Thermal infrared imaging of the atmosphere : the infrared cloud imager
MetadataShow full item record
Clouds play an important role in regulating the radiation budget of the atmosphere and are responsible for altering or maintaining the global climate. Understanding the interaction between clouds and radiation is necessary for accurate models that predict future climate and weather changes. The present cloud instruments are either zenithviewing, single-pixel instruments that calculate cloud amount using Taylor's hypothesis (spatial statistics are equal to temporal statistics), which do not work under certain atmospheric conditions, or are imagers that operate in the visible spectrum, thereby operating only in the daytime. Other imaging instruments use completely different techniques for daytime and nighttime detection of clouds and have problems detecting clouds at sunrise and sunset. The Infrared Cloud Imager (ICI), a ground based, thermal infrared imager was developed with funding from the Communications Research Laboratory (CRL), Japan, to provide continuous 24-hour cloud measurement without difference in daytime or nighttime sensitivity. The ICI is a spatially resolving instrument that measures the downwelling atmospheric radiance in the 8-14 aem region of the electromagnetic spectrum. The data collected are used to compute spatial and temporal cloud statistics. Although the ICI was initially developed to measure Arctic clouds, it has been successfully deployed in both Barrow, Alaska in the Spring of 2002 and in Lamont, Oklahoma (mid-latitude plains) in Spring 2003. I have developed algorithms for processing the ICI data based on Microwave Radiometer (MWR) water vapor data, for identifying and classifying clouds, and have demonstrated that ICI is capable of determining cloud statistics. From the radiative transfer analysis of the ICI data, weekly and monthly cloud statistics show a general trend of mostly clear and/or mostly cloudy skies with very small transition periods. Sky cover histograms from Oklahoma show predominantly thick clouds in March and thin clouds in April with a significant amount of variability in sky condition. The comparison of cloud statistics obtained from the entire ICI image and from a single pixel in the ICI image illustrates that zenith-viewing instruments generally tend to underestimate cloudiness. The correlation between ICI percent cloudiness (using different thresholds for Oklahoma data) with the Micro Pulse Lidar (MPL) cloudiness has also been analyzed.