• Babu, S. S., , S. K. Satheesh, , and K. K. Moorthy, 2002: Aerosol radiative forcing due to enhanced black carbon at an urban site in India. Geophys. Res. Lett., 29 .1880, doi:10.1029/2002GL015826.

    • Search Google Scholar
    • Export Citation
  • Bates, T. S., , B. J. Hubert, , J. L. Gras, , F. B. Griffiths, , and P. A. Durkee, 1998: International Global Atmospheric Chemistry (IGAC) project’s First Aerosol Characterisation Experiment (ACE-1): An Overview. J. Geophys. Res., 103 , 1629716318.

    • Search Google Scholar
    • Export Citation
  • Chandra, S., , S. K. Satheesh, , and J. Srinivasan, 2004: Can the state of mixing of black carbon aerosols explain the mystery of “excess” atmospheric absorption? Geophys. Res. Lett., 31 .L19109, doi:10.1029/2004GL020662.

    • Search Google Scholar
    • Export Citation
  • Charlson, R. J., , S. E. Schwartz, , J. M. Hales, , R. D. Cess, , J. A. Coakley, , J. E. Hansen, , and D. J. Hoffmann, 1992: Climate forcing by anthropogenic aerosols. Science, 255 , 423430.

    • Search Google Scholar
    • Export Citation
  • Eck, T. F., , B. N. Holben, , O. Dubovik, , A. Smirnov, , I. Slutsker, , J. M. Lobert, , and V. Ramanathan, 2001: Column-integrated aerosol optical properties over the Maldives during the northeast monsoon for 1998-2000. J. Geophys. Res., 106 , 2855528566.

    • Search Google Scholar
    • Export Citation
  • Formenti, P., and Coauthors, 2002a: STAAARTE-MED 1998 summer airborne measurements over the Aegean Sea. 1. Aerosol particles and trace gases. J. Geophys. Res., 107 .4450, doi:10.1029/2001JD001337.

    • Search Google Scholar
    • Export Citation
  • Formenti, P., and Coauthors, 2002b: STAAARTE-MED 1998 summer airborne measurements over the Aegean Sea. 2. Aerosol scattering and absorption, and radiative calculations. J. Geophys. Res., 107 .4451, doi:10.1029/2001JD001536.

    • Search Google Scholar
    • Export Citation
  • Hess, M., , P. Koepke, , and I. Schult, 1998: Optical properties of aerosols and clouds: The software package OPAC. Bull. Amer. Meteor. Soc., 79 , 831844.

    • Search Google Scholar
    • Export Citation
  • Holben, B. N., and Coauthors, 1998: AERONET-A federated instrument network and data archive for aerosol characterization. Remote Sens. Environ., 66 , 116.

    • Search Google Scholar
    • Export Citation
  • Houghton, J. T., , L. G. Meira Filho, , J. P. Bruce, , H. Lee, , B. A. Callander, , and E. F. Haites, 1995: Climate Change 1994: Radiative Forcing of Climate Change and an Evaluation of the IPCC. Cambridge University Press, 347 pp.

    • Search Google Scholar
    • Export Citation
  • Jacobson, M., 2001: Strong radiative heating due to the mixing state of black carbon in atmospheric aerosols. Nature, 409 , 695697.

  • Kaufman, Y. J., and Coauthors, 1998: Smoke, clouds, radiation-Brazil (SCAR-B) experiment. J. Geophys. Res., 103 , 3178331808.

  • Liley, B. J., and Coauthors, 2003: Black carbon in aerosol during BIBLE B. J. Geophys. Res., 108 .8399, doi:10.1029/2001JD000845.

  • Lubin, D., , S. K. Satheesh, , G. Macfarquar, , and A. Heymsfield, 2002: The longwave radiative forcing of Indian Ocean tropospheric aerosol. J. Geophys. Res., 107 .8004, doi:10.1029/2001JD001183.

    • Search Google Scholar
    • Export Citation
  • Moorthy, K. K., , and S. K. Satheesh, 2000: Characteristics of aerosols over a remote island, Minicoy, in the Arabian Sea: Optical properties and retrieved size characteristics. Quart. J. Roy. Meteor. Soc., 126 , 81109.

    • Search Google Scholar
    • Export Citation
  • Moorthy, K. K., , S. K. Satheesh, , and B. V. Krishna Murthy, 1997: Investigations of marine aerosols over tropical Indian Ocean. J. Geophys. Res., 102 , 1882718842.

    • Search Google Scholar
    • Export Citation
  • Moorthy, K. K., and Coauthors, 1999: Aerosol climatology over India. 1 - ISRO GBP MWR network and database. ISRO GBP SR-03-99.

  • O’Neill, N. T., , T. F. Eck, , B. N. Holben, , A. Smirnov, , A. Royer, , and Z. Li, 2002: Optical properties of boreal forest fire smoke derived from sun photometry. J. Geophys. Res., 107 .4125, doi:10.1029/2001JD000877.

    • Search Google Scholar
    • Export Citation
  • Prospero, J. M., and Coauthors, 1983: The atmospheric aerosol system. Rev. Geophys. Space Phys., 21 , 16071629.

  • Quinn, P. K., and Coauthors, 2000: A comparison of aerosol chemical and optical properties from the 1st and 2nd Aerosol Characterization Experiments. Tellus, 52B , 239257.

    • Search Google Scholar
    • Export Citation
  • Ramanathan, V., and Coauthors, 2001: Indian Ocean experiment: An integrated analysis of the climate forcing and effects of the great Indo-Asian haze. J. Geophys. Res., 106 , 2837128398.

    • Search Google Scholar
    • Export Citation
  • Ricchiazzi, P., , S. Yang, , C. Gautier, , and D. Sowle, 1998: SBDART: A research and teaching tool for plane-parallel radiative transfer in the earth’s atmosphere. Bull. Amer. Meteor. Soc., 79 , 21012114.

    • Search Google Scholar
    • Export Citation
  • Russell, P. B., , P. V. Hobbs, , and L. L. Stowe, 1999: Aerosol properties and radiative effects in the United States East Coast haze plume: An overview of the Tropospheric Aerosol Radiative Forcing Observational Experiment (TARFOX). J. Geophys. Res., 104 , 22132222.

