Development of a dust detection algorithm based on sky radiometer measurements in Hong Kong

Abstract

In this thesis a new algorithm to determine a dust index based on aerosol measurements using a sky radiometer is presented. The sky radiometer measurements at City University of Hong Kong were carried out from September 2008 to May 2011. This type of measurement allows for the retrieval of aerosol size distributions as well as aerosol optical depths. In order to validate the measurement data, the optical depths were compared to MODIS satellite and a nearby AERONET data. The HYSPLIT backward trajectory model was used to trace the origin of air mass during the measurement period, and the paths of air mass correlated with the measured AODs were studied. Comparisons between the nearby AERONET AOD measurement in wavelength channel 500nm, 675nm, 870nm and 1020nm showed linear correlation with correlation coefficients of 0.865, 0.936, 0.943 and 0.886. The fine mode volume concentration retrieved from both size distribution measurements shows a linear correlation of R2 = 0.799, while the coarse mode volume concentration shows a lower R2 value of 0.603. As an additional validation study, the data had been interpolated to 550nm and compared to AOD data from the MODIS instruments on board of the TERRA and AQUA satellite, showing a linear correlation coefficient of R2 = 0.745 and R2 = 0.851 respectively. Monthly mean AOD at 500nm showed maximum in March 2010 and minimum in July 2010, while the single scattering albedo showed maximum in September 2008 and minimum in July 2009. The aerosol size distributions derived from the sky radiometer measurements were decomposed into typical size distributions from different aerosol types. One of the components representing the dust aerosol concentration was used in combination with the single scattering albedo to calculate a dust index. As a case study, an event of unusually high dust concentrations in Hong Kong in March 2010, caused by dust transported from the Gobi desert, has been analyzed.

Date of Award	14 Feb 2014
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Mark Oliver WENIG (Supervisor) & Chung Leung Johnny CHAN (Co-supervisor)