Systems approach to developing a climate data record from satellite observations

A Climate Data Record (CDR) consists of a body of information of some observable of the Earth's climate that is of sufficient information content and accuracy to allow climate science to be performed with this record now and in the distant future. We examine the generation of a hyperspectral infrared CDR for the Atmospheric Infrared Sounder (AIRS) instrument as good example. For the information to be accurate, extensive pre-flight and in-flight calibration and characterization is required for spatial, spectral, polarimetric and radiometric performance. Maintaining the accuracy in orbit requires either inherent stability, or the ability to stabilize the calibration using additional information such as an ocean buoy network, or ground calibration sites. In addition to having excellent data quality, documentation and archiving are also critical. Details of the instrument design and test procedures must be recorded. All data acquired during the preflight testing must be archived in a format and medium that will be accessible by computers several decades into the future. A Systems Engineering approach is used to define the requirements for the AIRS hyperspectral infrared climate data record, for performance, characterization, and documentation. Examples are given from the AIRS project activities on how the record can be created including a comprehensive drawing database, a document archive for all pre-flight and in-flight procedures and reports, software and data archiving, and instrument performance verification and validation for compliance with climate science requirements.