    • Search Google Scholar
    • Export Citation
  • Satheesh, S. K., 2002: Aerosol radiative forcing by Indian Ocean aerosols: Effect of cloud and surface reflection. Ann. Geophys., 20 , 21052109.

    • Search Google Scholar
    • Export Citation
  • Satheesh, S. K., , and K. Krishna Moorthy, 1997: Aerosol characteristics over coastal regions of the Arabian Sea. Tellus, 49B , 417428.

  • Satheesh, S. K., , and V. Ramanathan, 2000: Large differences in the tropical aerosol forcing at the top of the atmosphere and Earth’s surface. Nature, 405 , 6063.

    • Search Google Scholar
    • Export Citation
  • Satheesh, S. K., , and J. Srinivasan, 2002: Enhanced aerosol loading over Arabian Sea during the pre-monsoon season: Natural or anthropogenic? Geophys. Res. Lett., 29 .1874, doi:10.1029/2002GL015687.

    • Search Google Scholar
    • Export Citation
  • Satheesh, S. K., , V. Ramanathan, , X. L. Jones, , J. M. Lobert, , I. A. Podgorny, , J. M. Prospero, , B. N. Holben, , and N. G. Loeb, 1999: A model for the natural and anthropogenic aerosols for the tropical Indian Ocean derived from Indian Ocean experiment data. J. Geophys. Res., 104 , 2742127440.

    • Search Google Scholar
    • Export Citation
  • Satheesh, S. K., , V. Ramanathan, , B. N. Holben, , K. Krishna Moorthy, , N. G. Loeb, , H. Maring, , J. M. Prospero, , and D. Savoie, 2002: Chemical, microphysical, and radiative effects of Indian Ocean aerosols. J. Geophys. Res., 107 .4725, doi:10.1029/2002JD002463.

    • Search Google Scholar
    • Export Citation
  • Seinfeld, J. H., and Coauthors, 2004: ACE–ASIA: Regional, climatic, and atmospheric chemical effects of Asian dust and pollution. Bull. Amer. Meteor. Soc., 85 , 367380.

    • Search Google Scholar
    • Export Citation
  • Shaw, G. E., , J. A. Regan, , and B. M. Herman, 1973: Investigations of atmospheric extinction using direct solar radiation measurements made with a multiple wavelength radiometer. J. Appl. Meteor., 12 , 374380.

    • Search Google Scholar
    • Export Citation
  • Sumanth, E., , K. Mallikarjuna, , J. Stephen, , M. Mahesh, , V. Vinoj, , S. K. Satheesh, , and K. Krishna Moorthy, 2004: Measurements of aerosol optical depths and black carbon over Bay of Bengal during post-monsoon season. Geophys. Res. Lett., 31 .L16115, doi:10.1029/2004GL020681.

    • Search Google Scholar
    • Export Citation
  • Vinoj, V., , S. S. Babu, , S. K. Satheesh, , K. Krishna Moorthy, , and Y. J. Kaufman, 2004: Radiative forcing by aerosols over the Bay of Bengal region derived from shipborne, islandbased, and satellite (Moderate-Resolution Imaging Spectroradiometer) observations. J. Geophys. Res., 109 .D05203, doi:10.1029/2003JD004329.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    SSA and Ångström wavelength exponent of various aerosol species (sea salt, soot, sulfate, and mineral dust).

  • View in gallery

    The effect of varying the relative dominance of various aerosol species on the spectral variation of composite aerosol optical depth: (a) effect of increasing soot concentration, (b) effect of increasing sulfate concentration, and (c) effect of increasing sea salt concentration.

  • View in gallery

    As in Table 4 but for the variation of Ångström coefficients αs (UV–visible) and αl (near-IR) in response to an increase in relative dominance of various aerosol species.

  • View in gallery

    Various stages of iteration showing how iteration is converged. This also shows that the ultimate solution is independent of the initial assumption.

  • View in gallery

    Comparison of measured (average of February–March 1998) and estimated (using the proposed method) aerosol optical depths. The measurements were made over the tropical Indian Ocean during INDOEX.

  • View in gallery

    Comparison of measured (average of February–March 1998) and estimated (using the proposed method) aerosol composition. The measurements were made over the tropical Indian Ocean during INDOEX.

  • View in gallery

    Comparison of measured (during February–March 1998) and estimated (composition derived by the proposed method) surface radiative fluxes.

  • View in gallery

    Comparison (a) of measured and estimated surface radiative forcing; (b) mean difference between measured and simulated aerosol surface forcing.

  • View in gallery

    Comparison of optical depths (absorption) derived from AERONET observations (Ascension Island, Mauna Loa, and Nauru) and those estimated using the approach proposed here.

  • View in gallery

    Comparison of optical depths (absorption) derived from AERONET observations (Bermuda, Barbados, and Reunion) and those estimated using the approach proposed here.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 46 46 1
PDF Downloads 41 41 3

A Method to Estimate Aerosol Radiative Forcing from Spectral Optical Depths

View More View Less
  • 1 Centre for Atmospheric and Oceanic Sciences, Indian Institute of Science, Bangalore, India
© Get Permissions
Full access

Abstract

Radiative forcing of aerosols is much more difficult to estimate than that of well-mixed gases due to the large spatial variability of aerosols and the lack of an adequate database on their radiative properties. Estimation of aerosol radiative forcing generally requires knowledge of its chemical composition, which is sparse. Ground-based sky radiance measurements [e.g., aerosol robotic network (AERONET)] can provide key parameters such as the single-scattering albedo, but in shipborne experiments over the ocean it is difficult to make sky radiance measurements and hence these experiments cannot provide parameters such as the single-scattering albedo. However, aerosol spectral optical depth data (cruise based as well as satellite retrieved) are available quite extensively over the ocean. Spectral optical depth measurements have been available since the 1970s, and spectral turbidity measurements (carried out at meteorological departments all over the world) have been available for several decades, while long-term continuous chemical composition information is not available. A new method to differentiate between scattering and absorbing aerosols is proposed here. This can be used to derive simple aerosol models that are optically equivalent and can simulate the observed aerosol optical properties and radiative fluxes, from spectral optical depth measurements. Thus, aerosol single-scattering albedo and, hence, aerosol radiative forcing can be estimated. Note that the proposed method is to estimate clear-sky aerosol radiative forcing (over regions where chemical composition data or sky radiance data are not available) and not to infer its exact chemical composition. Using several independent datasets from field experiments, it is demonstrated that the proposed method can be used to estimate aerosol radiative forcing (from spectral optical depths) with an accuracy of ±2 W m−2.

