L.Willard Richards

PM10 source apportionment in California's San Joaquin valley

Abstract A PM10 (particulate matter with aerodynamic diameter equal to or less than 10 μm) aerosol study was carried out at six sites in Californias San Joaquin Valley (SJV) from 14 June 1988 to 9 June 1989, as part of the 1988–1989 Valley Air Quality Study (VAQS). Concentrations of PM10 and PM2.5 (particles with aerodynamic diameters equal to or less than 2.5 μm) mass, organic and elemental carbon, nitrate, sulfate, ammonium and elements were determined in 24-h aerosol samples collected at three urban (Stockton, Fresno, Bakersfield) and three non-urban (Crows Landing, Fellows, Kern Wildlife Refuge) locations during this period. The sources which contributed to ambient concentrations of PM10 were determined by applying the Chemical Mass Balance (CMB) receptor model using the source profiles determined specifically for that study area. The VAQS data indicates the federal 24-h PM10 standard of 150 μg m−3 was exceeded at four out of the six sites and for reasons which differ by season and by spatial region of influence. The annual average source contributions to the PM10 at Bakersfield, the site with the highest annual average, were 54% from primary geological material, 15% from secondary ammonium nitrate, 10% from primary motor vehicle exhaust, 8% from primary construction; the remaining 4% was unexplained. The results of the source apportionment at all sites show that geological contributions (fugitive dust from tilling, roadways and construction) are largest in summer and fall months, while secondary ammonium nitrate contributions (deriving from direct emissions of ammonia and oxides of nitrogen from agricultural activities and engine exhaust) are largest during winter months.

Size-segregated fine particle measurements by chemical species and their impact on visibility impairment in Denver☆

Abstract In the winter of 1987–1988, a field study was undertaken to identify the principal causes of visibility impairment in Denver so that a plan to improve air quality could be designed. One focus of the field study was the measurement of the mass of sulfate, nitrate, organic and elemental carbon in particles as a function of particle size. The analysis of those measurements led to the following conclusions: (1) emissions of ammonia in areas northeast of Denver contribute to the mass of fine particle ammonium nitrate in Denver; (2) soot is much less efficient at scattering light than the other constituents of the Denver brown cloud, and much more efficient at absorbing light; (3) meteorological conditions that produce episodes of visibility impairment in Denver can be distinguished by the associated size distribution of airborne fine particles and (4) a significant fraction of the particle mass of organic carbon measured in the 1987–1988 Denver Brown Cloud Study could have been adsorbed gases.

Comments on the oxidation of NO2 to nitrate—day and night

It has long been suspected that there are two important mechanisms for the formation of atmospheric paniculate nitrates. Here it is suggested that different mechanisms operate in the day and night. Formation of nitrate (gaseous nitric acid and paniculate nitrate) in the daytime appears to be predominantly caused by the hydroxyl radical, but other reaction pathways also participate. This reaction becomes much less important at night because the hydroxyl radical is primarily formed photochemically. At night, ozone aloft reacts with NO2 to form NO3, which rapidly reacts with NO2 to form N2O5. The N2O5 reacts with water in droplets or adsorbed on surfaces to form nitrate, sometimes in large particles. This pathway is unimportant in the daytime because N2O5 is in equilibrium with NO3, which is photolyzed as well as rapidly destroyed by NO, which in turn is present whenever there is NOx and sunlight. These mechanisms can account for the observation of high concentrations of large particle nitrate in the morning as well as nitrate in smaller particles in the afternoon in Los Angeles. They also predict that NOx is very rapidly oxidized to nitrate aloft at night by the ozone that is typically present, and this has been observed in a power plant plume. Since NOx is typically oxidized to nitrate more rapidly than SO2 is oxidized to sulfate in the daytime, and NOx oxidation continues aloft at night while SO2 oxidation typically does not, the daily average rate of oxidation of NOx to nitrate is much greater than that for the oxidation of SO2 to sulfate, especially in the winter. Because gaseous nitric acid is removed from the atmosphere much more rapidly than paniculate sulfate, and some fine particle nitrates can volatilize to release nitric acid, it is expected that the geographic region over which it is possible for NOx emissions to have an impact on regional haze and deposition chemistry is much smaller than for sulfur dioxide emissions.

