SCIAMACHY (SCanning Imaging Absorption spectroMeter for Atmospheric CHartographY) on-board the European Environmental Satellite (Envisat) provided spectrally resolved measurements in the wavelength range from 0.24 to 2.4 µm by looking into the Earth’s atmosphere using different viewing geometries (limb, nadir, solar, and lunar occultation). These observations were used to derive a multitude of parameters, in particular atmospheric trace gas amounts. In addition to radiance measurements solar spectral irradiances (SSIs) were measured on a daily basis. The instrument was operating for nearly a decade, from August 2002 to April 2012. Due to the harsh space environment, it suffered from continuous optical degradation. As part of recent radiometric calibration activities an optical (physical) model was introduced that describes the behaviour of the scanner unit of SCIAMACHY with time (Krijger et al., 2014). This model approach accounts for optical degradation by assuming contamination layers on optical surfaces in the scanner unit. The variation in layer thicknesses of the various optical components is determined from the combination of solar measurements from different monitoring light paths available for SCIAMACHY. In this paper, we present an optimisation of this degradation correction approach, which in particular improves the solar spectral data. An essential part of the modification is the use of measurements from SCIAMACHY’s internal white light source (WLS) in combination with direct solar measurements. The WLS, as an independent light source, therefore, gives an opportunity to better separate instrument variations and natural solar variability. However, the WLS emission depends on its burning time and changes with time as well. To use these measurements in the optimised degradation correction, the change in the WLS emission in space needs to be characterised first. The changes in the WLS with accumulated burning time are in good agreement with detailed laboratory lamp studies by Sperling et al. (1996). Although the optimised degradation-corrected SCIAMACHY SSIs still show some instrumental issues when compared to SSI measurements from other instruments and model reconstructions, our study demonstrates the potential for the use of an internal WLS for degradation monitoring.