An Accurate Parameterization of the Infrared Radiative Properties of Cirrus Clouds for Climate Models

Qiang Fu Atmospheric Science Program, Departments of Oceanography and Physics, Dalhousie University, Halifax, Nova Scotia, Canada

Search for other papers by Qiang Fu in
Current site
Google Scholar
PubMed
Close
,
Ping Yang Department of Meteorology/CARSS, University of Utah, Salt Lake City, Utah

Search for other papers by Ping Yang in
Current site
Google Scholar
PubMed
Close
, and
W. B. Sun Atmospheric Science Program, Department of Oceanography, Dalhousie University, Halifax, Nova Scotia, Canada

Search for other papers by W. B. Sun in
Current site
Google Scholar
PubMed
Close
Full access

We are aware of a technical issue preventing figures and tables from showing in some newly published articles in the full-text HTML view.
While we are resolving the problem, please use the online PDF version of these articles to view figures and tables.

Abstract

An accurate parameterization is presented for the infrared radiative properties of cirrus clouds. For the single-scattering calculations, a composite scheme is developed for randomly oriented hexagonal ice crystals by comparing results from Mie theory, anomalous diffraction theory (ADT), the geometric optics method (GOM), and the finite-difference time domain technique. This scheme employs a linear combination of single-scattering properties from the Mie theory, ADT, and GOM, which is accurate for a wide range of size parameters. Following the approach of Q. Fu, the extinction coefficient, absorption coefficient, and asymmetry factor are parameterized as functions of the cloud ice water content and generalized effective size (Dge). The present parameterization of the single-scattering properties of cirrus clouds is validated by examining the bulk radiative properties for a wide range of atmospheric conditions. Compared with reference results, the typical relative error in emissivity due to the parameterization is ∼2.2%. The accuracy of this parameterization guarantees its reliability in applications to climate models. The present parameterization complements the scheme for the solar radiative properties of cirrus clouds developed by Q. Fu for use in numerical models.

Corresponding author address: Prof. Qiang Fu, Dept. of Oceanography, Dalhousie University, Halifax, NS B3H 4J1, Canada.

Abstract

An accurate parameterization is presented for the infrared radiative properties of cirrus clouds. For the single-scattering calculations, a composite scheme is developed for randomly oriented hexagonal ice crystals by comparing results from Mie theory, anomalous diffraction theory (ADT), the geometric optics method (GOM), and the finite-difference time domain technique. This scheme employs a linear combination of single-scattering properties from the Mie theory, ADT, and GOM, which is accurate for a wide range of size parameters. Following the approach of Q. Fu, the extinction coefficient, absorption coefficient, and asymmetry factor are parameterized as functions of the cloud ice water content and generalized effective size (Dge). The present parameterization of the single-scattering properties of cirrus clouds is validated by examining the bulk radiative properties for a wide range of atmospheric conditions. Compared with reference results, the typical relative error in emissivity due to the parameterization is ∼2.2%. The accuracy of this parameterization guarantees its reliability in applications to climate models. The present parameterization complements the scheme for the solar radiative properties of cirrus clouds developed by Q. Fu for use in numerical models.

Corresponding author address: Prof. Qiang Fu, Dept. of Oceanography, Dalhousie University, Halifax, NS B3H 4J1, Canada.

1. Introduction

Cirrus clouds, primarily present in the upper troposphere and lower stratosphere, are globally distributed and are composed almost exclusively of nonspherical ice crystals. Through their radiative effects, these clouds may modulate climate systems on all scales (e.g., Randall et al. 1989; Ramanathan and Collins 1991; Ramanathan et al. 1983; Ramaswamy and Ramanathan 1989;Fu et al. 1995; Starr and Cox 1985; Stephens 1983). Recent studies have shown that variations in assumed cirrus radiative properties can significantly alter the results of climate models (e.g., Ramanathan et al. 1983; Mitchel et al. 1988; Stephens et al. 1990). In his review article, Liou (1986) articulates that cirrus clouds are one of the most important yet least understood components in the climate system. The importance of understanding the radiative properties of cirrus clouds in weather and climate research has been recognized in view of several major field campaigns such as ICE (Raschke 1994), FIRE I (see Monthly Weather Review, Vol. 118, No. 11, 1990), and FIRE II (see Journal of the Atmospheric Sciences, Vol. 52, No. 23, 1995). One of the main conclusions drawn from these field experiments is that the asymmetry factor, which directly affects the reflection and transmission of cirrus clouds, is significantly smaller than that predicted from Mie theory in the solar spectral region (e.g., Francis 1995). However, it is still not clear if Mie theory, with its assumption of equivalent ice spheres, is adequate to describe the absorption and scattering associated with nonspherical particles in thermal infrared wavelengths.

Modeling and prediction of climate require that the parameterization of the radiative effects of cirrus clouds be as accurate as possible. However, to date, there is no exact theoretical solution for scattering and absorption by nonspherical ice particles (Liou and Takano 1994). In the solar spectrum, the geometric optics method (GOM) (e.g., Takano and Liou 1989; Macke et al. 1996), including an improved version (Yang and Liou 1996a), has been widely used to evaluate single-scattering properties of cirrus clouds. This method is an asymptotic approach that becomes accurate as the size parameter increases. In view of the observed ice crystal sizes in cirrus clouds, the geometric ray-tracing technique should be valid for solar wavelengths (Takano and Liou 1989). Based on optical parameters derived from the improved geometric ray-tracing method, Fu (1996) has developed an accurate parameterization of the solar radiative properties of cirrus clouds. However, this method may not be appropriate for thermal infrared wavelengths where ice crystals with small size parameters can be important in determining cirrus cloud radiative properties.