Corresponding author address: S. K. Satheesh, Centre for Atmospheric and Oceanic Sciences, Indian Institute of Science, Bangalore 560 012, India. Email: satheesh@caos.iisc.ernet.in

Abstract

Radiative forcing of aerosols is much more difficult to estimate than that of well-mixed gases due to the large spatial variability of aerosols and the lack of an adequate database on their radiative properties. Estimation of aerosol radiative forcing generally requires knowledge of its chemical composition, which is sparse. Ground-based sky radiance measurements [e.g., aerosol robotic network (AERONET)] can provide key parameters such as the single-scattering albedo, but in shipborne experiments over the ocean it is difficult to make sky radiance measurements and hence these experiments cannot provide parameters such as the single-scattering albedo. However, aerosol spectral optical depth data (cruise based as well as satellite retrieved) are available quite extensively over the ocean. Spectral optical depth measurements have been available since the 1970s, and spectral turbidity measurements (carried out at meteorological departments all over the world) have been available for several decades, while long-term continuous chemical composition information is not available. A new method to differentiate between scattering and absorbing aerosols is proposed here. This can be used to derive simple aerosol models that are optically equivalent and can simulate the observed aerosol optical properties and radiative fluxes, from spectral optical depth measurements. Thus, aerosol single-scattering albedo and, hence, aerosol radiative forcing can be estimated. Note that the proposed method is to estimate clear-sky aerosol radiative forcing (over regions where chemical composition data or sky radiance data are not available) and not to infer its exact chemical composition. Using several independent datasets from field experiments, it is demonstrated that the proposed method can be used to estimate aerosol radiative forcing (from spectral optical depths) with an accuracy of ±2 W m−2.

Corresponding author address: S. K. Satheesh, Centre for Atmospheric and Oceanic Sciences, Indian Institute of Science, Bangalore 560 012, India. Email: satheesh@caos.iisc.ernet.in

1. Introduction

Radiative forcing due to aerosols is one of the largest sources of uncertainties in estimating anthropogenic climate perturbations (Charlson et al. 1992; Houghton et al. 1995). Aerosols are produced by various sources that are highly inhomogeneous in both time and space (Shaw et al. 1973; Prospero et al. 1983; Bates et al. 1998; Russell et al. 1999; Quinn et al. 2000). Thus, estimating aerosol radiative forcing is much more complicated than estimating radiative forcing due to well-mixed greenhouse gases (Houghton et al. 1995). To estimate aerosol radiative forcing knowledge of the chemical composition is generally required. The data on aerosol physical and optical characteristics (such as aerosol optical depth and size distribution) are more readily available than data on aerosol chemical composition. This is because the determination of chemical composition requires dedicated field experiments and expensive instrumentation. In this paper we use the term aerosol radiative forcing to represent the change in radiative fluxes due to the presence of aerosols. In other words, aerosol radiative forcing is the difference between radiative fluxes with aerosols and those without aerosols. The effect of aerosols on top of the atmosphere (TOA) radiative flux is TOA radiative forcing. Similarly, the effect of aerosols on surface radiative flux is surface radiative forcing. The difference between TOA forcing and surface forcing is atmospheric radiative forcing.

There are two common approaches to estimate aerosol forcing. In the first method, aerosol samples collected on paper or fiber filters are chemically analyzed to obtain the mass concentration of different aerosol species (Prospero et al. 1983). These are then converted to number distribution and subsequently to optical depths using Mie scattering theory (Satheesh et al. 1999; Satheesh and Srinivasan 2002). Here, either size-segregated measurements or assumption of size distribution are required. At submicron size range the aerosol mass available for chemical analysis is small and hence can lead to errors in the estimation of chemical composition. Moreover, data on chemical composition are sparse. In this case, sampling uncertainties are the major source of error. Moreover, surface-measured properties need to be converted to column properties by making assumptions about vertical profiles. In many cases, surface aerosol properties are entirely different from column aerosol properties owing to aerosol layers present aloft (Ramanathan et al. 2001; Satheesh et al. 2002; Lubin et al. 2002). Thus the assumption of the same vertical profile of aerosols for different days can result in large errors [as much as a factor of 2, Satheesh (2002)]. In the second method, the radiative flux is measured and, by subtracting it from the flux for an aerosol-free condition (simulated), an estimate of aerosol forcing is obtained (Satheesh and Ramanathan 2000). In this case, calibration uncertainties can cause large errors. Moreover, measurements covering all zenith angles are required for accurate estimation of diurnally averaged forcing. These are often not available, especially over the Tropics, due to cloudy skies (Satheesh et al. 1999). Ground-based sky radiance measurements (e.g., AERONET) can provide key parameters such as the single-scattering albedo (SSA; Holben et al. 1998). In shipborne experiments over the ocean, however, it is difficult to make sky radiance measurements, and hence one cannot derive parameters such as the single-scattering albedo. However, aerosol spectral optical depth data (ship based as well as satellite retrieved) are available quite extensively over the oceans. Spectral optical depth measurements have been available since the 1970s, and spectral turbidity measurements (carried out by meteorological departments all over the world) have been available for several decades, while long-term continuous chemical composition information is not readily available.

In this paper, we propose a new method to distinguish between scattering and absorbing aerosols. This can be used to derive simple aerosol models that are optically equivalent and can simulate the observed aerosol optical properties and radiative fluxes, from spectral optical depth measurements. Thus aerosol single-scattering albedo and hence aerosol radiative forcing can be estimated. The purpose of the proposed method is to estimate clear-sky aerosol radiative forcing and not to obtain the exact chemical composition of the aerosols. By incorporating this information in radiative transfer models, aerosol radiative forcing can be estimated. This method is not a substitute for the measurement of aerosol chemical composition, but a method to estimate aerosol radiative forcing over regions where aerosol chemical composition or sky radiance data are not available.

