Aerosol Workshop — June 2-3, 1997

Session 2: Aerosol Distributions and Properties - Current Understanding

(Facilitator: John Seinfeld; Recorder Joe Prospero)

Aerosol Climatologies: What Modelers Need Most

Joyce Penner, University of Michigan

We now have the capability to simulate the full spectrum of aerosol types within global models: sulfate, organic carbon, black carbon, nitrate, ammonium, sea salt and dust, but there are significant differences between simulated and observed concentrations, particularly in the case of organics. It is crucial to determine the source of the organics measured in recent and past campaigns, because it introduces considerable uncertainty into the estimates of indirect forcing. If the concentration of background, natural organics is small, large estimates of indirect forcing result, but even modest increases in the background, natural aerosol can significantly reduce estimates of indirect forcing.

Direct forcing may be considerably smaller than indirect forcing, but is still not insignificant. A series of sensitivity tests with our model have shown that the uncertainty associated with the distribution of relative humidity can change our estimate of forcing (for sulfate aerosols) by 35%, while uncertainties associated with size distribution and composition (for biomass aerosols) can change estimates of forcing by 50%. There is a similar uncertainty associated with black (absorbing carbon), depending on whether it is assumed to be externally or internally mixed with other aerosol components.

We can begin to make progress in determining the total forcing by anthropogenic aerosols through a combination of comparison of these simulations with satellite data for aerosol and cloud properties and black, absorbing aerosols. Where differences are large, we will need a combination of ground-based and aircraft missions in order to determine the reasons for differences. Ground based and aircraft measurements together with modeling are needed to help understand the source of the discrepancy in predicted organics.

Carbonaceous Aerosols

Tica Novakov, Lawrence Berkeley Laboratory

The carbonaceous part of aerosol material is composed of two components: black (BC) and organic (OC) carbon. Light absorbing BC results entirely from incomplete combustion of fossil and biomass fuels. OC (principally light scattering) can be either directly produced by combustion sources or by gas-to-particle conversion of reactive anthropogenic and biogenic hydrocarbons. Information on concentration, distributions, and chemical composition of carbonaceous aerosols is sparse. Most measurement emphasized either inorganic or carbonaceous component - very few dealt with both. From surface based and recent airborne (TARFOX) measurements it is known that: OC is ubiquitously present at all locations; OC mass concentrations are comparable to and sometimes exceed sulfate concentrations; aerosol mass cannot be reconciled with inorganic species only; the carbon mass fraction may increase with altitude; OC and BC are significant contributors to light extinction; and, OC contributes to both natural and anthropogenic CCN concentrations.

This information, although fragmentary, strongly suggests that the carbonaceous aerosol component must play a role in both the direct and indirect aerosol forcing. Finally, available in-situ data are insufficient to assess temporal changes in anthropogenic carbonaceous aerosols from 1979 to present. However, known temporal changes in regional (fossil and biomass) fuel consumption might be used as a surrogate for emission modeling. Furthermore, effects of regional fuel use changes (e.g. in China and India) may be detectable in existing satellite data.

Implications of TARFOX Data for Aerosol Climate Forcing

Peter Hobbs, University of Washington

The simple paradigm that sulfate dominates aerosol column optical depth was not verified on the East Coast of the United States, where one would have thought it most likely to hold. Instead, after condensed water, organic aerosols dominated the optical depth. Similar data sets are needed for other regions to check this conclusion.

In view of the likely importance of organic aerosols, more information is needed on the vertical distribution of total (and organic and inorganic) carbonaceous aerosols, natural versus anthropogenic sources of carbonaceous aerosols, and the physical and chemical properties of these aerosols.

The following tabulated summary of initial TARFOX findings was provided after the meeting by Larry Stowe and Peter Hobbs:

Initial TARFOX Findings

  • Closure obtained between computed and observed optical depth from C-131 in situ and sun-photometer data [Hegg et al.]
  • Closure obtained between total aerosol mass and carbon and sulfate masses [Hegg, Hobbs and Novakov]
  • Mean carbon mass fraction is about 0.5; increases with altitude [Novakov et al.]
  • 66% of dry aerosol light scattering, on average, is from carbonaceous species [Hegg et al.]
  • Mean albedo of single scattering of dry aerosol is 0.9 [Hegg et al.]
  • Condensed water is dominant contributor to ambient aerosol optical depth. Sulfate is third, behind carbonaceous aerosol [Hegg et al.]
  • Humidification factor exceeds value expected for sulfate aerosol in half the cases observed [Kotchenruther and Hobbs]
  • Closure obtained between Wallops Island Raman lidar and sun-photometer measurements of optical depth at 355 nm [Ferrare et al.]
  • Closure obtained between AVHRR and Wallops Island AERONET sunphotometer optical depths at 0.63 microns over six days in July, but discrepancies exist between C-131 sunphotometer and AVHRR on July 25th using the NESDIS algorithm [Ignatov et al.]
  • Reasonable agreement obtained between C-131 sunphotometer and AVHRR estimates of aerosol optical depth at 0.63 microns on July 25th using the NPGS algorithm [Durkee et al.]
  • Reasonable agreement also obtained with C-131 sunphotometer optical depth at 0.62 microns using the NPGS algorithm applied to GOES-8 IMAGER reflectance data, using aerosol size distribution models derived from NOAA/14 AVHRR [Durkee et al.]
  • Closure obtained between computed and observed downward shortwave forcing vertical profiles from C-130 data [Hignett and Taylor]
  • Upward shortwave aerosol radiative forcing sensitivity, Fsw, observed by AVHRR at top of atmosphere on July 25th is significantly lower than theoretical computations [Stowe et al.]

Soil Dust Aerosols

Ina Tegen, NASA Goddard Institute for Space Studies

Soil dust aerosol dominates aerosol optical thickness over large continental areas where the highest optical thickness values are measured. Our knowledge of soil dust aerosol is still limited. The global annual source strength is estimated by various authors to lie between 500 and 5000 Mt/yr, the contribution of anthropogenically disturbed soils vs. natural soils is estimated to lie between 20 and 70%, and atmospheric lifetimes can vary between hours and 14 days depending on particle size.

Knowledge of dust optical properties is crucial for evaluation of its radiative forcing and the climate effect. Different assumptions for effective size and refractive indices can cause a change in the sign of the dust forcing. Calculation of the climate effect of dust with the GISS GCM produces statistically significant cooling of the order of 1 C over large areas, but it remains to be evaluated how these values would change for different radiative parameters of dust.

Another problem is to evaluate the relationship between interannual changes in the dust load and climate modes like NAO and ENSO. We also need to separate natural variability of dust and changes in dust load by anthropogenic disturbance of soil surfaces.

Aerosol Single Scatter Albedo: Report on Leipzig Workshop

John Ogren, NOAA CMDL

A small workshop was held in Leipzig, Germany in March, 1997, to assess the current state of knowledge of aerosol single-scattering albedo (SSA) relevant to climate models. A report from the workshop has been submitted to "Contr. to Atmos. Phys." (Heintzenberg et al., 1997). Some of the main conclusions from the workshop are:

  1. Source strengths used in global models of black carbon are highly uncertain, primarily because the emission factors were never intended for use in climate studies. Critically needed information on the mass fraction of submicrometer particles, the mass fraction of black carbon, and the mass absorption efficiency of black carbon in various combustion sources is greatly lacking.
  2. Observational data sets on SSA are inadequate for climate studies, primarily because of uncertainties in measurements of aerosol light absorption coefficient (Bap). These uncertainties largely result from lack of a common calibration standard for Bap, and a complete lack of knowledge of the dependence of Bap on relative humidity.
  3. Climate studies require knowledge of SSA to two significant figures over the range 0.8-1.0.

In the second part of my talk, I summarized the results of NOAA/CMDL's measurements of SSA. Values vary widely among the various airmass types studied (remote, marine, clean and polluted continental, and free troposphere), and even at any given site, the 75th percentile of the single-scattering co-albedo (1-SSA) was typically a factor of two greater than the 25th percentile. Especially over the continents, with their higher surface reflectivity compared to the oceans, the observed variations in SSA induce large variability in aerosol radiative forcing. As a result, it is inappropriate to represent SSA with a single value in climate studies, or in satellite data retrieval algorithms. A more appropriate approach is to consider a range of values, and calculate the sensitivity of the calculations to uncertainties in the appropriate value in SSA.

Workshop Homepage * Agenda * Summary
Sessions: 1, 2, 3, 4, 5, 6
Panels: A, B, C, D * Participants