It is common practice to use Mie theory to approximate the absorption and scattering by nonspherical ice crystals in the infrared (e.g., Stephens et al. 1990; Ebert and Curry 1992; Sun and Shine 1995). The radiative properties of cirrus clouds in the infrared are dominated by absorption although the effects of scattering cannot be neglected (Fu et al. 1997). For this reason, the use of Mie theory to evaluate the scattering coefficient and asymmetry factor for ice crystals may not present a serious problem (Fu et al. 1998, manuscript submitted to J. Atmos. Sci.). However, using Mie theory to specify the absorption coefficient, which is the key parameter describing the infrared radiative properties of cirrus clouds, is questionable. Mitchell (1995) argued that not all of the physics that Mie theory embraces is applicable to nonspherical particles.

An alternative method of calculating extinction and absorption coefficients for ice crystals is the anomalous diffraction theory (ADT) (e.g., Ackerman and Stephens 1987; Chylek and Klett 1991; Mitchell and Arnott 1994). Chylek and Videen (1994) found that the accuracy of the ADT depends on the real part of the ice refractive index. Mitchell et al. (1996a) claimed that the ADT is a much better approximation than Mie theory for nonspherical particles. However, by comparing results from the ADT with those from the discrete dipole approximation, Maslowska et al. (1994) concluded that the ADT cannot be used for scattering and absorption by nonspherical particles without verification with rigorous methods. Although the ADT can be improved by considering the effects of refraction and reflection (Ackerman and Stephens 1987; Mitchell et al. 1996a), the resulting absorption coefficients should not be better than those calculated from the geometric ray-tracing method (Fu et al. 1998, manuscript submitted to J. Atmos. Sci.).

Fu and Liou (1993) developed a parameterization of cirrus cloud radiative properties. In the infrared spectrum, the optical parameters for nonspherical ice crystals with size parameters greater than 30 were computed using the geometric ray-tracing technique. For size parameters less than 30, Mie-type solution for spheroids was used. Although spheroids may be a better approach than the spherical approximation of nonspherical ice crystals, further inquiry into this problem appears necessary.

Recently, Yang and Liou (1996b) employed the finite-difference time domain (FDTD) method for light scattering by small ice crystals. The FDTD technique solves the Maxwell equations numerically in the time domain by using the finite-difference analog, which can be applied to nonspherical particles accurately. This numerical approach, however, can be used only for particles with small size parameters in practice because of significant computational efforts. Yang and Liou (1996b) showed that the FDTD can achieve an accuracy of ∼1% for size parameters smaller than 10.

In the present study, a composite method to determine single-scattering properties of hexagonal ice crystals in the infrared is developed by comparing results from the Mie, ADT, GOM, and FDTD. Following Fu (1996), an accurate parameterization for the infrared radiative properties of cirrus clouds is developed based on optical parameters derived from this composite method. The present parameterization complements the scheme for the solar radiative properties of cirrus clouds developed by Fu (1996) for use in climate models. In section 2, the calculation of single-scattering properties of ice crystals, including the development of the composite scheme, is discussed. The parameterization of these properties following Fu (1996) is shown in section 3. In section 4, the present parameterization is validated by examining bulk radiative properties. Summary and conclusions are given in section 5.

2. Determination of single-scattering properties of ice crystals

The radiative properties of cirrus clouds required for radiative transfer calculations in climate models are the scattering and absorption coefficients and the asymmetry factor. The computation of these single-scattering properties involves detailed information of ice particle shapes and their size distributions, knowledge of the refractive indices of ice, and a light-scattering program.

a. Ice crystal size distributions

The 28 ice crystal size distributions adopted in Fu (1996) are used in the present study, which were based on in situ aircraft observations from both midlatitude and tropical regions.

To include the possible effects of small ice crystals that may exist in cirrus clouds (Heymsfield and Platt 1984) on the infrared, all 28 ice crystal size distributions used in Fu (1996) are extrapolated to 1.0 μm following Eq. (2) of Mitchell et al. (1996b). For scattering and absorption calculations, these size distributions are discretized into 38 bins to resolve the structure of the spectra. The midpoint ice crystal lengths in the first 10 bins are 2, 4, 6, 8, 10, 12, 16, 20, 25, and 37.5 μm, whereas the lengths for the next 28 bins are the same as those in Table 1 of Fu (1996). The width of an ice crystal, D, may be derived from the length, L, following the aspect ratio used in Fu (1996) as follows:
i1520-0442-11-9-2223-e2-1
By assuming that ice crystals are hexagonal, the ice water content (IWC) and generalized effective size (Dge) are defined as (Fu 1996)
i1520-0442-11-9-2223-e2-2a
where n(L) denotes the ice crystal size distribution, Lmin and Lmax are the minimum and maximum lengths of ice crystals, respectively, and ρi is the density of ice. Here an ice density of 0.9167 g cm−3 is used. The resulting IWC and Dge range from 0.66 × 10−3 to 0.28 g m−3 and from 11.0 to 129.6 μm, respectively, for the 28 size distributions.

b. A composite light-scattering program

For calculations of the scattering and absorption properties of nonspherical ice crystals, a composite scheme is developed by comparing results based on the Mie, ADT, GOM, and FDTD.

To apply the Mie theory, nonspherical particles are usually converted into spheres with either equivalent projected area or equivalent volume. Letting P and V be the projected area and volume of a nonspherical particle, respectively, the radius of the equivalent sphere is given as
rpPπ1/2
or
rvVπ1/3
where rp and rv are for spheres with equivalent projected area and equivalent volume, respectively.
For a nonspherical particle, spheres with radius rvp can also be defined with both equivalent projected area and volume by letting P = πr2vpn and V = 4/3πr3vpn, where n is the number of spheres (n here is not necessarily integer). So we have
rvpVP.
In this case, the particle size distribution may no longer be preserved; the particle size distribution for equivalent spheres becomes nN where N is the size distribution for nonspherical particles. However, the ratio of volume to projected area is conserved for each particle. As pointed out by Mitchell et al. (1996a), this ratio is a fundamental parameter in determining the absorption efficiency of a nonspherical particle. The effective radius derived in this case for ice clouds would be the same as that defined by Foot (1988). Using Eq. (2.5), the total cross-sectional area and volume of the ice cloud particles per unit volume are both conserved. It can be shown that rvprvrp. In this paper, we will use the radius rvp to define the size parameter, α, for nonspherical particles as
απrvpλ,
where λ is the wavelength.

