Publications

Paszkuta, M; Zapadka, T; Krezel, A (2019). Assessment of cloudiness for use in environmental marine research. INTERNATIONAL JOURNAL OF REMOTE SENSING, 40(24), 9439-9459.

Abstract
A very simple algorithm with which to assess cloudiness over the Baltic Sea has been developed. The algorithm involves analysis of cloud longwave emissivity, supplemented with available information on shortwave emissions. The simple classification method is based on the split-window technique and compares a signal recorded in the neighbouring spectral bands with results of cloud-free atmosphere modelling. The algorithm was tested using data from the SEVIRI (Spinning Enhanced Visible and InfraRed Imager) satellite radiometer, but the method can be easily implemented for other data sources. Compatibility with instant shipborne radiation measurements was obtained. The effectiveness of the comparison was determined from the Hanssen-Kuiper score. The CM SAF (Satellite Application Facility on Climate Monitoring) data comparison shows that the system correctly classified areas in 88%, the satellite map accuracy was 46% with a 54% overvaluation error. The accuracy of the detection system was 44%, and the underestimation error was 56%. The comparison of modelling results showed 71% of correctly classified areas, the accuracy of the satellite map was 35%, and the repricing error was 65%. The accuracy of the detection system was 36%, and the underestimation error was 64%. The concordance of satellite measurements (Cohen's kappa coefficient) was at a level of 98% and 99% between the satellite and model. Statistically, the analysis is based on the results of measurements for the Baltic Sea basin in 2015. The obtained data on the monthly cloudiness was compared with the operational, model and lidar data. A difference between the solution proposed and the approaches derived from prognostic models is demonstrated. The regional scale solution is shown to be a better representation of the true state of the atmosphere, compared to global-scale models. The results obtained show cloudiness assessments to differ depending on the model applied (CM SAF, UMPL - Unifed Model for Poland). The average annual cloudiness of the Baltic Sea is around 58.5%. It is a repeatable value by other solutions based on various methods and sources (MODIS - Moderate Resolution Imaging Spectroradiometer), including lidar ones (CALIPSO - Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation). In addition, a potential for applying the regional cloudiness assessment to estimate the top-down components of the radiation budget is illustrated.

DOI:
10.1080/01431161.2019.1633697

ISSN:
0143-1161