| description abstract | ncertainty requirements for climate observations are more stringent than for weather observations because of the scale dependency of natural variation. At present there is no space-based climate observing system, so weather observations have to be aggregated for the study of large-scale change. The management and minimization of uncertainty sources in weather observations are, therefore, a high priority. This work is a first attempt at investigating if a single long-term record can be assembled with temperature retrievals from three hyperspectral satellite sounders in polar orbit: the Atmospheric Infrared Sounder (AIRS) on Aqua in afternoon orbit since 2002, the Infrared Atmospheric Sounding Interferometer (IASI) on MetOp-A in morning orbit since 2006, and the Cross-track Infrared Sounder (CrIS) on board the Suomi National Polar-Orbiting Partnership in afternoon orbit since 2011. These instruments measure not only the vertical atmospheric structure but also atmospheric composition, thus providing coincident observations of many essential climate variables. Two main sources of known systematic differences, namely local sampling time (LST) and instrument type (i.e., interferometer vs grating spectrometer), are characterized. For a 5-yr record it is shown that differences among these temperature soundings are height dependent; a multi-instrument data record could be assembled from measurements of the stratosphere but this is not yet possible for the troposphere, where cloud effects are shown to be dominant and dependent on instrument type. Moreover, LST differences are limited to regional systematic effects in the lower troposphere and boundary layer. This research is a step toward the development of transparent and geophysically consistent methods with which to assemble weather measurements into climate observations. | |
