Improved solar Lyman α irradiance modeling from 1947 through 1999 based on UARS observations

The solar Lyman alpha radiation is the brightest solar vacuum ultraviolet (VUV: lambda < 200 nm) emission, and this radiation is deposited in Earth's atmosphere above 70 km. The Lyman <alpha> irradiance and its variability are therefore important for many studies of the Earth's upper atmosphere. A long-term data set of the solar Lyman alpha irradiance from 1947 through 1999 is constructed using the measurements from the Atmospheric Explorer E (AE-E), the Solar Mesospheric Explorer (SME), and the Upper Atmosphere Research Satellite (UARS) along with predictions from proxy models to fill in data gaps and to extrapolate back to 1947. The UARS measurement is used as the reference, and the AE-E and SME measurements and the proxy models are adjusted to agree with the UARS values. The estimated I-a uncertainty for this long-term Lyman alpha time series is 10%. The average solar rotation (27-day) variability in Lyman alpha is 9% from this composite times series, and the solar rotation variability averaged over 2 years at solar minimum and maximum conditions is 5 and 11%, respectively. The average solar cycle (11-year) variability is a factor of 1.5 when the data are smoothed over 2 years, and the extreme Lyman alpha variability is a factor of 2.1. The Lyman alpha irradiances averaged over 2 years during solar minimum and maximum conditions are 3.7 and 5.6 x 10(11) photons s(-1) cm(-2), respectively. The proxy models include three components to better fit the UARS measurements; nonetheless, there remain differences between the proxy models and the observed Lyman alpha irradiance, which are related to the source of the Lyman alpha radiation being different than that for the proxies. The available proxies are primarily chromospheric and coronal emissions, whereas the Lyman alpha variability is manifested more in the transition region. It is shown that emissions throughout the chromosphere, transition region, and corona vary differently mainly because their contrasts for active network and plage components are different. A transition region proxy is needed to improve the empirical proxy models of solar irradiance, and this composite Lyman alpha time series could serve as a proxy for other transition region emissions.