James W. Fitzgerald

Marine aerosols : a review

The background aerosol in the boundary layer over the remote oceans is not aged continental aerosol but, rather, is largely of marine origin. Total particle concentrations are quite uniform throughout the tropical trade wind regions and normally are in the range of 100–300 cm−3. Precipitation reduces particle concentrations, but there is apparently an in situ source of small particles which allows particle concentrations to recover to their normal background level. The fine particle mode (r < 0.3 μm), which comprises 90–95% of the particles but only about 5% of the total mass, cosists primarily of non-sea-salt sulfate (nss-sulfate). There is considerable evidence that nss-sulfate, which is present in concentrations ranging from about 0.2 to 1.5 μg m−3, is formed by gas-to-particle conversion of the oxidation products of organosulfur gases (principally DMS) emitted by the ocean. The principal gas-to-particle conversion mechanisms are particle formation by homogeneous nucleation of low-volatility gas-phase reaction products, condensation of these products on existing particles, and SO2-to-sulfate conversion in cloud droplets. The submicron portion of the particle size distribution is bimodal with peaks at 0.03 μm and 0.1 μm radius. The peak at 0.1 μm is believed to be due to the growth of CCN-sized particles as a result of incloud SO2-to-sulfate conversion. It has been speculated that the sea-to-air flux of DMS affects the number of CCN and thereby affects cloud droplet size, cloud albedo and, consequently, climate. Coarse particles (r > 0.5 μm) are composed primarily of sea salt. The concentration of sea salt shows a strong dependence of wind speed and ranges from about 2 μg m−3 to as much as 50 μg m−3 or more at wind speeds in excess of 15 m s−1. The background coarse mode also contains smaller amounts of nitrate and mineral dust. The concentration of each of these components is normally less than 5% of the mass of sea salt, although dust concentrations can occasionally equal the sea salt loading during fresh intrusions of continental dust. Nitrate is formed by gas-to-particle conversion but the relative importance of the ocean, the stratosphere and lightning as a source of the nitrogen-containing precursor gases remains uncertain. Since nitrate is not found on the fine mode particles, it probably does not result from condensation of gas-phase reaction products or from aqueous-phase oxidation of NOx in cloud droplets.

Marine boundary layer measurements of new particle formation and the effects nonprecipitating clouds have on aerosol size distribution

Measurements of aerosol size distributions (0.005 < r < 20 μm), cloud droplet spectra, SO2, O3, CN, and other supporting quantities were made in the cloud-topped and clear marine boundary layer (MBL) from an airship operating within about 50 km of the Oregon coast. Comparison of size distribution of interstitial aerosol within the cloud with the size distribution below the cloud clearly indicates that the processing of the aerosol through (nonprecipitating) stratus can lead to increased mass of the subset of particles which had served as cloud condensation nuclei (CCN). This increase in mass in the CCN results in a distinct “cloud residue” mode in the size distribution measured below the cloud. In all cases the aerosol mass in the cloud residue mode greatly exceeded the mass in the interstitial mode, even though the number concentration of interstitial particles sometimes exceeded the CCN concentration. Evidence of new particle formation in clear air was also found on numerous occasions. Analyses of the data indicate that the growth of newly formed particles into the observed size range is consistent with gas phase oxidation of SO2 to sulfate and subsequent condensation on the aerosol. However, the exact nucleation process, whether by homogeneous nucleation, ion-assisted nucleation, or heterogeneous nucleation on precursor embryos, is still an open question.

The Size and Scattering Coefficient of Urban Aerosol Particles at Washington, DC as a Function of Relative Humidity

Abstract The relative humidity dependence of the size and scattering coefficient of atmospheric aerosol particles was measured at Washington, DC during the period 26–31 July 1979. Particle growth curves (i.e., curves of the ratio r/r0, of particle radius to particle dry radius, versus relative humidity) were calculated using measured values of the particle composition parameter B ≡ ν¯ϕϵMwρ0/ρwMs, where ρ0 is the density of the particle in dry state, ρw and Mw the density and molecular weight of water, ϵ the mass fraction of soluble material in the particle. Ms the molecular weight of the soluble material, ν¯ the mean number of moles of ions per mole of solute, and ϕ¯ the mean value of the practical osmotic coefficient. To determine B, a mobility analyzer was used to transmit dry particles of nearly uniform size to a thermal gradient diffusion cloud chamber where the supersaturation (Sc,) necessary to activate the particles was measured. Then, a relationship between Sc, r0 and B was employed to determine...

A cloud chamber study of the effect that nonprecipitating water clouds have on the aerosol size distribution

When an air parcel in the atmosphere passes through a nonprecipitating cloud cycle, a subset of the aerosol population called cloud condensation nuclei (CCN) is activated and forms cloud droplets. During the cloud phase, trace gases, such as SO2, are dissolved into the droplets and undergo aqueous phase chemical reactions, forming low-volatility products, such as sulfates, that remain as residue when the cloud droplets evaporate. The resulting increase in residual mass can have a dramatic effect on the aerosol size distribution, causing the CCN to grow relative to the smaller particles (interstitial aerosol) which were not activated in the cloud. This process was graphically demonstrated in a series of experiments carried out in the Calspan 600-m3 environmental chamber, under conditions where the precloud reactants could be carefully controlled. Size distributions taken before and after a cloud cycle showed significant conversion of SO2 to H2SO4 and a dramatic change in the aerosol size distribution. Subs...