The single-scattering calculations are performed for randomly oriented hexagonal ice crystals. The Mie code is from Wiscombe (1979). The geometric ray tracing program and finite-difference time domain program are based on Yang and Liou (1996a and 1996b). We have developed the anomalous diffraction scheme for randomly oriented hexagonal ice crystals (Fu et al. 1998b, manuscript submitted to J. Atmos. Sci.). In the single-scattering calculations, the refractive indices for ice from Warren (1984) and Gosse et al. (1995) are used.

Figure 1 shows the comparison of absorption efficiency from different methodologies as a function of the size parameter for the wavelengths of 3.7 μm in (a) and 12.99 μm in (b). The refractive indices are 1.40 + 0.0072i and 1.47 + 0.389i for the wavelengths of 3.7 and 12.99 μm, respectively. The different dependence of absorption efficiency on size parameters between Fig. 1a and 1b is due to the difference in the refractive indices. In Fig. 1a, the aspect ratio D/L used is ⅓ and the FDTD results are taken from Yang and Liou (1996a). In Fig. 1b, the aspect ratio used follows Eq. (2.1); the hexagonal particle with a length of 300 μm corresponds to the size parameter of ∼28. From Fig. 1, we see that the absorption efficiency is not sensitive to the particle shape for large size parameters for which the Mie results using different equivalent spheres converge to GOM solutions. The absorption efficiency from ADT approaches one for large size parameters. This is because ADT does not consider the external reflection. The difference between ADT and GOM for smaller size parameters is evident because ADT neglects refraction and internal reflection. At 3.7 μm where ice absorption is quite weak (Fig. 1a), the absorption efficiency is overestimated by using spheres with the equivalent projected area or volume; in both cases, the ratio of V/P is overestimated. Figure 1a also shows that GOM and Mie theory using rvp obtain the best results compared with reference calculations from FDTD. Significant errors are produced by ADT because of neglecting reflection and refraction. At 12.99 μm, where the absorption is quite strong (Fig. 1b), the results from FDTD show that the absorption efficiency of nonspherical ice crystals can be significantly larger than one. This is the so-called tunneling effect, that is, the incident radiation outside the projected area of particles can be absorbed (Guimaraes and Nussenzveig 1992). The tunneling effect can be regarded as the difference between exact results and those from GOM. The Mie results seem to overestimate the tunneling effect associated with nonspherical particles while the ADT and GOM totally ignore this physics. Numerical results demonstrate that the tunneling effect strongly depends on refractive indices including both the real part, mr, and imaginary part, mi. The tunneling effect increases as mr or mi increases. It vanishes when mi approaches zero. However, the tunneling effect does not necessarily vanish when mr = 1.0, which is different from the result predicted using complex angular momentum theory (Guimaraes and Nussenzveig 1992).

A further comparison of results from the Mie, ADT, and GOM with those from FDTD at different wavelengths suggests a composite method to determine the single-scattering properties of hexagonal ice crystals in the infrared. The absorption efficiency Qa can be obtained by a linear combination of results from GOM and the Mie theory as follows:
i1520-0442-11-9-2223-e2-7
where QaGOM is the absorption efficiency from GOM for randomly oriented hexagonal ice crystals, and QaMie(vp), QaMie(v), and QaMie(p) are those from Mie theory. The subscripts Mie(vp), Mie(v), and Mie(p) denote the Mie results using both volume and area equivalent, volume equivalent, and area equivalent spheres, respectively. The coefficients for different wavelength intervals are empirically determined. The wavelength dependence of these coefficients and the use of different Mie results for different intervals reflect the dependence of tunneling effects for nonspherical ice crystals on the refractive index. It is found that for wavelengths with a strong tunneling effect, the absorption efficiency follows Mie results more closely; for weak tunneling effects, the absorption efficiency is largely determined by GOM results.

For hexagonal ice crystals with size parameters smaller than 10, the asymmetry factor can be obtained from Mie theory for volume equivalent spheres. For size parameters larger than 10, we use the asymmetry factor from GOM. The extinction efficiency is simply derived as the mean value of results from ADT for hexagonal ice crystals and Mie in which both the projected area and volume are preserved.

Figure 2 shows the absorption efficiency for randomly oriented hexagonal ice crystals determined from the composite scheme as a function of size parameter for eight different wavelengths. Also shown is the absorption efficiency predicted from FDTD. The relative differences are generally smaller than ∼3%. There are two points at the wavelength of 18.18 μm and one point at 12.2 μm having the relative differences larger than 5% when the absorption efficiency is small. On the basis of these comparisons and considering the wide range of refractive indices used, we conclude that the composite scheme is generally valid for calculations of the absorption efficiency of hexagonal ice crystals.

Figures 3 and 4 provide comparison of the composite method and FDTD for calculations of asymmetry factor and extinction efficiency, respectively. The results are shown for three wavelengths. For the asymmetry factor, the relative differences are generally smaller than 2% when the asymmetry factor is larger than 0.5. The relative differences in extinction efficiency are usually smaller than 15%.

c. Single-scattering properties of ice clouds

For ice clouds, we need to incorporate the ice crystal size distribution in the computation of the single-scattering parameters. Based on observations, the aspect ratio, D/L, of ice crystals may be related to the crystal length L [see Eq. (2.1)]. For given ice crystal size distribution n(L), the single-scattering properties of ice clouds, including the extinction coefficient, β, absorption coefficient, βa, and the asymmetry factor, g, may be obtained from
i1520-0442-11-9-2223-e2-8
where Q, Qa, and Qs are the extinction, absorption, and scattering efficiencies, respectively, for a randomly oriented hexagonal ice crystal, gL is the asymmetry factor, and P is the projected area.

