PASSIVE THERMAL RETRIEVALS OF ICE AND LIQUID WATER PATH, EFFECTIVE SIZE AND OPTICAL DEPTH and their dependence on particle and size distribution shape

While there are many methods for retrieving the ice water path (IWP) from ice clouds using ground-based active remote sensing, few methods exist using passive remote sensing from satellites. While the new A-Train satellites may provide estimates of IWP using active methods, historical IWP information is desirable. This information is needed to test simulations of the water-energy cycle in global climate models (GCMs) and to evaluate the hydrologic state of the atmosphere over decadal time scales.

A new IWP retrieval method has been developed based on radiances from a moderate absorbing and a strongly absorbing wavelength channel, such as channels centered near 8.5 µm and 11 µm, or 3.74 µm and 11 µm for nighttime retrievals. Such channels are common on past and current satellites. First, the effective diameter (De) is retrieved using ratios of the absorption optical depth (Tabs) retrieved for either of these wavelength pairs. Second, a simple equation gives the IWP as a function of De, cloud emissivity, the area-weighted absorption efficiency of the size distribution (Qabs) and the satellite zenith angle.

Surprisingly, the retrieval of De depends on the degree of bimodality of the ice particle size distribution (SD). Moreover,Qabs depends on De and the bimodality of the SD. Fortunately, the SD bimodality is known a priori provided it is known whether the ice cloud was associated with deep convection. Two parameterization schemes describing the SD for tropical anvil cirrus and “non-convective” mid-latitude cirrus in terms of IWC and temperature have been developed and are part of the retrieval scheme, and they provide the temperature dependence of SD bimodality.

These SD schemes were developed from cirrus in situ data and validated with radiances from the Atmospheric Emitted Radiance Interferometer (AERI) along with lidar measurements during field campaigns. The AERI/lidar/sounding data provide Tabs ratios as a function of temperature. This same information is predicted by the SD schemes (conjoined with a scheme predicting optical properties), thus allowing them to be validated against the AERI/lidar/sounding observations.

The retrieval of IWP appears relatively accurate based on uncertainty estimates and radiation transfer model retrieval simulations, with accuracy of about ± 15% over an IWP range of 1 to 100 g m-2. Provided De generally increases with IWP, IWP up to 200 g m-2 may be retrieved. The retrieval does not perform well under conditions of multiple cloud layers, or very thin cirrus having relatively large De. Retrieved IWPs will be compared with IWPs estimated from in situ measurements of ice water content based on the Counterflow Virtual Impactor (CVI) during several DOE ARM IOPs.