A Simple Neural Network for Estimating Emission Rates of Hydrogen Sulfide and Ammonia from Single Point Sources

Abstract

Neural networks have shown tremendous promise in modeling complex problems. This work describes the development and validation of a neural network for the purpose of estimating point source emission rates of hazardous gases. This neural network approach has been developed and tested using experimental data obtained for two specific air pollutants of concern in West Texas, hydrogen sulfide and ammonia. The prediction of the network is within 20% of the measured emission rates for these two gases at distances of less than 50 m. The emission rate estimations for ground level releases were derived as a function of seven variables: downwind distance, crosswind distance, wind speed, downwind concentration, atmospheric stability, ambient temperature, and relative humidity. A backpropagation algorithm was used to develop the neural network and is also discussed here. The experimental data were collected at the Wind Engineering Research Field Site located at Texas Tech University in Lubbock, Texas. Based on the results of this study, the use of neural networks provides an attractive and highly effective tool to model atmospheric dispersion, in which a large number of variables interact in a nonlinear manner.     

Methodology for Siting Ambient Air Monitors at the Neighborhood Scale

Abstract

In siting a monitor to measure compliance with U.S. National Ambient Air Quality Standards (NAAQS) for par-ticulate matter (PM), there is a need to characterize variations in PM concentration within a neighborhood-scale region to achieve monitor siting objectives. A simple methodology is provided here for the selection of a neighborhood-scale site for meeting either of the two objectives identified for PM monitoring. This methodology is based on analyzing middle-scale (from 100 to 500 m) data from within the area of interest. The required data can be obtained from widely available dispersion models and emissions databases.

The performance of the siting methodology was evaluated in a neighborhood-scale field study conducted in Hudson County, NJ, to characterize the area’s inhalable particulate (PM₁₀) concentrations. Air monitors were located within a 2- by 2-km area in the vicinity of the Lincoln Tunnel entrance in Hudson County. Results indicate the siting methodology performed well, providing a positive relationship between the predicted concentration rank at each site and the actual rank experienced during the field study. Also discussed are factors that adversely affected the predictive capabilities of the model.     

Assessing Sorbents for Mercury Control in Coal-Combustion Flue Gas

Abstract

Sorbent injection for Hg control is one of the most promising technologies for reducing Hg emissions from power-generation facilities, particularly units that do not require wet scrubbers for SO₂ control. Since 1992, EPRI has been assessing the performance of Hg sorbents in pilot-scale systems installed at full-scale facilities. The initial tests were conducted on a 5000-acfm (142-m³/min) pilot baghouse. Screening potential sorbents at this scale required substantial resources for installation and operation and did not provide an opportunity to characterize sor-bents over a wide temperature range.

Data collected in the laboratory and in field tests indicate that sorbents are affected by flue gas composition and temperature. Tests carried out in actual flue gas at a number of power plants also have shown that sorbent performance can be site-specific. In addition, data collected at a field site often are different from data collected in the laboratory, with simulated flue gas mixed to match the major components in the site’s gas. To effectively estimate the costs of Hg sorbent systems at different plants, a measure of sorbent performance in the respective flue gases must be obtained. However, injection testing at multiple facilities with large pilot systems is not practical.

Over the past five years, fixed-bed characterization testing, modeling studies, and bench-scale injection testing have been undertaken to develop a low-cost technique to characterize sorbent performance in actual flue gas and subsequently to project normalized costs for Hg removal prior to full-scale demonstration. This article describes the techniques used and summarizes field-testing results from two plants burning Powder River Basin (PRB) coal for commercial activated carbon and several other sorbent types. Full-scale projections based on the results and data collected on larger-scale systems also are included.     

The Steubenville Comprehensive Air Monitoring Program (SCAMP): Overview and Statistical Considerations

