This site describes current results from the above project, funded by the Department of Energy's Atmospheric Radiation Measurement (ARM) Program. The results herein cover a broad range of activities, including: (1) development of an ice cloud radiation code accounting for all scattering/absorption processes. (2) testing of a new cirrus radiation code with microphysical and reflectance measurements from a CEPEX tropical anvil case study, (3) development of a GCM parameterization for ice particle size spectra, based on ice water mixing ratio and cloud temperature, (4) ice crystal replicator analysis and comparison of size spectra from the replicator, 2DC and FSSP probes during the ARM '97 and '98 Cloud IOPs and, (5) an interpretation of Martin Platt's visible extinction/IR absorption ratios obtained from lidar-radiometer (LIRAD) measurements, indicating the amount of small ice crystals in tropical cirrus during the PROBE experiment. Highlights from these 5 activities are indicated below.

Ice Cloud Radiation Scheme
Testing the Radiation Scheme with a Tropical Cirrus Case Study
Size Distribution Parameterization in GCM's
Comparison Between Ice Particle Measuring Probes
Ice Particle Size Spectra in Tropical Cirrus

Ice Cloud Radiation Scheme       {top}

• This scheme now calculates the size distribution absorption and extinction coefficient (b_abs and b_ext) for any solar or terrestrial wavelength, due to the incorporation of reflection/refraction and photon tunneling effects, and the extension of the scheme into the Rayleigh regime. Tunneling is the process by which photons beyond a particle's physical cross-section can be absorbed, and it also contributes to extinction through scattering by large-angle diffraction. Examples of the scheme's performance with and without tunneling effects are given for a size distribution of hexagonal columns in Fig. 1 - 3.

• Comparisons with Mie theory using ice spheres indicate b_abs and b_ext are generally accurate to within 10%. The schemes advantage is that it can treat any ice particle shape. An unresolved issue is the degree of tunneling exhibited by ice crystals.

• The scheme was improved to treat bimodal size spectra, such as the example given in Fig. 4. Such high concentrations of small crystals are often observed by the FSSP, which may yield realistic ice particle concentrations in cirrus as discussed under (4).

Ice Particle Size Spectra in Tropical Cirrus       {top}

• Using bimodal size spectra similar to the spectrum in Fig. 4, the scheme was used to generate theoretical a ratios of visible extinction to thermal absorption efficiency, which were compared to those measured by Martin Platt in tropical cirrus during the ARM PROBE experiment. The predicted a ratios, shown in Fig. 5, were consistent with those observed, due to lower absorption efficiencies produced by the small crystals, as shown in Fig. 6.

Testing the Radiation Scheme with a Tropical Cirrus Case Study       {top}

• The only tropical cirrus microphysics/radiation experiment to date, a CEPEX case study, was used to test the radiation scheme. Measured size spectra were parameterized as shown in Fig. 7 in order to estimate the concentrations of the unmeasured small crystals. From the parameterized spectra, reflectances were predicted in the visible and near infrared for comparison with those measured during an ER-2 overflight using the Modis Airborne Simulator (MAS). As results in Fig. 8 and 9 show, predicted and observed reflectances were in agreement when scattering phase functions for hexagonal columns were used, but not for the randomized fractal ice particle surrogate. This suggests that the asymmetry parameter (g) need not be < 0.77. Similar results were obtained regarding g when testing this scheme with two other mid-latitude case studies.

• No evidence of excess solar absorption was found for the CEPEX case study.

Size Distribution Parameterization in GCMs       {top}

• The mean ice particle size (maximum dimension) of the large particle size distribution (D > 100 mm), D_l, was related to cloud temperature using all size spectra from three CEPEX case studies, as shown in Fig. 10 (slope = 1/D_l). The same was also done for the small particle mode (D < 100 mm), as shown in Fig. 11. The size distribution ice water content (IWC) was partitioned between the large and small particle modes solely in terms of D_l, as shown in Fig. 12. Given the GCM IWC and temperature, bimodal size spectra can thus be estimated.

• A comparison with a temperature - D_l relationship found for mid-latitude cirrus by Martin Platt, shown in Fig. 10, suggests that for a given temperature, D_l for mid-latitude cirrus are about a factor of two larger than for tropical anvil cirrus. If true, this could have a significant global radiative impact.

Comparison Between Ice Particle Measuring Probes       {top}

• Dr. Arnott has compared the performance of various ice particle measuring probes flown concurrently during the '97 ARM Cloud IOP. An example of measured size spectra is shown in Fig. 13. Although the Cloudscope collects particles through impaction while the FSSP is based on optical principles, both probes agree relatively well. The replicator appears to underestimate the concentrations of small crystals with D < 50 mm. Another 7 size spectra comparing the FSSP and cloudscope show similar behavior. If this agreement between the Cloudscope and FSSP continues, our understanding of cirrus microphysical and radiative properties could experience a major overhaul, since numerous other FSSP measurements in cirrus indicate very high concentrations of small crystals. Such measurements have not been taken too seriously in the past, since the FSSP was not designed to measure nonspherical particles.

• Three different measurements now indicate very high concentrations of small ice crystals (e.g. Fig. 4 and Fig. 13) in cold cirrus: LIRAD a ratios, FSSP measurements and recent Cloudscope measurements. In light of these measurements, the sign of the radiative forcing by cold cirrus may need to be reevaluated under some conditions.

• Dr. Arnott has been involved with an interesting cirrus case study from 26 September 1997. He operated the DRI replicator to capture images of ice crystals from the outflow of Hurricane Nora. This case study has been fairly extensively documented from many perspectives including satellite, radar, lidar, and in situ with the University of North Dakota Citation aircraft. The cirrus crystals were unusually pristine and the corresponding optical displays were of excellent quality such as a circumscribed halo around the normal 22 deg halo that was a bright as a common rainbow.

• Dr. Arnott also has been involved with replicator data analysis of the 1994 DOE cloud IOP. An example of the varied crystals shows regions of sublimation, polycrystals of the plate variety, and single crystals.