Fig. 1. The map of Yulin Industrial Park and Mai-Liao separation waterway. (Lines cutting across the waterway and streams are model transects at 0.5 km intervals; numbers are model segment numbers).

Fig. 2. Observed time series of (a) tidal current and tidal height; and (b) salinity and tidal height (positive current is ebb).

Fig. 3. Time-series tidal height at the Mai-Liao gauge station in 1998.

Fig. 4. Historical data of dissolved oxygen and biochemical oxygen demand in: (a) Ho-An Bridge (Shu-Chu-Liao Stream); (b) Wen-Kung Bridge (Hsin-Hu-Wei Stream); and (c) S1, S2, and S3 stations.

Fig. 5. The schematic diagram of interacting water quality state variables.

Fig. 6. Grid pattern, locations and indexing of variables.

Fig. 7. Model calibration results for the 1998 simulation: (a) current; (b) residual current at YLRA station; (c) current; and (d) residual current at YLRC station. (Field measurements were made at 1 and 2.5 m below surface for YLRA and YLRC stations, respectively).

Fig. 8. Model verification results for the 1999 simulation: (a) current; (b) residual current at YLRA station; (c) current; and (d) residual current at YLRC station. (Field measurements were made at 1 and 2.5 m below surface for YLRA and YLRC stations, respectively).

Fig. 9. Model calibration results for the 1998 simulation: (a) salinity at YLRA station; and (b) salinity at YLRC station.

Fig. 10. Model verification results for the 1999 simulation: (a) salinity at YLRA station; and (b) salinity at YLRC station.

Fig. 11. The result of model sensitivity run for the 1998 simulation: (a) salinity at YLRA station; and (b) salinity at YLRC station.

Fig. 12. The result of model sensitivity run for the 1999 simulation: (a) salinity at YLRA station; and (b) salinity at YLRC station.

Fig. 13. Water quality model calibration results at the S3 station in 1998: (a) dissolved oxygen; (b) carbonaceous biochemical oxygen demand; (c) ammonium nitrogen; (d) nitrite-nitrate nitrogen; and (e) total phosphorus.

Fig. 14. Water quality model calibration results at the S2 station in 1998: (a) dissolved oxygen; (b) carbonaceous biochemical oxygen demand; (c) ammonium nitrogen; (d) nitrite-nitrate nitrogen; and (e) total phosphorus.

Fig. 15. Water quality model verification results at the S2 station in 1999: (a) dissolved oxygen; (b) carbonaceous biochemical oxygen demand; (c) ammonium nitrogen; (d) nitrite-nitrate nitrogen; and (e) total phosphorus.

Fig. 16. The model domain with both the Yulin and Hsin-Hsin Industrial Parks completed. (Lines cutting across the waterways are model transects at 0.5 km intervals; number are model segment numbers).

Fig. 17. Boundary conditions for flooding simulations: (a) Water surface elevation at downstream boundaries (points A and B of Fig. 16); and (b) freshwater discharges at upstream boundaries.

Fig. 18. Model predictions of 200 m separation waterway (a) water surface elevation; and (b) current velocity at the top layer. (Segments 10, 14, and 17 represent upriver, middle-river, and downriver of the Mai-Liao separation waterway, respectively).

Fig. 19. Model results as function of the width of separation waterway: (a) maximum water surface elevation; and (b) maximum velocity.

Fig. 20. Model simulation results of 200m separation waterway: (a) dissolved oxygen; and (b) CBOD. (Segments 10, 14, and 17 represent upriver, middle-river, and downriver of the Mai-Liao separation waterway, respectively).

Fig. 21. Model predicted water quality conditions as functions of waterway width: (a) dissolved oxygen; and (b) CBOD.

Table 1. Comparison of harmonic analyses with various numbers of constituents

Table 2. The amplitudes and phases of the primary nine tidal constituents

Table 3. Coefficients and constants used in the model calibration

Table 4. The water quality conditions at upstream and downstream boundaries for model application runs