Search

During the SAFARI 2000 field campaign, both smoke aerosols from savanna fires and haze aerosols in the boundary layer and in the free troposphere were collected from an aircraft in southern Africa. These aerosol samples were analyzed for their water-soluble chemical components, particularly the organic species. A novel technique, electrospray ionization-ion trap mass spectrometry, was used concurrently with an ion chromatography system to analyze for carbohydrate species. Seven carbohydrates, seven organic acids, five metallic elements, and three inorganic anions were identified and quantified. On the average, these 22 species comprised 36% and 27% of the total aerosol mass in haze and smoke aerosols, respectively. For the smoke aerosols, levoglucosan was the most abundant carbohydrate species, while gluconic acid was tentatively identified as the most abundant organic acid. The mass abundance and possible source of each class of identified species are discussed, along with their possible formation pathways. The combustion phase of a fire had an impact on the chemical composition of the emitted aerosols. Secondary formation of sulfate, nitrate, levoglucosan, and several organic acids occurred during the initial aging of smoke aerosols. It is likely that under certain conditions, some carbohydrate species in smoke aerosols, such as levoglucosan, were converted to organic acids during upward transport.

In order to verify performance, aerosol transport models (ATM) compare aerosol columnar mass (ACM) with those derived from satellite measurements. The comparison is inherently indirect since satellites derive optical depths and they use a proportionality constant to derive the ACM. Analogously, ATMs output a four dimensional ACM distribution and the optical depth is linearly derived. In both cases, the proportionality constant requires a direct intervention of the user by prescribing the aerosol composition and size distribution. This study introduces a method that minimizes the direct user intervention by making use of the new aerosol products of MODIS. A parameterization is introduced for the derivation of columnar aerosol mass (AMC) and CCN concentration (CCNC) and comparisons between sunphotometer, MODIS Airborne Simulator (MAS) and in-measurements are shown. The method still relies on the scaling between AMC and optical depth but the proportionality constant is dependent on the MODIS derived r$_{eff}$,\eta (contribution of the accumulation mode radiance to the total radiance), ambient RH and an assumed constant aerosol composition. The CCNC is derived fkom a recent parameterization of CCNC as a function of the retrieved aerosol volume. By comparing with in-situ data (ACE-2 and TARFOX campaigns), it is shown that retrievals in dry ambient conditions (dust) are improved when using a proportionality constant dependent on r$ {eff}$ and \eta derived in the same pixel. In high humidity environments, the improvement inthe new method is inconclusive because of the difficulty in accounting for the uneven vertical distribution of relative humidity. Additionally, two detailed comparisons of AMC and CCNC retrieved by the MAS algorithm and the new method are shown. The new method and MAS retrievals of AMC are within the same order of magnitude with respect to the in-situ measurements of aerosol mass. However, the proposed method is closer to the in-situ measurements than the MODIS retrievals. The retrievals of CCNC are also within the same order of magnitude for both methods. The new method is applied to an actual MODIS retrieval and although no in-situ data is available to compare, it is shown that the proposed method yields more credible values than the MODIS retrievals. In addition, recent data available from the PRIDE (Puerto Rico Dust Experiment, July 2000) will be shown by comparing sunphotometer, MODIS and in-situ data.

As part of the second Aerosol Characterization Experiment (ACE-2), European urban-marine and African mineral-dust aerosols were measured aboard the Pelican aircraft, the Research Vessel Vodyanitskiy from the ground and from satellites.

We report on clear-sky column closure experiments (CLEARCOLUMN) performed in the Canary Islands during the second Aerosol Characterization Experiment (ACE-2) in June/July 1997. We present CLEARCOLUMN results obtained by combining airborne sunphotometer and in-situ (a differential mobility analyzer, three optical particle counters, three nephelometers, and one absorption photometer) measurements taken aboard the Pelican aircraft, space-borne NOAA/AVHRR data and ground-based lidars. A wide range of aerosol types was encountered throughout the ACE-2 area, including background Atlantic marine, European pollution-derived, and (although less frequently than expected) African mineral dust. During the two days discussed here, vertical profiles flown in cloud free air masses revealed three distinctly different layers: a marine boundary layer (MBL) with varying pollution levels, an elevated dust layer, and a very clean layer between the MBL and the dust layer. Based on size-resolved composition information we have established an aerosol model that allows us to compute optical properties of the ambient aerosol using the optical particle counter results. In the dust, the agreement in layer AOD (lambda=380-1060 nm) is 3-8%. In the MBL there is a tendency for the in-situ results to be slightly lower than the sunphotometer measurements (10-17% at lambda=525 nm), but these differences are within the combined error bars of the measurements and computations. Aerosol size-distribudon closure based on in-situ size distributions and inverted sunphotometer extinction spectra has been achieved in the MBL (total surface area and volume agree within 0.2, and 7%, respectively) but not in the dust layer. The fact that the three nephelometers operated at three different relative humidities (RH) allowed to parameterize hygroscopic growth and to therefore estimate optical properties at ambient RH. The parameters derived for different aerosol types are themselves useful for the aerosol modeling community. The fact that the nephelometers and the absorption photometer sampled the aerosol through a cyclone make those measurements less useful for thee closure study carried out here. Large corrections (especially in the dust) had to be applied. Therefore, it is not surprising that closure with the sunphotometer was not always achieved. Agreement within 0.02 in AOD was achieved in the dust layer when the airorne sunphotometer extinction or AOD was compared to ground-based lidar measurements. We found that the presence of the elevated dust layers removes the good agreement between satellite and sunphotometer AOD usually found in the absence of the dust layer. We still compare the scattering phase functions used in the satellite retrieval with those obtained from the aerosol or the sunphotometer measurements.

