Herman Sievering

Factor analysis of the MAP3S/RAINE precipitation chemistry network: 1976―1980

Abstract This paper presents a factor analysis of the MAP3S/RAINE precipitation chemistry data for the period 1976–1980. The purpose of the analysis is to determine the underlying source and/or processes accounting for the variance in the data. Three general sources, ‘acid,’ sea, and ‘ammonium/road dust/soil’ account for most of the variance in the data for the entire network for the aggregated three-and-one-half-year period. Both sulfate and nitrate are found to be significant components of the acid factor. Seasonal and spatial (individual site) analyses were able to distinguish other, less encompassing factors. Seasonal results showed a summer maximum in contribution for the acid factor, while individual site results showed that the contribution of the acid factor increases in going from the midwest to the northeast. Seasonal results show an ‘ammonium/soil’ factor appearing in spring and summer, while an ammonium sulfate factor appears in fall and winter. The results of this study should serve as a baseline for comparison with factor analyses of subsequent MAP3S data sets and factor analysis of other northeastern U.S. precipitation chemistry networks.

Trace element loading of southern Lake Michigan by dry deposition of atmospheric aerosol

Aerosol samples and meteorological data were collected at a mid-southern Lake Michigan site (87° 00′ W, 42° 00′ N) from May through September 1977. Hi-volume samplers with cellulose fiber filters and a digital meteorological data recording system were operated on board the U.S. EPAs R/VRoger R. Simons during four intensive sampling periods. Aerosol samples were analyzed by atomic absorption spectroscopy for seventeen trace elements. A diabatic drag coefficient method was used to determine aerosol deposition velocity (vd) overlake. A meanvd of 0.5 cm s−1 was found for the 0.1<diameter<2.0 μm aerosol. By relating the observed trace element concentrations andvd to a long-term climatological record, annual dry deposition loadings to the southern basin for nine elements were estimated. For four elements, Fe, Mn, Pb, and Zn, dry deposition loadings to the southern basin alone of at least 500, 30, 250, and 100 (×103 kg yr−1) were found. For Fe and Mn, these loadings represent about 15% of the total of all inputs to Lake Michigan. for Zn and Pb, about one-third to one-half of the annual loading from all sources is from dry deposition of atmospheric aerosol.

Profile measurements of particle mass transfer at the air-water interface

Abstract In the context of a study to determine trace element loading to Lake Michigan, a limited profile method was used to compare particle mass transfer with momentum transfer. Four profile data sets are considered in detail. Nine additional data sets taken during moderate wind speeds of 2.4–8.2 m s−1 resulted in a ratio of the profile measured particle deposition velocity vdp to the momentum transfer velocity v d of 0.48 ± 0.22. It was also found that this ratio, averaged across these nine data sets, varied by only two-fold in the particle diameter range 0.1–2.0 μm. Nine remaining moderate wind speed data sets showed no deposition trend. The overall mean ratio of v dp to v d for a total of 20 moderate wind speed profile data sets is 0.28. This rather large ratio is not in agreement with expectations. One explanation is that fluctuating wind speed and direction may cause the waters roughness elements to protrude through the viscous sublayer with the resultant surface resistance to particle transfer substantially reduced. In any event, fairly large transfer rates of atmospheric particles to open water bodies are suggested by these limited profile measurements.

Impact of particle characterization in confounding reported particle deposition velocities

Abstract Chemical and physical particulate matter data were collected at the site of previously reported particle dry deposition velocity (vd) measurements. These data suggest the reported vd of 0.38 ±0.29cms−1 for particles of diameter 0.15–0.3μm may be less than the true value for anthropogenic particles at this Boulder Atmospheric Observatory site. A small biological particle surface-derived upflux may have reduced particle profile gradients resulting in understated vd. In addition, the present data suggest these vd values are rather site specific and that measured particle vd at other air/land interface sites may also be site specific.

Trace element concentrations over midlake michigan as a function of meteorology and source region

Abstract During the period May–September 1977 aerosol samples were collected over southern Lake Michigan. Concentrations of Pb, Zn, Mn, Mo, Fe, Ti, Al, Ca and Mg in these samples were determined by emission spectroscopy analysis. The measurements also included several meteorological parameters. The results show that midlake concentrations are a strong function of temperature stability and linearly independent of wind speed in the lowest few meters above the lakes surface. There is also a strong dependence upon nearshore source regions. It is estimated that over 95% of the Pb, 75% of the Zn and Mn and more than half of the Fe midlake atmospheric concentrations are anthropogenic in origin and that fine particulates with high Pb, Zn and Mn content control the variability in midlake mass concentration.

