contribution was 1.6 μg day⁻¹ per person. We took samples up trunk sewers in the Tama River and finally found at the highest PFOS concentration (58,000 ng L⁻¹) from one of the wastewater of the electronic parts manufacturing facilities.

A simple algorithm for merging aerosol number particle size distribution (AMANpsd) data from multiple instruments is presented. The code uses a multilognormal model for the evaluation of merged aerosol particle size distribution data. AMANpsd is able to recalculate the aerodynamic diameter to the mobility diameter in the cases when, e.g., the Aerodynamic Particle Sizer data need to be connected to the data obtained by the Scanning Mobility Particle Sizer. The only assumption is the lognormality of the particle size distribution. As a result, the algorithm gives the parameters of the multilognormal model (particle number concentration, geometric mean diameter, and geometric standard deviation of each single lognormal mode). The algorithm was tested on a set of data from intensive indoor/outdoor measurement campaigns performed in Prague (Czech Republic), Oslo (Norway), and Milan (Italy). The particle size distribution data set covers the range of particle mobility diameters from 2 nm up to 8 μm. The agreement between the multilognormal model and the available experimental data from multiple instruments was satisfactory.

This paper presents experimental data on particulate matter (PM₁₀ and PM₂.₅) in the atmosphere of Candiota--South Brazil. Samples were collected using stacked filter units equipped with polycarbonate filters, which separate particles into two fractions: coarse 10-2.5 μm and fine <2.5 μm. The particulate matter was collected from January 2003 to September 2004 at three sampling sites: Acegua, 8 de Agosto, and Três Lagoas. The collected material was analyzed for Na, Mg, Al, Si, S, Cl, K, Ca, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Se, Br, Mo, Hg, and Pb, using particle-induced X-ray emission technique. Seasonal variation of the chemical elements with the size fraction was assessed. The elements K, Ca, Si, Ti, Fe, and Mn showed similar seasonal variability in the fine and coarse fractions with a low enrichment factor (<1.0) in both fractions, except for Mn and K, indicating common sources (soil dust, fly ash). Chemical species showing seasonal variability and high values of enrichment factor in fine fractions may indicate an anthropogenic origin.

The lability and mobility of Zn(II)-, Cd(II)-, Pb(II)-, and Cu(II)-humic acid complexes were studied using diffusive gradients in thin films (DGT). A unique feature of this research was (1) the use of DGTs with diffusive layer thicknesses ranging from 0.4 to 2.0 mm to study lability and mobility of Zn(II)-, Cd(II)-, Pb(II)-, and Cu(II)-humic acid complexes, combined with (2) the application of a competing ligand exchange (CLE) method using Chelex 100, the same chelating resin that is used in DGT, to study the kinetic speciation. The CLE experiments were run immediately after the completion of the DGT experiments, thereby allowing effects of the competing ligand to be separated from the effects introduced by the use of the polyacrylamide gel that is used in DGT. The results indicate that Zn(II) and Cd(II) tend to form more labile and more mobile complexes with humic acid than Pb(II) or Cu(II). The dissociation rate constants of Zn(II), Cd(II), and Pb(II) were found to increase with the ionic potential of the metal, suggesting that the binding between some trace metals and humic acid has a significant covalent component. Furthermore, the results suggest that the Eigen mechanism may not be strictly obeyed for metals such as Cu(II) which have high rate constants of water exchange, k w. Consequently, the markedly slow kinetics of Cu(II)-HA species suggests that the usual equilibrium assumption may not be valid in freshwaters.

A submerged membrane bioreactor (MBR) was used to treat two types of petrochemical wastewater: olefin process wastewater and total petrochemical wastewater. Various operational MBR conditions, such as influent pH and hydraulic retention time (HRT), were tested while these wastewaters were treated. The MBR treatment of olefin process wastewater reduced chemical oxygen demand (COD) and total organic carbon (TOC) by around 90% in both cases, and more than 90% of the suspended solids (SS) were removed. When MBR was used, the reduction in COD and TOC was also high for the total petrochemical wastewater, and was compared with the reduction obtained when the conventional activated sludge treatment plant was used (CASP). However, MBR effluent wastewaters showed high conductivity so for some reuse purposes subsequent reverse osmosis (RO) treatment would be needed. We characterized the MBR influent and MBR effluent wastewaters using a sequential solid phase extraction (SSPE) followed by gas chromatography-mass spectrometry (GC-MS) analyses. The main compounds that were tentatively identified were hydrocarbons, alkyl benzenes, phenols, acidic acids and esters, almost all of which could be partially or completely removed by MBR treatment.

In this study, a high-rate fibre filter was used as a pre-treatment to stormwater in conjunction with in-line flocculation. The effect of operating the fibre filter with different packing densities (105, 115 and 125 kg/m³) and filtration velocities (20, 40, 60 m/h) with and without in-line flocculation was investigated. In-line flocculation was provided using 5, 10 and 15 mg/L of ferric chloride (FeCl₃·6H₂O). The filter performance was studied in terms of pressure drop (ΔP), solids removal efficiency, heavy metals (total) removal efficiency and total organic carbon (TOC) removal efficiency. It is found that the use of in-line flocculation at a dose of 15 mg/L improved the performance of fibre filter as measured by turbidity removal (95%), total suspended solids reduction (98%), colour removal efficiency (99%), TOC removal (reduced by 30-40 %) and total coliform removal (93%). The modified fouling index reduced from 750-950 to 12 s/L² proving that fibre filter can be an excellent pre-treatment to membrane filtration that may be consider as post-treatment. The removal efficiency of heavy metal was variable as their concentration in raw water was small. Even though the concentration of some of these metals such as iron, aluminium, copper and zinc were reduced, others like nickel, chromium and cadmium showed lower removal rates.

Lake Koumoundourou is a shallow meromictic lake located ∼11 km NW of Athens. It is surrounded by various industries and oil refineries, which contaminate the lake by oil spills and leakage. Moreover, the lake receives freshwater from underwater springs, plus drainage from industrialized catchment and Athens landfill. Potential contamination of the lakes' sediments in heavy metals and metalloids was assessed by means of enrichment factors (EFs) estimated against local preindustrial core sediment; elements were normalized to loss on ignition. EFs revealed that surface sediments were enriched in Pb (×10.2), Cu (×6.7), V (×5.1), Ni (×4.1), and other heavy metals. The use of EFs is recommended as a reliable method for heavy metal contamination assessment, provided that (1) element contents are corrected following a careful normalization procedure, (2) local preindustrial sediment is used as reference, and (3) reference sediment should be unaffected by diagenetic alterations.

The Laucala Bay coastal system was investigated from 2003 to 2005 to evaluate the changes in water quality (nutrients) after improvements in sewage effluent quality from Kinoya sewage treatment plant (KSTP), whose output is discharged into the bay. The oxidized nitrogen (NO x -N) values averaged 1.31 µM, but varied from 0.76 to 3.77 µM, and the filterable reactive phosphate or orthophosphate (FRP) averaged 0.77 µM, ranging from 0.24 to 2.37 µM. The mean concentrations of NO x -N and FRP, respectively, were two and ten times higher than concentrations found in unpolluted sites in Fiji. Based on these values and NH₃ concentrations from parallel studies, Redfield ratio estimates suggest that primary productivity of the bay is N limited. No significant decrease in concentration of both NO x -N and FRP was found in Laucala Bay waters during the present investigation compared to concentrations found in two previous studies. This indicates that KSTP effluent may be only a minor source of nutrients into the bay, and to protect the Laucala Bay ecosystem from increasing nutrient pollution, there is a need to control nutrient input from other major point and nonpoint sources.