Analyses of aerosol optical depth (AOD) and colurnmn water vapor (CWV) measurements acquired with NASA Ames Research Center's 6-channel Airborne Tracking Sunphotometer (AATS-6) operated aboard the R/V Professor Vodyanitskiy during the 2nd Aerosol Characterization Experiment (ACE-2) are discussed. Data are compared with various in situ and remote measurements for selected cases. The focus is on 10 July, when the Pelican airplane flew within 70 km of the ship near the time of a NOAA-14/AVHRR satellite overpass and AOD measurements with the 14-channel Ames Airborne Tracking Sunphotometer (AATS-14) above the marine boundary layer (MBL) permitted calculation of AOD within the MBL from the AATS-6 measurements. A detailed column closure test is performed for MBL AOD on 10 July by comparing the AATS-6 MBL AODs with corresponding values calculated by combining shipboard particle size distribution measurements with models of hygroscopic growth and radiosonde humidity profiles (plus assumptions on the vertical profile of the dry particle size distribution and composition). Large differences (30-80% in the mid-visible) between measured and reconstructed AODs are obtained, in large part because of the high sensitivity of the closure methodology to hygroscopic growth models, which vary considerably and have not been validated over the necessary range of particle size/composition distributions. The wavelength dependence of AATS-6 AODs is compared with the corresponding dependence of aerosol extinction calculated from shipboard measurements of aerosol size distribution and of total scattering mearured by a shipboard integrating nephelometer for several days. Results are highly variable, illustrating further the great difficulty of deriving column values from point measurements. AATS-6 CWV values are shown to agree well with corresponding values derived from radiosonde measurements during 8 soundings on 7 days and also with values calculated from measurements taken on 10 July with the AATS-14 and the University of Washington Passive Humidigraph aboard the Pelican.

We report on clear-sky column closure experiments (CLEARCOLUMN) performed in the Canary Islands during the second Aerosol Characterization Experiment (ACE-2) in June/July 1997. We present CLEARCOLUMN results obtained by combining airborne sunphotometer and in-situ (optical particle counter, nephelometer, and absorption photometer) measurements taken aboard the Pelican aircraft, space-borne NOAA/AVHRR data and ground-based lidar and sunphotometer measurements. During both days discussed here, vertical profiles flown in cloud-free air masses revealed 3 distinctly different layers: a marine boundary layer (MBL) with varying pollution levels, an elevated dust layer, and a very clean layer between the MBL and the dust layer. A key result of this study is the achievement of closure between extinction or layer aerosol optical depth (AOD) computed from continuous in-situ aerosol size-distributions and composition and those measured with the airborne sunphotometer. In the dust, the agreement in layer AOD (lambda = 380-1060 nm) is 3-8%. In the MBL there is a tendency for the in-situ results to be slightly lower than the sunphotometer measurements (10-17% at lambda = 525 nm), but these differences are within the combined error bars of the measurements and computations.

While Hegg et al. (1993) accepts the criticism of Anderson et al. (1994) in principle, this involves the adoption of an aerosol composition model and the model that they propose to reconcile these observations with the assertion of Charlson et al. (1992) does not agree with many observations, particularly those made over the North Atlantic Ocean. Although the use of a gain factor (i.e. the partial derivative of aerosol mass with respect to the sulfate ion), proposed by Anderson et al., may be valid for particular cases where a proposed composition model really reflects the actual aerosol composition, this procedure is considered questionable in general. The use of sulfate as a tracer for nonsulfate aerosol mass is questionable, because in the present authors' data set, sulfate averaged only about 26% of the dry aerosol mass. The ammonium mass associated with sulfate mass is not analogous to that betwen the oxygen mass and sulfur mass in the sulfate ion. Strong chemical bonds are present between sulfur and oxygen in sulfate, whereas ammonium and sulfate in haze droplets are ions in solution that may or may not be associated with one another. Thus, there is no reason to assume that sulfate will act as a reliable tracer of ammonium mass. Hegg et al. expresses the view that their approach used for estimating sulfate light scattering efficiency is appropriate for the current level of understanding of atmospheric aerosols.

Analysis of previously published, simultaneous measurements of cloud condensation nucleus number concentration and sulfate mass concentration suggest a nonlinear relationship between the two variables. This nonlinearity reduces the sensitivity of cloud albedo to changes in the sulfur cycle.

As part of NASA's Tropospheric Experiment Chemical Instrumentation Test and Evaluation (GTE/CITE 3) Sulfur Gas Intercomparison, we conducted measurements of dimethyl sulfide (DMS) and SO2 using two techniques well suited for sampling from an aircraft due to their simplicity of design. DMS was collected by preconcentration on gold wire preceded by a KOH-impregnated filter oxidant scrubber, and analyzed by gas chromatography with flame photometric detection. SO2 was collected on K2CO3/glycerol-impregnated filters and analyzed by ion chromatography. In blind tests, both techniques produced excellent agreement with National Institutes of Standards and Technology (NIST) standards. For field measurements, the DMS technique produced excellent correlation with the mean of the six different techniques intercompared. For SO2, the five techniques intercompared were rather poorly correlated, but correlations between the three techniques which passed NIST standards tests were somewhat better. Our SO2 filter measurements exhibited rather large uncertainties due to higher than normal variabiltiy of the filter blanks, which we believe was caused by extended storage in the field. In measurements conducted off the coast of Natal, Brazil, a diurnal afternoon minimum in DMS concentrations accompanied by a corresponding maximum in SO2 concentrations was observed. However, due to rather large uncertainties in the SO2 measurements, any conclusions about the SO2 trend must by considered tentative.