2. Effect of aerosol composition on spectral optical depths

Measurements of spectral optical depth are simple and they are the most accurate data available about aerosols (Shaw et al. 1973; Moorthy et al. 1997). These kinds of data are the most widely available compared to any other type of aerosol data (Kaufman et al. 1998; Holben et al. 1998; Moorthy et al. 1999; Moorthy and Satheesh 2000; Eck et al. 2001). The column optical depth estimated from sun photometers or satellite data is the sum of contributions from various aerosol species present in a column of the atmosphere. The purpose of this section is to demonstrate the impact of aerosol chemical composition on spectral optical depth.

To describe a wide range of possible aerosol compositions, Hess et al. (1998) modeled aerosols as containing different components, each of them meant to be representative of aerosols of certain origin. These components can be mixed together to form various aerosol mixtures (such as urban, continental clean, marine polluted, etc.) (Hess et al. 1998; Satheesh 2002). The physical (mode radii, standard deviation, density, etc.) and optical properties (single-scattering albedo) of the individual aerosol components are shown in Table 1. The components of some of the aerosol mixtures are given in Table 2 (Hess et al. 1998). The number density of different aerosol components and Ångström coefficients are different in different aerosol mixtures. The single-scattering albedo and Ångström wavelength exponent of various aerosol species are shown in Fig. 1.

The composite aerosol optical depth (τa) measured using a sun photometer is described as the sum of optical depths owing to the individual aerosol species
i1520-0469-63-3-1082-e1
where Qext is the extinction efficiency factor, which is a function of refractive index (m), radius (r), and wavelength (λ), and n(r) is the size distribution function. The subscript n denotes different aerosol species. The parameters S1, S2, etc., are scaling factors for the number density.
The aerosol size distribution function in Eq. (1) is represented by a lognormal distribution function. This type of function has the form
i1520-0469-63-3-1082-e2
where rmi and σi are the mode radii and standard deviation, respectively, and i represents each mode (Hess et al. 1998).

If n(r) in Eq. (1) is a normalized size distribution [with integral of n(r)dr equal to unity], then Sn are not only scaling factors, but also the total number densities of the respective aerosol components. By varying the scaling factors Sn in an iterative manner, different aerosol mixtures can be generated.

In Eq. (1),
i1520-0469-63-3-1082-e3
In other words,
i1520-0469-63-3-1082-e4
Here we make use of the fact that different aerosol species have different sizes and spectral optical properties (such as single-scattering albedo). Thus, a given change in any individual species would change the spectral variation of aerosol optical depths differently. For example, a change in submicron-sized aerosols would have more impact in visible regions compared to the near-infrared region. The effect of varying the relative dominance of various aerosol species is shown in Figs. 2a–c. For this purpose, we have assumed a four-component aerosol system containing sulfate, soot, sea salt, and dust. The optical depth for individual aerosol species was computed using Mie calculations. The parameters of aerosol size distributions [Eq. (2)] used here are given in Table 1 and refractive indices are given in Table 2. In Fig. 2a, case 2, case 3, and case 4 represent the change in spectral optical depth due to change in soot abundance, whereas in Figs. 2b and 2c, different cases (2 to 4) represent the change in optical depth due to change in abundance of sulfate and sea salt, respectively. Case 1 is the same for all three panels. The mass fractions in four cases (in Fig. 2) of soot are 1.6%, 3.14%, 5.14%, and 7.98%, respectively; those of sulfate are 64.4%, 70.7%, 75.1%, and 78.4%, respectively; and those of sea salt are 25.4%, 68.4%, 80.8%, and 87.4%, respectively. The total aerosol mass in four cases shown in three panels of Fig. 2 is different, and hence mass fractions of different cases in the three panels will not add up to 100%. Note that an increase in submicron aerosol (such as sulfate) abundance has a greater impact on shorter visible wavelengths than on near IR, and an increase in supermicron aerosol (such as sea salt) abundance has a greater impact on near-IR wavelengths than on visible wavelengths (Fig. 2).

A simple way of representing the spectral variation of aerosol optical depth (τa) is by using the Ångström power law (Shaw et al. 1973; Satheesh and Moorthy 1997). Here we define Ångström coefficients for two wavelength bands, one in the UV–visible (0.3 to 0.7 μm) and one near-IR band (0.7 to 1.2 μm). This is because the slope of the spectral optical depth may be different in UV–visible and near-IR regions depending on the size distribution. The Ångström wavelength exponent was calculated from optical depths in the two wavelength bands described above.

Thus
i1520-0469-63-3-1082-e5
and
i1520-0469-63-3-1082-e6
where α is the Angstrom wavelength exponent, β is the turbidity parameter, and λ is the wavelength in μm. The αs and αl correspond to wavelength exponents at short (0.3 to 0.7 μm) and long wavelength (0.7 to 1.2 μm), respectively. The value of α depends on the ratio of the concentration of large to small aerosols and β depends on the total aerosol loading in the atmosphere. The variation of αs and αl with mass fractions of different aerosol components is shown in Figs. 3a–c and Table 3. The four points shown in this figure correspond to the four cases shown in Figs. 2a–c. The first point is same in all three panels (Figs. 2a–c). It can be seen that a change in particular aerosol species (keeping the other species concentration the same) affects the spectral variation of aerosol optical depth differently. The variation of βs and βl with mass fractions of different aerosol components is shown in Table 4. Here, since β depends on the total aerosol loading, the value of β per unit mass is also shown in Table 4 as total mass changes in different cases.

The above result demonstrates that different aerosol types have different influences on spectral optical depths. This suggests that a method based on minimizing the error between the estimated and observed aerosol spectral optical depths should work well. This is discussed in the next section.

3. Proposed method

The first step in our approach is to estimate the Angstrom wavelength exponent αs and αl from spectral optical depth measurements. We make an initial assumption of the aerosol composition (zero order) [such as continental clean, marine polluted, desert, urban etc., in the model of Hess et al. (1998)]. The assumed initial composition has no impact on the final result, but minimizes the number of iterations required. The scaling factors (Si) in Eq. (3) are varied iteratively in order to make the absolute values of spectral optical depths simulated by aerosol mixtures consistent with those estimated from observations. The mean square difference between the observed (τaobs) and estimated (τaest) spectral optical depth is given by
i1520-0469-63-3-1082-e7
where p is the number of wavelengths.

