Junk Bay Research Articles

Remotely sensed estimates of long-term biochemical oxygen demand over Hong Kong marine waters using machine learning enhanced by imbalanced label optimisation

In many coastal cities around the world, continuing water degradation threatens the living environment of humans and aquatic organisms. To assess and control the water pollution situation, this study estimated the Biochemical Oxygen Demand (BOD) concentration of Hong Kong's marine waters using remote sensing and an improved machine learning (ML) method. The scheme was derived from four ML algorithms (RBF, SVR, RF, XGB) and calibrated using a large amount (N > 1000) of in-situ BOD5 data. Based on labeled datasets with different preprocessing, i.e., the original BOD5, the log10(BOD5), and label distribution smoothing (LDS), three types of models were trained and evaluated. The results highlight the superior potential of the LDS-based model to improve BOD5 estimate by dealing with imbalanced training dataset. Additionally, XGB and RF outperformed RBF and SVR when the model was developed using log10(BOD5) or LDS(BOD5). Over two decades, the BOD5 concentration of Hong Kong marine waters in the autumn (Sep. to Nov.) shows a downward trend, with significant decreases in Deep Bay, Western Buffer, Victoria Harbour, Eastern Buffer, Junk Bay, Port Shelter, and the Tolo Harbour and Channel. Principal component analysis revealed that nutrient levels emerged as the predominant factor in Victoria Harbour and the interior of Deep Bay, while chlorophyll-related and physical parameters were dominant in Southern, Mirs Bay, Northwestern, and the outlet of Deep Bay. LDS provides a new perspective to improve ML-based water quality estimation by alleviating the imbalance in the labeled dataset. Overall, the remotely sensed BOD5 can offer insight into the spatial-temporal distribution of organic matter in Hong Kong coastal waters and valuable guidance for the pollution control.

Adaptive-weight water quality assessment and human health risk analysis for river water in Hong Kong.

The water quality of Hong Kong's four water control zones (Tolo Harbour and Channel, Port Shelter, Victoria Harbour, and Junk Bay) is of vital importance and has attracted much attention. This study aims to more objectively and comprehensively assess the water quality and its health impact based on the four-year monitoring data of 21 parameters collected from four zones. First, physicochemical characteristics of the water system were investigated based on multivariate statistical approaches, including Kruskal-Wallis test, hierarchical cluster analysis, and Mann-Kendall test. Then, water quality levels over space and time and the element sources were analyzed using adaptive-weight water quality index (AWQI) method, and factor analysis, respectively. Finally, the potential harm of trace elements for humankind was identified based on the health risk assessment model. The results revealed that (1) the values of more than half of the water quality parameters exhibited significant interannual changes, and the values of all parameters distinctly varied over space; (2) The water quality status in four water control zones showed a steady and long-term improvement trend from 2016 to 2019; (3) The sources of pollution elements impacting water quality status were related to the comprehensive influence of human activities and natural processes; (4) The carcinogenic risks of all trace elements were negligible or acceptable, while Mn and As may cause noncarcinogenic harm to humankind.

Cross Bay Link - Public Engagement in Planning Urban Infrastructure

The Cross Bay Link (CBL) will be a bridge across Junk Bay in Tseung Kwan O (TKO) New Town, Hong Kong. It will connect the south-eastern part of TKO with the territorial-wide strategic road network of Hong Kong in order to cater for the projected traffic flow generated from the future developments in this growing new town. In addition to a dual two-lane carriageway, CBL will also carry a footwaay and cycle path, which is the first of its kind for a marine viaduct in Hong Kong.The road link will be approximately 1.8km long and the sspan of the main bridge is approximately 2000m. Because of the prominent location, there is a strong public aspiration to build a landmark structure. The Civil EEngineering and Development Department (CEDD) decided to conduct a series of public engagement activities to soliicit the public's support on the preferred design option of the bridge.Two stages of public engagement works for the featurre bridge design were completed from July 2009 to June 2010. In the first stage, a design ideas invitation evennt was organized to invite professionals and the public to contribute design ideas and concepts on the form, appearance and aesthetics of the CBL. Thhe results were incorporated into six design options prepared by the project engineering consultants for the second stage public engagement. Apart from the technical evaluation, the seelection of the preferred option included a public survey to gauge their preference on the six design options.The paper wiill describe the public engagement process, the development of the design options, as well as the further evaluation, which was carried out to determine the preferred option for preliminary design development.