The chemistry, aerosol physics, and optical properties of a western coal-fired power plant plume☆

Abstract Data obtained from the airborne measurements in the plume of the Navajo Generating Station (NGS) in June–July and December 1979 as part of the EPA project VISTTA are reported. Source test and airborne data for the ratios SO2/NOx, particulates/SO2 and the size distribution of the primary particles agreed to within the variability of the emissions. NO2 concentrations in the plume were in agreement with the photostationary steady state relations; there was no evidence for additional oxidant formation at distances up to 115 km. The formation of sulfate and nitric acid was strongly suppressed in the concentrated plume, where ozone is depleted. Sulfate conversions of less than 0.1 % were typically observed for plume ages of 2 to 3 h in the morning sunlight. The highest SO2 oxidation rates observed in the dilute plume were 0.8%h in the summer between 59 and 89 km in the late morning and 0.2 %/h in the winter between 93 and 108 km in the afternoon. Nitric acid was always observed in the plume and the rate of conversion of NOx to nitric acid was 3 to 10 times the rate of conversion of SO2 to sulfate. Ammonia concentrations were adequate to neutralize the secondary sulfate, but not to saturate the plume with ammonium nitrate. Particulate nitrates were not observed. New aerosol was reliably detected only in the 0.01 to 0.1 μm size range, which is ineffective at scattering light. Growth of particles larger than 0.1 μm was hard to detect in the presence of the variations in the background aerosol concentration. The emissions which affect the plume visibility are NOx and fly ash, which was predominantly in the 2–7 μm size range. Secondary aerosol formation in the NGS plume can be neglected in visibility models for distances up to 100km, but significant amounts of NOx were removed by oxidation. On the average, extinction due to fly ash and NO2 were equal for blue light for 7% conversion of NOx to NO2, and for green light at 26 % conversion. Fly ash always dominated extinction in the red. Telephotometer sight path flight data were obtained for plume visibility model validation.

Regional haze case studies in the southwestern U.S—I. Aerosol chemical composition

Abstract Aerosol chemical composition as a function of particle size was determined in the southwestern U.S.A. during four weeks of sampling in June, July and December, 1979 as a part of project VISITA. Samples were collected at two ground stations about 80 km apart near Page (AZ) and in two aircraft flying throughout the region. Several different size separating aerosol samplers and chemical analysis procedures were intercompared and were used in determining the size distribution and elemental composition of the aerosol. Sulfur was shown to be in the form of water soluable sulfate, highly correlated with ammonium ion, and with an average [NH + 4 ]/[SO 2− 4 ] molar ratio of 1.65. During the summer sampling period, three distinct regimes were observed, each with a different aerosol composition. The first, 24 h sampling ending 30 June, was characterized by a higher than average value of light scattering due to particles (b sp ) of 24 × 10 −6 m −1 and a fine particulate mass ( M f ) of 8.5 μg m −1 . The fine particle aerosol was dominated by sulfate and carbon. Aircraft measurements showed the aerosol was homogeneous throughout the region at that time. The second regime, 5 July, had the highest average b sp of 51 × 10 −6 m −1 during the sampling period with M f of 3.2 μgm −3 . The fine particle aerosol had nearly equal concentrations of carbon and ammonium sulfate. For all three regimes, enrichment factor analysis indicated fine and coarse particle Cu, Zn, Cl, Br, and Pb and fine particle K were enriched above crustal concentrations relative to Fe, indicating that these elements were present in the aerosol from sources other than wind blown dust. Particle extinction budgets calculated for the three regimes indicated that fine particles contributed most significantly, with carbon and (NH 4 ) 2 SO 4 making the largest contributions. Fine particle crustal elements including Si did not contribute significantly to the extinction budget during this study. The December sampling was characterized by very light fine particle loading with two regimes identified. One regime had higher fine mass and sulfate concentrations while the other had low values for all species measured.

Comparability between PM2.5 and Particle Light Scattering Measurements

Particle light scattering and PM2.5 (particles with aerodynamic diameters less than 2.5 μm) concentration data from air quality studies conducted over the past ten years wereexamined. Fine particle scattering efficiencies were determinedfrom statistical relationships among measured light scattering and fine and coarse mass concentrations. The resulting fine particle scattering efficiencies ranged from 1.7 m2 g-1at Meadview in the Grand Canyon to over 5 m2 g-1 in Mexico City. Most of the derived fine scattering efficiencieswere centered around 2 m2 g-1, which is considerablylower than most values reported from previous studies.

