Inter-comparison between GNSS RO and hyperspectral infrared soundings and combined retrieval results
Presenter:
Shu-peng Ho
NOAA/NESDIS/STAR
ESSIC/CISESS-MD, University of Maryland; NOAA/NESDIS/STAR; NOAA/NESDIS/STAR
Talk
The global navigation satellite system (GNSS) radio occultation (RO) data provide accurate, stable, and high vertical resolution profiles of atmospheric conditions and do not require calibration in the traditional sense. On the other hand, hyperspectral infrared (IR) sounders, such as the Cross-track Infrared Sounder (CrIS) on the Suomi National Polar-orbiting Partnership (NPP) and NOAA-20, and the Infrared Atmospheric Sounding Interferometer (IASI) on MetOp satellites, also provide stable and accurate measurements from which atmospheric properties can be derived.
Calibration of IR sounders is performed using onboard calibration targets such as high quality International System of Units (SI) traceable blackbodies. In turn, IR observations from hyperspectral sounders can be used as an on-orbit calibration reference for other broad- or narrow-band IR observations. Validation of these IR hyperspectral observations can be performed using well-understood reference data including GNSS-RO measurements. Because both sets of observations are vital to NWP and validation of other measurements, it is desirable to establish the link between GNSS RO and IR sounder observations.
In this study, we compare the IR sounder measurements from the CrIS on-board NOAA-20 to GNSS RO measurements from various satellites including COSMIC, KOMPSAT-5, and MetOp-A,-B,-C. To do so, we simulate CrIS spectral radiances near the 15 um CO2 absorption bands (in the low Stratosphere) by using 1) thermodynamic profile data retrieved from RO observations and 2) radiative transfer models (e.g. Community Radiative Transfer Model (CRTM)) . We compare the differences (i.e. biases) between the radiances from CrIS observations and simulated radiances based on GNSS RO in order to characterize the monthly, latitudinal, and scan-angle dependence of the biases. This information is expected to provide a better understanding of the potential root causes of the identified differences. The following questions are addressed as part of this study: Are there any differences between radiative transfer calculations based on GPS RO observations and CrIS observations? What are the root causes of these differences? Are the differences attributable to the GNSS RO observations, RTM models, or potential radiometric biases on the CrIS observations?
Additionally, the tools developed as part of this work are planned to support the calibration and validation effort of the data derived from the FORMOSAT-7/COSMIC-2 constellation. The findings from this study are not only useful for data assimilation (DA) into NWP models, but also to provide direct support to the COSMIC-2 and NOAA-20 CrIS calibration and validation efforts.
Presentation in PDF:
