Accurately measuring long-term solar UV variability is an experimental
challenge because instrument response degradations are typically larg
e enough to obscure solar change. For satellite instruments, one solut
ion is a series of regular comparisons with a well-calibrated referenc
e. The NOAB 11 Solar Backscatter Ultraviolet, model 2 (SBUV/2) instrum
ent made solar spectral irradiance measurements between 170 and 400 nm
from December 1988 to October 1994, covering the maximum and most of
the decline of solar cycle 22. The NOAA 1 1 irradiance data were corre
cted for long-term instrument sensitivity changes using comparisons wi
th coincident flights of the Shuttle SBUV (SSBUV) instrument. The NOAA
11 data show a decrease of 7.0(+/-1.8)% in smoothed 200-208 nm irradi
ance from Cycle 22 maximum in mid-1989 to October 1994, near solar min
imum. The long-term decrease in solar-irradiance at 250 nm was similar
to 3.5(+/-1.8)%. Longward of 300 nm, no solar variations were observe
d to within the 1% accuracy of the data. The NOAA 11 measurements over
lap observations from the Upper Atmosphere Research Satellite (UARS) S
olar Stellar Irradiance Comparison Experiment (SOLSTICE) and Solar Ult
raviolet Spectral Irradiance Monitor (SUSIM) instruments from October
1991 to October 1994, providing the first opportunity to compare three
coincident long-term solar UV irradiance data sets. We find reasonabl
e agreement between the NOAA 11, SOLSTICE, and SUSIM results at all wa
velengths in the 170-400 nm region. Power spectral analysis gives cons
istent results for all three instruments on solar rotational timescale
s, and reveals the evolution of solar rotation periodicity and strengt
h during a solar cycle. We find significant differences between instru
ments in both period and spectral location when the spectral irradianc
e data are analyzed on intermediate (50-250 days) timescales, The NOAA
11 spectral irradiance data provide a valuable complement to the UARS
solar data, and capture the entire maximum of solar cycle 22.