Criteria for AURA calibration/validation
We consider first criteria that must be met for AURA calibration/validation and for AURA collaborative science that proposed instruments must satisfy. In situ instruments must address the resolution of structure, the achievement of accuracy, and establishment of collaborative strategy and the accomplishment of critically accepted validation of satellite observations. These criteria set in place important requirements:
- Sufficient number of species measured with coordinated and repetitive observations. The Earth-observing satellites scheduled to be launched in the next few years will produce a wealth of new measurements of atmospheric gases. On the AURA satellite alone, four instruments (HIRDLS, MLS, OMI, and TES) will monitor concentrations of more than twenty species. Each of these requires validation by in situ measurements, under a variety of conditions, locations, and concentration levels. Tracer-tracer correlations obtained with simultaneous observations of different transport tracers are valuable for assuring overlap between the aircraft flight track and the satellite footprint, providing recognition that the same airmass is in fact being sampled. The inhomogeneity of air within the satellite footprint also mandates multiple simultaneous observations, to provide a check that the in situ profiles can be accurately used to infer the mean concentrations a satellite instrument would observe. This is especially important in regions where pollutant transport occurs in vertically thin layers. In general, in situ validations are both more efficient and more accurate if multiple measurements can be made with the same or similar instrument.
- Sufficient accuracy and precision. In situ instruments used for comparison with remote observations must have accuracy and precision exceeding those of the remote instruments in order to produce a meaningful validation. An intercomparison that merely concludes that the instruments agree to within their respective error bars is insufficient, and a remote instrument with no better validation cannot produce scientific conclusions that will withstand the scrutiny of public policy debate. The AURA validation document emphasizes this point strongly, stating that remote instruments must be validated against “similar data products of high quality that have already been validated themselves” (emphasis theirs). It is critical that these accuracies can be substantiated to the satisfaction of the entire community.
- Cost-effective operation over a wide range of the lower and middle atmosphere. The science questions ESE is addressingspecifically the transport and effect of pollutants, ozone depletion, climate change, and the sources and sinks of atmospheric carbonrequire measurements from the boundary layer to the upper troposphere and lower stratosphere (UT/LS). This requirement in turn mandates a much broader type of measurement campaign than that needed for purely stratospheric research. In comparison to the stably-stratified stratosphere, the troposphere contains large inhomogeneities, and is characterized by characteristic timescales for chemical production/loss on the same order as that from flux divergence, so that measurements must be taken over a far broader area to obtain a representative view. The number of species relevant to tropospheric chemistryespecially pollution issuesis much greater than for stratospheric issues, meaning that each platform used must carry instrumentation for a more comprehensive set of species measurements than in a typical stratospheric campaign.
- Powerful complementary observational and modeling strategy that opens avenues for collaborative science. A true validation of a remote instrument entails not only confirming the accuracy of the measurement, but also confirming that the remote measurement is in fact meaningful for the science in question. Measurements of CO profiles from space with several km altitude resolution, for example, may not be sufficient to determine the chemical effects of pollutant plumes in the troposphere. Different pollutant distributions, with very different consequences for tropospheric chemistry, might still produce the same remove measurement. Local, high-resolution, in situ studies are required to determine the robustness of the scientific interpretation of remote observations.