Synergistic controls of water column stability and groundwater phosphate on coastal algal blooms.

Algal blooms have been identified as one major threat to coastal safety and marine ecosystem functioning, but the dominant mechanism regulating the formation of algal blooms remains controversial, ranging from physical control (via water column stability), the chemical control (via coastal nutrients) to joint control. Here we leveraged the unique data collected in the Hong Kong water over the annual cycle and past three decades, including direct observations of algal blooms and coastal nutrients and process model output of water column stability, and evaluated the differential competing hypotheses in regulating algal blooms. Our results demonstrate that the joint mechanism rather than the single mechanism effectively predicts all algal blooms. Meanwhile, we observed that the adequate nutrients (phosphate, PO43-) significantly originate from coastal groundwater. The production and fluctuation of PO43- in beach aquifers are primarily governed by groundwater temperature, leading to a sustained and sufficient supply of PO43- in a low groundwater temperature environment. Furthermore, along with submarine groundwater discharge (SGD), the ongoing release of PO43- in groundwater enters coastal waters and serves as sufficient nourishment for promoting algal blooms in coastal areas. These results highlight the importance of both physical and chemical mechanisms, as well as SGD, in regulating coastal algal blooms. These findings have practical implications for the prevention of coastal algal blooms and provide insights into mariculture, water security, and the sustainability of coastal ecosystems.

A low-cost edge AI-chip-based system for real-time algae species classification and HAB prediction.

Harmful Algal Blooms (HAB) are damaging to ecosystem functions and pose challenges to environmental and fisheries management. The key to HAB management and understanding the complex algal growth dynamics is the development of robust systems for real-time monitoring of algae populations and species. Previous algae classification studies mainly rely on the combination of an in-situ imaging flow cytometer and an off-site lab-based algae classification model such as Random Forest (RF) for the analysis of high-throughput images. An on-site AI algae monitoring system on top of an edge AI chip embedded with the proposed Algal Morphology Deep Neural Network (AMDNN) model is developed to achieve real-time algae species classification and HAB prediction. Based on a detailed examination of real-world algae images, dataset augmentation is first performed: consisting of orientation, flipping, blurring, and Resizing with Aspect ratio Preserved (RAP). The dataset augmentation is shown to significantly improve classification performance which is superior to that of the competitive RF model. And the attention heatmaps show that for relatively regular-shaped algal species (e.g., Vicicitus), the model weights the color and texture information heavily; while the shape-related features are more important for complex-shaped algae (e.g., Chaetoceros). The AMDNN is tested on a dataset of 11,250 algae images containing the 25 most common HAB classes in Hong Kong subtropical waters with 99.87% test accuracy. Based on the fast and accurate algae classification, the AI-chip-based on-site system is applied to a one-month dataset in February 2020; the predicted trends of total cell counts and targeted HAB species counts are in good agreement with observations. The proposed edge AI algae monitoring system provides a platform for the development of practical HAB early warning systems that can effectively support environmental risk and fisheries management.

A remotely controlled automated field measurement system for light extinction in coastal waters.

The Secchi disk depth (SD) is an important parameter in aquatic ecosystem monitoring. As algal growth depends on solar irradiation, the SD - a measure of light extinction - gives an indirect indication of the chlorophyll concentration. However, most SD measurements are manually based and too sparse to resolve water quality variations during algal blooms. A remotely controlled automatic system for field measurement of light extinction has been developed and installed in three marine fish culture zones in Hong Kong. The visual images of the disk at different prescribed depths and the surrounding water are taken. Based on the contrast theory and image analysis, the recorded light intensity distributions can be analyzed to give the SD and the light extinction coefficient. The method has been extensively verified by field data over a wide range of water quality and hydro-meteorological conditions. The proposed system enables high frequency SD measurements on demand for environmental management and emergency response.

Uncertainty evaluation and performance assessment of water quality model for mariculture management.