Using the composite light-scattering and absorption scheme, the single-scattering properties of cirrus clouds are derived at 36 wavelengths from 4 to 100 μm (see Table 1). This spectral resolution is fine enough for the calculation of broadband radiative fluxes and heating rates. The 28 ice crystal size distributions are incorporated into the computation of the single-scattering parameters.

Figure 5 shows the single-scattering properties as a function of wavelength for six particle size distributions. The single-scattering properties have strong variation with both wavelength and cloud microphysical properties.

3. Parameterization of the single-scattering properties of ice clouds

Based on the physical reasoning in Fu (1996), the single-scattering properties of ice clouds can be parameterized in terms of IWC and Dge as
i1520-0442-11-9-2223-e3-1
where the coefficients are obtained by numerical fitting to the single-scattering properties computed from the composite scheme for the 28 ice crystal size distributions. The single-scattering albedo, ˜ω, is defined as 1 − ˜ω = βa/β. In the present study we parameterize the absorption coefficient instead of the single-scattering albedo because the bulk radiative properties of cirrus clouds in the infrared are most sensitive to the absorption coefficient. For the scattering phase function in the thermal infrared wavelengths, halo and delta-transmission peak features are largely suppressed due to strong absorption. For this reason and to a good approximation, we may use the asymmetry factor to represent the phase function via the Henyey–Greenstein function (Fu and Liou 1993).

The single-scattering properties of cirrus clouds are parameterized at 36 wavelengths in the infrared spectrum. The infrared spectrum is also divided into 12 bands as in the previous study (Fu and Liou 1993). In this case, the cirrus single-scattering properties obtained from the composite scheme are weighted with the Planck function (T = −40°C) to obtain average values in each spectral band (Fu 1991). Tables 1a–c list the wavelengths and band limits and values of the coefficients in Eqs. (3.1)–(3.3), respectively. In the present parameterization, the units for IWC and Dge are in grams per cubic meter and microns, respectively, and the units for β are inverse meters.

Figure 6 shows the fitting for β/IWC, βa/IWC × Dge, and g in the spectral intervals 1250–1100 cm−1 (upper panel), 980–800 cm−1 (middle panel), and 670–540 cm−1 (lower panel). The averaged relative errors shown in the figure are defined as (100%/n) Σn |(xfxe)/xe|, where n = 28 and xe and xf are “exact” values and those from the numerical fitting, respectively. The maximum absolute value of relative error is also shown in the figure. The averaged relative differences between the fitted and reference calculations are less than ∼5%, ∼5%, and 1% for β/IWC, βa/IWC × Dge, and g, respectively. It should be noted that the linear relationship between β/IWC and 1/Dge, which is a very good approximation in the solar spectrum (Fu 1996), is no longer valid for the infrared.

The present parameterization is based on single-scattering calculations of hexagonal columns. From in situ observations, cirrus clouds are composed of nonspherical ice crystals of various shapes, such as columns, plates, bullet rosettes, and aggregates, etc. For cirrus clouds containing various nonspherical particles, we may define the Dge as (Fu 1996)
i1520-0442-11-9-2223-e3-4
where Ac is the total cross-sectional area of cloud particles per unit volume. For hexagonal ice crystals, Dge from Eq. (3.4) is equivalent to that from Eq. (2.2b). Recently, Heymsfield and McFarquhar (1996) parameterize the Ac as a function of the IWC and temperature based on observations. Using IWC derived from measurements along with Dge from Eq. (3.4), the single-scattering properties can be properly predicted from the present parameterization for measured cirrus clouds.

The scattering of thermal infrared radiation by clouds is often neglected in climate studies. This would introduce an increase of the outgoing longwave fluxes at the top of the atmosphere by ∼6 W m−2 in cloudy atmospheres (Fu et al. 1997). The present parameterization can be directly incorporated into radiation models to consider multiple scattering processes associated with cirrus clouds.

4. Accuracy of bulk radiative properties

The parameterization of single-scattering properties of ice clouds can be evaluated by examining the bulk cloud radiative properties, that is, the absorptivity (emissivity), transmissivity, and reflectivity. The broadband infrared absorptivity, transmissivity, and reflectivity are defined by
i1520-0442-11-9-2223-e4-1
where wi is the fraction of the blackbody flux (T = −40°C) for an individual band, and ai, ti, and ri denote the absorptivity, transmissivity, and reflectivity, respectively, in this band. The bulk cloud radiative properties are derived by an application of the delta-four-stream solution (Liou et al. 1988) to a homogeneous layer in the present study. In the calculation, an isotropic incident radiation is used.

The comparisons have been performed between the bulk broadband radiative properties obtained from the present parameterization (both for 36 wavelengths and 12 bands) of single-scattering parameters and those from the composite method at 36 wavelengths. Here 28 types of ice clouds are used with different thicknesses (0.5, 1.0, 2.0, and 4.0 km) to cover a wide range of atmospheric conditions.

Figure 7 shows the absolute values of relative errors due to the 36-wavelength parameterization versus the reference broadband absorptivity, transmissivity, and reflectivity. Each figure contains 112 points, representing 28 types of ice clouds with four different thicknesses. The averaged relative errors in absorptivity, transmissivity, and reflectivity are 2.2%, 3.3%, and 8.8%, respectively. Since thermal emission from the cloud layer is the dominating radiative process, the magnitude of absorptivity, that is, emissivity, must be determined accurately. Figure 7 shows that the relative errors in the absorptivity are less than about 5%. For transmissivity, the large relative errors occur when the transmissivity is small so that the errors in radiative energy may not be large. Since the reflectivity, as shown in Fig. 7c, is always smaller than ∼0.06, the accuracy of ∼10% is satisfactory.

Figure 8 is the same as Fig. 7 except for the 12-band parameterization of single-scattering parameters. From Figs. 7 and 8, it is clear that the accuracy of the 12-band parameterization is similar to that obtained with the 36-wavelength parameterization. This suggests that the accuracy of the parameterization is insensitive to the spectral resolution.

