Comparison of Sun photometer calibration by use of the Langley technique and the standard lamp.

Asix-channel Sun photometer has been calibrated by means of two different methods: Langley plots and standard irradiance lamps. A 4-month calibration campaign was carried out at a high mountain site, Jungfraujoch (3580 m above sea level), in the Swiss Alps. Calibration constants V(0)(λ) determined on clear and stable days by means of a refined Langley-plot technique scatter by less than 0.25% (rms) for wavelengths outside of strong gaseous absorption bands. Inside the 0.94-µm water-vapor absorption band, the V(0)(λ) values retrieved by means of modified Langley plots scatter by 1.0% (rms). Repeated calibrations of the Sun photometer by means of irradiance standard lamps were performed at the World Radiation Center in Davos. The comparison of both methods ranges from perfect agreement to a deviation of 4.9% for the different channels. A discussion of the errors introduced by both methods shows that the Langley-plot calibration, when performed under very clear atmospheric conditions, is superior. However, by means of the standard-lamp calibrations a temporal degradation of the instrument's response up to 4.6% per year was found, implying that a single calibration campaign as done here is not sufficient. Thus we recommend the use of a combination of both methods for maintaining an accurate calibration.