Deducing droplet concentration and supersaturation in marine boundary layer clouds from surface aerosol measurements

Air parcels in the marine boundary layer (MBL) are mixed up through nonprecipitating clouds at the top of the MBL many times (on average) before they can be removed by precipitation scavenging. The equivalent dry size of the particles (cloud condensation nuclei, CCN) upon which droplets are formed increases because of liquid phase oxidation of soluble trace gases during the cloud processing. The observed separation of the submicron size distribution into an interstitial mode and cloud droplet residue mode makes it possible to infer the effective MBL cloud supersaturation and cloud droplet concentrations from surface measurements of the aerosol size distribution during periods when nonprecipitating MBL clouds are present in the back trajectory and the MBL is well mixed. The effect of particle composition on the accuracy of the inferred cloud supersaturations is evaluated. A large database of hundreds of size distributions taken on an Atlantic and a Pacific cruise and an airship flight off the Oregon coast are used to calculate the range of effective MBL cloud supersaturations and droplet concentrations encountered during these expeditions. The inferred droplet concentrations on the Pacific cruise were mostly in the 25 to 150 cm -3 range, whereas they were mostly in the 50 to 500 cm -3 range for the Atlantic cruise. The inferred effective supersaturation in the tropical MBL clouds was typically in the 0.15% to 0.25% range. Recent work of Tang and Munkelwitz [1994] would indicate that particles consisting of mixtures of ammonium sulfate and sulfuric acid would not have recrystalized in the differential mobility analyzer (DMA) within the range of relative humidities (45% to 60%) at which the DMA was operated. At these humidities the hydrated size can be as much as 20% greater than the dry size. Corrections for the hydrated size within the DMA at the time of measurement are included and are also used to correct previous measurements of the relationship between dry size and critical supersaturation made using the Naval Research Laboratory (NRL) DMA and NRL thermal gradient CCN counter.

Model of the aerosol extinction profile in a well-mixed marine boundary layer

A simple model is described which predicts the vertical profile of aerosol extinction and the aerosol optical thickness at visible wavelengths in a well-mixed marine boundary layer, given the relative humidity and aerosol extinction coefficient at the surface (shipboard level) and the height of the boundary layer. The model is presented in the form of both analytical approximation formulas and nomograms.

Effect of relative humidity on the aerosol backscattering coefficient at 0.694- and 10.6-μm wavelengths

The effect of relative humidity on the backscattering of 0.694- and 10.6-μm radiation by aerosol particles in the lower troposphere is modeled. Two models of particle composition are considered: (1) all particles are composed of a uniform mixture of water-soluble material, dustlike material, and soot (uniform internal mixture) and (2) pure soot particles coexist with particles which are mixtures of water-soluble and dustlike materials (external mixture of soot). The amount of soot ranges from 1% to 20% of the volume of the aerosol. Changes in relative humidity have a greater effect on the backscattering coefficient, βπ, at 0.694 μm than at 10.6 μm. If soluble material accounts for 30% of the volume of mixed particles and if an urban type aerosol size distribution is assumed, an increase in relative humidity from 0% to 99% results in an increase in βπ at 0.694 μm ranging from a factor of 5.7 for an external mixture containing 20% soot by volume to a factor of 15.6 in the case of a uniform internal mixture containing 20% soot. At 10.6 μm the increase in βπ ranges from a factor of 2.1 to a factor of 2.8. The backscatter-to-extinction relation for 0.694-μm radiation propagating through a region of varying relative humidity is also investigated.

A Sound Velocity Method for Determination of Molecular Weight of Liquid Polymers

A new method is presented for determining number average molecular weights of liquid polymers, by means of easily performed measurements. The molecular weight is an explicit function M=Bdv13−A(n2−1n2+2) of the sound velocity v, density d, refractive index n, and the two empirical constants A and B. Accuracy of about 2 parts per hundred was attained for the lower polyethylene glycols. The method seems suitable for molecular weights up to several thousand. Sound velocities were measured by an acoustic interferometer, at one megacycle frequency.

A one-dimensional sectional model to simulate multicomponent aerosol dynamics in the marine boundary layer: 3. Numerical methods and comparisons with exact solutions

We present the theoretical framework and computational methods that were used by Fitzgerald et al. [this issue (a), (b)] describing a one-dimensional sectional model to simulate multicomponent aerosol dynamics in the marine boundary layer. The concepts and limitations of modeling spatially varying multicomponent aerosols are elucidated. New numerical sectional techniques are presented for simulating multicomponent aerosol growth, settling, and eddy transport, coupled to time-dependent and spatially varying condensing vapor concentrations. Comparisons are presented with new exact solutions for settling and particle growth by simultaneous dynamic condensation of one vapor and by instantaneous equilibration with a spatially varying second vapor.

The relative contribution of fluctuations in relative humidity and particulate concentrations to the variability of the scattering coefficient over the North Atlantic

Abstract Measurements of relative humidity, dry aerosol characteristics and the aerosol scattering coefficient in the visible were made in the marine boundary layer of the North Atlantic during a cruise of the USNS Hayes in Ma2-June 1977. An analysis of these measurements showed that fluctuations in the concentration of dry aerosol particles made a greater contribution to the variability of the scattering coefficient than did fluctuations in relative humidity. An aerosol growth curve was inferred from the variation in the scattering coefficient with relative humidity over a 2-h period during which the dry aerosol characteristics were quite constant This curve was found to be in reasonable agreement with the growth curves of large and giant marine aerosol particles obtained by Winkler (1977).