Ecosystem based water quality models are important tools for prognostic site assessment and evaluation of ecosystem-performance of marine fish farms. We present the development and application of a comprehensive Fish Culture Zone Water Quality Model using continuous bi-weekly field data over a six-year period (2012-2017). The model simulates five interacting subsystems: phytoplankton, phosphorus and nitrogen cycles, and the dissolved oxygen (DO) and particulate organic carbon balance. The application of the model to two fish culture zones in Hong Kong shows the model captures the trends of nutrient and DO variation and the performance in quantitative prediction of algal biomass is challenging. The effect of errors in the specification of primary model inputs are evaluated using dimensionless sensitivity coefficients based on First Order Variance Analysis reveals the relative importance of fish stock (loading), physical size (volume), tidal flushing rate and boundary conditions in the prediction of key water quality variables.

Prediction of lead leaching from galvanic corrosion of lead-containing components in copper pipe drinking water supply systems.

Galvanic corrosion is one of the main reasons for pipe degradation and lead contamination in drinking water systems. The electrical connection of dissimilar metals in corrosive tap water accelerates the dissolution rate of lead from leaded materials. This paper reports an electrochemistry based model to predict lead leaching from a copper pipe fitted with leaded connections. The corrosion of lead at the metal-electrolyte interface depends on the charge transfer and the electric field across the interface. The electric potential field and the mass transport process are dynamically coupled for corrosion propagation in stagnant water; they are respectively governed by the conservation of charge and reactant mass. Using polarization parameters for the electrodes as a function of concentration of oxidizing agents, a dynamic electrochemical model is developed to predict lead leaching from galvanic corrosion. The predicted lead and copper leaching curves are in good agreement with the experimental data for a lead-soldered coupled copper pipe, a brass valve coupled copper pipe, and a pure copper pipe. The findings offer a quantitative understanding on galvanic corrosion in drinking water supply systems and a practical modeling framework for prediction of lead leaching in tap water as a function of stagnation time.

Development of Predictive Models for "Very Poor" Beach Water Quality Gradings Using Class-Imbalance Learning.

Statistical water quality forecast models are useful tools to assist with beach management. In particular, multiple linear regression (MLR) models have been successfully developed for prediction of fecal indicator bacteria concentrations for beaches in river, lake, and marine environments. Nevertheless, an unresolved challenging issue is the reliable prediction of infrequent events of high bacterial concentrations to inform beach closure decisions to protect public health. The number of field data available for the infrequent events is typically an order of magnitude less than that for days when the water quality criterion is met-MLR models often perform poorly in predicting bacterial concentrations on days when the beaches should be closed. For beach management in Hong Kong, MLR models have been developed to predict beach water quality indices in terms of four gradings (BWQI-1 to 4) based on Escherichia coli (E. coli) concentrations. In this study, we propose an artificial intelligence (AI)-based binary classification (EasyEnsemble) model using class-imbalance learning to predict "very poor" occasions (BWQI-4)-when E. coli concentration exceeds 610 counts/100 mL. Models are developed for three marine beaches with different hydrographic and pollution characteristics using a 30 year data set spanning three periods with different water quality status. The model-data comparison over a wide range of conditions shows that the proposed method results in a significant improvement in the prediction of "very poor" water quality. The proposed class-imbalance method for predicting rare events has an F-score of 0.84, and it significantly outperforms MLR and classification tree (CT) models with corresponding F-scores of 0.39 and 0.69. A robust beach water quality forecast system can hence be developed using hybrid MLR-binary classification modeling.

A real time data driven algal bloom risk forecast system for mariculture management.

In eutrophic coastal waters, harmful algal blooms (HAB) often occur and present challenges to environmental and fisheries management. Despite decades of research on HAB early warning systems, the field validation of algal bloom forecast models have received scant attention. We propose a daily algal bloom risk forecast system based on: (i) a vertical stability theory verified against 191 past algal bloom events; and (ii) a data-driven artificial neural network (ANN) model that assimilates high frequency data to predict sea surface temperature (SST), vertical temperature and salinity differential with an accuracy of 0.35oC, 0.51oC, and 0.58 psu respectively. The model does not rely on past chlorophyll measurements and has been validated against extensive field data. Operational forecasts are illustrated for representative algal bloom events at a marine fish farm in Tolo Harbour, Hong Kong. The robust model can assist with traditional onsite monitoring as well as artificial-intelligence (AI) based methods.

