Publications

Barbulescu, A (2022). On the Spatio-Temporal Characteristics of Aerosol Optical Depth in the Arabian Gulf Zone. ATMOSPHERE, 13(6), 857.

Abstract
The article investigates some of the available measurements (Terra MODIS satellite data) of the aerosol optical depth (AOD) taken in the Arabian Gulf, a zone traditionally affected by intense sand-related (or even sand-driven) meteorological events. The Principal Component Analysis (PCA) reveals the main subspace of the data. Clustering of the series was performed after selecting the optimal number of groups using 30 different methods, such as the silhouette, gap, Duda, Dunn, Hartigan, Hubert, etc. The AOD regional and temporal tendency detection was completed utilizing an original algorithm based on the dominant cluster found at the previous stage, resulting in the regional time series (RTS) and temporal time series (TTS). It was shown that the spatially-indexed time series (SITS) agglomerates along with the first PC. In contrast, six PCs are responsible for 60.5% of the variance in the case of the temporally-indexed time series (TITS). Both RTS and TTS are stationary in trend and fit the studied data series set well.

DOI:
10.3390/atmos13060857

ISSN:
2073-4433