Post-processing to remove residual clouds from aerosol optical depth retrieved using the Advanced Along Track Scanning Radiometer

摘要

Cloud misclassification is a serious problem in theretrieval of aerosol optical depth (AOD), which might considerably bias theAOD results. On the one hand, residual cloud contamination leads to AODoverestimation, whereas the removal of high-AOD pixels (due to theirmisclassification as clouds) leads to underestimation. To removecloud-contaminated areas in AOD retrieved from reflectances measured withthe (Advanced) Along Track Scanning Radiometers (ATSR-2 and AATSR), usingthe ATSR dual-view algorithm (ADV) over land or the ATSR single-viewalgorithm (ASV) over ocean, a cloud post-processing (CPP) scheme has beendeveloped at the Finnish Meteorological Institute (FMI) as described inKolmonen et al. (2016). The application of this scheme results in theremoval of cloud-contaminated areas, providing spatially smoother AOD mapsand favourable comparison with AOD obtained from the ground-based referencemeasurements from the AERONET sun photometer network. However, closerinspection shows that the CPP also removes areas with elevated AOD not dueto cloud contamination, as shown in this paper. We present an improved CPPscheme which better discriminates between cloud-free and cloud-contaminatedareas. The CPP thresholds have been further evaluated and adjusted accordingto the findings. The thresholds for the detection of high-AOD regions ( 60 % of the retrieved pixels should be high-AOD ( 0.6)pixels), and cloud contamination criteria for low-AOD regions have been accepted asthe default for AOD global post-processing in the improved CPP. Retainingelevated AOD while effectively removing cloud-contaminated pixels affectsthe resulting global and regional mean AOD values as well as coverage.Effects of the CPP scheme on both spatial and temporal variation for theperiod 2002–2012 are discussed. With the improved CPP, the AOD coverageincreases by 10–15 % with respect to the existing scheme. The validationversus AERONET shows an improvement of the correlation coefficient from 0.84to 0.86 for the global data set for the period 2002–2012. The globalaggregated AOD over land for the period 2003–2011 is 0.163 with the improvedCPP compared to 0.144 with the existing scheme. The aggregated AOD overocean and globally (land and ocean together) is 0.164 with the improved CPPscheme (compared to 0.152 and 0.150 with the existing scheme, for oceanand globally respectively). Effects of the improved CPP scheme on the10-year time series are illustrated and seasonal and temporal changes arediscussed. The improved CPP method introduced here is applicable to otheraerosol retrieval algorithms. However, the thresholds for detecting the high-AODregions, which were developed for AATSR, might have to be adjusted to theactual features of the instruments.