Unraveling the Causes of Excess Lead in Drinking Water Supply Systems of Densely Populated High-Rise Buildings in Hong Kong.

Excess concentrations of lead (Pb) were found in tap water from drinking water supply systems of high-rise buildings in 11 public rental housing (PRH) estates in Hong Kong, posing threats to public health. The copper supply lines are fitted with lead-soldered connections and brass fixtures and faucets. The causes of excess lead are studied through field sampling on occupied households, experiments on prototype supply chains, and 3D numerical modeling. The tap water lead concentration of 129 households in the PRH estates was sampled using a specially designed protocol, revealing the highly variable lead concentration variations induced by sources along the supply chain. Lead concentration variation at consumer tap and its relation with various lead sources are studied in a full-scale test rig. A 3D computational fluid dynamics (CFD) model is successfully developed to interpret the time variation of lead concentrations at the consumer tap. Model predictions of the complex variation of dissolved lead are in good agreement with data and confirm lead solder in copper pipe connections as a major cause of the "lead water" episode in Hong Kong. The CFD calculations demonstrate the importance of turbulent diffusion and shear flow dispersion in the modeling of lead; the use of a "plug flow" approximation can result in significant overestimation of lead concentration. The findings provide a basis for lead risk assessment of different water sampling strategies in densely populated high-rise buildings in Megacities.

Remote sensing of coastal algal blooms using unmanned aerial vehicles (UAVs).

The explosive growth of phytoplankton under favorable conditions in subtropical coastal waters can lead to water discolouration and massive fish kills. Traditional water quality monitoring relies on manual field sampling and laboratory analysis of chlorophyll-a (Chl-a) concentration, which is resources intensive and time consuming. The cloudy weather of Hong Kong also precludes using satellite images for algal blooms monitoring. This study for the first time demonstrates the use of an Unmanned Aerial Vehicle (UAVs) to quantitatively map surface water Chl-a distribution in coastal waters from a low altitude. An estimation model for Chl-a concentration from visible images taken by a digital camera on a UAV has been developed and validated against one-year field data. The cost-effective and robust technology is able to map the spatial and temporal variations of Chl-a concentration during an algal bloom. The proposed method offers a useful complement to traditional field monitoring for fisheries management.

Mystery of the high chlorine consumption in disinfecting a chemically enhanced primary saline sewage.

Stonecutters Island Sewage Treatment Works is one of the largest sewage treatment plants in the world and consists mainly of a chemically enhanced primary treatment (CEPT) unit and a disinfection unit. It has long been realized that most of the dosed chlorine (15 mg/L) is lost at the beginning part of the disinfection unit during disinfection of the CEPT effluent. Lab-scale tests were therefore conducted in this study to determine the causes. Because ferric chloride is used in CEPT, ferrous iron in the CEPT effluent (from the reduction of ferric iron) was initially thought to be the main chlorine consumer. However, the chlorine consumption by ferrous iron was found to be 1.2 mg/L at most. Suspended solids and sulfide also did not contribute significantly to the chlorine consumption. Batch tests were therefore conducted to evaluate the effects of mixing condition and chlorine stock solution concentration on the chlorine consumption. Less chlorine was consumed upon increased mixing. Using a high-concentration chlorine stock solution (25000 mg/L) resulted in a 3-times-higher chlorine consumption in the absence of mixing than using a low-concentration chlorine stock solution (2500 mg/L). By correlating the losses of ammonia and total nitrogen with the chlorine consumption, we hypothesized that the use of a high-concentration chlorine stock solution under poor mixing leads to a localized high ratio of chlorine to ammonia, resulting in breakpoint chlorination and an unusually excessive chlorine consumption. A novel apparatus was developed to quantify the nitrogen gas generated during chlorination of a simulated wastewater, and the mass balance of nitrogen-containing species (i.e., ammonia, nitrogen gas, nitrite and nitrate) during the chlorination was inspected. The good fit between the measured chlorine consumption and that back-calculated from nitrogen-containing species verified our hypothesis. Finally, it needs mentioning that the high chlorine consumption and the breakpoint chlorination may occur during chlorine disinfection of any sewage effluents with relatively high ammonia levels; thus it is suggested that either not-too-high concentrations of chlorine stock solutions or sufficient mixing should be applied during disinfection of the sewage effluents.