Deposition of sulfate during stable atmospheric transport over lake michigan

Abstract Sulfate concentration measurements were made during very stable atmospheric conditions at-a Lake Michigan sampling site 55 km downwind of Chicago. Chicago sulfate concentrations are compared with the midlake concentrations. Dilution and transformation effects upon downwind sulfate concentrations are considered to be small during the very stable atmospheric conditions encountered. Depositional loss to the lake is then assumed to explain essentially all the differences between Chicago and midlake concentrations. The deposition velocity of sulfate to the lake is found to be 0.2 ± 0.16 cms −1 despite the very stable conditions prevailing.

Dry deposition loading of Lake Michigan by airborne particulate matter

A transport model as a function of particle size is presented which couples Chicago area meteorological data through the mixed layer with laboratory and field data on water surface dry deposition. The model is used to (a) estimate a minimum mass transfer efficiency by dry deposition from the Chicago area to Lake Michigan, (b) compare this minimum with the maximum wet deposition likely, and (c) estimate the transfer efficiency of trace metals.Dry deposition is found to cause 15% or more mass transfer efficiency of aerosols to the Lake. This value is considered to be equal to or greater than the efficiency of precipitation scavenging mass transfer. However, dry deposition as a function of particle size is such that the transfer of trace metals by this mechanism appears to be a few percent or less. Consequently, precipitation scavenging appears to be the dominant mechanism for trace metal transfer to Lake Michigan.

Atmospheric aerosol mass concentration and scattering coefficient in a midlake region

Abstract The relationship between aerosol mass concentration ( M ) and scattering coefficient ( B scat ) in urban and rural land-based environments has been well documented. In this study, comparison of M and B scat in the mid-Lake Michigan region has revealed a linear regression line B scat = (0.298 ± 0.036) + (0.0138 ± 0.0029) M with correlation coefficient 0.97. The slope of this regression is substantially less than the values reported previously in land-based settings. This difference is accounted for by the midlake aerosol size distribution, which exhibits a distinctly greater number concentration of secondary aerosols as measured by an active scattering aerosol spectrometer. The integrating nephelometer is confirmed to be a good instrument for in situ measurement of mass concentration. However, the slope of the linear regression line between M and B scat is dependent on the ambient aerosol size distribution and, therefore, on the sampling environment.

Trace element pass-through for cellulose filters when used for aerosol collection

Abstract Filter papers and impaction substrates made of cellulose fibers, such as Whatman 41 and Misco P810/252, are of considerable utility in the collection of aerosol for subsequent trace elemental analysts. This experiment evaluated the performance of Misco P810/252 in collecting trace elements, relative to a co-located standard glass fiber filter hi-vol collection. Sampling was conducted in varying meteorological conditions, so that results might be expressed in terms of environmental variables such as temperature and relative humidity. The pass-through factors presented here were derived from a series of environmental samples collected over land and over water. Overall, the Misco impactor/filter failed to collect 38% of Pb and 32% of Zn.

Nutrient Loading of Southern Lake Michigan by Dry Deposition of Atmospheric Aerosol

Aerosol samples and micrometeorological data were collected at 87°00’W 42°00’N in the mid southern Lake Michigan basin from May through September, 1977, to determine total phosphorus and nitrate-nitrite nitrogen particulate (TP + N) loadings. Hi-volume samplers with cellulose fiber filters and a meterological data collection system were operated on board the USEPAs R/V Roger R. Simons. A diabatic drag coefficient method was used to estimate aerosol deposition velocity (Vd), which had a mean value of 0.65 cm/s. TP + N concentration and Vd were determined for 42 distinct sampling periods of at least three to four hours each. A climatological data base was used to compute weighted average loading rates for TP + N. Dry deposition loading for the southern basin was found to be 0.15 to 0.18 × 106 kg/year for P; N loading was 3.5 to 5.1 × 106 kg/year. TP + N inputs via dry deposition account for 15% or more of all atmospheric nutrient inputs, and are significant nutrient sources to midlake waters.