The errors in bulk cloud radiative properties due to the present parameterization are small compared with those produced by other error sources (Fu 1996). These error sources include radiative transfer approximations and uncertainties in refractive indices and atmospheric input parameters. The accuracy of this parameterization guarantees its reliability in the applications to climate studies. The present parameterization complements the scheme for the solar radiative properties of ice clouds developed by Fu (1996) for use in numerical models.

5. Summary and conclusions

In the present study, a composite methodology to determine single-scattering properties of hexagonal ice crystals in the infrared is developed by comparing results from the Mie theory, anomalous diffraction theory, geometric optics method, and the finite-difference time domain technique. The absorption efficiency from this scheme yields relative accuracies within ∼3%. It is found that for size parameters smaller than ∼30, Mie theory using equivalent ice spheres tends to overestimate the absorption efficiency of nonspherical ice crystals, while the ADT and GOM underestimate it.

An accurate parameterization of the infrared radiative properties of cirrus clouds is developed based on improved light-scattering calculations. Here 28 ice crystal size distributions from in situ aircraft observations in both tropical and midlatitude regions are employed. These size distributions are extrapolated to 1.0 μm to include the effects of small ice crystals. Thirty-eight size bins are used to resolve the detailed structure of the observed particle size distributions. The single-scattering properties of hexagonal ice crystals are calculated by using the composite scheme that can produce accurate results for a wide range of size parameters.

Following Fu (1996), a generalized effective size, Dge, is defined to account for the ice crystal size distribution in the radiative calculations. Based on physical principles, the single-scattering properties have been parameterized in terms of ice water content and Dge. The generalized effective size can be related to the total cross-sectional area of ice particles per unit volume, a quantity directly measured by the 2D optical probe in in situ microphysical observations of cirrus clouds.

The present parameterization of single-scattering properties of cirrus clouds is evaluated by examining the bulk radiative properties. Here 28 types of cirrus clouds with four different thicknesses are used to cover a wide range of atmospheric conditions. Compared with reference results, the typical relative error due to the parameterization is ∼2.2% in cirrus cloud emissivity. This error is small when compared with those produced by other error sources. The accuracy of this parameterization guarantees its reliability in the applications to climate studies. The present parameterization complements the scheme for the solar radiative properties of ice clouds developed by Fu (1996) for use in numerical models.

Acknowledgments

The authors wish to thank Drs. P. Chylek, G. Videen, and G. Lesins for useful discussions. Many thanks to M. Cribb for help in preparing the manuscript. The research work contained herein has been supported by DOE Grant DE-FG02-97ER62363, and in part by an NSERC operating grant, an AES Science Subvention grant, and an NSERC collaborative grant.

REFERENCES

  • Ackerman, S. A., and G. L. Stephens, 1987: The absorption of solar radiation by cloud droplets: An application of anomalous diffraction theory. J. Atmos. Sci.,44, 1574–1588.

  • Chylek, P., and J. D. Klett, 1991: Extinction cross sections of nonspherical particles in the anomalous diffraction approximation. J. Opt. Soc. Amer.,8A, 274–281.

  • ——, and G. Videen, 1994: Longwave radiative properties of polydispersed hexagonal ice crystals. J. Atmos. Sci.,51, 175–190.

  • Ebert, E. E., and J. A. Curry, 1992: A parameterization of ice cloud optical properties for climate models. J. Geophys. Res.,97, 3831–3836.

  • Foot, J. S., 1988: Some observations of the optical properties of clouds. Part II: Cirrus. Quart. J. Roy. Meteor. Soc.,114, 145–164.

  • Francis, P. N., 1995: Some aircraft observations of the scattering properties of ice crystals. J. Atmos. Sci.,52, 1142–1154.

  • Fu, Q., 1991: Parameterization of radiative processes in vertically nonhomogeneous multiple scattering atmospheres. Ph.D. dissertation, University of Utah, 259 pp. [Available from University Microfilm, 305 N. Zeeb Rd., Ann Arbor, MI 48106.].

  • ——, 1996: An accurate parameterization of the solar radiative properties of cirrus clouds for climate models. J. Climate,9, 2058–2082.

  • ——, and K. N. Liou, 1993: Parameterization of the radiative properties of cirrus clouds. J. Atmos. Sci.,50, 2008–2025.

  • ——, S. K. Krueger, and K. N. Liou, 1995: Interactions between radiation and convection in simulated tropical cloud clusters. J. Atmos. Sci.,52, 1310–1328.

  • ——, K. N. Liou, M. C. Cribb, T. P. Charlock, and A. Grossman, 1997: Multiple scattering parameterization in thermal infrared radiative transfer. J. Atmos. Sci.,54, 2799–2812.

  • Gosse, S., D. Labrie, and P. Chylek, 1995: Refractive index of ice in the 1.4 to 7.8 μm spectral range. Appl. Opt.,34, 6582–6586.

  • Guimaraes, L. G., and H. M. Nussenzveig, 1992: Theory of Mie resonances and the ripple fluctuations. Opt. Commun.,89, 363–369.

  • Heymsfield, A. J., and C. M. R. Platt, 1984: A parameterization of the particle size spectrum of ice clouds in terms of the ambient temperature and the ice water content. J. Atmos. Sci.,41, 846–856.

  • ——, and G. M. McFarquhar, 1996: High albedos of cirrus in the tropical pacific warm pool: Microphysical interpretations from CEPEX and from Kwajalein, Marshall Islands. J. Atmos. Sci.,53, 2424–2451.

  • Liou, K. N., 1986: Influence of cirrus clouds on weather and climate processes: A global perspective. Mon. Wea. Rev.,114, 1167–1199.

  • ——, and Y. Takano, 1994: Light scattering by nonspherical particles:Remote sensing and climate implications. Atmos. Res.,31, 271–298.