Tracking major endocrine disruptors in coastal waters using an integrative approach coupling field-based study and hydrodynamic modeling.

Many of the world's large coastal cities discharge partially treated wastewater effluents containing various endocrine disrupting chemicals (EDCs) to coastal environments. Nonylphenols (NP) and bisphenol A (BPA) were found to be the most abundant EDCs in sewage effluents in Hong Kong. The environmental fate and ecological risk of these two EDCs remains largely unknown, particular for coastal systems with complex hydrodynamic flows. Based on a validated three-dimensional (3D) multiple-scale hydrodynamic model, a field-based study was conducted to track the two EDCs from potential sources to the only marine reserve in Hong Kong. The two compounds were detected in all seawater, suspended particle, and sediment samples, with higher aqueous concentrations in wet season than in dry season. High concentrations in sediments suggest sediment is a sink, posing an ecological risk to the benthos. The fate and transport of the two EDCs was predicted using a 3D near-field Lagrangian jet model seamlessly coupled with a 3D shallow water circulation model. The results suggested the NP and BPA in the marine reserve cannot be solely attributed to the nearby submarine sewage outfall, but likely concurrently contributed by other sources. This study calls for more effective measures of reducing the use and release of these EDCs, and research to investigate their impacts on the marine benthos.

Environmental fate and ecological risks of nonylphenols and bisphenol A in the Cape D'Aguilar Marine Reserve, Hong Kong.

Nonylphenols (NPs) and bisphenol A (BPA) are the most common endocrine disruptors detected in the coastal waters of Hong Kong. The Cape D'Aguilar Marine Reserve (CAMR), the only marine reserve in Hong Kong is close to urbanized areas, thus the resident marine organisms are inevitably influenced by partially treated wastewater from adjacent sewage treatment plants (STPs). Elevated levels of NPs and BPA were detected in all seawater, sediment and biota samples collected from the CAMR. Estrogenic activities of seawater from the CAMR, and sludge and sewage from a nearby STP were assessed using yeast estrogen screen assay. We found aromatase, estrogen receptor and vitellogenin genes in the marine medaka fish Oryzias melastigma were significantly up-regulated after exposure to the reserve's seawater. According to a tissue-residue-based probabilistic risk assessment, the marine species living in the CAMR are having 35% and 21% of chance to be at risk due to exposure to NPs and BPA, respectively.

An integrated environmental risk assessment and management framework for enhancing the sustainability of marine protected areas: the Cape d'Aguilar Marine Reserve case study in Hong Kong.

Marine protected areas (MPAs), such as marine parks and reserves, contain natural resources of immense value to the environment and mankind. Since MPAs may be situated in close proximity to urbanized areas and influenced by anthropogenic activities (e.g. continuous discharges of contaminated waters), the marine organisms contained in such waters are probably at risk. This study aimed at developing an integrated environmental risk assessment and management (IERAM) framework for enhancing the sustainability of such MPAs. The IERAM framework integrates conventional environmental risk assessment methods with a multi-layer-DPSIR (Driver-Pressure-State-Impact-Response) conceptual approach, which can simplify the complex issues embraced by environmental management strategies and provide logical and concise management information. The IERAM process can generate a useful database, offer timely update on the status of MPAs, and assist in the prioritization of management options. We use the Cape d'Aguilar Marine Reserve in Hong Kong as an example to illustrate the IERAM framework. A comprehensive set of indicators were selected, aggregated and analyzed using this framework. Effects of management practices and programs were also assessed by comparing the temporal distributions of these indicators over a certain timeframe. Based on the obtained results, we have identified the most significant components for safeguarding the integrity of the marine reserve, and indicated the existing information gaps concerned with the management of the reserve. Apart from assessing the MPA's present condition, a successful implementation of the IERAM framework as evocated here would also facilitate better-informed decision-making and, hence, indirectly enhance the protection and conservation of the MPA's marine biodiversity.

