The presence of clouds in the High Arctic plays a fundamental role in the energy balance of the region, particularly in relation to the seasonal prolonged periods of darkness and daylight. To investigate the properties of Arctic clouds, an Extended-range Atmospheric Emitted Radiance Interferometer (E-AERI) was installed at the Polar Environment Atmospheric Research Laboratory (PEARL) at Eureka, Canada (80°N, 86°W) in October 2008. The E-AERI is a moderate resolution (1 cm−1) Fourier Transform Infrared Spectrometer that measures the absolute downwelling infrared spectral radiance from the atmosphere between 400 and 3000 cm−1. Spectra are recorded every 7 minutes year round, including during the polar night.
Using observations from the E-AERI, an iterative optimal nonlinear inverse method based on the Levenberg–Marquardt algorithm, the CLoud and Atmospheric Radiation Retrieval Algorithm (CLARRA) is used to retrieve cloud properties. This retrieval uses radiances within selected microwindows in the E-AERI observation spectrum, particularly the 400-600 and 750-1300 cm-1 regions, which are optically sensitive to the retrieval variables. A preliminary compilation of the retrieved microphysical properties (cloud height, temperature, optical depth, phase, and particle effective radii) at Eureka will be presented. A second retrieval approach based on statistical machine learning techniques is also presented, evaluated, and used to perform cloud microphysical property retrievals using E-AERI data as with the CLARRA retrieval. Comparisons will also be made between these two datasets, and a previous study of the retrievals of such cloud properties using observations from a precursor instrument to the E-AERI (the Polar-AERI), which was located at Eureka between 2006-2009.