  • ——, Q. Fu, and T. P. Ackerman, 1988: A simple formulation of the delta-four-stream approximation for radiative transfer parameterizations. J. Atmos. Sci.,45, 1940–1947.

  • Macke, A., J. Mueller, and E. Raschke, 1996: Single scattering properties of atmospheric ice crystals. J. Atmos. Sci.,53, 2813–2825.

  • Maslowska, A., P. J. Flatau, and G. L. Stephens, 1994: On the validity of the anomalous diffraction theory to light scattering by cubes. Opt. Commun.,107, 35–40.

  • Mitchel, J. F. B., C. A. Senior, and W. J. Ingram, 1988: CO2 and climate: A missing feedback? Nature,341, 132–134.

  • Mitchell, D. L., 1995: How appropriate is Mie theory for predicting the radiative properties of atmospheric particles? GEWEX News,5 (1), 7–11.

  • ——, and W. P. Arnott, 1994: A model predicting the evolution of ice particle size spectra and radiative properties of cirrus clouds. Part II: Dependence of absorption and extinction on ice crystal morphology. J. Atmos. Sci.,51, 817–832.

  • ——, A. Macke, and Y. G. Liu, 1996a: Modeling cirrus clouds. Part II: Treatment of radiative properties. J. Atmos. Sci.,53, 2967–2988.

  • ——, S. K. Chai, and Y. G. Liu, 1996b: Modeling cirrus clouds. Part I: Treatment of bimodal size spectra and case study analysis. J. Atmos. Sci.,53, 2952–2966.

  • Ramanathan, V., and W. Collins, 1991: Thermodynamic regulation of ocean warming by cirrus clouds deduced from observations of the 1987 El Niño. Nature,351, 27–32.

  • ——, E. J. Pitcher, R. C. Malone, and M. L. Blackmon, 1983: The response of a general circulation model to refinements in radiative processes. J. Atmos. Sci.,40, 605–630.

  • Ramaswamy, V., and V. Ramanathan, 1989: Solar absorption by cirrus clouds and the maintenance of the tropical upper troposphere thermal structure. J. Atmos. Sci.,46, 2293–2310.

  • Randall, D. A., Harshvardhan, D. A. Dazlich, and T. G. Corsetti, 1989: Interactions among radiation, convection, and large-scale dynamics in a general circulation model. J. Atmos. Sci.,46, 1943–1970.

  • Raschke, E., 1994: Results on cirrus properties from the European Cloud and Radiation Experiment (EUCREX). Preprints, Eighth Conf. on Atmospheric Radiation, Nashville, TN, Amer. Meteor. Soc., 276–278.

  • Starr, D. O’C., and S. K. Cox, 1985: Cirrus clouds. Part II: Numerical experiments on the formation and maintenance of cirrus. J. Atmos. Sci.,42, 2682–2694.

  • Stephens, G. L., 1983: The influence of radiative transfer on the mass and heat budgets of ice crystals falling in the atmosphere. J. Atmos. Sci.,40, 1729–1739.

  • ——, S.-C. Tsay, P. W. Stackhouse Jr., and P. J. Flatau, 1990: The relevance of the microphysical and radiative properties of cirrus clouds to climatic feedback. J. Atmos. Sci.,47, 1742–1753.

  • Sun, Z., and K. P. Shine, 1995: Parameterization of ice cloud radiative properties and its application to the potential climatic importance of mixed-phase clouds. J. Climate,8, 1874–1888.

  • Takano, Y., and K. N. Liou, 1989: Solar radiative transfer in cirrus clouds. Part I: Single scattering and optical properties of hexagonal ice crystals. J. Atmos. Sci.,46, 3–19.

  • Warren, S. G., 1984: Optical constants of ice from ultraviolet to the microwave. Appl. Opt.,23, 1206–1225.

  • Wiscombe, W., 1979: Mie scattering calculations. NCAR Tech. Note TN-140+STR, NCAR, 66 pp. [Available from the National Center for Atmospheric Research, Boulder, CO 80307.].

  • Yang, P., and K. N. Liou, 1996a: Geometric-optics-integral-equation method for light scattering by nonspherical ice crystals. Appl. Opt.,35, 6568–6584.

  • ——, and ——, 1996b: Finite-difference time domain method for light scattering by small ice crystals in three-dimensional space. J. Opt. Soc. Amer.,13A, 2072–2085.

Fig. 1.
Fig. 1.

Comparison of absorption efficiencies for randomly oriented hexagonal ice crystals derived from different scattering programs: Mie theory for sphere with equal projected area (········),equal volume (—), both equal area and volume (____), anomalous diffraction theory(–·–), geometric optics method (–··–), and finite-difference time domain technique (×). The results are shown as a function of size parameter for wavelengths of 3.7 μm (a) and 12.99 μm (b). In (a) the aspect ratio, D/L, used is ⅓; in (b) the aspect ratio follows Eq. (2.1).

Citation: Journal of Climate 11, 9; 10.1175/1520-0442(1998)011<2223:AAPOTI>2.0.CO;2

Fig. 2.
Fig. 2.

Comparison of the absorption efficiency derived from the composite method with the result from the finite-difference time domain technique for randomly oriented hexagonal ice crystals at different wavelengths. The results are shown as a function of size parameter and the aspect ratio used follows Eq. (2.1).

Citation: Journal of Climate 11, 9; 10.1175/1520-0442(1998)011<2223:AAPOTI>2.0.CO;2

Fig. 3.
Fig. 3.

As in Fig. 2 but for the asymmetry factor and results are only shown for three wavelengths.

Citation: Journal of Climate 11, 9; 10.1175/1520-0442(1998)011<2223:AAPOTI>2.0.CO;2

Fig. 4.
Fig. 4.

As in Fig. 3 but for the extinction efficiency.

Citation: Journal of Climate 11, 9; 10.1175/1520-0442(1998)011<2223:AAPOTI>2.0.CO;2

Fig. 5.
Fig. 5.

