Annual and Inter-annual Variability Coupled with Comparison of MODIS-AERONET Retrieved Aerosol Optical Depth over a Rural Site in the Central Indo-Gangetic Basin

Abstract

The long-term (2000–2015) MODIS EOS-Terra/Aqua multi-algorithm (DT, DB and combined DTB) retrieved AOD_{550 nm} and AERONET measured AOD_{550 nm} data at Gandhi College (24.87° N, 84.19° E; 60 m amsl), a rural site in the central IGB, have been employed to assess the performance of MODIS AOD products against AERONET AOD and to examine their annual and inter-annual variability. For both MODIS Terra/Aqua sensors, the linear regression data statistics reveals that the values of slopes for MODIS Terra-EOS sensor lie in the range 1.04 ± 0.03 [DB (QA = 2,3)] to 1.11 ± 0.02 [DT (QA = (2,3)] which are slightly higher than 1. Also, similar observation is noticed for MODIS Aqua-EOS sensor for which slopes of linear regression fit span over 0.99 ± 0.03 [DB (QA = 2,3)] – 1.16 ± 0.03 [DT (QA = 3)]. The intercept, however, approach zero values for both MODIS Terra/AQUA–EOS sensor at DT (QA = 3), DT (QA = 2,3), DB (QA = 2,3) and DTB combined retrieval algorithms. The evaluation/performance analysis, therefore, exhibits the observed near-perfect match of MODIS Terra/Aqua-EOS sensors derived AOD_550nm from all algorithms with AERONET measured AOD_{550 nm} as a result of magnitudes of the slope and intercept of the linear regression fit to the scatter diagrams of MODIS AOD₅₅₀ nm against AERONET AOD_{550 nm}. Results, thus indicate that at Gandhi College, the DT, and DTB combined retrieval algorithms satisfactorily estimate AOD products which can then be used to build aerosol climatology over Gandhi College. The MODIS and AERONET derived AOD_{550 nm} values over the Gandhi College indicate a distinct annual pattern with maximum AOD_{550 nm} during the winter season and minimum AOD_{550 nm} during monsoon season and winter-summer transition period. Analysis revealed that the aerosol loading starts building up over the study region from March to June of the pre-monsoon season mainly due to the high convective activity and long-range mineral dust transport from western arid regions. An increase in AOD values during the post-monsoon and winter season is primarily due to the influx of aerosols from biomass burning processes and stable atmospheric conditions, the MODIS Terra retrieved AOD_{550 nm} using DT and combined DTB algorithm showed higher values in the month of July of monsoon season as compared to MODIS Aqua retrieved AOD_{550 nm}. The inter-annual AOD trend analysis reveals that for MODIS Terra satellite there exits an increasing AOD trend during post-monsoon and winter seasons for DT (0.0233 year^–1), DB (0.0239 year^–1) and DTB (0.0246 year^–1), retrieval algorithms. For MODIS Aqua satellite also, there exists an increasing AOD_{550 nm} trend with differing but higher AOD year^–1 trend value. On the contrary, for monsoon and pre-monsoon seasons, for MODIS Terra and Aqua satellite, the AOD trends are found to be statistically insignificant.