Information content analysis in support of a new Infrared Cloud Ice Radiometer for SIRICE

Quantifying the global distribution of ice mass and its role in the global energy and water cycles represents an important unresolved challenge in the satellite community. Through a combination of measurements from the first submillimeter wavelength radiometer flown in space and a new Infrared Cloud Ice Radiometer (IRCIR) the proposed Submillimeterwave Infrared Radiometer Ice Cloud Experiment (SIRICE) mission promises to address this shortcoming in the observational record. SIRICE observations in combination with precipitation measurements from other platforms and supporting modeling studies offer great potential for significantly advancing our understanding of ice-phase microphysical processes and their representation in current numerical models. The IRCIR is an important component of the SIRICE mission needed to complement the submillemeter observations in regions of thin cirrus that are not only important radiatively but may also represent a significant fraction of the global ice mass due to their wide extent.

This study presents an objective analysis of the information content of measurements across the infrared spectrum for the retrieval of thin cirrus to establish an optimal channel combination and associated instrument noise requirements for the IRCIR. Using an analogue to the thermodynamic entropy as a metric for information content the subset of wavelengths between 7 and 14 ìm that offer the greatest amount of independent information for cloud ice water path and mass-weighted median diameter is determined. The technique explicitly represents the sensitivity of each channel to the geophysical retrieval parameters and the inherent uncertainties in both the observations and the physical model used to map these parameters into observation space. The impact of different levels of measurement noise, different instrument bandwidths, and the importance of cloud height information provided by novel stereo viewing techniques proposed for the IRCIR will be presented.