A multispectral retrieval method is developed on the 100 km regional scale to extract
the temperature, particle size, fractional cover and 11-μm emissivity of clouds which
may be semitransparent in the infrared based on emitted radiances. The scheme utilizes
the nonlinear relationship between emitted radiances when clouds are semitransparent and
form a single-layered system. The retrieval method has the limitation that the particle size
must be sufficiently small that the extinction and absorption cross sections at the different
wavelengths are significantly different. For combinations of emission at 11 and 12 μm,
the droplets must have radii less than 20 μm for both ice and water; for 3.7 and 12 μm
the droplets must have radii less than 60 μm for ice and less than 100 μm for water; for
3.7 and 8 μm the droplets must have radii less than l50 μm for both ice and water.
The retrieval scheme together with the spatial coherence method is applied to the
analysis of NOAA-11 4-km Advanced Very High Resolution Radiometer 3.7, 11 and
12-μm observations obtained during the First ISCCP Regional Experiment Intensive
Field Observations, Kansas, 1991. The results indicate that clear skies, single-layered
and multi-layered cloud systems constitute equal proportions of 100-km scale regions.
For the upper-level, single-layered clouds, 100-km scale emissivity and fractional cloud
cover are correlated and the average effective radius of ice particles in semitransparent
clouds is about 10 μm.
Numerical simulations are performed to determine the sensitivity of the retrieved
results to errors in observations and model assumptions and to compare the results of the
current algorithm to those obtained with a threshold algorithm like that used by ISCCP.
For effective radii less than 15 μm, errors in the retrieved effective radius are less than
1.5 μm and errors in cloud temperature are less than 5 K. The fractional cloud cover
may be underestimated by 0.15. The 11-μm emissivity may be overestimated by 0.30.
For upper-level, semitransparent/broken clouds, the current algorithm is superior to the
threshold algorithm in the determination of cloud temperature and effective radius. / Graduation date: 1996
| Identifer | oai:union.ndltd.org:ORGSU/oai:ir.library.oregonstate.edu:1957/28696 |
| Date | 23 April 1996 |
| Creators | Lin, Xijian |
| Contributors | Coakley, James A. Jr |
| Source Sets | Oregon State University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
Page generated in 0.2393 seconds