An Assessment of Reprocessed GPS/MET Observations Spanning 1995-1997
Presenter:
Anthony J Mannucci
Jet Propulsion Laboratory, California Institute of Technology
(1)Jet Propulsion Laboratory, California Institute of Technology, (2)PlanetIQ, Boulder, CO, (3)European Centre for Medium-Range Weather Forecasts, Reading, UK, (4)University Corporation for Atmospheric Research
Talk
The GPS radio occultation (RO) remote sensing technique is now spanning nearly 25 years, having begun in 1995 with the GPS/MET proof-of-concept mission. The presence of multi-year gaps in the RO temperature record has not prevented the use of GPS/MET data in climate trend studies that extend from 1995 to the years 2010 and beyond, encompassing the COSMIC era of observations. These studies have established that temperatures derived from different RO instruments are well calibrated and do not require bias adjustment using periods of overlapping time series. Climate trend studies using GPS/MET data have used only a small fraction of atmospheric profiles currently available from the COSMIC Data Analysis and Archiving Center (CDAAC) spanning 1995-1997. The profiles used thus far correspond to four 2-week periods when signal encryption (“anti-spoofing” or AS) was turned off enabling standard dual-frequency processing. A much larger reprocessed GPS/MET data set exists with a modified dual-frequency technique that can accommodate AS-on periods. Recently, 18 days of GPS/MET profiles from this more extensive data set were used in assimilation experiments with a new ECMWF reanalysis. The results showed that the reprocessed GPS/MET data improves reanalysis fits to radiosondes despite their noisier characteristics compared to AS-off retrievals. In this talk, we provide a preliminary analysis of the reprocessed GPS/MET retrievals spanning 1995-1997, comparing them to NASA’s MERRA2 reanalysis and to a single-frequency processing method developed at JPL. Strengths and limitations of single-frequency processing will be discussed, including its potential application to data sets where encryption of the GPS L2 signal frequency prevents full use of dual frequency processing.
Presentation in PDF:
