Online monitoring of pollutants exporting into coastal waters is one of the key technique to realize marine pollution control and coordination of monitoring and management. In this study, based on salinity and fluorescence dissolved organic matter (FDOM) sensor data between 2014−2018 from a multi-parameter buoy deployed in Jiulong River Estuary, a rapid inversion model of chemical oxygen demand (COD) was established. Daily river end member COD concentrations were then extrapolated by effective concentration method. Combined with daily runoff data, daily estuarine export fluxes of COD between 2014 and 2017 were estimated. The factors controlling the flux variations were quantified by flux decomposition model. The results indicated: (1) The average concentration deviation of the inversion was (10.4±8.8)%, indicating the good reliability and stability of inversion model. (2) On seasonal scale, COD concentration in dry season of 2016 was lower than the other dry seasons, which was regulated by 2015−2016 super El Niño events. However, the sharp increase of runoff still significantly increased the COD export flux. (3) On annual scale, the COD export flux in 2016 from the Jiulong River 4.4 ×10⁴ t/a was significantly higher than the other three years (3.0−3.2)×10⁴ t/a. The abnormal precipitation in 2016 caused by 2015−2016 El Niño event was the major reason for such yearly variation. This study highlighted the application of online FDOM sensor system for high frequency monitoring of COD export. This would help to achieve long-term continuous high-frequency monitoring of land-based pollutants and their fluxes into the sea and their regulatory factors. This would provide important technical support for marine ecological environmental protection and management.

multi-parameter buoys  /  fluorescence dissolved organic matter  /  chemical oxygen demand  /  inversion  /  flux to the sea  /  El Niño event