Abstract

Average concentrations of particulate matter with an aerodynamic diameter less than or equal to 2.5 μm (PM_2.5) in Steubenville, OH, have decreased by more than 10 μg/m³ since the landmark Harvard Six Cities Study¹ associated the city’s elevated PM_2.5 concentrations with adverse health effects in the 1980s. Given the promulgation of a new National Ambient Air Quality Standard (NAAQS) for PM_2.5 in 1997, a current assessment of PM_2.5 in the Steubenville region is warranted. The Steubenville Comprehensive Air Monitoring Program (SCAMP) was conducted from 2000 through 2002 to provide such an assessment. The program included both an outdoor ambient air monitoring component and an indoor and personal air sampling component. This paper, which is the first in a series of four that will present results from the outdoor portion of SCAMP, provides an overview of the outdoor ambient air monitoring program and addresses statistical issues, most notably autocorrelation, that have been overlooked by many PM_2.5 data analyses. The average PM_2.5 concentration measured in Steubenville during SCAMP (18.4 μg/m³) was 3.4g/m³ above the annual PM_2.5 NAAQS. On average, sulfate and organic material accounted for ～31% and 25%, respectively, of the total PM_2.5 mass. Local sources contributed an estimated 4.6 μg/m³ to Steubenville’s mean PM_2.5 concentration. PM_2.5 and each of its major ionic components were significantly correlated in space across all pairs of monitoring sites in the region, suggesting the influence of meteorology and long-range transport on regional PM_2.5 concentrations. Statistically significant autocorrelation was observed among time series of PM_2.5 and component data collected at daily and 1-in-4-day frequencies during SCAMP. Results of spatial analyses that accounted for autocorrelation were generally consistent with findings from previous studies that did not consider autocorrelation; however, these analyses also indicated that failure to account for autocorrelation can lead to incorrect conclusions about statistical significance.     

Second generation anticoagulant rodenticides in predatory birds: Probabilistic characterisation of toxic liver concentrations and implications for predatory bird populations in Canada

Second-generation anticoagulant rodenticides (SGARs) are widely used to control rodent pests but exposure and poisonings occur in non-target species, such as birds of prey. Liver residues are often analysed to detect exposure in birds found dead but their use to assess toxicity of SGARs is problematic. We analysed published data on hepatic rodenticide residues and associated symptoms of anticoagulant poisoning from 270 birds of prey using logistic regression to estimate the probability of toxicosis associated with different liver SGAR residues. We also evaluated exposure to SGARs on a national level in Canada by analysing 196 livers from great horned owls (Bubo virginianus) and red-tailed hawks (Buteo jamaicensis) found dead at locations across the country. Analysis of a broader sample of raptor species from Quebec also helped define the taxonomic breadth of contamination. Calculated probability curves suggest significant species differences in sensitivity to SGARs and significant likelihood of toxicosis below previously suggested concentrations of concern (<0.1mg/kg). Analysis of birds from Quebec showed that a broad range of raptor species are exposed to SGARs, indicating that generalised terrestrial food chains could be contaminated in the vicinity of the sampled areas. Of the two species for which we had samples from across Canada, great horned owls are exposed to SGARs to a greater extent than red-tailed hawks and the liver residue levels were also higher. Using our probability estimates of effect, we estimate that a minimum of 11% of the sampled great horned owl population is at risk of being directly killed by SGARs. This is the first time the potential mortality impact of SGARs on a raptor population has been estimated.     

Speciation and transport of arsenic in an acid sulfate soil-dominated catchment, eastern Australia

Factors controlling the transport of geogenically-derived arsenic from a coastal acid sulfate soil into downstream sediments are identified in this study with both solid-phase associations and aqueous speciation clearly critical to the mobility and toxicity of arsenic. The data from both sequential extractions and X-ray adsorption spectroscopy indicate that arsenic in the unoxidised Holocene acid sulfate soils is essentially non-labile in the absence of prolonged oxidation, existing primarily as arsenopyrite or as an arsenopyrite-like species, likely arsenian pyrite. Anthropogenically-accelerated pedogenic processes, which have oxidised this material over time, have greatly enhanced the potential bioavailability of arsenic, with solid-phase arsenic almost solely present as As(V) associated with secondary Fe(III) minerals present. Analyses of downstream sediments reveal that a portion of the arsenic is retained as a mixed As(III)/As(V) solid-phase, though not at levels considered to be environmentally deleterious. Determination of arsenic speciation in pore waters using high performance liquid chromatography/Inductively Coupled Plasma-Mass Spectrometry shows a dominance of As(III) in upstream pore waters whilst an unidentified As species reaches comparative levels within the downstream, estuarine locations. Pore water As(V) was detected at trace concentrations only. The results demonstrate the importance of landscape processes to arsenic transport and availability within acid sulfate soil environments.     

