BASELINE STUDY OF WATER QUALITY IN SINGAPORE COASTAL WATERS

Hin-Fatt Cheong¹, N.J.Shankar², Chor-Eong Ong³ and M. K. Huda⁴

^1,2Professors, Department of Civil Engineering

³Managing Director and

⁴Research Engineer, Applied Research Corporation

National University of Singapore, Singapore 117566

Abstract: Field measurements of the bulk descriptors of water quality such as BOD, DO, TN and FC in the Johor Straits were monitored and found to show little variation with depth. Nutrient levels at certain tidal inlet areas with treated sewage effluent indicate the likelihood of algal bloom. This paper also describes the application of a depth-averaged tidal circulation model coupled with the depth-averaged water quality model for the baseline calibration of bulk parameters for water quality in Singapore coastal waters.

1    INTRODUCTION

Singapore is a small island nation where a thriving port, an international airport in the east, shipbuilding industries in the west and northeast, naval bases in the east and west and recreational beaches in the south east compete for the use of the limited stretches of coastline.  Treated sewage effluents in the Seletar (near WQ1), Punggol (near WQ2) and Serangoon (near WQ4) areas (Fig. 1), together with the rivers and mangroves on the Malaysian side are sources of nutrients. High levels of nutrients stimulate rapid growth of algae, known as blooms. There are plans to develop a deep tunnel sewerage system wherein the entire sewage production in Singapore is treated and released at two offshore outfalls. There are also plans for further reclamation, including the amalgamation of offshore islands into larger ones. Hence, there is the need for baseline studies of the water quality in the surround coastal waters of Singapore in order to assess the impact of all these future developments on the ecology of the coastal waters. This paper describes the baseline studies that have been undertaken over the past several years in anticipation of major reclamation projects in the north east Tekong area (Fig.1). The depth-integrated tidal circulation model described by Cheong et al (1992) is used to derive the tidal circulation and the water quality model described herein is used for calibration and validation over different field measurement periods covering spring and neap conditions in November 1999.

2    WATER QUALITY MODEL DESCRIPTION

The ambient factors that influence the water quality variables are advection by current streams and diffusion by ambient turbulence. The governing depth averaged advective-diffusive equation is as follows:

                      (1)

where,

=concentration of N water quality variables, =concentration of effluent source with N water quality variables, Q_e=effluent discharge, = chemical reaction description of N water quality variables, h=water depth, u,v=depth-averaged velocity components in the lateral x and y directions, E_x, E_y=turbulent eddy coefficient in x and y directions, Dx, Dy=grid sizes along x and y directions. In total, five water quality variables have been taken into account in the model, through the kinetic models (Gin and Tkalich 1998).

      [For Ammoniacal Nitrogen (NH₃-N) ]

   [For Nitrate Nitrogen (NO₃-N) ]

 [For Carbonaceous BOD (CBOD)]

For Dissolved Oxygen (DO)

 14.6244 – 0.367134 T + 0.0044972 T² – 0.0966 S + 0.00205 S T + 0.0002739 S²

     [For Fecal Coliform (FC)]

where c₁, c₂, c₃, c_4, c₅=concentration of NH₃-N, NO₃-N, CBOD, DO and fecal coliform (FC) respectively, c_sat=saturation concentration of DO; S=salinity, k₁₂=nitrification rate; k_nit=oxygen limitation of nitrification, g(O₂)/m³; k_2D=denitrification rate; k_NO3=oxygen limitation of denitrification, g(O₂)/m³; k_D=deoxygenation rate, 1/s; k_BOD=oxygen limitation, g(O₂)/m³; k_a= reaeration rate, 1/s; P_syn =algal photosynthesis, g (O₂)/m³; R_sp=algal respiration, g (O₂)/m³;  k_b= bacterial die-off rate, 1/s; T=temperature (⁰C); q₁₂, q_D, q_2D, q_a =temperature coefficients.

These equations are solved using an implicit finite difference scheme with an upwind finite difference formulation and a multi-stage implicit scheme (Douglas and Rachford, 1956). Most of the kinetic coefficients are taken from the Potomac Estuary study by Thomann and Fitzpatrick (1982). The background concentrations are mainly obtained from field measurements conducted in the coastal waters of Singapore in the period from 1996 to 1998 (Gu, 1998; Lim, 1997). Table1 and Table2 show kinetic coefficients and averaged background concentrations of water quality variables.