The occurrence and ecological risks of endocrine disrupting chemicals in sewage effluents from three different sewage treatment plants, and in natural seawater from a marine reserve of Hong Kong.

We determined the concentrations of 12 endocrine disrupting chemicals (EDCs) in sewage effluents collected from three different sewage treatment plants (STPs) in Hong Kong, and found 4-nonylphenol (NP) and bisphenol A (BPA) were the most abundant EDCs. Effluent concentrations of NP and BPA were higher in dry season than in wet season, but opposite seasonal changes of NP were observed in receiving waters, probably due to the surface runoff. The two secondary STPs showed higher removal efficiency for these compounds than the preliminary STP, while having higher removal efficiency in wet season. Therefore, it is necessary to upgrade the preliminary STP and improve the EDC removal efficiency in dry season. Seawaters from the Cape D' Aguilar Marine Reserve adjacent to these STPs also exhibited elevated NP levels with a hazard quotient >1. Furthermore, diluted effluents from the STPs elicited significant transcriptional responses of EDC-related genes in the marine medaka fish.

Environmental response to sewage treatment strategies: Hong Kong's experience in long term water quality monitoring.

In many coastal cities around the world, marine outfalls are used for disposal of partially treated wastewater effluent. The combined use of land-based treatment and marine discharge can be a cost-effective and environmentally acceptable sewage strategy. Before 2001, screened sewage was discharged into Victoria Harbour through many small outfalls. After 2001, the Hong Kong Harbour Area Treatment Scheme (HATS) was implemented to improve the water quality in Victoria Harbour and surrounding waters. Stage I of HATS involved the construction of a 24 km long deep tunnel sewerage system to collect sewage from the densely populated urban areas of Hong Kong to a centralized sewage treatment plant at Stonecutters Island. A sewage flow of 1.4 million m3 d(-1) receives Chemically Enhanced Primary Treatment (CEPT) followed by discharge via a 1.2 km long outfall 2 km west of the harbor. The ecosystem recovery in Victoria Harbour and the environmental response to sewage abatement after the implementation of HATS was studied using a 21-year data set from long term monthly water quality monitoring. Overall, the pollution control scheme has achieved the intended objectives. The sewage abatement has resulted in improved water quality in terms of a significant reduction in nutrients and an increase in bottom DO levels. Furthermore, due to the efficient tidal mixing and flushing, the impact of the HATS discharge on water quality in the vicinity of the outfall location is relatively limited. However, Chl a concentrations have not been reduced in Victoria Harbour where algal growth is limited by hydrodynamic mixing and water clarity rather than nutrient concentrations. Phosphorus removal in the summer is suggested to reduce the risk of algal blooms in the more weakly-flushed and stratified southern waters, while nutrient removal is less important in other seasons due to the pronounced role played by hydrodynamic mixing. The need for disinfection of the effluent to reduce bacterial (E. coli) concentrations to acceptable levels is also confirmed and has recently been implemented.

Stable isotopes as a useful tool for revealing the environmental fate and trophic effect of open-sea-cage fish farm wastes on marine benthic organisms with different feeding guilds.

