Jb. Kumer et al., COMPARISON OF CLAES PRELIMINARY N2O5 DATA WITH CORRELATIVE DATA AND AMODEL, JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES, 101(D6), 1996, pp. 9657-9677
The cryogenic limb etalon array spectrometer (CLAES) aboard the Upper
Atmosphere Research Satellite (UARS) has made near-global measurements
of N2O5. Data for 388 days have been processed to version 7 (V7) for
the period from January 9, 1992, to April 25, 1993. Results from UARS
instruments, including CLAES and the improved stratospheric and mesosp
heric sounder (ISAMS) provide the first near-global N2O5 measurements.
Retrieval below 3.16 mbar is adversely affected by aerosols and above
1.47 mbar by lack of signal and possible instrument effects, so data
usage is recommended for just the three ''UARs: pressure surfaces'' 3.
16, 2.15, and 1.47 mbar. A comparison of the diurnal data variation wi
th the model suggests there are offsets in the data that are to first
order diurnally independent. These offsets are tabulated to facilitate
subtraction, which is recommended for most data applications. Candida
te mechanisms for the offsets are discussed. Comparisons of CLAES data
with the offsets subtracted, with profiles obtained by the shuttle-de
ployed Atmospheric Trace Molecule Spectroscopy (ATMOS) experiment and
concurrent ISAMS profiles, and with a profile obtained by the balloon-
borne NASA Jet Propulsion Laboratory (JPL) MARK IV Instrument, show po
orest agreement in equatorial regions at 3.15 mbar where CLAES values
are larger by about 30 to 40%. At higher altitudes and latitudes the c
omparison improves and tends toward consistency with systematic error
estimates that are based on instrument and retrieval process character
ization and range from 14% at 3.16 mbar to 21% at 1.47 mbar. A similar
estimate of random CLAES error ranges from 7% at 3.16 mbar 26% at 1.4
7 mbar. By comparison, the average values of the error estimates gener
ated by the production processing algorithm at 3.16 and 1.47 mbar are
8 and 36%, respectively, and the average values derived from the obser
ved data variability are 19 and 24%. Confidence is enhanced by the goo
d global scale agreement and correlation of CLAES and ISAMS during an
N2O5 enhancement event in early-mid-January 1992 polar winter, in whic
h values >5.5 parts per billion at 3.16 mbar by volume are observed. A
description of artifacts that may occur at 3.16 mbar and much less fr
equently at 2.15 mbar, during this and other enhancement conditions, a
nd the demonstrated approach to eliminate these in future versions, is
given in the text.