A23P-02:
Validating Satellite-Retrieved Cloud Properties for Weather and Climate Applications
Tuesday, 16 December 2014: 1:55 PM
Patrick Minnis1, Kristopher M Bedka2, William Smith Jr3, Christopher R Yost4, Sarah T Bedka4, Rabindra Palikonda4, Douglas Spangenberg4, Sunny Sun-Mack4, Qing Trepte4, Xiquan Dong5 and Baike Xi5, (1)Nasa Larc, Hampton, VA, United States, (2)NASA Langley Research Center, Climate Science Branch, Hampton, VA, United States, (3)NASA Langley Research Ctr, Hampton, VA, United States, (4)Science Systems and Applications, Inc. Hampton, Hampton, VA, United States, (5)University of North Dakota, Grand Forks, ND, United States
Abstract:
Cloud properties determined from satellite imager radiances are increasingly used in weather and climate applications, particularly in nowcasting, model assimilation and validation, trend monitoring, and precipitation and radiation analyses. The value of using the satellite-derived cloud parameters is determined by the accuracy of the particular parameter for a given set of conditions, such as viewing and illumination angles, surface background, and cloud type and structure. Because of the great variety of those conditions and of the sensors used to monitor clouds, determining the accuracy or uncertainties in the retrieved cloud parameters is a daunting task. Sensitivity studies of the retrieved parameters to the various inputs for a particular cloud type are helpful for understanding the errors associated with the retrieval algorithm relative to the plane-parallel world assumed in most of the model clouds that serve as the basis for the retrievals. Real world clouds, however, rarely fit the plane-parallel mold and generate radiances that likely produce much greater errors in the retrieved parameter than can be inferred from sensitivity analyses. Thus, independent, empirical methods are used to provide a more reliable uncertainty analysis. At NASA Langley, cloud properties are being retrieved from both geostationary (GEO) and low-earth orbiting (LEO) satellite imagers for climate monitoring and model validation as part of the NASA CERES project since 2000 and from AVHRR data since 1978 as part of the NOAA CDR program. Cloud properties are also being retrieved in near-real time globally from both GEO and LEO satellites for weather model assimilation and nowcasting for hazards such as aircraft icing. This paper discusses the various independent datasets and approaches that are used to assessing the imager-based satellite cloud retrievals. These include, but are not limited to data from ARM sites, CloudSat, and CALIPSO. This paper discusses the use of the various datasets available, the methods employed to utilize them in the cloud property retrieval validation process, and the results and how they aid future development of the retrieval algorithms. Future needs are also discussed.
