INTRODUCTION

Professor Tom Vonder Haar and Dr. G. Garrett Campbell (campbell@cira.colostate.edu) of the Cooperative Institute for Research in the Atmosphere at Colorado State University have participated in International Satellite Cloud Climatology Project (ISCCP) of the World Meteorological Organization since its inception in 1982. Researchers and students from several different countries have collected and processed satellite data from geosynchronous and polar orbiting meteorological satellites over the years in support of ISCCP. This process has contributed to a multi-decade global cloud analysis which has contributed to our understanding of the Earth's climate and climate change mechanisms.

RESULTS AND ACCOMPLISHMENTS:

A new product we are preparing with the GOES 8 and later data is an estimate of the variation of the radiance inside the 4 km pixel. This provides an indication of cloud edges and places with partly cloudy situations. This will provide more information about the clouds than available in the older data sets. A sample time series with 6 of the 7 parameters was constructed from ISCCP data processed from GOES 8 or GOES 9 data. These products may not be on all tapes processed at CIRA. Some ISCCP tapes were constructed with data from channel 1 (visible) and channel 4 (infrared).

The collection of information from these meteorological satellites by CIRA and other countries contributes to the global cloud analysis of monthly mean cloud amounts. Special algorithms developed by members of ISCCP are used to specify the radiative properties of the surface and atmosphere.

The ISCCP products are now being used by many researchers in the climate modeling community for model development and validation. It is well suited for studies of climate fluctuations like El Nino or volcanic events.

Our own studies have included comparisons to other satellite data sets like NASA's Earth Radiation Budget and GOES Pathfinder (Campbell et al. 1993, Campbell and Vonder Haar, 1994) and earlier datasets. This page also provides a link to an interactive time series viewer which allows the user to review clouds in the Earth's climate over the last 20 years.

In 2003 we have been examining the time series of ISCCP observations in comparison to other data sets. It is apparent that the ISCCP estimates have a distinct decreasing trend in cloudiness. We think a significant fraction of this trend arises from changes in the satellite observation view point. Click to enlarge image A map of the trend shows obvious effects of satellite changes and view angles, such as in the Indian Ocean (no geostationary coverage) and the central Atlantic (varying Meteosat coverage): Click to enlarge image The trend is discussed in this Power Point presentation (~1.4 MB).

DATA COLLECTION AND ANALYSIS:

Our data collection effort has involved the collection from many satellites over the western hemisphere and many parts of the world.

• GOES 6 July 1983 to January 1989 -- Eastern Pacific and North America
• Meteosat 3 May 1992 to January 1995 -- North and South America and Atlantic
• GOES 7 December 1994 to January 1996 -- Eastern Pacific and North America
• GOES 8 December 1994 to February 1996 -- North and South America and Atlantic
• GOES 9 January 1996 to July 1998 -- Eastern Pacific and North America
• GOES 10 July 1998 to present -- Eastern Pacific and North America
• GOES 12 March 2003 to present -- Western Atlant and North and South America

ISCCP processes data at the resolution of the infrared sensors which is 4 km for GOES 8, GOES-9, and GOES 10. The pixels are sampled to about 32 km resolution which adequately samples the data in the 280 km analysis ISCCP grid. This sampling also provides a reasonable data volume for processing the global data set. Digital images are collected every 3 hours from all the geosynchronous satellites. In addition, the data is sampled at 8 km resolution for archival and potential later analysis. These files are stored at NOAA in the ISCCP Archive Center at NCDC.

Over the years, the CIRA's Earthstation used a variety of satellite dish antennas ranging from 15 m to 4 m in diameter. Currently, 4 m mesh satellite dishes are used to collect the signal from the geostationary satellites operated by NOAA. The satellite radio signals are processed with specialized software and hardware into McIDAS/OS2 formatted files.

