NOAA/NOS/Coast Survey Development Laboratory Port of New York/New Jersey Operational Forecast System (NYOFS) and Tracer Simulation Eugene Wei Coastal Survey.

NOAA/NOS/Coast Survey Development Laboratory

Port of New York/New Jersey Port of New York/New Jersey Operational Forecast System Operational Forecast System

(NYOFS)(NYOFS)and Tracer Simulationand Tracer Simulation

Eugene WeiCoastal Survey Development Laboratory

NOAA/National Ocean Service

Ted CaplowPeter Schlosser

Earth and Environmental Engineering Department

Columbia University

2003 Terrain-Following Ocean Models Users Workshop

4-6 August, 2003, PMEL/NOAA, Seattle WA


Port of New York/New Jersey Port of New York/New Jersey Operational Forecast System Operational Forecast System

(NYOFS)(NYOFS) and Tracer Simulation and Tracer Simulation

ObjectivesObjectives

Promote Safe NavigationPromote Safe Navigation Provide mariners and port managers Provide mariners and port managers

with with near real-time and forecastnear real-time and forecast water levels and currents water levels and currents throughout the harborthroughout the harbor

Tracer dispersion study and Tracer dispersion study and predictionprediction


Physical Oceanographic Real-Time System

Center for Operational Oceanographic Products and Services


NeedsNeeds

Near real-time water level and Near real-time water level and current information at many non-current information at many non-gage locationsgage locations

Short term water level and current Short term water level and current forecast guidance informationforecast guidance information

Detail current information (Kill Detail current information (Kill Van Kull) for Coast Guard ship Van Kull) for Coast Guard ship “right of way” decision“right of way” decision



Technical Technical ApproachApproach

3-D barotropic Princeton Ocean 3-D barotropic Princeton Ocean Model (POM)Model (POM)

Coarse grid: 73X134 cells, 7 levels, Coarse grid: 73X134 cells, 7 levels, 130 m – 1 km130 m – 1 km

Higher resolution nested fine grid: Higher resolution nested fine grid: 126X34 cells, 7 levels, 70 m – 130 m126X34 cells, 7 levels, 70 m – 130 m

One-way couplingOne-way coupling


- One-way coupling

The fine grid (FG) models obtain their open boundary conditions from the coarse grid (CG) models and FG model through a common block. Two models run as a single model.


Model Nowcast/Forecast Model Nowcast/Forecast FlowchartFlowchart



Model and System Skill Assessments

* Astronomical tide simulation* Hindcast* Semi-operational nowcasts/forecasts* Operational nowcasts/forecasts


http://co-ops.nos.noaa.gov/NYOFS/nyofs.shtml









* Near real-time model input - Boundary conditions (water levels) at Sandy Hook and Kings Point

Operational Model Forecast System Challenge




* System operational flags - CORMS (Continuous Operational Real-time Monitoring

System)





* System operational flags - CORMS (Continuous Operational Real-time Monitoring

System)

* System failure automatic recovery procedure - what if a network interruption ? - what if a system server shutdown or reboot ?



SF6 Tracer Experiments and Simulations

Earth and Environmental Engineering Department Columbia University

• July 14-25, 2002 at Newark Bay

• injected at 1600 UTC July 14• ~ 0.94 mol dissolved through water column• automatic sampling at 1.2 m depth• gas extractor and gas chromatograph

• passive tracer concentration simulation using NYOFS model • initial concentration and gas exchange rate


Daily average concentration agrees well between model and data. Model shows dispersion into lower bay even on Day 1

Daily average concentration agrees well between model and data. Model shows dispersion into lower bay even on Day 1



Measured and Simulated Total SF6 Inventory

Initial Injection: 0.9 molModel is forced with exponential decay gas exchange

Model is forced with exponential decay gas exchange


Longitudinal SF6 Concentration Model Data Comparison Along Arthur Kill

Distance inland from Arthur Kill Mouth

Model seems underestimate tracer dispersion probably due to less current speed along AK

Model seems underestimate tracer dispersion probably due to less current speed along AK


Longitudinal SF6 Concentration Model Data Comparison Along Kill Van Kull

Distance seaward from Bergen Pt.

Model better representing current speed in KVK, straight channel

Model better representing current speed in KVK, straight channel


SF6 Tracer Experiments and Simulations

Earth and Environmental Engineering Department Columbia University

• June 17-July 5, 2003 at East River

• ~ 1.0 mol dissolved through water column• concentration dispersion predictions to assist planning


Move the injection location north because the prediction showing dispersion to upper bay faster than expected

Move the injection location north because the prediction showing dispersion to upper bay faster than expected


NYOFS is running operationally at NOS’s CO-OPS with hourly nowcast and four times a day 30-hour forecasts

Most model water level and current skills meet NOS standard

Data validated model for tracer concentration prediction to assist field work planning

http://co-ops.nos.noaa/NYOFS/nyofs.shtml

SummarySummary



NOAA/NOS/Coast Survey Development Laboratory Port of New York/New Jersey Operational Forecast System (NYOFS) and Tracer Simulation Eugene Wei Coastal Survey.

Documents

prediction slide

way decision slide

s ervices slide

forecast water levels

seattle wa slide

current information

realtime water level

single model