The effects of nutrients and their ratios on phytoplankton abundance in Junk Bay, Hong Kong

The effects of nutrients and their ratios on phytoplankton abundance in Junk Bay, Hong Kong

Barnacles as biomonitors of trace metal availabilities in Hong Kong coastal waters: changes in space and time

The use of selected organisms as biomonitors of trace metal bioavailabilities allows comparisons to be made over space and time. The concentrations of 11 trace metals (arsenic, cadmium, chromium, cobalt, copper, iron, lead, manganese, nickel, silver, zinc) were measured in the bodies of two barnacle species, Balanus amphitrite and Tetraclita squamosa, from up to 18 littoral sites from Hong Kong coastal waters in April 1998. These data provide evidence on the geographical variation in metal bioavailabilities at this time, and are compared selectively against historical data sets for 1986 and 1989. Geographical variation in bioavailabilities is clear for several metals, with hotspots for arsenic, copper, nickel and silver at Chai Wan Kok, and for lead in Junk Bay. Victoria Harbour sites head the rankings for silver and arsenic, and Tolo Harbour sites exhibit relatively elevated cobalt, manganese and zinc. Many bioavailabilities of trace metals to barnacles are lower in Hong Kong coastal waters in 1998 than in 1986. The two barnacle species are widespread and the extensive data set presented is a benchmark which can be compared to the results of similar biomonitoring programmes elsewhere in the Indo-Pacific and beyond.

Some Observations on the Changes of Physico-Chemical and Biological Factors in Victoria Harbour and Vicinity, Hong Kong, 1988–1996

This study analysed monthly physico-chemical and biological data collected from 18 marine monitoring stations in Victoria Harbour and its vicinity in Hong Kong, from 1988 to 1996. Cluster analysis based on all water quality parameters measured shows that the 18 monitoring stations can be grouped into four clusters: Cluster I consists of stations located in the Harbour proper; Cluster II consists of stations located west of the Harbour and along the Rambler Channel; Cluster III consists of stations located east of the Harbour near Junk Bay and Cluster IV consists of stations located west of the Harbour and near the Ma Wan, Kap Shui Mun and Western Fairways. Factor analysis shows high positive loadings for nutrients in the first two factors of the four clusters. This suggests that effluents from the 11 outfalls of sewage screening plants influence the water quality of Victoria Harbour and its vicinity. Other factors such as storm water runoff, marine traffic, construction and industrial activities and the Pearl River discharges also appear to play an important role in determining local water quality. Five stations located along an east–west transect across the Harbour were selected for trend analysis. The three stations located in the Harbour exhibit an increasing trend for temperature and levels of total phosphorus (TP), ortho-phosphate phosphorous (PO 4-P) and faecal bacteria and a decreasing trend for pH and levels of total nitrogen (TN), total kjeldahl-nitrogen (TKN), 5-day biochemical oxygen demand (BOD 5) and chlorophyll a. For the station located east of the Harbour, an increasing trend is observed for levels of TP, PO 4-P, but no decreasing trend in TN and TKN is detected. For the station located west of the Harbour, no decreasing trend in TN, TKN and chlorophyll a is observed. Changes in levels of phosphorus and nitrogen in Victoria Harbour and the immediate vicinity have led to significant increases in the ratios of Total Silica (TSi) to TN, as well as a decrease in TN to TP and TSi to TP in most stations. Results of the present study show that Victoria Harbour and its immediate vicinity remain polluted.

