QPE Pilot Study Ocean Physics Forecast

This web page presents the output from two 6 day free surface simulations forced with COAMPS (wind stress) and NOGAPS (heat-flux, E-P) atmospheric forcing, initialized with a combination of the Taiwanese data (see http://mseas.mit.edu/Sea_exercises/QPE/Pilot2008/index_init_data.html) and of a summer climatology created using June-August profiles and the HydroBase2 software. The bathymetry used is the NCOR bathymetry.

In our simulations so far, the biggest uncertainty seems again to be the transport though the Taiwan Strait between Taiwan and mainland China. To illustrate this point, we decided to issue 2 forecasts, one with a northward transport and one with a southward transport through the Strait. We believe that some sampling just north of Taiwan would substantial reduce this uncertainty (e.g. some drifter release and/or CTD data). If we let our initialization algorithm compute this initial transport (i.e. free transport initialization constrained by geostrophy in the interior and a first-guess level of no motion), our initialization algorithm leads to a southward 2.5 Sv transport between China & Taiwan. Of course, this boundary transport changes with time, based on the interior flow and open boundary conditions. The other set of simulations shown corresponds to a 1Sv Northward barotropic transport imposed initially between China & Taiwan. Even though these differences are only set in the initial conditions, they lead to an opposite flow north of Taiwan which remains after 7 days of simulation.

The run with a southward 2.5Sv transport has a westward flow along the northern Taiwanese coast, with a tongue of deep Kuroshio water that has upwelled and extends north-northeast, advected by the Kuroshio which has intruded a bit onto the shelf. On the other hand, the run with a northward 1Sv transport has an eastward flow along the northern Taiwanese coast, with a weaker cold Dome and the creation of a weak cold cyclonic eddy just north-northeast of the eastward coastal flow along Taiwan. Based on the Japanese SST and Taiwanese data, and different water masses present (T/S properties), it appears that the run with a westward coastal flow north of Taiwan (run with southward flow through the Taiwan Strait) is closer to the truth, but it is not yet certain.

Of course, we are still tuning a large set of numerical and physical model parameters. The two runs above are only illustrative of one of the main uncertainty issues we are dealing with so far. To see a series of images for each case, click on the link for the "Complete set of Plots" below. Note that more than 50 simulations were carried out for sensitivity studies on the numerical and physical parameters. Here, we only provide the results of the likely better simulation we have achieved so far.

Acoustic simulations from each of these two dynamical simulations, with vertical sections of transmission loss (TL) and sound speed, can be found here.

Nowcasts and Forecasts of Environmental Conditions
Initial Transport	50m Temperature		50m Salinity		50m Velocity		Complete set of Plots
	Nowcast - 5 September	Forecast - 6 September	Nowcast - 5 September	Forecast - 6 September	Nowcast - 5 September	Forecast - 6 September
2.5Sv S							X
1.0Sv N							X

Available NetCDF Files
2.5Sv S	Along Grid	peout_free04sep2008.nc.gz
	Geographic	peout_geo_free04sep2008.nc.gz
1.0Sv N	Along Grid	peout_fix04sep2008.nc.gz
	Geographic	peout_geo_fix04sep2008.nc.gz