Environmental fate of fish farm wastes (FFW) released from an open-sea-cage farm at Kat O, Hong Kong was examined by measuring carbon and nitrogen stable isotope (SI) ratios in selected benthic organisms collected along a 2000 m transect from the farm. Our results showed that FFW significantly influenced the energy utilization profile of consumers near the fish farm. Although nitrogen enrichment effect on δ15N was anticipated in biota near the farm, the predicted patterns did not consistently occur in all feeding guilds. Two species of suspension-feeders, which relied on naturally δ15N-depleted sources, were δ15N-enriched near the fish farm. In contrast, both species of benthic grazer and deposit-feeder, which relied on naturally δ15N-enriched algal sources, were δ15N-depleted under the influence of FFW. The SI signatures of biota can, therefore, serve as feasible biomarkers for FFW discharges only when the trophic structure of the receiving environment is fully elucidated.

The extended Kalman filter for forecast of algal bloom dynamics.

A deterministic ecosystem model is combined with an extended Kalman filter (EKF) to produce short term forecasts of algal bloom and dissolved oxygen dynamics in a marine fish culture zone (FCZ). The weakly flushed FCZ is modelled as a well-mixed system; the tidal exchange with the outer bay is lumped into a flushing rate that is numerically determined from a three-dimensional hydrodynamic model. The ecosystem model incorporates phytoplankton growth kinetics, nutrient uptake, photosynthetic production, nutrient sources from organic fish farm loads, and nutrient exchange with a sediment bed layer. High frequency field observations of chlorophyll, dissolved oxygen (DO) and hydro-meteorological parameters (sampling interval Deltat=1 day, 2h, 1h, respectively) and bi-weekly nutrient data are assimilated into the model to produce the combined state estimate accounting for the uncertainties. In addition to the water quality state variables, the EKF incorporates dynamic estimation of algal growth rate and settling velocity. The effectiveness of the EKF data assimilation is studied for a wide range of sampling intervals and prediction lead-times. The chlorophyll and dissolved oxygen estimated by the EKF are compared with field data of seven algal bloom events observed at Lamma Island, Hong Kong. The results show that the EKF estimate well captures the nonlinear error evolution in time; the chlorophyll level can be satisfactorily predicted by the filtered model estimate with a mean absolute error of around 1-2 microg/L. Predictions with 1-2 day lead-time are highly correlated with the observations (r=0.7-0.9); the correlation stays at a high level for a lead-time of 3 days (r=0.6-0.7). Estimated algal growth and settling rates are in accord with field observations; the more frequent DO data can compensate for less frequent algal biomass measurements. The present study is the first time the EKF is successfully applied to forecast an entire algal bloom cycle, suggesting the possibility of using EKF for real time forecast of algal bloom dynamics.

Integrated stochastic environmental risk assessment of the harbour area treatment scheme (HATS) in Hong Kong.

Submarine ocean outfalls are commonly used for the disposal of partially treated effluents in coastal cities. Typically, the greatest environmental risk caused by toxic substances occurs in the near field of the outfall discharge. The ecological impact of the effluent varies greatly under different discharge and environmental conditions that are characterized by both regular and stochastic variations. For a comprehensive environmental risk assessment of a coastal discharge, it is necessary to determine both the likelihood and severity of the adverse effects on the biological community. We present the first integrated stochastic (Monte Carlo) environmental risk assessment of a major coastal sewage outfall discharge--the Stonecutters Island outfall of the Harbour Area Treatment Scheme (HATS) in Hong Kong. Unionized ammonia (NH3) is used as the target pollutant. To accurately envisage the ambient concentrations of NH3, a Lagrangian jet model (JETLAG/VISJET) is used to analyze pollutant concentrations in the nearfield of the outfall. The environmental conditions are simulated from 3D hydrodynamic model simulations over a 4 month period for typical wet and dry seasons. Statistical characteristics of the effluent discharge and receiving water temperature are derived from field data. The probability distribution of predicted exposure concentrations (EC) is generated from this integrated simulation. A species sensitivity distribution, which represents a statistical distribution of threshold sublethal effects levels or benchmark concentrations (BC) for various marine organisms is constructed using available chronic toxicity data. The environmental risk of NH3 on the marine community is characterized by computing statistical distributions of Hazard Quotient (HO = EC/BC) using Monte Carlo simulation. It is found that the probability of HO > for HATS Stage 1 (1.6 million m3/day sewage treated with chemically enhanced primary treatment) is around 0.11 for wet season but just about 0.06 for the dry season. The risk increases by about 10% to 0.08-0.13 with additional sewage loads of 0.8 million m3/day at the same level of treatment (HATS Stage 2A). With an upgrade to secondary treatment (HATS Stage 2B), the probability will be reduced to 0.03-0.05. Compared to the use of "worst case" scenarios or point pollution threshold estimates, the present method offers a more holistic ecological assessment, and is much less sensitive to arbitrary choice of model parameters. The present risk assessment approach can be readily extended to the accurate determination of mixing zones based on statistical evaluation of ecological risks.

