A 43 B-0077 Improved Automated Cloud Classification and Cloud Property Continuity Studies for the Visible/Infrared Imager/Radiometer Suite (VIIRS) Michael J. Pavolonis*, Andrew K. Heidinger*, and Keith Hutchison# *NOAA/NESDIS Center for Satellite Applications and Research - Madison, WI #The University of Texas at Austin, Center for Space Research; Northrop Grumman Corp. , Space Technology 1. Introduction 3. New VIIRS Cloud Typing - Demonstration A new automated cloud typing algorithm for the Visible/Infrared Imager/Radiometer Suite (VIIRS) is presented. The new algorithm, which is a part of the Integrated Data Processing System (IDPS), builds upon the heritage Advanced Very High Resolution Radiometer (AVHRR) and Moderate Resolution Imaging Spectroradiometer (MODIS) operational cloud-top phase algorithms. This poster will highlight the VIIRS cloud typing algorithm enhancements such as multilayered cloud detection. Further, in order to compare various VIIRS cloud algorithms to heritage AVHRR and MODIS algorithms on a global basis, a processing system called the Global Multi-imager Cloud Algorithm Testbed (GMCAT) was developed. In this poster, GMCAT is used to illustrate the cloud typing algorithm enhancements and some preliminary cloud top temperature/emissivity comparisons are shown. A schematic illustration of GMCAT. Note the multi-algorithm processing capability. Daytime New VIIRS MODIS • Knowledge of multilayered clouds can help in the development of more advanced cloud property retrieval algorithms. • Certain multilayered clouds tend to be called mixed or ambiguous by MODIS. Single Layer Thin Cirrus Thicker Cirrus Overlapping Lower Cloud • Note, the MODIS Team has recently developed an overlap detection method an improved daytime phase algorithm; however, neither are part of the operational cloud phase product at this time. Thin Cirrus Overlapping Low Cloud “Validation” NEW VIIRS Nighttime MODIS • The VIIRS algorithm is able to identify most of the multilayered cloud in this scene. Low Cloud • MODIS returns ambiguous when thin cirrus overlaps low cloud. Cirrus Overlapping Low Cloud 4. Cloud Property Comparisons AVHRR Thin Cirrus MODIS Bulk cloud phase (day) Thin cirrus identification (night) Volcanic Cloud Low Cloud Algorithm Classifies Clouds as: • Liquid water • Supercooled water/mixed • Opaque Ice • Cirrus • Multilayered cloud Detection (day) Multilayered cloud Detection (night) • Comparisons with ground-based instruments show that the VIIIRS algorithm improves upon the heritage AVHRR algorithm. • MODIS comparisons will be added in the future. 5. Summary • The new VIIRS cloud typing algorithm presented here is unique in that thin cirrus and multilayered clouds can be accurately identified. • Thin cirrus clouds and certain multilayered clouds tend to be classified as mixed and ambiguous by the heritage MODIS cloud phase algorithm, but are often correctly classified by the new VIIRS algorithm. 2. New VIIRS Cloud Typing Algorithm Bulk cloud phase (day/night) • Many thin cirrus clouds are called “ambiguous” by MODIS, but are correctly identified by the VIIRS algorithm. The GMCAT tool allows us to compare various cloud property algorithms including cloud top temperature and emissivity. In this example, the heritage AVHRR algorithm is shown on the left and the MODIS CO 2 slicing technique is on the right. Since the VIIRS will not have CO 2 absorption channels, the AVHRR algorithm is a reasonable representation of the VIIRS capability for this product. As expected, this simple comparison illustrates that the biggest differences are associated with thin high clouds. • A new MODIS processing tool was developed to aid in the comparison of various VIIRS cloud property algorithms with heritage MODIS and AVHRR algorithms. Several detailed comparisons will be performed in the future. • References: • Pavolonis, M. J. , A. K. Heidinger, and T. Uttal, 2005: Daytime global cloud typing from AVHRR and VIIRS: Algorithm description, validation, and comparisons. J. Appl. Meteor. , 44, 804 -826. • Pavolonis, M. J. , and A. K. Heidinger, 2004: Daytime cloud overlap detection from AVHRR and VIIRS. J. Appl. Meteor. , 43 (5), 762 -778. • Pavolonis, M. J. and A. K. Heidinger, 2005: Advancements in identifying cirrus and multilayered cloud systems from operational satellite imagers at night. International Asia-Pacific Environmental Remote Sensing Symposium, 4 th: Remote Sensing of the Atmosphere, Ocean, Environment, and Space, Honolulu, Hawaii, 8 -11 November 2004. Applications with Weather Satellites II (proceedings). International Society for Optical Engineering, (SPIE), Bellingham, WA, 2005, pp. 225 -234. • Heidinger, A. K. , and M. J. Pavolonis, 2005: Global daytime distribution of overlapping cloud from NOAA's Advanced Very High Resolution Radiometer. J. Climate, 18 (22), 4772 -4784. Contact the author at Mike. Pavolonis@noaa. gov
Скачать презентацию A 43 B-0077 Improved Automated Cloud Classification and
6b8a9a9d23880d101934f579d25b4ed0.ppt