Impacts of atmospheric pollution on the plant communities of British acid grasslands

Air pollutants are recognised as important agents of ecosystem change but few studies consider the effects of multiple pollutants and their interactions. Here we use ordination, constrained cluster analysis and indicator value analyses to identify potential environmental controls on species composition, ecological groupings and indicator species in a gradient study of UK acid grasslands. The community composition of these grasslands is related to climate, grazing, ozone exposure and nitrogen deposition, with evidence for an interaction between the ecological impacts of base cation and nitrogen deposition. Ozone is a key agent in species compositional change but is not associated with a reduction in species richness or diversity indices, showing the subtly different drivers on these two aspects of ecosystem degradation. Our results demonstrate the effects of multiple interacting pollutants, which may collectively have a greater impact than any individual agent.     

Partitioning of metals in a degraded acid sulfate soil landscape: influence of tidal re-inundation

The oxidation and acidification of sulfidic soil materials results in the re-partitioning of metals, generally to more mobile forms. In this study, we examine the partitioning of Fe, Cr, Cu, Mn, Ni and Zn in the acidified surface soil (0-0.1 m) and the unoxidised sub-soil materials (1.3-1.5 m) of an acid sulfate soil landscape. Metal partitioning at this acidic site was then compared to an adjacent site that was previously acidified, but has since been remediated by tidal re-inundation. Differences in metal partitioning were determined using an optimised six-step sequential extraction procedure which targets the “labile”, “acid-soluble”, “organic”, “crystalline oxide”, “pyritic” and “residual” fractions. The surficial soil materials of the acidic site had experienced considerable losses of Cr, Cu, Mn and Ni compared to the underlying parent material due to oxidation and acidification, yet only minor losses of Fe and Zn. In general, the metals most depleted from the acidified surface soil materials exhibited the greatest sequestration in the surface soil materials of the tidally remediated site. An exception to this was iron, which accumulated to highly elevated concentrations in the surficial soil materials of the tidally remediated site. The “acid-soluble”, “organic” and “pyritic” fractions displayed the greatest increase in metals following tidal remediation. This study demonstrates that prolonged tidal re-inundation of severely acidified acid sulfate soil landscapes leads to the immobilisation of trace metals through the surficial accumulation of iron oxides, organic material and pyrite.     

Hydrogen peroxide stabilization in one-dimensional flow columns

Rapid hydrogen peroxide decomposition is the primary limitation of catalyzed H(2)O(2) propagations in situ chemical oxidation (CHP ISCO) remediation of the subsurface. Two stabilizers of hydrogen peroxide, citrate and phytate, were investigated for their effectiveness in one-dimensional columns of iron oxide-coated and manganese oxide-coated sand. Hydrogen peroxide (5%) with and without 25 mM citrate or phytate was applied to the columns and samples were collected at 8 ports spaced 13 cm apart. Citrate was not an effective stabilizer for hydrogen peroxide in iron-coated sand; however, phytate was highly effective, increasing hydrogen peroxide residuals two orders of magnitude over unstabilized hydrogen peroxide. Both citrate and phytate were effective stabilizers for manganese-coated sand, increasing hydrogen peroxide residuals by four-fold over unstabilized hydrogen peroxide. Phytate and citrate did not degrade and were not retarded in the sand columns; furthermore, the addition of the stabilizers increased column flow rates relative to unstabilized columns. These results demonstrate that citrate and phytate are effective stabilizers of hydrogen peroxide under the dynamic conditions of one-dimensional columns, and suggest that citrate and phytate can be added to hydrogen peroxide before injection to the subsurface as an effective means for increasing the radius of influence of CHP ISCO.     

Preparation of activated carbons from raw and biotreated agricultural residues for removal of volatile organic compounds

Activated carbons with diverse physical and chemical properties were produced from four agriculture residues, including raw barley husk, biotreated barley husk, rice husk, and pistachio shell. Results showed that with adequate steam activation (30-90 min, 50% H2O(g),/50% N2), activated carbons with surface areas between 360 and 950 m2 g(-1) were developed. Further increases in the activation time destroyed the pore structure of activated carbons, which resulted in a decrease in the surface area and pore volume. Biotreated agricultural residues were found to be suitable precursors for producing mesoporous activated carbons. The oxygen content of activated carbons increased with increasing activation time. Results from X-ray photoelectron spectroscopy examination further suggested that H2O molecules react with the carbon surface, enhancing the deconvoluted peak area of carbonyl and carboxyl groups. Equilibrium adsorption of toluene indicated that the adsorption capacities increased with an increase in the inlet toluene concentration and a decrease in temperature. The adsorption isotherms were successfully fitted with Freundlich, Langmuir, and Dubinin-Radushkevich equations. Activated carbons derived from agricultural residues appear to be more applicable to adsorb volatile organic compounds at a low concentration and high-temperature environment.