Numerical Forecasting Of Wave In Mirs Bay Of Shenzhen

Mirs Bay of Shenzhen is situated in the north of the South China Sea, the east of the Pearl River estuary, and the west of Dapeng Peninsula. There are ports, swimming resorts, marine aquiculture zones and deep-water channels along the coast in the bay.Marine activities are very busy, therefore accurate forecasting of waves is of great significance for the safety of operation and marine disaster warning.The government of Shenzhen sets up the buoys network for marine environment monitoring in the Shenzhen waters. Based on the monitoring network, this paper tried to set up the wave forecasting modeling in Mirs Bay, which includes: data-driven wave forecasting model,and regional wave forecasting based on atmosphere-ocean coupled model.Firstly, the data-driven model was developed for on-line wave forecasting using the buoy on-line monitoring data, in which nonlinear autoregressive network(NAR) and non-linear autoregressive network with exogenous inputs(NARX) were applied to forecast the real-time wave height at bouys in Mirs Bay. The 3, 6 and 12 h wave heights at three buoy stations were forecasted. The result shows that, the errors of 3, 6h forecasting of wave height at Dameisha and Wankou Station are less than 0.10 m.; the errors of 3, 6h, 12 h forecasting at Dongchong Station are less than 0.10 m, and the correlation coefficients are more than 0.904.Secondly, a ocean-atmosphere coupled model was established for the regional wave forecasting in Mirs bay. This coupled model consists of the smulation waves nearshore(SWAN) model and the weather research and forecasting model(WRF).Several parameters such as the vertical dimensions, top pressure, micropysical process are determined by comparative analysis. The land use type dataset of WRF was replaced by Globcover2009 dataset based on the spatial analysis capacity of GIS, which aims to study the impacts of land use evolution on wind simulation. The coefficient of whitecapping dissipation rate in SWAN model is adjusted. Based on the model parameters calibration, the mean errors of significant wave height at buoy stations are within 0.29 m.The ocean-atmosphere coupled model is validated to hindcast the wind and wave processes of Kalmaegi typhoon in 2014. The simulated error is within 0.32 m.Eventually, the model is used to forecast the 72 h wave characteristics in Mirs Bay based on the forecast product provided by the global forecast system(GFS). The simulated error of forecasting at Wankou Station is 0.23 m. The application shows that the ocean-atmosphere coupled model is applicable to wave forecasting in Mirs Bay.