Intervention analysis on the effect of two sewage screening plants on the BOD5 levels in the Hong Kong Victoria Harbour and vicinity

In this paper, we applied the technique of intervention analysis to assess the changes in BOD5 levels in marine water of Victoria Harbour and its vicinity in Hong Kong, after the commissioning of two sewage screening plants located in the central and western part of the Harbour. Data for this analysis were collected from 10 stations along an east-west transect from Junk Bay to Tsing Yi Island during the period from January 1988 to December 1995. Our analysis shows that the BOD5 levels at four of the five stations near the To Kwa Wan Sewage Screening Plant decreased immediately after the commissioning of the Plant in July 1990. The reduction ranged from 6.76% in Station VM5 to 15.44% in Station VM1. For the five stations located near the Stonecutters Island Sewage Screening Plant, their respective BOD5 levels also decreased by amounts ranging from 6.5% in Station WM2 to 21.54% in Station VM8, immediately after the commissioning of the Plant in August 1992. The stations closest to the outfalls of the two respective Plants showed a greater reduction in BOD5 level compared with the stations that are further away from the outfalls. Moreover, the stations near the larger Stonecutters Island Plant outfall also showed a greater decrease in BOD5 level compared with the stations near the smaller To Kwa Wan Plant outfall. These results are consistent with those obtained by a simple descriptive analysis.

EFFECTS OF LANDFILL GAS ON GROWTH AND NITROGEN FIXATION OF TWO LEGUMINOUS TREES (ACACIA CONFUSA, LEUCAENA LEUCOCEPHALA)

A study was made on the effects of landfill gas on ARA (acetylene reducing activity) of nodules of two woody legumes (Acacia confusa and Leucaena leucocephala) widespread on landfill sites in Hong Kong. The effects of the three main components of landfill gas, O2, CO2 and CH4, were first measured separately over a 1-hr period. Maximum ARA was found at 20% O2 (close to atmospheric partial pressure) and ARA decreased as the O2 decreased in the range of 16–1%. Acacia confusa nodular ARA was significantly inhibited at 30–50% CO2, but not Leucaena leucocephala nodular ARA. CH4 had no significant effect on ARA of either species. As the landfill gas concentrations in the landfill topsoil were mostly > 10% O2 and < 10% CO2, root nodules should fix N2 effectively over these ranges of gases. A four-week test was conducted to assess the long-term influence of landfill gas on seedlings of the two legumes. Landfill gas and elevated CO2 both suppressed their growth and their nodular ARA. Even under the influence of the gases, however, seedlings with nodules formed a higher biomass than seedlings lacking nodules. The growth of the two legumes under actual landfill conditions was investigated by transplanting non-inoculated and pre-inoculated seedlings to two landfill sites in Hong Kong: Junk Bay and Shuen Wan Landfill. After six months, most of the non-inoculated seedlings became infected: Acacia confusa 63 and 70%, Leucaena leucocephala 17 and 89%, respectively, at the test sites. The results indicate that there were free rhizobia at these landfill sites to infect the legumes and they had formed effective nodules to fix N2 under landfill conditions.

An Analysis of the Representative Composition of Rainwater at Six Locations in Hong Kong

A 3-year weekly data set on the ionic composition of rainwater from 6 sampling sites in Hong Kong representing the 1990-92 period was resolved employing factor analysis. Two-to-three factors explaining cumulative variances of 78-93% of the total in the 9 ionic parameters studied are utilised in the interpretation. The procedure is repeated on the screened data sets obtained on the basis, among other constraints, of a cut-off ion balance of 15%. The results indicate that: (i) the 'mixed' (residential-industrial) area, Central-Western has a 7- membered 'degenerate' first factor which fails to resolve into smaller groups; (ii) the quasi-rural site, Junk Bay has a partially neutralised acid (PNA) and a pure sea salt (SS) component; (iii) two other mainly residential areas, Hong Kong South - whose factors resemble those for industrial Kwun Tong - and Tai Po have a third factor containing potassium, and additional chloride, respectively. The latter factor, V/W, for Tai Po believed to be due to burning vegetation/waste incineration is similar only to that for the industrial Kwai Chung. On the other hand, the process of data screening has no effect on the factorial sequence or make-up, except in the cases of the two industrial locations, where the order of the resolved factors in two of their sets are swapped among two different factors; and, in the case of one of them, a variable (ion) shifts its loading to a different factor upon screening. Finally, stepwise multiple regression corroborates our earlier provisional conclusion that the free acidity of rainwater in Hong Kong is also contributed by acids other than H2SO4 and HNO3, possibly by oxalic, formic and acetic acids, among some others.

