Hypoxia and nutrient dynamics affected by marine aquaculture in a monsoon-regulated tropical coastal lagoon

Abstract

The Laoyehai (lagoon) is located at the east coast of Hainan Island in the South China Sea and has been subject to perturbations from human activities, notably marine aquaculture, and has eutrophic surface and hypoxic near-bottom waters. A lack of knowledge of hydrodynamic and biogeochemical processes is a challenge to the sustainable management of lagoon at the ecosystem level in science. Five field campaigns, including three during the southwest monsoon and two in the northeast monsoon periods, were carried out at the Laoyehai in 2008–2011. The aim of this study is to investigate the impacts of dynamic processes of hydrography and human activities on nutrient geochemistry and their relationships to the system eutrophication and hypoxia in the lagoon. In this coastal system, high levels of ammonium relative to nitrate are found, elevated phosphate skews the DIN/DIP relative to the Redfield ratio, and the dissolved silicate concentration is high because of submarine groundwater discharge. The organic fraction in the Laoyehai accounts for a large proportion of the total nutrients associated with the release of wastes from marine aquaculture. The hypoxia of near-bottom waters in the Laoyehai is created and maintained by heterotrophic processes that are fueled by organic matter, which are exacerbated by poor water exchange as a consequence of the geomorphology and weak tidal circulation.

Appendix 1. Numerical model simulation for the Laoyehai

The 3-D ELCIRC hydrodynamic model solves for the free surface elevation, water velocity, and distribution of hydrographic parameters (e.g., salinity and temperature), using a set of six hydrostatic equations based on the Boussinesq approximation, which represent mass conservation in both 3-D and depth-integrated forms (Zhang and Baptista 2004). The governing equations are given below (A1–A4):

$$ \frac{\partial u}{\partial x}+\frac{\partial v}{\partial y}+\frac{\partial w}{\partial z}=0 $$

(A1)

$$ \frac{\partial \eta }{\partial t}+\frac{\partial }{\partial x}{\int}_{H_{R-\eta}}^{H_{R+\eta }} udz+\frac{\partial }{\partial y}{\int}_{H_{R-\eta}}^{H_{R+\eta }} udz=0 $$

(A2)

$$ \frac{Du}{Dt}= fv-\frac{\partial }{\partial x}\left\{g\left(\eta -\alpha \overset{\frown }{\psi}\right)+\frac{P_a}{\rho_0}\right\}-\frac{g}{\rho_0}{\int}_z^{H_{R+\eta }}\frac{\partial \rho }{\partial x} dz+\frac{\partial }{\partial z}\left({K}_{mv}\frac{\partial u}{\partial z}\right)+{F}_{mx} $$

(A3)

$$ \frac{Dv}{Dt}=- fu-\frac{\partial }{\partial y}\left\{g\left(n-\alpha \overset{\frown }{\psi}\right)+\frac{P_a}{\rho_0}\right\}-\frac{g}{\rho_0}{\int}_z^{H_{R+\eta }}\frac{\partial \rho }{\partial y} dz+\frac{\partial }{\partial z}\left({K}_{mv}\frac{\partial v}{\partial z}\right)+{F}_{my} $$

(A4)

where (x, y) are the horizontal Cartesian coordinates (m), ϕ, λ are the latitude and longitude, z is the vertical coordinate with upward being positive (m), t is time (s), H_R is the z-coordinate at the reference level (i.e., mean sea level, MSL), η(x, y, t) is the free surface elevation (m), and h(x, y) is the bathymetric depth (m); $\overrightarrow{u}\left(\overrightarrow{x},t\right)$ is the water velocity at $\overrightarrow{x}\left(x,y,z\right)$, having Cartesian components (u, v, and w), in m/s; f is the Coriolis factor (per s), G is the acceleration under gravity (m/s), ψ(ϕ, λ) is the tidal potential (m), and α is the effective Earth elasticity factor; $\rho \left(\overrightarrow{x},t\right)$ is the water density, for which the default reference value ρ₀ is 1025 kg/m³; and P_a(x, y, t) is the atmospheric pressure at the free surface (N/m), K_mv is the vertical eddy viscosity (m²/s), and F_mx, F_my is the horizontal diffusion for momentum and transport equations.

The differential system for the six primary variables (η, u, v, w, T, S) in Eqs. A1, A2, A3, and A4 is closed with the equation of state, the definition of the tidal potential and the Coriolis factor, parameterizations for vertical mixing, and initial and boundary conditions. Initial conditions require problem-dependent specification of pre-simulation fields for all primary variables and for any turbulence parameters required by the vertical mixing parameterization (Zhang and Baptista 2004).

In implementation, the model was firstly validated by time-series data of tidal level, current and hydrographic parameters (e.g., salinity and temperature) from anchor stations of 26 h for both spring and neap tides and the simulation results were compared to the section profiles of hydrography in different seasons (cf. HNMDDI 2011). Then the numerical simulations were applied to understand the hydrodynamic processes of Laoyehai using Lagrangian particles as passive tracer.