Pearl River Estuary Design Current Speed Derivation Using Numerical Model Simulation

Seabed erosion by strong bottom currents can cause severe damage to the oil or gas pipelines laid on the seabed. Extreme current speeds of various return periods are therefore the most important oceanographic parameters that are required for engineering of the pipelines. It is often unpractical to derive the extreme current speeds using measured data as current measurements are costly and take longer time than allowed by the pipeline project's schedule. Extreme current speeds occur during typhoons, thus, it can be assumed that the yearly extreme current speeds are the results of interaction between the storm surges and the astronomical tides. Coupled storm surge and astronomical tide numerical models would be a quick and economic method to estimate return period extreme current speed values.To accurately simulate the characteristics of the currents in the Pearl River Estuary, this experiment uses the ECOM (Estuarine Coastal Ocean Model) current model. Firstly, a astronomical tide and storm surge interaction model with coarse grid is built covering the northern South China Sea. Then, using the calculation results of the coarse grid model, a fine grid model is built with horizontal grid intervals of 1' x 1', covering the estuary. Comparison between the model results and the observations indicates that the model can accurately simulate the characteristics of currents and water levels in the area. Accuracy of the simulation results is sufficient for offshore engineering.Hindcast simulations are performed for each typical typhoon per year that hit the project area. Yearly extreme current speed values for the grid points are obtained from the simulations, which are then used to calculate return period extreme current speeds based on Gumbel distribution function. For each calculation point, Kolomogrov assumption test is carried out for the distribution function fittings and the standard deviations and the maximum errors are estimated. The maximum fitting errors for the extreme surface and bottom current speeds are less than 0.15 and the standard deviations less than 0.08. The Kolomogrov statistics is less than 0.8. For a sample of 30 year time series, all calculation points pass the Kolomogrov assumption test with 0.05 confidence level. Based on the numerical simulation, this paper gives distribution of extreme surface and bottom current speeds for the return periods of 2, 5,10,25, 50, and 100 years for the area.