HR: 0800h
AN: IN31A-1141 [Abstracts]
TI: The Synergistic Use of MISR and MODIS Observations from the Terra Satellite to Observe Thin Cirrus
Clouds
AU: * Garay, M J
EM: garay@atmos.ucla.edu
AF: Department of Atmospheric & Oceanic Sciences, University of California, Los Angeles, 405 Hilgard Ave.,
Los Angeles, CA 90095
United States
AU: Mazzoni, D
EM: Dominic.Mazzoni@jpl.nasa.gov
AF: Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Drive, Pasadena, CA 91101
United States
AU: Horvath, A
EM: Akos.Horvath@jpl.nasa.gov
AF: Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Drive, Pasadena, CA 91101
United States
AU: Davies, R
EM: Roger.Davies@jpl.nasa.gov
AF: Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Drive, Pasadena, CA 91101
United States
AB:
Optically thin cirrus clouds have attracted a great deal of attention in recent years because of their effect on the
radiation budget of the Earth and their poor representation in global climate models. These clouds have mainly been studied
using satellite limb sounding instruments, which do not provide detailed information about the horizontal extent of cirrus
cloud fields. Better horizontal resolution is available from nadir-looking instruments, but thin cirrus clouds are
notoriously difficult to detect in nadir observations. To overcome this difficulty we have begun investigating the
synergistic use of the Multi-angle Imaging SpectroRadiometer (MISR) and the Moderate Resolution Imaging Spectroradiometer
(MODIS) instruments on the EOS Terra satellite platform to detect and characterize thin cirrus clouds. Preliminary
comparisons with ground-based lidar have shown that the MODIS 1.38 and 11 micron channels are sensitive to the presence of
thin cirrus, while MISR is capable of observing thin cirrus directly due to the increase in the optical path through the
clouds in its off-nadir cameras. We will compare the sensitivity of MISR and MODIS to thin cirrus at horizontal scales
ranging from 275 m for MISR to 1 km for MODIS.
In addition, we have been developing machine learning approaches that are able to exploit the massive amounts of geophysical
information available from satellite instruments to yield new retrieval methodologies. A thin cirrus cloud detection
algorithm for MISR is being tested which uses Support Vector Machines (SVMs), a form of supervised learning algorithm similar
to neural networks. We will explore the potential of fusing MISR and MODIS observations together to provide superior thin
cirrus detection and characterization at a global scale.
DE: 0321 Cloud/radiation interaction
DE: 0394 Instruments and techniques
DE: 0555 Neural networks, fuzzy logic, machine learning
DE: 0594 Instruments and techniques
DE: 3360 Remote sensing
SC: Earth and Space Science Informatics [IN]
MN: Fall Meeting 2005