3.2. Model calibration

3. Results and discussion

76 Applications in Water Systems Management and Modeling

3.1. Hydrological and meteorological characteristics

Figure 3. Rainfall amount and wind speed in three lakes in 2016.

Wind speed and rainfall pattern vary between the three lakes (Figure 3). Rainfall amount and wind speed in 2016 was highest in Durian Tunggal Lake compared to Bera and Bukit Merah lakes. Highest daily rainfall amount recorded was 151, 50, and 33 mm in Durian Tunggal, Bukit Merah, and Bera lakes, respectively. Wind speed was also highest in Durian Tunggal Lake followed by Bukit Merah and Bera lakes. Mean daily wind speeds were relatively higher

> In this chapter, evaluation of the model performance was based on the visual inspection of observed and simulated trends and correlation coefficients of determination (r 2 ). Flow field of east, north, and vertical directions was measured at one location in Bukit Merah using ADCP and used to calibrate the hydrodynamic model. The results show good agreement between measured and simulated velocities at 1-m depth (Figure 4a). However, proper agreement of velocities at different depths and water level was difficult to achieve due to many uncertainties connected with hydrology and water demand management. Additionally, temperature sensor was also deployed to calibrate the hydrodynamic model. Simulated temperature yields very good agreement with measured temperature (Figure 4b). Based on the coefficients of determination (r 2 ) between model output and observed data, the simulated current has higher correlation with the observed values at 1 m level. The coefficient of determination for temperature was 0.742. In Durian Tunggal and Bera Lake, temperature-depth profiling was carried out at few sites for calibration purposes. The coefficient of determination for temperature was 0.762 and 0.6 for Durian Tunggal and Bera Lake, respectively.