Ammonia stripping as a pretreatment for landfill leachate

The effectiveness of ammonia stripping at different air flow rates (0, 1 and 5 L min−1) and lime dosages (0 and 10 000 mg L−1 calcium hydroxide) was investigated in aeration tanks in a laboratory as a pretreatment to remove ammoniacal-nitrogen and organic load (COD) in landfill leachate. Ammoniacal-nitrogen removal at 20°C after one day was 70% for 0 L min−1, 81% for 1 L min−1 and 90% for 5 L min−1 regardless of the origin of leachate. Ammonia loss was mainly due to desorption through water surface. The levels of phosphorus and COD were only reduced by lime precipitation, with 85% and 93% phosphorus removal and 24% and 47% COD removed for leachate from the Junk Bay Landfill (JB) and Gin Drinkers' Bay Landfill (GDB) respectively. The highly significant difference (P<0.05) of COD removal between JB and GDB might be due to the different age of the two landfills studied. Leachate quality and configuration of the treatment reactor were important factors affecting the efficiency of ammonia removal by stripping processes.

Algal purification of pretreated landfill Leachate

Abstract The efficiency of microalgae in removing various pollutants in landfill leachate after pretreatment by free stripping or air stripping was tested by a laboratory batch system. The results showed that Chlorella pyrenoidosa and Scenedesmus sp. had similar removal efficiencies with regard to ammoniacal‐nitrogen, oxidized‐nitrogen, orthophosphate and chemical oxygen demand (COD). The COD removal by algal treatment was better in Junk Bay (JB) leachate than in Gin Drinkers’ Bay (GDB) leachate (14–21% and 0.4–7% respectively). No significant difference (P > 0.05) was found in removing other nutrients including ammoniacal‐nitrogen, oxidized‐nitrogen and orthophosphate between leachate from the two landfills, regardless of the pretreatment used. The removal efficiencies of ammoniacal‐nitrogen and phosphorus were found to be higher in air‐stripped leachate than in free‐stripped one. Removal of ammoniacal‐nitrogen and phosphorus in air‐stripped leachate was 30% and 87% respectively. Poor removal of ammoniac...

Variations in the chemical properties of landfill leachate

Landfill leachates were collected and their chemical properties analyzed once every two months over a ten-month period from the Gin Drinkers' Bay (GDB) and Junk Bay (JB) landfills. The contents of solids, and inorganic and organic components fluctuated considerably with time. In general, the chemical properties of the two leachates correlated negatively (P<0.05) with the amounts of rainfall prior to the sampling periods. However, magnesium and pH of the leachates remained relatively constant with respect to sampling time. The JB leachate contained higher average contents of solids and inorganic and organic matter than those of GDB with the exception of trace metals. Trace metals were present in the two leachates in trace quantities (<1.0 mg/liter). The concentrations of average ammoniacal nitrogen were 1040 and 549 mg/liter, while chemical oxygen demand (COD) values were 767 and 695 mg/liter for JB and GDB leachates, respectively. These results suggest that the leachates need further treatment before they can be discharged to the coastal waters.