We process the satellite data further to reduce the 10-bit GVAR data from GOES into 8-bit pixel values and create files with pixel resolutions of 10 km and 32 km data for ISCCP. The final ISCCP B1 (10 km) and B2 (32km) products created here at CIRA are "McIDAS" like files interleaved by pixels to handle all seven image products for each ISCCP hour. The data files are then written to tape and distributed to the ISCCP Archive Center (IAC) at NCDC and the ISCCP Global Processing Center (GPC) . The ISCCP B1 products collected since 1995 have been sent to NCAR to be place in the MSS for use by the UCAR community. A complete listing of the ISCCP B1 products can be found at the NCAR's Data Support Section (DSS). There is a HTML file and text file available that describes the ISCCP B1 and B2 file format, navigation, and calibration with source code in FORTRAN, C, and IDL.

A sample ISCCP B2 (test.b2d) file processed from GOES-9 data collected here at CIRA is available to help you construct programs to handle this format. To verify your results, JPEG images and header dump are available.

The GOES 12 satellite has some changes in the sensors, but the basic data used by ISCCP is still channels 1, the visible reflected channel and channel 4, the 11 um radiances. Since there is a delay at the Atmospheric Envirnment Center in Canada, CIRA is collecting GOES 12 data for ISCCP as the interim GOES East Sector Processing Center.

The ISCCP calibration is performed at the Centre de Meteorologie Spatiale, Lannion, France by comparing AVHRR (infrared and visible) or HIRS (water vapor) data to the geosynchronous data collected by the ISCCP Special Processing Centers around the globe to prepare intersatellite normalization paramters. For the GOES data collected at CSU, an image is collected and processed into the B1 data format. A sector is extracted from this B1 file and overlaped with an AVHRR image. In addition tables are constructed for the nominal GOES calibration to convert counts to radiances and temperatures. These tables are used by the ISCCP Satellite Calibration Center (SCC) for the comparisons of temperature between AVHRR/HIRS and GOES for channels with matching spectral responses. These tables can be used as an approximate calibration of the ISCCP B1 or B2 data products created by CIRA.

CIRA prepares IBM 3480 tapes with data from GOES-W for ISCCP Global Processing Center (GPC) to process into the final ISCCP D1 and D2 formats and for IAC to archive for future research requests of satellite data. The ISCCP B1 and B2 image files are written as "Unlabeled" tapes. Due to the variations in the transmission schedule from the geostationary satellite, full disk images are not available all the time for ISCCP and the Northern Hemisphere sector are often used as an replacement for the full disk or an abbreviated full disk maybe transmitted instead. These variations in the transmission results in B1 and B2 files to be in various record lengths. Special processing is done to the B2 images to place the smaller images into the full disk projection before the image file is sent to tape. This allows the images on the B2 tape to be the same size thoughout the tape. Check the ISCCP B2 tape page for further information.

PRESENTATIONS:

• Presented to NOAA Climate and Global Change Program - Climate Change Data and Detection Element
  Presented by Prof. Tom Vonder Haar, June 21, 2000

  • HTML Version - Works best in Microsoft Internet Explorer 5.0+
  • PowerPoint Version ~ 1.4 Mb
  • The apparent decreasing trend in global cloudiness is discussed in this Power Point presentation (~1.4 MB).

REFERENCES:

Campbell, G.G. and T.H. Vonder Haar, 1995, Multispectral GOES 8 Data for Climate Studies, IUGG, Boulder, CO.

Campbell, G.G. and T.H. Vonder Haar, 1994, GOES Pathfinder Cloud Retrievals, Seventh Conference on Satellite Meteorology and Oceanography, American Meteorological Society

Campbell, G.G., D.L. Randel and T.H. Vonder Haar, 1993, View of the Monthly Mean Climate with Many Parameters: Precipitable Water, Precipitation, Cloudiness, Radiation Budget and SST, Fourth Symposium on Global Change Studies of the AMS.

G.Garrett Campbell, July, 1999, Global Satellite Cloud Observation Data Sets, International Union of Geodesy and Geophysics, Birmingham, England

Campbell, G.G. and Thomas. H. Vonder Haar, 1997, Comparison of Surface Temperature Minimum and Maximum and Satellite Measured Cloudiness and Radiation Budget, J. Geophy. Res. Atmospheres., 102 No. D14, p 16639-16645.

Moore, R.W. and T.H. Vonder Haar, 2000, Interannual variations in Northern Hemisphere Winter Energy Transport Related to the El Nino-Southern Oscillation Phenomenon, CIRA Science Report, ISSN 0737-5352-45, p 116.