The goal is to assess the feasibility of an ice crystal icing detection algorithm based exclusively on remote sensing data. Active measurements are used to train and validate a newly developed random forest algorithm that is applied to passive satellite imagery to estimate the ice crystal icing conditions probability. 83 % of ice crystal icing conditions are correctly detected, showing potential for an operational implementation to mitigate its negative effects on the fleet.

Solar irradiance monitoring is an integral part of climate research. But the major drawback was the larger uncertainties associated with the measurements. To minimize the uncertainties, improved calibration standards, measurement techniques and sensors were developed. In this study we are validating the a newly developed Bi-Tec sensor spectroradiometer with an average uncertainty of 0.53 % with world radiometric reference (WRR) and international system of units (SI).