Single-scattering properties including the extinction coefficient, absorption coefficient, and asymmetry factor as a function of wavelength for six cirrus clouds.

Citation: Journal of Climate 11, 9; 10.1175/1520-0442(1998)011<2223:AAPOTI>2.0.CO;2

Fig. 6.
Fig. 6.

Extinction coefficient/IWC, absorption coefficient/IWC × Dge, and asymmetry factor vs the generalized mean effective size for the spectral intervals 1250–1100 cm−1 (upper panels), 980–800 cm−1 (middle panels), and 670–540 cm−1 (lower panels). The “○” points represent reference results from the light-scattering program for 28 measured ice crystal size distributions, and the curves are from the parameterization.

Citation: Journal of Climate 11, 9; 10.1175/1520-0442(1998)011<2223:AAPOTI>2.0.CO;2

Fig. 7.
Fig. 7.

Absolute values of relative errors due to the 36-wavelength parameterization of single-scattering parameters in absorptivity (a), transmissivity (b), and reflectivity (c) versus reference calculations. The 112 points in each figure represent results of 28 cirrus clouds with four thicknesses (0.5, 1.0, 2.0, 4.0 km). The reference absorptivity, transmissivity, and reflectivity are based on single-scattering properties from the composite method.

Citation: Journal of Climate 11, 9; 10.1175/1520-0442(1998)011<2223:AAPOTI>2.0.CO;2

Fig. 8.
Fig. 8.

As in Fig. 7 but for absolute values of relative errors due to the 12-band parameterization of single-scattering parameters.

Citation: Journal of Climate 11, 9; 10.1175/1520-0442(1998)011<2223:AAPOTI>2.0.CO;2

Table 1a. Values of coefficients in Eq. (3.1) for the parameterization of the extinction coefficient (m−1). (The units for Dge and IWC are in μm and gm−3, respectively.)

i1520-0442-11-9-2223-t01

Table 1b. Values of coefficients in Eq. (3.2) for the parameterization of the absorption coefficient (m−1). (The units for Dge and IWC are in μm and gm−3, respectively.)

i1520-0442-11-9-2223-t102

Table 1c. Values of coefficients in Eq. (3.3) for the parameterization of the asymmetry factor. (The units for Dge are in μm.)

i1520-0442-11-9-2223-t103
Save
  • Ackerman, S. A., and G. L. Stephens, 1987: The absorption of solar radiation by cloud droplets: An application of anomalous diffraction theory. J. Atmos. Sci.,44, 1574–1588.

  • Chylek, P., and J. D. Klett, 1991: Extinction cross sections of nonspherical particles in the anomalous diffraction approximation. J. Opt. Soc. Amer.,8A, 274–281.

  • ——, and G. Videen, 1994: Longwave radiative properties of polydispersed hexagonal ice crystals. J. Atmos. Sci.,51, 175–190.

  • Ebert, E. E., and J. A. Curry, 1992: A parameterization of ice cloud optical properties for climate models. J. Geophys. Res.,97, 3831–3836.

  • Foot, J. S., 1988: Some observations of the optical properties of clouds. Part II: Cirrus. Quart. J. Roy. Meteor. Soc.,114, 145–164.

  • Francis, P. N., 1995: Some aircraft observations of the scattering properties of ice crystals. J. Atmos. Sci.,52, 1142–1154.

  • Fu, Q., 1991: Parameterization of radiative processes in vertically nonhomogeneous multiple scattering atmospheres. Ph.D. dissertation, University of Utah, 259 pp. [Available from University Microfilm, 305 N. Zeeb Rd., Ann Arbor, MI 48106.].

  • ——, 1996: An accurate parameterization of the solar radiative properties of cirrus clouds for climate models. J. Climate,9, 2058–2082.

  • ——, and K. N. Liou, 1993: Parameterization of the radiative properties of cirrus clouds. J. Atmos. Sci.,50, 2008–2025.

  • ——, S. K. Krueger, and K. N. Liou, 1995: Interactions between radiation and convection in simulated tropical cloud clusters. J. Atmos. Sci.,52, 1310–1328.

  • ——, K. N. Liou, M. C. Cribb, T. P. Charlock, and A. Grossman, 1997: Multiple scattering parameterization in thermal infrared radiative transfer. J. Atmos. Sci.,54, 2799–2812.

  • Gosse, S., D. Labrie, and P. Chylek, 1995: Refractive index of ice in the 1.4 to 7.8 μm spectral range. Appl. Opt.,34, 6582–6586.

  • Guimaraes, L. G., and H. M. Nussenzveig, 1992: Theory of Mie resonances and the ripple fluctuations. Opt. Commun.,89, 363–369.

  • Heymsfield, A. J., and C. M. R. Platt, 1984: A parameterization of the particle size spectrum of ice clouds in terms of the ambient temperature and the ice water content. J. Atmos. Sci.,41, 846–856.

  • ——, and G. M. McFarquhar, 1996: High albedos of cirrus in the tropical pacific warm pool: Microphysical interpretations from CEPEX and from Kwajalein, Marshall Islands. J. Atmos. Sci.,53, 2424–2451.

  • Liou, K. N., 1986: Influence of cirrus clouds on weather and climate processes: A global perspective. Mon. Wea. Rev.,114, 1167–1199.

  • ——, and Y. Takano, 1994: Light scattering by nonspherical particles:Remote sensing and climate implications. Atmos. Res.,31, 271–298.

  • ——, Q. Fu, and T. P. Ackerman, 1988: A simple formulation of the delta-four-stream approximation for radiative transfer parameterizations. J. Atmos. Sci.,45, 1940–1947.

  • Macke, A., J. Mueller, and E. Raschke, 1996: Single scattering properties of atmospheric ice crystals. J. Atmos. Sci.,53, 2813–2825.

  • Maslowska, A., P. J. Flatau, and G. L. Stephens, 1994: On the validity of the anomalous diffraction theory to light scattering by cubes. Opt. Commun.,107, 35–40.

