Numerical simulation on estuarine and coastal circulation in the Caribbean and the Virgin Islands using ROMS M. Solano 1 , J. Capella 3 , M. Canals 2 , J. Morell 3 , S. Leonardi 1 1 University of Puerto Rico Mayagüez Campus, Mechanical Engineering Department 2 University of Puerto Rico Mayagüez Campus, General Engineering Department 3 Caribbean Coastal Ocean Observing System Sponsored by: Caribbean Coastal Ocean Observing System University of Puerto Rico 1
Outline • Background • Motivation • Objectives • Setting • Numerical Setup – Numerical model description – Grid generation – Initial/Boundary conditions – Forcing • Regional Model Results – San Juan Bay – Charlotte Amalie • Conclusions • Future Work 2
Background • Our initiative consist of a team of mechanical engineers graduates from the CFD UPRM lab to create an operational high resolution coastal observing system for Puerto Rico and the Virgin Islands, sponsored by CariCOOS and the UPR. • CariCOOS, the United States Caribbean Coastal Ocean Observing System , with CaRA guidance, operates a network of observing assets including data buoys, coastal meteorological stations, vessels, instruments and radars. CaRA is one of eleven regional associations (RAs) devoted to coastal ocean observing throughout the United States coastal oceans and the Great Lakes. • Efforts to create this high resolution observing system consist in downscaling existing operational models in the Caribbean to provide reliable high resolution information for various coastal and estuarine applications. • Past studies of the department of marine sciences at UPRM include surf-zone hydrodynamics (wave induced currents) of “ Tres Palmas” marine reserve, sediment transport and sand composition in the near-shore of Rincon and the Mona passage using a non-linear shallow water wave model (BOUSS2D). • Our efforts began just after the last Miami coordination workshop and since then we have successfully implemented ROMS in the San Juan bay, testing its diverse range of applications such as online and offline nesting capabilities, river discharge and particle tracking in hindcast mode. 3
Motivation • Puerto Rico and St. Thomas have some of the busiest ports in the Caribbean, including touristic cruise and cargo ships. • Several estuarine regions with national reservoirs filled with diverse wildlife and unique bio luminescent microorganisms make coastal regions rare and special. • Particle tracking algorithms have a lot of potential in ocean related applications such as spill containment, sediment transport and coastal security. 4
Objectives • Implement ROMS numerical model in Puerto Rico and the US virgin Islands to study coastal and estuarine ocean dynamics and validate results against hydro- graphic observed measurements. • Develop more efficient particle tracking algorithms and test it in real life applications (Passive drifter) • Create a high resolution, operational ocean-atmosphere observing system for any potential application in Puerto Rico and the US Virgin Islands. 5
Setting • Puerto Rico is the smallest of the greater Antilles, located between the Caribbean sea and the Atlantic Ocean. The U.S. Virgin Islands are located east of Puerto Rico, these include: St. John, St. Thomas and St. Croix. • The region is characterized by numerous islands on the continental shelf extending all throughout the Antilles, with the Puerto Rico trench in the North and the Caribbean sea in the south. • Several estuarine and coastal regions. 6
Numerical Set Up • Regional Ocean Modeling System ( ROMS ) – Free Surface, terrain following, primitive equations ocean model. – Hydrostatic – Boussinesq – Mixing schemes (horizontal and vertical) • Grid: Arakawa C-grid – Horizontal: Arakawa C-grid – Vertical: Sigma terrain following • Time stepping: – Barotropic – Baroclinic 7
Initial and Boundary Conditions • Sea state variables used to generate initial and boundary conditions are interpolated from the AmSeas NCOM model. • The Naval Oceanographic Office (NAVOCEANO) operational ocean prediction system for the Gulf of Mexico and Caribbean is based on the NRL-developed Navy Coastal Ocean Model (NCOM). • The AmSeas model has a 3km resolution with 40 vertical levels. The model assimilates SST, altimetry (SSH) as well as profile temperature and salinity nested into the 1/8 degree global NCOM model. • The nonlocal closure scheme is based on the K-profile, boundary layer formulation by Large et al. (1994). The K-profile scheme has been expanded to include both surface and bottom oceanic boundary layers. • Open lateral boundary conditions. – Flather condition: Barotropic variables (2D). – Orlansky condition: Baroclinic variables (3D). • Forcing: – Tides – Winds – Rivers http://www.northerngulfinstitute.org/edac/NCOM_AmSeas.php 8
Input parameters • Grid – Horizontal resolution: • (SJU) 90 m [496 x 446] || (STT) 110 m [299 x 202] – Vertical: 24 sigma layers – Stretching parameters: theta_s=5, theta_b=0.1, t_cline=1 m • Time stepping – Simulation time: SJU (5 months ) | STT ( 1 month) – Time stepping: SJU (1 sec) | STT (2 sec) • Boundary Conditions – Free surface: Chapman – Open lateral conditions at open sea: Flather (2D), Radiation (3D) – Closed boundary condition for land areas • Forcing: – Tides (OTIS) – River 9
San Juan Bay, Puerto Rico 10
Velocity Vectors at the Surface Superimposed to Velocity Magnitude Contours Area of interest 11
Time Series of Sea Surface Height and the U-comp. of the Velocity (Jan- Feb 2011) 12
Tidal Components and River Discharge • Tidal components were extracted using Point of Fresh Water the Oregon Tidal Inversion Software . At the site of river ‘Rio Piedras ’ • Point source of fresh water from the National Water Information System from USGS – Simulation 1 • Amseas NCOM • Tidal Forcing • Wind Forcing – Simulation 2 • AmSeas NCOM • Tidal Forcing • Wind Forcing • River Forcing 13
Velocity Vectors Superimposed to Color Contours of Velocity Magnitude w/o river w/ river Velocity Vectors at Sea Surface Superimposed to Color Contours of Salinity w/o river w/ river 14
Currents Direction Rose - Simulation with River Simulation Experimental Data Depth of 2.3 meters -> Depth of 7.3 meters -> 15
Vertical Section of Salinity Contour Inside the San Juan Bay - Simulation with River 16
San Juan Bay Velocity Vectors Superimposed to Color Contours of Barotropic Velocity Magnitude (25-Jan-2011) w/ river w/o river 17
Charlotte Amalie, St. Thomas 18
St. Thomas Time Series of Sea Surface Height and Depth Integrated U and V
Velocity Contours with Superimposed Vectors Scale: -0.5 m/s to 1.3 m/s Errors in bathymetry High velocity tidal currents
Temperature and Salinity Contours Scale: 27.5 C to 30.5 C Scale: 34.4 g/kg to 35.9 g/kg
Conclusions and Future Work • In just under a year the hindcast models developed for the San Juan Bay and the Virgin Islands show promising results. • In the San Juan Bay, computed velocity agree well with the data collected with the ADCP at mid levels. Current velocity and characteristic tidal current directions are well reproduced by de model. • In Charlotte Amalie modelled results agreed reasonably well with sea surface height and velocity direction measured by CariCOOS buoy VI102 located south of St. John. The velocity magnitude presents discrepancies from experimental observation. Errors are probably due the approximation in bathymetry. • Future work includes coupling of (real time) atmospheric and wave models to validate results and move towards the goal of establishing a high resolution operational model. 22
Comments and Questions 23
Recommend
More recommend