Toxic effect of landfill leachate on microalgae

Leachates from two landfills, namely Junk Bay (JB) and Gin Drinkers' Bay (GDB), were assessed for their acute toxicity using four green algal species, Chlorella pyrenoidosa, C. vulgaris, Scenedesmus sp. and Dunaliella tertiolecta. JB leachate was more toxic to the four algal species tested than leachate from GDB landfill. The growth rates of all four species in 50% JB leachate, and two species (C. vulgaris and D. tertiolecta) in 50% GDB leachate were significantly lower (P<0.05) than those in the control (Bristol Medium only). Values of 96h-EC50 of JB leachate for all four species were lower than those of GDB. The high contents of ammoniacal-nitrogen and organic compounds (such as volatile fatty acids) seemed to be the factors governing the toxicity of leachate on algae. There were differential sensitivities to leachate exhibited by the tested algal species. Susceptibility to leachates in terms of cell number were in the ascending order of C. pyrenoidosa, Scenedesmus sp., C. vulgaris and D. tertiolecta.

Polychlorinated biphenyls (PCBs) in sediments in Hong Kong: A congener-specific approach to the study of coplanar PCBs in aquatic ecosystems

Patterns of contamination by polychlorinated biphenyls (PCBs) were investigated in fourteen samples of coastal sediments from Hong Kong. Congener-specific analyses revealed nine sediment samples from Junk Bay to contain PCBs at concentrations ranging from 31 to 2200 ng g −1 dry weight, concentrations generally increasing with distance north in the Bay. By contrast, five sediments from the Tolo area to the north-east of Hong Kong exhibited total PCB levels of only 6·6 to 45 ng g − dry weigth. The patterns of relative abundance of PCB congeners in the northern Junk Bay sediments suggested the existence of ongoing source(s) of PCBs in this area; biphenyls of lower chlorination were present at high concentration in these samples. Three coplanar PCBs (3′, 4, 4′-tetrachlorobiphenyl; 3,3′,4,4′,5-pentachlorobiphenyl; and 3,3′,4,4′,5,5′-hexachlorobiphenyl) were found to be present in Junk Bay sediments; these are highly toxic and are cause for concern in aquatic environments. The abundance of the three coplanar PCBs in the sediments studied was similar to that in commercial PCB mixtures, suggesting that these congeners are not enriched over other PCBs by the sediments of coastal ecosystems. It is concluded that the preferential enrichment of coplanar PCBs occurs in the biosphere, rather than in sediments.

Measurements of atmospheric sulfur dioxide concentrations at three sites in junk bay

Abstract As part of an environmental impact assessment for building a new town in Junk Bay, sulfur dioxide (SO2) concentrations were measured from 18 May to 9 June 1981 at three locations in Junk Bay. Hourly averaged samples were collected with an automatic sequential air sampler over the hours from 1100 to 2300 and analyzed with flame photometric detection technique for sulfur dioxide. It is found that mean SO2 concentrations at the three sites range from 18.6 μg.m−3 (7 ppb) to 45.0 μg.m−3 (17 ppb). Maximum hourly averaged value is 126 μg.m−3 (48 ppb). The SO2 level is approximately inverse proportional to the square root of wind speed. Analysis of the polar isopleth diagram suggests that the recorded high SO2 concentrations were likely due to emissions from a nearby ship‐breaking factory equipped with iron reheating furnaces. Concentrations of SO2 usually increased during the day and were highest in the evening.

Continuous measurement of particulate and sulfur dioxide concentrations in Junk Bay, Hong Kong

As part of an environmental impact assessment for building a new town in Junk Bay, continuous measurement of SO 2 and particulate concentrations was carried out from October 1981 to June 1982 at three sites in Junk Bay to study the air quality in the area. Flame photometric SO 2 analyzers were used to measure ambient SO 2 level, whereas tape monitors were used to measure the ambient suspended particulate level (in terms of soiling index, or coefficient of haze per 1000 feet). It was found that the mean SO 2 concentrations at the three monitoring sites ranged from 5μg m −3 to 35μg m −3. Maximum daily values up to about 250 μg m −3 and hourly values up to 800 μg m −3 had been recorded on occasion. Comparison of the hourly meteorological data and the hourly SO 2 concentrations in four high-SO 2-level days suggested that the ‘sulfur dioxide episodes’ were all associated with very light wind speeds and local sources. The mean coefficient of haze level in Junk Bay was less than one, which corresponded to very slight particulate pollution.