  • Mitchel, J. F. B., C. A. Senior, and W. J. Ingram, 1988: CO2 and climate: A missing feedback? Nature,341, 132–134.

  • Mitchell, D. L., 1995: How appropriate is Mie theory for predicting the radiative properties of atmospheric particles? GEWEX News,5 (1), 7–11.

  • ——, and W. P. Arnott, 1994: A model predicting the evolution of ice particle size spectra and radiative properties of cirrus clouds. Part II: Dependence of absorption and extinction on ice crystal morphology. J. Atmos. Sci.,51, 817–832.

  • ——, A. Macke, and Y. G. Liu, 1996a: Modeling cirrus clouds. Part II: Treatment of radiative properties. J. Atmos. Sci.,53, 2967–2988.

  • ——, S. K. Chai, and Y. G. Liu, 1996b: Modeling cirrus clouds. Part I: Treatment of bimodal size spectra and case study analysis. J. Atmos. Sci.,53, 2952–2966.

  • Ramanathan, V., and W. Collins, 1991: Thermodynamic regulation of ocean warming by cirrus clouds deduced from observations of the 1987 El Niño. Nature,351, 27–32.

  • ——, E. J. Pitcher, R. C. Malone, and M. L. Blackmon, 1983: The response of a general circulation model to refinements in radiative processes. J. Atmos. Sci.,40, 605–630.

  • Ramaswamy, V., and V. Ramanathan, 1989: Solar absorption by cirrus clouds and the maintenance of the tropical upper troposphere thermal structure. J. Atmos. Sci.,46, 2293–2310.

  • Randall, D. A., Harshvardhan, D. A. Dazlich, and T. G. Corsetti, 1989: Interactions among radiation, convection, and large-scale dynamics in a general circulation model. J. Atmos. Sci.,46, 1943–1970.

  • Raschke, E., 1994: Results on cirrus properties from the European Cloud and Radiation Experiment (EUCREX). Preprints, Eighth Conf. on Atmospheric Radiation, Nashville, TN, Amer. Meteor. Soc., 276–278.

  • Starr, D. O’C., and S. K. Cox, 1985: Cirrus clouds. Part II: Numerical experiments on the formation and maintenance of cirrus. J. Atmos. Sci.,42, 2682–2694.

  • Stephens, G. L., 1983: The influence of radiative transfer on the mass and heat budgets of ice crystals falling in the atmosphere. J. Atmos. Sci.,40, 1729–1739.

  • ——, S.-C. Tsay, P. W. Stackhouse Jr., and P. J. Flatau, 1990: The relevance of the microphysical and radiative properties of cirrus clouds to climatic feedback. J. Atmos. Sci.,47, 1742–1753.

  • Sun, Z., and K. P. Shine, 1995: Parameterization of ice cloud radiative properties and its application to the potential climatic importance of mixed-phase clouds. J. Climate,8, 1874–1888.

  • Takano, Y., and K. N. Liou, 1989: Solar radiative transfer in cirrus clouds. Part I: Single scattering and optical properties of hexagonal ice crystals. J. Atmos. Sci.,46, 3–19.

  • Warren, S. G., 1984: Optical constants of ice from ultraviolet to the microwave. Appl. Opt.,23, 1206–1225.

  • Wiscombe, W., 1979: Mie scattering calculations. NCAR Tech. Note TN-140+STR, NCAR, 66 pp. [Available from the National Center for Atmospheric Research, Boulder, CO 80307.].

  • Yang, P., and K. N. Liou, 1996a: Geometric-optics-integral-equation method for light scattering by nonspherical ice crystals. Appl. Opt.,35, 6568–6584.

  • ——, and ——, 1996b: Finite-difference time domain method for light scattering by small ice crystals in three-dimensional space. J. Opt. Soc. Amer.,13A, 2072–2085.

  • Fig. 1.

    Comparison of absorption efficiencies for randomly oriented hexagonal ice crystals derived from different scattering programs: Mie theory for sphere with equal projected area (········),equal volume (—), both equal area and volume (____), anomalous diffraction theory(–·–), geometric optics method (–··–), and finite-difference time domain technique (×). The results are shown as a function of size parameter for wavelengths of 3.7 μm (a) and 12.99 μm (b). In (a) the aspect ratio, D/L, used is ⅓; in (b) the aspect ratio follows Eq. (2.1).

  • Fig. 2.

    Comparison of the absorption efficiency derived from the composite method with the result from the finite-difference time domain technique for randomly oriented hexagonal ice crystals at different wavelengths. The results are shown as a function of size parameter and the aspect ratio used follows Eq. (2.1).

  • Fig. 3.

    As in Fig. 2 but for the asymmetry factor and results are only shown for three wavelengths.

  • Fig. 4.

    As in Fig. 3 but for the extinction efficiency.

  • Fig. 5.

    Single-scattering properties including the extinction coefficient, absorption coefficient, and asymmetry factor as a function of wavelength for six cirrus clouds.

  • Fig. 6.

    Extinction coefficient/IWC, absorption coefficient/IWC × Dge, and asymmetry factor vs the generalized mean effective size for the spectral intervals 1250–1100 cm−1 (upper panels), 980–800 cm−1 (middle panels), and 670–540 cm−1 (lower panels). The “○” points represent reference results from the light-scattering program for 28 measured ice crystal size distributions, and the curves are from the parameterization.

  • Fig. 7.

    Absolute values of relative errors due to the 36-wavelength parameterization of single-scattering parameters in absorptivity (a), transmissivity (b), and reflectivity (c) versus reference calculations. The 112 points in each figure represent results of 28 cirrus clouds with four thicknesses (0.5, 1.0, 2.0, 4.0 km). The reference absorptivity, transmissivity, and reflectivity are based on single-scattering properties from the composite method.

  • Fig. 8.

    As in Fig. 7 but for absolute values of relative errors due to the 12-band parameterization of single-scattering parameters.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1443 465 60
PDF Downloads 761 195 16