Observations of "hysteresis'' in backscattered ultraviolet ozone data

Satellite measurements using the backscattered ultraviolet technique provid e a powerful method for the observation of stratospheric ozone. However, ra pid input signal variations over three to four orders of magnitude in sever al minutes can lead to problems with instrument response. Inflight data hav e recently been used to characterize a "hysteresis'' problem on the NOAA-9 SBUV/2 instrument, which affects measurements made shortly after emerging f rom darkness. Radiance values observed under these conditions can be up to 2%-3% lower than expected. A correction has been derived for NOAA-9 data th at is solar zenith angle dependent and varies in amplitude and time. Typica l changes to affected polar total ozone values are on the order of 1% but c an reach 5% in some cases. Profile ozone changes are altitude dependent, wi th maximum values of 4%-5% at 1 hPa. The NOAA-11 and NOAA-14 SBUV/2 instrum ents have a much smaller hysteresis effect than that observed for NOAA-9 SB UV/2 due to a change in photomultiplier tubes. The Nimbus-7 SBUV instrument also shows a hysteresis effect, which has not been fully characterized at this time.