The purpose of iteration is to minimize ε. The aerosol composition thus obtained should be able to simulate the observed spectral optical depths (and Ångström coefficients). In some cases, an overestimate in shorter wavelengths may be compensated by an underestimate in longer wavelengths and result in a minimum ε. To avoid this, in addition to Eq. (6), we impose a constraint that the difference between (τasτal)meas and (τasτal)est should be a minimum.

Figure 4 shows how the iteration is converged. Here, open circles (large) represent measured aerosol optical depths. We have used two different aerosol mixtures to initiate iteration. The aerosol optical depths corresponding to these mixtures are shown as open diamonds and open squares. As iteration proceeds, it can be seen that the final solution (cross and open triangle) converges close to the observed data (open circles).

In summary, when ε is at a minimum, we get an aerosol mixture (of various aerosol species) that can simulate the observed spectral optical depths. From individual size distributions weighted by the scaling factors, Si [following Eq. (1)], the spectral optical depths (and hence the Ångström coefficients) due to individual aerosol species can be estimated from the corresponding size distributions. Thus,
i1520-0469-63-3-1082-e8
where τai are the measured spectral optical depths, and αj and βj (j = 1, 4) are the Ångström coefficients of the four aerosol components. In this way we can estimate the contribution of major aerosol species to composite optical depth. The single scattering albedo of the composite aerosol is the weighted average (weighted by extinction coefficient) of the single-scattering albedo of individual aerosol species. By incorporating the measured aerosol optical depths, derived (using this method) values of the single-scattering albedo and other optical properties such as phase function in radiative transfer models, aerosol radiative forcing can be estimated.

4. Validation

The proposed method was validated using Indian Ocean Experiment (INDOEX) data, which comprise the only database available over the Indian region with simultaneous observations of radiative fluxes, aerosol optical depth, and chemical composition (Satheesh et al. 1999; Satheesh and Ramanathan 2000). From measurements at Kaashidhoo Climate Observatory (KCO), the Maldives, during February and March 1998, Satheesh et al. (1999) have derived an aerosol model for the tropical Indian Ocean. Aerosol samples were chemically analyzed at the University of Miami, Florida. To test the proposed method, the spectral optical depths measured at KCO were used. The aerosol components were varied iteratively so as to satisfy Eq. (6). Thus an optically equivalent aerosol mixture was derived using the proposed iterative method. The estimated aerosol spectral optical depth has been compared with observations in Fig. 5 (average of February–March 1998) where the vertical bars represent the error in observed aerosol optical depths. The quantity
i1520-0469-63-3-1082-e9
is 0.024 in this case, which is smaller than the errors in the observed optical depth (∼0.02 to 0.04).

A comparison of the INDOEX aerosol composition (average of February–March 1998) and derived aerosol composition is shown in Fig. 6. Aerosol chemical composition measurements need samples gathered for at least 24 h to get sufficient samples for analysis. Hence it is not possible to show the difference between measured and derived composition as a function of time or zenith angle. We compare the measured surface fluxes with that simulated by the radiative transfer model by incorporating the derived aerosol model (using the proposed method). We have used a Discrete Ordinate Radiative Transfer (RT) model developed by the University of Santa Barbara (SBDART) (Ricchiazzi et al. 1998). The radiative transfer calculations were made at full spectral resolution. The pyranometers (ventilated) and pyrheliometer used at KCO to measure the ground reaching solar (global, direct, and diffuse) fluxes are from the Kipp & Zonen group. There were two broadband ventilated Kipp & Zonen pyranometers, one shaded from the direct solar beam to measure the diffuse radiation and the other an unshaded instrument that measures the total (direct and diffuse) radiation (Satheesh et al. 1999, 2002). Thus, we determined the global radiation or the irradiance as the sum of direct solar radiation (pyrheliometer value multiplied by cosine of the solar zenith angle) and diffuse radiation (the shaded pyranometer). Since the accuracy of the pyrheliometer is better than the pyranometer by at least a factor of 2 and since the pyrheliometer does not have angular response errors (it is always normal to the incident solar radiation), this method of estimating global flux is more accurate than that measured by the unshaded pyranometer (Satheesh et al. 2002). All three instruments were calibrated at Kipp & Zonen as well as at the NOAA Global Monitoring Division (formerly Climate Monitoring and Diagnostic Laboratory) and the calibration coefficients provided by both of them agree within ±1% (Satheesh et al. 1999). A comparison of measured and simulated fluxes shows agreement within measurement accuracy and is shown in Fig. 7. The correlation coefficient is ∼0.97. We have used spectral optical depths here to derive the composition and independently measured radiative fluxes for validation. Both measurements are completely independent. While spectral optical depths are narrow band, radiative fluxes are broadband (0.25 to 4.0 μm). The estimated single scattering albedo is ∼0.866 compared to INDOEX measured column single-scattering albedo of ∼0.88 ± 0.04. When incorporated in the radiative transfer model, the estimated aerosol forcing efficiency (using the proposed approach) is −77 W m−2 at the surface and −22 W m−2 at the TOA, compared to −73 ± 5 and −25 ± 3 W m−2, respectively, observed during INDOEX, estimated using five independent radiometers and satellite data.

The comparison described above used average aerosol optical depths and composition. Next, we have used 35 sets of daily aerosol spectral optical depths measured during INDOEX to validate the method. For each day, aerosol forcing was estimated using the method described in this paper. A comparison of estimated and observed aerosol surface forcing is shown in Fig. 8a. The correlation coefficient is ∼0.95. The corresponding mean difference (between observed and estimated) is also shown as a function of aerosol optical depth (Fig. 8b). It can be seen that the mean difference between measured and observed forcing is ±2 W m−2, which is about 6% of the aerosol forcing reported over the tropical Indian Ocean (Satheesh and Ramanathan 2000). Aerosol radiative forcing values at the surface corresponding to urban aerosol and desert dust are −59 and −21.5 W m−2, respectively. These values were estimated using typical aerosol optical depths of 0.643 for urban aerosols and 0.286 for desert dust, following Hess et al. (1998). To demonstrate how sensitive aerosol radiative forcing is to different aerosol mixtures we have estimated the radiative forcing of different types of aerosol mixtures corresponding to an optical depth of 0.3 (which is typical for Northern Hemisphere continents and ocean). Aerosol radiative forcing values at the surface corresponding to urban, desert dust, marine clean and marine polluted are −27.6, −22.5, −12, and −17.4 W m−2, respectively.

The AERONET sites make simultaneous measurements of direct solar radiance and sky radiance (Holben et al. 1998). This enables partitioning of the aerosol optical depth into scattering and absorbing components. As mentioned earlier, sky radiance measurements during cruises in the ocean are difficult due to the ship’s movement. We have used AERONET data from six island locations to further validate our approach. The data used are from the following sites: (a) Ascension Island (7.98°S, 14.42°W; Atlantic), (b) Mauna Loa (19.54°N, 155.58°W; Pacific), (c) Nauru (0.52°S, 166.92°E; Pacific), (d) Bermuda (32.37°N, 64.7°W; Atlantic), (e) Barbados (13.17°N, 59.5°W; Atlantic), and (f) Reunion (20.88°S, 55.48°E; Indian Ocean). Comparison of absorption optical depths from these sites derived using sky radiance measurements with those estimated using the proposed method here is shown in Figs. 9 and 10 (correlation coefficients are provided in Table 5). The single-scattering albedo (ratio of scattering to extinction optical depths) reported from sky radiance measurements at four AERONET sites (where the database on absorption optical depth was sufficient) and that derived using the proposed method is shown in Table 6.

Shipborne measurements of aerosol black carbon and composite aerosol mass over the Bay of Bengal region have shown that the black carbon mass fraction over the Bay of Bengal reaches its maximum value during April/May (∼5.8%) and minimum during October/November (∼2.9%) (Vinoj et al. 2004; Sumanth et al. 2004). The corresponding values estimated following the proposed method (using spectral optical depths measured during the same period) were 5.6% and 3.4%, respectively. Airborne measurements over the Aegean Sea during the STAAARTE-MED experiment have shown aerosol optical depths of ∼0.39 (with single scattering albedo of ∼0.89) attributed to transport of polluted air masses from western and eastern Europe (Formenti et al. 2002a, b). The estimate of the single scattering albedo using the proposed method was 0.91. The optical properties of biomass burning in south central Africa (Zambia) were measured during Zambian IBBE (Eck et al. 2001; O’Neill et al. 2002) and a single scattering albedo of 0.82 was reported. The estimate of the single-scattering albedo using the proposed method was 0.79. The black carbon mass fraction reported over the tropical western Pacific was about 25% (Liley et al. 2003) and our estimate was 21%. During the field campaign at Bangalore (13°N, 77°E) in India, a large amount of black carbon was observed, both in absolute terms and fraction of composite aerosol mass (∼11%; Babu et al. 2002). We have used the spectral optical depths measured simultaneously and estimated a black carbon mass fraction of 13%.

5. Summary and conclusions

A new method to differentiate between scattering and absorbing aerosols has been proposed here. This method can be used to derive simple aerosol models from spectral optical depth measurements. These models are optically equivalent and can simulate the observed aerosol optical properties and radiative fluxes. Thus the aerosol single-scattering albedo and hence aerosol radiative forcing can be estimated over regions where chemical composition data or sky radiance data are not available.

The major advantage of the proposed method is that radiative forcing can be estimated with reasonable accuracy using spectral optical depth measurements, which are widely available. This method provides column aerosol properties (such as single-scattering albedo, percent contribution to optical depth, etc.). Satellite maps of spectral optical depths can be used to estimate regional forcing.

The major disadvantage is that aerosols with similar size and optical properties (e.g., sulfates and organics) cannot be distinguished and will be grouped together. Moreover, aerosol composition inferred using this approach is not unique, but provides an optically equivalent aerosol model that can simulate the observed spectral optical properties and radiative fluxes. This method assumes a constant size distribution for individual aerosol species as given in Hess et al. (1998). When aerosol optical depth is available at only one wavelength, this method is not applicable. The purpose of the proposed method is primarily to estimate aerosol radiative forcing from the spectral aerosol optical depth and not to infer the exact chemical composition. A 1-μm sea salt aerosol and a 1-μm sulfate aerosol (both having single-scattering albedo of close to unity at visible wavelengths) would have similar optical effects. Hence aerosols with similar size and refractive index cannot be distinguished using the method. However, as far as radiative forcing is concerned, this would not make any difference as aerosols with similar size and refractive index would have the same radiative forcing.

Investigations made as part of the Aerosol Characterization Experiments (ACE-Asia) have demonstrated that the dust we observe may not be just dust, but dust mixed with other aerosols (Seinfeld et al. 2004). Recent studies have suggested that, if one aerosol species is in a mixed state with another species in a core-shell structure, then the radiative impact can be significantly different than those of the externally mixed case or internally mixed cases (Jacobson 2001). Chandra et al. (2004), using direct observations and modeling, demonstrated that aerosol radiative properties change significantly corresponding to change in the state of mixing of aerosol. Data on aerosol mixing state, however, are sparse. The method proposed here has used the assumption of externally mixed aerosol. The method can be modified for various mixed aerosols if the state of mixing is known. In this case, we need to use Mie theory for layered spheres instead of spheres.

The main conclusions are as follows:

  1. A new method to distinguish between scattering and absorbing aerosols has been proposed. This can be used to derive simple aerosol models that are optically equivalent from spectral optical depth measurements.
  2. By incorporating these models in radiative transfer models, the aerosol radiative forcing can be estimated with an accuracy of ±2 W m−2.
  3. This method can be used to derive aerosol radiative forcing from spectral optical depths obtained from satellites.

Acknowledgments

The authors thank the Chairman, ISRO for supporting this work as part of the ISRO-RESPOND Programme. We thank B. N. Holben of NASA/GSFC for providing AERONET data for Ascension Island, Mauna Loa, Nauru, Bermuda, Barbados, and Reunion. We also thank K. Krishna Moorthy of SPL, VSSC for valuable suggestions and discussions.

REFERENCES

  • Babu, S. S., , S. K. Satheesh, , and K. K. Moorthy, 2002: Aerosol radiative forcing due to enhanced black carbon at an urban site in India. Geophys. Res. Lett., 29 .1880, doi:10.1029/2002GL015826.

    • Search Google Scholar
    • Export Citation
  • Bates, T. S., , B. J. Hubert, , J. L. Gras, , F. B. Griffiths, , and P. A. Durkee, 1998: International Global Atmospheric Chemistry (IGAC) project’s First Aerosol Characterisation Experiment (ACE-1): An Overview. J. Geophys. Res., 103 , 1629716318.