3    FIELD MEASUREMENTS

The locations of the tide gauges, current meters and water sampling points and the hydrographic survey boundaries are shown in Fig. 1. The  recording of the tide is carried out continuously over a period of at least 32 days and cover a spring-neap tidal cycle completely. The tide recording is carried out concurrently during the period when the field sampling and water current measurements are carried out. The samples collected are immediately sealed, stored in a cold box  and delivered to an approved laboratory for analysis.

4    RESULTS, DISCUSSION & CONCLUSIONS

The field results for the spring and neap tidal conditions are shown in Table 3. Since the presence of NH₃-N is indicative of possibly incomplete nitrification, locations of comparatively high values of NH₃-N may be expected to be associated with lower values of NO₃-N. Nitrogen is often the limiting component in coastal waters in terms of eutrophication, certain areas of the coastal waters near the sewage outfalls at Seletar (WQ1) and Serangoon (WQ4) show potential for eutrophication with the nitrogen inputs occurring there. While the PO₄-P levels in these areas are already at levels which can support algal growth, the data would suggest that should nitrogen inputs into these areas increase or accumulate, then that area is likely to see a deterioration of water quality due to such plant growth. When algae decays, an oxygen demand is exerted and this may result in low DO or even anaerobic conditions (odour) in shallow water conditions. WQ1, near the sewage outfall, showed the highest NO₃-N suggesting the presence of aerobically treated sewage. DO in the tidal inlets with treated sewage outfalls could drop to levels which cause distress to fish life. Away from these inlets towards the main straits, conditions improve considerably.

Figs. 2, 3 and 4 show the distribution patterns of DO, TN and FC during spring conditions. The numerical monitoring locations gave average values which are in reasonable agreement with the measured. The spring period validation can be judged to be reasonably successful whilst the neap period provides the calibration. The numerical model therefore is a useful EIA planning tool for future developments in the area.

References

[1] Cheong, H F, N J Shankar and C T Chan, Numerical modelling of tidal motion in the Southern islands of Singapore. In Computer Modelling of Seas and Coastal Regions, edited by P W Partridge, pp. 175-196. Southampton: Computational Mechanics Publications, Southampton and Elsevier Applied Science, London, April 1992.

[2] Churchill MA, Elmore HL and Buckingham RA. Prediction of stream reaeration rates. J. of Sanitary Eng. Div., ASCE SA4il, Proceeding Paper 3199, 1962.

[3] Douglas, J., Jr. and Rachford, H. H., Jr. (1956). On the Numerical Solution of Heat Conduction Problems in Two and Three Space Variables. Trans. Amer. Math. Soc., 82, pp. 421-439.

[4] Gin KYH and P Tkalich. A three dimensional eutrophication model for Singapore coastal waters. Proceedings of the 11^th Congress of the Asia-Pacific Division of the IAHR, Indonesia 1998.

[5] Gu, G. (1998). Phytoplankton Dynamics in Singapore’s Coastal Waters. M.Eng. Thesis, National University of Singapore, Singapore.

[6] Lim, H. L. (1997). Baseline Water Quality Monitoring in the Singapore Straits. B.Eng. Thesis, National University of Singapore, Singapore.

[7] Thomann, R.V. and Fitzpatrick, J.J. (1982). Calibration and verification of a model of the Potomac Estuary. Hydroqual, Inc., Final Report to D.C. Dept. of Environmental Services, Washington, D.C., pp. 500.

Table1    Description and typical values of kinetic coefficients used in the model

Table2    Average background concentrations of water quality variables used in the model

 Table 3    Measured field results on water quality parameters in the Tekong area

Fig.1    Computational Domain of Local Water Quality Model ( Baseline Year 2000 : 481×731×75 m) with 75 m×75 m grid showing outfall locations of sewage treatment plants, other loading locations and outfall locations of numerical monitorina stations.

Fig.2    Concentration of Dissolved Oxygen (DO) after 465 hrs simulation period, LEst 75m / WQ 01 / Baseline / Tekong Study / Spring Tide Condition/.

Fig.3    Concentration of Total Nitrogen (TN) after 465 hrs simulation period. LEst 75m / WQ 01 / Baseline / Tekong Study / Spring Tide Condition/.

Fig.4    Concentration of Fecal Coliform (FC) after 459 hrs simulation period. LEst 75m / WQ 01 / Baseline / Tekong Study / Spring Tide Condition/.