Extension of atmospheric dispersion models to incorporate fast reversible reactions

Abstract Atmospheric dispersion with simultaneous chemical reaction is important with many air pollutants. A method is presented for separating the chemical reaction calculations from the atmospheric dispersion calculations in a model where the chemistry can be represented by one or a series of stoichiometric reactions. It is shown that when the chemical reactions are fast and reversible, the concentration of all chemical species can be calculated from a system of algebraic equations along with dispersion calculations for an inert species, rather than by the integration of differential equations. The conditions under which this separation is valid are discussed. Practical applications of this method include modeling the conversion of emitted NO to NO 2 by reaction with ambient ozone, and modeling the reaction of emitted sulfuric acid with ambient ammonia.

Characterization of the regional haze in the southwestern United States

Abstract As part of project VISTTA, aircraft sampling flights over non-urban portions of Arizona and Utah were conducted on 6 days in June and July 1979. The spatial distribution of “regional haze” is mapped in terms of the particle light scattering coefficient. Fine particle elemental composition was obtained from impactor and filter samples integrated over each flight path. One sampling day, 6/29/79, is characterized by reduced visibility throughout the sampling region. Possible sources are discussed.

Airborne chemical measurements in nighttime stratus clouds in the Los Angeles Basin

Abstract An aircraft was used to collect cloudwater, aerosol, and gas samples and to make air quality and meteorological measurements in stratus clouds during five sampling sessions in the Los Angeles Basin between November 1981 and June 1985. The clouds form at night in a thin layer, so most samples were collected at altitudes between 600 and 750 m msl. There were strong vertical gradients in the air quality, but the cloudwater composition was horizontally quite uniform and the pH was typically just above 3.0. H2O2 was found in all cloudwater samples analyzed for this species. No analyses were performed for organic species other than aldehydes. All data below are averages for the May–June 1985 samples and except for S(IV) are similar to earlier data. The average mass composition of the species measured in the cloudwater was 48% NO3−, 23% SO42−, 7.6% NH4+ , 6.9% Cl−, 5.4% Na, 2.7% H2O2,1.3% Ca, 1.2% HCHO, 0.8% Mg, 0.7% H+, 0.5% K, 0.5% Pb, 0.4% Fe, 0.3% CH3CHO, 0.03% S(IV), and 0.02% Mn. Stated as equivalents, the cations averaged 42% H+, 27% NH4+, 20% Na+, and 11% other metals, and the anions averaged 51% NO3− , 32% SO42−, and 17% Cl−. Cloudwater species concentrations multiplied by the cloud liquid water content gave mean ambient concentrations of 22 μgm−3 NO3−, 11 μgm−3 SO42−, and 3.4 μgm−3 NH4+. It was estimated that if enough SO2 to consume the H2O2 were made available, the SO42− concentrations would increase by an average of 5.2 μgm−3 or an average percentage increase of 65%. In 1983–1985, the conversion of SO2 to sulfate was essentially complete and the conversion of NOx to nitrate was typically half to three-quarters complete in the clouds. Flights which followed air parcels showed that vertical mixing and removal processes such as droplet settling were important. Droplets can evaporate below the clouds and form large aerosol particles during the night and morning. Cloudwater composition and droplet size data were used to estimate size distributions and composition of the aerosol formed when the clouds evaporate. Volatilization of the HNO3 would leave an aerosol with an average of 7 μg m−3 NO3 and 11 μg m−3 SO42− with most of the aerosol mass in 0.4–1.0 um diameter size range.

The optical effects of fine-particle carbon on urban atmospheres

Abstract The effects of fine-particle C, such as diesel soot, on the optical properties of urban haze in the visible wavelength range were explored to determine the dominant effects and to see if simple parameters (such as visual range in green) provide an adequate measure of these effects. It is known that fine-particle C absorbs more strongly in the blue than in the red, so that when it is mixed with a white pigment, the resulting gray can appear somewhat brown. The possibility of a similar effect in urban hazes was investigated, but found not to occur. When the sun is overhead, the near-horizon sky chromaticities caused by mixtures of fineparticle C and non-absorbing aerosol can also be produced by non-absorbing aerosols alone. It is shown that absorbing aerosol will darken the horizon sky, and a simple equation for the radiance of the horizon sky is derived. The effect of absorbing aerosol on the distance at which white and black objects can be seen is calculated.