    • Search Google Scholar
    • Export Citation
  • Chandra, S., , S. K. Satheesh, , and J. Srinivasan, 2004: Can the state of mixing of black carbon aerosols explain the mystery of “excess” atmospheric absorption? Geophys. Res. Lett., 31 .L19109, doi:10.1029/2004GL020662.

    • Search Google Scholar
    • Export Citation
  • Charlson, R. J., , S. E. Schwartz, , J. M. Hales, , R. D. Cess, , J. A. Coakley, , J. E. Hansen, , and D. J. Hoffmann, 1992: Climate forcing by anthropogenic aerosols. Science, 255 , 423430.

    • Search Google Scholar
    • Export Citation
  • Eck, T. F., , B. N. Holben, , O. Dubovik, , A. Smirnov, , I. Slutsker, , J. M. Lobert, , and V. Ramanathan, 2001: Column-integrated aerosol optical properties over the Maldives during the northeast monsoon for 1998-2000. J. Geophys. Res., 106 , 2855528566.

    • Search Google Scholar
    • Export Citation
  • Formenti, P., and Coauthors, 2002a: STAAARTE-MED 1998 summer airborne measurements over the Aegean Sea. 1. Aerosol particles and trace gases. J. Geophys. Res., 107 .4450, doi:10.1029/2001JD001337.

    • Search Google Scholar
    • Export Citation
  • Formenti, P., and Coauthors, 2002b: STAAARTE-MED 1998 summer airborne measurements over the Aegean Sea. 2. Aerosol scattering and absorption, and radiative calculations. J. Geophys. Res., 107 .4451, doi:10.1029/2001JD001536.

    • Search Google Scholar
    • Export Citation
  • Hess, M., , P. Koepke, , and I. Schult, 1998: Optical properties of aerosols and clouds: The software package OPAC. Bull. Amer. Meteor. Soc., 79 , 831844.

    • Search Google Scholar
    • Export Citation
  • Holben, B. N., and Coauthors, 1998: AERONET-A federated instrument network and data archive for aerosol characterization. Remote Sens. Environ., 66 , 116.

    • Search Google Scholar
    • Export Citation
  • Houghton, J. T., , L. G. Meira Filho, , J. P. Bruce, , H. Lee, , B. A. Callander, , and E. F. Haites, 1995: Climate Change 1994: Radiative Forcing of Climate Change and an Evaluation of the IPCC. Cambridge University Press, 347 pp.

    • Search Google Scholar
    • Export Citation
  • Jacobson, M., 2001: Strong radiative heating due to the mixing state of black carbon in atmospheric aerosols. Nature, 409 , 695697.

  • Kaufman, Y. J., and Coauthors, 1998: Smoke, clouds, radiation-Brazil (SCAR-B) experiment. J. Geophys. Res., 103 , 3178331808.

  • Liley, B. J., and Coauthors, 2003: Black carbon in aerosol during BIBLE B. J. Geophys. Res., 108 .8399, doi:10.1029/2001JD000845.

  • Lubin, D., , S. K. Satheesh, , G. Macfarquar, , and A. Heymsfield, 2002: The longwave radiative forcing of Indian Ocean tropospheric aerosol. J. Geophys. Res., 107 .8004, doi:10.1029/2001JD001183.

    • Search Google Scholar
    • Export Citation
  • Moorthy, K. K., , and S. K. Satheesh, 2000: Characteristics of aerosols over a remote island, Minicoy, in the Arabian Sea: Optical properties and retrieved size characteristics. Quart. J. Roy. Meteor. Soc., 126 , 81109.

    • Search Google Scholar
    • Export Citation
  • Moorthy, K. K., , S. K. Satheesh, , and B. V. Krishna Murthy, 1997: Investigations of marine aerosols over tropical Indian Ocean. J. Geophys. Res., 102 , 1882718842.

    • Search Google Scholar
    • Export Citation
  • Moorthy, K. K., and Coauthors, 1999: Aerosol climatology over India. 1 - ISRO GBP MWR network and database. ISRO GBP SR-03-99.

  • O’Neill, N. T., , T. F. Eck, , B. N. Holben, , A. Smirnov, , A. Royer, , and Z. Li, 2002: Optical properties of boreal forest fire smoke derived from sun photometry. J. Geophys. Res., 107 .4125, doi:10.1029/2001JD000877.

    • Search Google Scholar
    • Export Citation
  • Prospero, J. M., and Coauthors, 1983: The atmospheric aerosol system. Rev. Geophys. Space Phys., 21 , 16071629.

  • Quinn, P. K., and Coauthors, 2000: A comparison of aerosol chemical and optical properties from the 1st and 2nd Aerosol Characterization Experiments. Tellus, 52B , 239257.

    • Search Google Scholar
    • Export Citation
  • Ramanathan, V., and Coauthors, 2001: Indian Ocean experiment: An integrated analysis of the climate forcing and effects of the great Indo-Asian haze. J. Geophys. Res., 106 , 2837128398.

    • Search Google Scholar
    • Export Citation
  • Ricchiazzi, P., , S. Yang, , C. Gautier, , and D. Sowle, 1998: SBDART: A research and teaching tool for plane-parallel radiative transfer in the earth’s atmosphere. Bull. Amer. Meteor. Soc., 79 , 21012114.

    • Search Google Scholar
    • Export Citation
  • Russell, P. B., , P. V. Hobbs, , and L. L. Stowe, 1999: Aerosol properties and radiative effects in the United States East Coast haze plume: An overview of the Tropospheric Aerosol Radiative Forcing Observational Experiment (TARFOX). J. Geophys. Res., 104 , 22132222.

    • Search Google Scholar
    • Export Citation
  • Satheesh, S. K., 2002: Aerosol radiative forcing by Indian Ocean aerosols: Effect of cloud and surface reflection. Ann. Geophys., 20 , 21052109.

    • Search Google Scholar
    • Export Citation
  • Satheesh, S. K., , and K. Krishna Moorthy, 1997: Aerosol characteristics over coastal regions of the Arabian Sea. Tellus, 49B , 417428.

  • Satheesh, S. K., , and V. Ramanathan, 2000: Large differences in the tropical aerosol forcing at the top of the atmosphere and Earth’s surface. Nature, 405 , 6063.

