Introduction
Introduction
As noted in the preceding news feature (Bates and Gras, IGACtivities 4), uncertainties regarding the climatic impacts of anthropogenic aerosols are unacceptably large, both for validating current climate models and for predicting future regional and global climate. An important step in reducing these uncertainties is to measure the direct radiative forcing by tropospheric aerosols over various regions of the globe while simultaneously measuring the properties of the responsible aerosols. Intensive field programs, when extended with satellite measurements and validated retrieval algorithms, can provide the necessary data on both the aerosols and their radiative effects. The Tropospheric Aerosol Radiative Forcing Observational Experiment (TARFOX) will focus on providing this information for one of the most polluted regions of the world, namely, the United States (US) eastern seaboard.
To illustrate the coverage possible with satellite remote sensing of aerosol properties, Figure 1 shows a map of the mean June/July/August aerosol optical thickness derived from the NOAA/AVHRR operational product. These are approximate values, since an aerosol model is used to retrieve aerosol optical thickness from the measured satellite reflectances. TARFOX will provide a critical assessment of the correctness of these retrievals. The data in Figure 1 show a well-defined plume of aerosol extending from the US east coast over the Atlantic Ocean. This plume is well separated from the plume that extends from the west coast of Africa, across the Atlantic, and over the Caribbean, which consists predominantly of mineral dust. Hence, the US east coast plume provides an excellent opportunity to isolate and study aerosols generated by industrial activity, and to determine the magnitude and uncertainty of the direct radiative forcing due to these aerosols.
Goals
The principal goals of TARFOX are to:
To achieve these goals, the following principal questions have to be answered:
Implementation
Answering the questions posed above will require an extensive theoretical radiative transfer modeling effort, an intensive field measurement program, and integrated analyses. The specific Tasks of TARFOX are discussed below.
An accurate theoretical determination of the wavelength-dependent, direct radiative forcing due to tropospheric aerosols in a vertically inhomogeneous, absorbing and scattering atmosphere, must be carried out. These investigations should employ standard meteorological profiles, together with a compilation of various natural and anthropogenic aerosol optical properties, to examine the sensitivity of the radiative fluxes at various atmospheric levels to:
These computations should use measured aerosol radiative properties. There may be limitations in their use, but identifying these will help define the types of measurements that must be carried out in field programs. These same computations will be used to complete column-closure experiments by comparing the radiation perturbations measured at the top, bottom, and other levels of the atmosphere with calculations based on in situ and remote sensing measurements of the properties of the aerosol in the atmospheric column.
The direct, clear-sky radiative forcing (i.e., change in net radiative flux) caused by aerosols is related to the aerosol optical thickness through such factors as the underlying surface albedo and the aerosol chemical composition and size distribution. This forcing can be measured by surface and airborne radiometers and derived from satellite-measured radiances using retrieval algorithms. Similarly, optical thickness can be measured by surface and airborne photometers and radiometers, and also derived from satellite-measured radiances. Regressing measured radiative flux changes versus measured optical depths will provide an empirical measure of the sensitivity of radiative forcing to aerosol optical thickness. An important TARFOX objective is to compare this empirical sensitivity to results of computations that use aerosol chemical, physical, and optical properties obtained from simultaneous measurements.
Figure 2 summarizes the platforms and measurements planned for TARFOX. Because the effects of aerosol scattering on upwelling radiative flux are generally stronger over low-albedo surfaces, the satellite and aircraft measurements will focus on ocean areas, while also tying in the more continuous measurements at nearby land sites. To observe the strongest possible effects of anthropogenic aerosols, the field study will be conducted off the US east coast, in the area of the aerosol plume in Figure 1. The intensive field period (IFP), July 10-31, 1996, was selected using several years of satellite imagery, which show that this is the period of minimum clouds and maximum haze optical depth. In addition to the polar-orbiting satellites listed in Figure 2, TARFOX will use half-hourly images from the geosynchronous GOES-8 satellite to define regions most likely to be cloud-free and hazy during polar satellite overpasses. This information will be used in directing TARFOX aircraft, in selecting periods for intensive ground observations, and in post-mission analyses. The primary aircraft base for TARFOX will be the NASA Facility at Wallops Island, Virginia, with the possibility of some flights to Bermuda.
Daily satellite measurements during the IFP may provide sufficient aerosol variability to infer the background (natural) tropospheric aerosol optical thickness from the minima observed. The difference between the IFP mean aerosol optical thickness and the background value, when multiplied by the sensitivity factor described above, will yield a regional estimate of tropospheric aerosol radiative forcing for these summertime conditions. However, ascertaining what portion of this forcing is due to anthropogenic aerosols will require careful analysis of other simultaneous data on aerosol properties, precursor gases, tracers, radiatively active gases (e.g., water vapor), and surface albedo. Also, independent measurements of optical depths and radiative fluxes will be required to test and improve the satellite retrievals.
The airborne and surface measurements are designed to provide this simultaneous information. The imaging spectrometers on the ER-2 (MAS and AVIRIS in Figure 2) will provide retrievals of aerosol and surface properties with finer spatial, temporal, and spectral resolution than the satellite measurements. The ER-2 lidar (LASE in Figure 2) will provide vertical profiles of water vapor, aerosols, and clouds from the surface to 20 km. Flights of the medium- and low-altitude aircraft (C-130, C-131A, Pelican) will be embedded in these lidar vertical profiles. The lidar profiles will show the vertical context (including any unsampled aerosol or cloud layers) of in situ samples and radiative flux and optical depth measurements.
Figure 2 summarizes the measurements made by each of the medium- and low-altitude aircraft. Flights will be directed to satellite-determined hazy, cloud-free regions, or will span contrasting hazy and clean areas. Careful flight coordination will permit: (1) simultaneous radiative flux measurements above and below aerosol layers while the layers themselves are sampled, (2) simultaneous in situ aerosol measurements at different heights, to document differences within the column, (3) simultaneous optical depth measurements from below and within or above sampled aerosol layers, to separate tropospheric haze from overlying optical depths, and (4) vertical profiles of in situ samples, to provide column properties for comparison to photometer- and radiometer-derived optical depths.
Surface measurements will provide additional information at fixed locations, with greater temporal coverage. Locations will include long-term monitoring sites operated by the Atmospheric Ocean Chemistry Experiment (AEROCE) network in Bermuda and by NOAA/ERL, NASA/GSFC and the WMO Global Atmospheric Watch (GAW) program at other North Atlantic sites, plus special TARFOX measurements at Wallops Island.
The combined surface, air, and space data sets will permit a wide variety of closure analyses. Specifically, in-situ measurements of aerosol light-scattering, absorption, and forward/backscatter ratios at a given height will be used to derive the basic quantities needed to determine the effects of aerosols on solar radiation, namely, extinction, single-scattering albedo, and asymmetry factor. "Internal closure" will be assessed by comparing the quantities thus derived with those deduced from the simultaneous in situ measurements of particle size distribution and chemical composition. Analyses will seek to apportion the measured extinction by chemical species and hence (although more problematically) by source. The main tool for this analysis will be multiple linear regression. At each altitude in the vertical profiles of aerosol properties, the measured aerosol scattering and absorption coefficients will be regressed onto the co-measured mass concentrations of the various chemical components of the aerosol. This will yield the chemically specific mass scattering efficiencies and hence the relative contribution of each component to the light scattering at that altitude. Weighted vertical integration of these scattering budgets will then yield a chemical "budget." This statistical approach has proven fruitful in visibility studies, which are similar in this aspect to TARFOX.
"External column closure" will be assessed by vertically integrating the two types of aircraft-determined extinction and comparing the resulting optical depths with those simultaneously derived from the airborne sunphotometers, satellite radiometers, and ER-2 imaging spectrometers. Another aspect of column closure will be comparisons of aerosol radiative forcing, or radiative flux changes, determined by: (1) airborne flux radiometer measurements, (2) satellite flux retrievals from radiance measurements, and (3) flux calculations from (a) in-situ measured aerosol scattering, absorption, and asymmetry factors, (b) the same properties derived from size distribution and composition measurements, and (c) sunphotometer- and satellite-derived optical depths with ancillary single-scattering albedos and asymmetry factors. The sensitivity of radiative forcing to changes in aerosol optical thickness will be derived from the detailed in situ measurements and compared to the empirical sensitivity obtained by regressing satellite-derived radiative forcing vs. satellite-derived optical depth.
Such closure analyses will yield critically needed assessments and reductions of the uncertainties in deriving anthropogenic aerosol radiative forcing for use in climate models. The closure analyses that use satellite optical depth and flux results will provide tests and, where necessary, improvements of the satellite retrieval algorithms. The resulting validated algorithms will permit extensions of the TARFOX results beyond the TARFOX period and to other areas dominated by similar aerosols (e.g. the European Atlantic coast). Lessons learned and remaining questions will provide a useful background for ACE-2, which will study both direct and indirect effects of not only the European industrial aerosol plume, but also the African mineral dust plume.
| Return to Main |