Antibiotics in the Hong Kong metropolitan area: Ubiquitous distribution and fate in Victoria Harbour.

We investigated the presence and fate of 16 antibiotics belonging to seven groups (beta-lactams, fluoroquinolones, macrolides, sulfonamides, tetracyclines, trimethoprim and amphenicols) in effluents of sewage plants and receiving waters in Hong Kong. Cefalexin, amoxicillin, ofloxacin and erythromycin-H(2)O were ubiquitous in sea water throughout Victoria Harbour, indicating continuous discharge to the environment. This is one of the few studies reporting the frequent occurrence of cefalexin and amoxicillin in sewage effluents and sea water (170-5070 and 64-1670 ng/L in sewage; 6.1-493 and 0.64-76 ng/L in sea water, respectively). Mass flows from seven sewage plants discharged an estimated total of 14.4 kg/day to the Harbour. Typhoon shelters also appeared to play an important role as sources of antibiotics, as evidenced by elevated concentrations within their boundaries. Mass balance estimations suggested significant quantities of antibiotics are discharged to the Harbour without passage through treatment plants.

Seasonal and spatial dynamics of nutrients and phytoplankton biomass in Victoria Harbour and its vicinity before and after sewage abatement.

This study investigated the seasonal and spatial dynamics of nutrients and phytoplankton biomass at 12 stations in Hong Kong (HK) waters during a three year period from 2004 to 2006 after upgraded sewage treatment and compared these results to observations before sewage treatment. Pearl River estuary (PRE) discharge significantly increased NO(3) and SiO(4) concentrations, particularly in western and southern waters when rainfall and river discharge was maximal in summer. Continuous year round discharge of sewage effluent resulted in high NH(4) and PO(4) in Victoria Harbour (VH) and its vicinity. In winter, spring and fall, the water column at all stations was moderately mixed by winds and tidal currents, and phytoplankton biomass was relatively low compared to summer. In summer, the mean surface phytoplankton chl biomass was generally > 9 microL(-1) in most areas as a result of thermohaline stratification, and high nutrients, light, and water temperature. In summer, the potential limiting nutrient is PO(4) in the most productive southern waters and it seldom decreased to limiting levels ( approximately 0.1 microM), suggesting that phytoplankton growth may be only episodically limiting. The mean bottom dissolved oxygen (DO) remained > 3.5 mg L(-1) at most stations, indicating that the eutrophication impact in HK waters was not as severe as expected for such a eutrophic area. After the implementation of chemically enhanced primary sewage treatment in 2001, water quality in VH improved as indicated by a significant decrease in NH(4) and PO(4) and an increase in bottom DO. In contrast, there were an increase in chl a and NO(3), and a significant decrease in bottom DO in southern waters in summer, suggesting that hypoxic events are most likely to occur in this region if phytoplankton biomass and oxygen consumption keep increasing and exceed the buffering capacity of HK waters maintained by monsoon winds, tidal mixing and zooplankton grazing. Therefore, future studies on the long-term changes in nutrient loading from PRE and HK sewage discharge will be crucial for developing future strategies of sewage management in HK waters.