systems such as satellites. At the moment, however, only sketchy information exists on the suitability of remotely sensed data as a source for climate change detection and attribution. Some studies have shown a positive impact of these data; others have clearly indicated substantial problems, two of which are calibration and poorly documented changes in the instrument performance from one satellite to the next.

This report notes the need to build climate observing requirements into the operational programs as a high priority. That need is especially important as observing networks rely even more on remote sensing. Climate requirements should be considered in the planning process. There should be sufficient funding to test and evaluate new data sources before implementation, and support should be given to developing the new data assimilation and related techniques needed to make the transition from current to future observing systems.