The absolute solar spectral irradiance from 200 to 2500 nm as measured by the SOLSPEC spectrometer with the ATLAS and EURECA missions

The SOLSPEC instrument measures the absolute solar spectral irradiance from 200 to 2500 nn. The instrument flew in March 1992, March 1993 and November 1994 with the three ATLAS missions during 10 days. SOSP is the spare unit of SOLSPEC. It flew on the EURECA platform from ii August 1992 to May 1993. SOLSPEC and SOSP are made of three spectrometers and contains several lamp s allowing to check in flight, the instrument stability and its wavelength scale. These two units have an identical design and are made of the same component s. However, their detectors have different performance leading to a SOSP re sponsivity smaller than the SOLSPEC responsivity. As during the ATLAS missi ons two other spectrometers were also observing in the UV and near visible domain, we took advantage of this situation by choosing SOLSPEC for the thr ee ATLAS missions. Afterward, it appeared that it was also the good choice for the IR channel of SOSP on board EURECA. The calibration of the instrument is performed with the blackbody of the He idelberg observatory (Mandel et al., 1998). Its temperature is adjusted as a function of the wavelength domain, ranging from 3000 K for UV calibration , down to 2700 K for the IR spectrometer calibration. The blackbody tempera ture is measured with a pyrometer calibrated by the Physikalisch-Technische Bundesanstalt (PTB) of Berlin (Germany). The mean accuracy of the solar ir radiance is estimated to be 2 to 3% in UV and 2% in the visible and infrare d domains. In details, it depends on the spectral interval being better in the middle of each spectral domain. We shall present the best data obtained during the ATLAS and EURECA mission s, e.g., UV and visible spectra from the three ATLAS missions and the IR re sults from EURECA. The duration of each ATLAS mission does not allow study of the solar variab ility. This is why the best spectra are selected. This selection is based o n several criteria including the thermal conditions, stability of the point ing, linearity of the measurements, stability of the wavelength scale, posi tion of the Sun in the field of view and tangent height of the line of sigh t observations greater than 100 km. The selected spectra are averaged after ward. A similar procedure was also applied to the SOSP infrared data becaus e the solar variability in this domain is very small and is not detectable with SOSP. Special care was dedicated to this channel due to the variable r esponsivity in the field of view. Spectra which were not measured in the co ndition of the laboratory calibration, were eliminated. (C) 2000 Elsevier S cience Ltd. All rights reserved.