Comparison of Solar UV Spectral Irradiance from SUSIM and SORCE

Knowledge of solar spectral irradiance (SSI) is important in determining the impact of solar variability on climate. Observations of UV SSI have been made by the Solar Ultraviolet Spectral Irradiance Monitor (SUSIM) on the Upper Atmosphere Research Satellite (UARS), the Solar-Stellar Irradiance Comparison Experiment (SOLSTICE), and the Solar Irradiance Monitor (SIM), both on the Solar Radiation and Climate Experiment (SORCE) satellite. Measurements by SUSIM and SORCE overlapped from 2003 to 2005. SUSIM and SORCE observations represent ～?20 years of absolute UV SSI. Unfortunately, significant differences exist between these two data sets. In particular, changes in SORCE UV SSI measurements, gathered at moderate and minimum solar activity, are a factor of two greater than the changes in SUSIM observations over the entire solar cycle. In addition, SORCE UV SSI have a substantially different relationship with the Mg ii index than did earlier UV SSI observations. Acceptance of these new SORCE results impose significant changes on our understanding of UV SSI variation. Alternatively, these differences in UV SSI observations indicate that some or all of these instruments have changes in instrument responsivity that are not fully accounted for by the current calibration. In this study, we compare UV SSI changes from SUSIM with those from SIM and SOLSTICE. The primary results are that (1) long-term observations by SUSIM and SORCE generally do not agree during the overlap period (2003?–?2005), (2) SUSIM observations during this overlap period are consistent with an SSI model based on Mg ii and early SUSIM SSI, and (3) when comparing the spectral irradiance for times of similar solar activity on either side of solar minimum, SUSIM observations show slight differences while the SORCE observations show variations that increase with time between spectra. Based on this work, we conclude that the instrument responsivity for SOLSTICE and SIM need to be reevaluated before these results can be used for climate-modeling studies.