A numerical analysis of biogeochemical controls with physical modulation on hypoxia during summer in the Pearl River estuary

Abstract. A three-dimensional (3-D) physical–biogeochemical coupled model was applied to explore the mechanisms controlling the dissolved oxygen (DO) dynamics and bottom hypoxia during summer in the Pearl River estuary (PRE). By using the numerical oxygen tracers, we proposed a new method (namely the physical modulation method) to quantify the contributions of boundary conditions and each source and sink process occurring in local and adjacent waters to the DO conditions. A mass balance analysis of DO based on the physical modulation method indicated that the DO conditions at the bottom layer were mainly controlled by the source and sink processes, among which the sediment oxygen demand (SOD) at the water–sediment interface and the re-aeration at the air–sea interface were the two primary processes determining the spatial extent and duration of bottom hypoxia in the PRE. The SOD could cause a significant decrease in the bottom DO concentrations (averaged over July–August 2006) by over 4 mg L⁻¹ on the shelf off the Modaomen sub-estuary, leading to the formation of a high-frequency zone of hypoxia (HFZ). However, the hypoxia that occurred in the HFZ was intermittent and distributed in a small area due to the combined effects of re-aeration and photosynthesis, which behaved as sources for DO and offset a portion of the DO consumed by SOD. The bottom DO concentrations to the west of the lower Lingdingyang Bay (i.e. the western shoal near Qi'ao Island) were also largely affected by high SOD, but there was no hypoxia occurring there because of the influence of re-aeration. Specifically, re-aeration could lead to an increase in the bottom DO concentrations by ∼ 4.8 mg L⁻¹ to the west of the lower Lingdingyang Bay. The re-aeration led to a strong vertical DO gradient between the surface and the lower layers. As a result, the majority (∼ 89 %) of DO supplemented by re-aeration was transported to the lower layers through vertical diffusion and ∼ 28 % reached the bottom eventually. Additional numerical experiments showed that turning off re-aeration could lead to an expansion of the hypoxic area from 237 to 2203 km² and result in persistent hypoxia (hypoxic frequency  >  80 %) to the west of the lower Lingdingyang Bay. Compared to re-aeration and SOD, photosynthesis and water column respiration had relatively small impacts on the DO conditions; turning off these two processes increased the hypoxic area to 591 km². In summary, our study explicitly elucidated the interactive impacts of physical and biogeochemical processes on the DO dynamics in the PRE, which is critical to understanding hypoxia in this shallow and river-dominated estuarine system.