Satellite multi-sensor mapping of suspended particulate matter in turbid estuarine and coastal ocean, China

In this work, five ocean-colour sensors, the Moderate Resolution Imaging Spectroradiometer aboard the Terra satellite (Terra MODIS), Moderate Resolution Imaging Spectroradiometer aboard the Aqua satellite (Aqua MODIS), Medium Range Imaging Spectrometer aboard the Environmental Satellite (Envisat MERIS), Medium Resolution Spectral Imager aboard the FY-3 satellite (FY-3 MERSI), and Geostationary Ocean Colour Imager (GOCI), were selected to examine the compatibility of an algorithm proposed for suspended particulate matter (SPM) retrieval and concordance of satellite products retrieved from different ocean-colour sensors. The results could effectively increase revisit frequency and complement a temporal gap of time series satellites that may exist between on-orbit and off-orbit. Using in situ measurements from 17 cruise campaigns between 2004 and 2012, the SPM retrieval algorithm was recalibrated so as to be universal and adapted for multi-sensor retrievals. An inter-comparison of multi-sensor-derived products showed that GOCI-derived SPM and Envisat MERIS-derived SPM had the best fitting on a 1:1 scatterplot, with a statistic regression slope of 0.9617 and an intercept of 0.0041 (in units of g l–1), respectively. SPM products derived from three sensors with nearly synchronous transit, Envisat MERIS, Terra MODIS, and FY-3 MERSI, exhibited excellent accordance with mean differences of 0.056, 0.057, and 0.013 g l–1 in three field fixed stations, respectively, in the Yangtze estuary. Terra MODIS-derived SPM with GOCI-derived SPM, except in the high SPM waters of Hangzhou Bay, and Aqua MODIS-derived SPM with GOCI-derived SPM, except in the moderate SPM waters of the South Branch and south of the Subei Coast, showed a good correspondence. Meanwhile, synchronous multi-sensor-derived SPM with concurrent in situ SPM time series observed in fixed field stations mostly displayed a good correspondence. Results suggest that the algorithm is feasible and compatible for SPM retrieval by multiple sensors.