    • Search Google Scholar
    • Export Citation
  • Satheesh, S. K., , and J. Srinivasan, 2002: Enhanced aerosol loading over Arabian Sea during the pre-monsoon season: Natural or anthropogenic? Geophys. Res. Lett., 29 .1874, doi:10.1029/2002GL015687.

    • Search Google Scholar
    • Export Citation
  • Satheesh, S. K., , V. Ramanathan, , X. L. Jones, , J. M. Lobert, , I. A. Podgorny, , J. M. Prospero, , B. N. Holben, , and N. G. Loeb, 1999: A model for the natural and anthropogenic aerosols for the tropical Indian Ocean derived from Indian Ocean experiment data. J. Geophys. Res., 104 , 2742127440.

    • Search Google Scholar
    • Export Citation
  • Satheesh, S. K., , V. Ramanathan, , B. N. Holben, , K. Krishna Moorthy, , N. G. Loeb, , H. Maring, , J. M. Prospero, , and D. Savoie, 2002: Chemical, microphysical, and radiative effects of Indian Ocean aerosols. J. Geophys. Res., 107 .4725, doi:10.1029/2002JD002463.

    • Search Google Scholar
    • Export Citation
  • Seinfeld, J. H., and Coauthors, 2004: ACE–ASIA: Regional, climatic, and atmospheric chemical effects of Asian dust and pollution. Bull. Amer. Meteor. Soc., 85 , 367380.

    • Search Google Scholar
    • Export Citation
  • Shaw, G. E., , J. A. Regan, , and B. M. Herman, 1973: Investigations of atmospheric extinction using direct solar radiation measurements made with a multiple wavelength radiometer. J. Appl. Meteor., 12 , 374380.

    • Search Google Scholar
    • Export Citation
  • Sumanth, E., , K. Mallikarjuna, , J. Stephen, , M. Mahesh, , V. Vinoj, , S. K. Satheesh, , and K. Krishna Moorthy, 2004: Measurements of aerosol optical depths and black carbon over Bay of Bengal during post-monsoon season. Geophys. Res. Lett., 31 .L16115, doi:10.1029/2004GL020681.

    • Search Google Scholar
    • Export Citation
  • Vinoj, V., , S. S. Babu, , S. K. Satheesh, , K. Krishna Moorthy, , and Y. J. Kaufman, 2004: Radiative forcing by aerosols over the Bay of Bengal region derived from shipborne, islandbased, and satellite (Moderate-Resolution Imaging Spectroradiometer) observations. J. Geophys. Res., 109 .D05203, doi:10.1029/2003JD004329.

    • Search Google Scholar
    • Export Citation
Fig. 1.
Fig. 1.

SSA and Ångström wavelength exponent of various aerosol species (sea salt, soot, sulfate, and mineral dust).

Citation: Journal of the Atmospheric Sciences 63, 3; 10.1175/JAS3663.1

Fig. 2.
Fig. 2.

The effect of varying the relative dominance of various aerosol species on the spectral variation of composite aerosol optical depth: (a) effect of increasing soot concentration, (b) effect of increasing sulfate concentration, and (c) effect of increasing sea salt concentration.

Citation: Journal of the Atmospheric Sciences 63, 3; 10.1175/JAS3663.1

Fig. 3.
Fig. 3.

As in Table 4 but for the variation of Ångström coefficients αs (UV–visible) and αl (near-IR) in response to an increase in relative dominance of various aerosol species.

Citation: Journal of the Atmospheric Sciences 63, 3; 10.1175/JAS3663.1

Fig. 4.
Fig. 4.

Various stages of iteration showing how iteration is converged. This also shows that the ultimate solution is independent of the initial assumption.

Citation: Journal of the Atmospheric Sciences 63, 3; 10.1175/JAS3663.1

Fig. 5.
Fig. 5.

Comparison of measured (average of February–March 1998) and estimated (using the proposed method) aerosol optical depths. The measurements were made over the tropical Indian Ocean during INDOEX.

Citation: Journal of the Atmospheric Sciences 63, 3; 10.1175/JAS3663.1

Fig. 6.
Fig. 6.

Comparison of measured (average of February–March 1998) and estimated (using the proposed method) aerosol composition. The measurements were made over the tropical Indian Ocean during INDOEX.

Citation: Journal of the Atmospheric Sciences 63, 3; 10.1175/JAS3663.1

Fig. 7.
Fig. 7.

Comparison of measured (during February–March 1998) and estimated (composition derived by the proposed method) surface radiative fluxes.

Citation: Journal of the Atmospheric Sciences 63, 3; 10.1175/JAS3663.1

Fig. 8.
Fig. 8.

Comparison (a) of measured and estimated surface radiative forcing; (b) mean difference between measured and simulated aerosol surface forcing.

Citation: Journal of the Atmospheric Sciences 63, 3; 10.1175/JAS3663.1

Fig. 9.
Fig. 9.

Comparison of optical depths (absorption) derived from AERONET observations (Ascension Island, Mauna Loa, and Nauru) and those estimated using the approach proposed here.

Citation: Journal of the Atmospheric Sciences 63, 3; 10.1175/JAS3663.1

Fig. 10.
Fig. 10.

Comparison of optical depths (absorption) derived from AERONET observations (Bermuda, Barbados, and Reunion) and those estimated using the approach proposed here.

Citation: Journal of the Atmospheric Sciences 63, 3; 10.1175/JAS3663.1

Table 1.

Properties of various aerosol components.

Table 1.
Table 2.

Refractive indices of various aerosol components [from Hess et al. (1998)].

Table 2.
Table 3.

Components of various aerosols: α is the Ångström wavelength exponent defined by Eqs. (5) and (6), and αs and αl correspond to short (0.3–0.7 μm) and long wavelength range (0.7–1.2 μm).

Table 3.
Table 4.

The variation of β with mass fractions of different aerosols: β is the Ångström turbidity parameter defined by Eqs. (5) and (6), and βs and βl correspond to short (0.3–0.7 μm) and long wavelength range (0.7–1.2 μm).

Table 4.
Table 5.

Correlation between measured and estimated absorption optical depths.

Table 5.
Table 6.

Comparison of measured and estimated SSAs (441 nm) for June 1999.

Table 6.
Save