UNDERSTANDING WIND/WAVE FORCING OF THE ST. JOHNS RIVER

1
UNDERSTANDING WIND/WAVE FORCING OF THE ST. JOHNS
RIVER

• Scott C. Hagen (UCF/CEE/CHAMPS Lab)
• Yuji Funakoshi (NOAA/NOS/CSDL)
• Andrew Cox (Oceanweather, Inc.)

2
Outline

• Overall Modeling Approach
• Forcing Mechanisms for the St. Johns River
• Short and Long-Wave Coupling
• Conclusions
• Future Work

3
Dr. Pedro RestrepoMs. Reggina Cabrera
4
NOAA/NWS/OHD Project Goals

• Development of a 2D model for the St. Johns
  River to predict flow tides (astronomic and
  meteorologic)
• Develop the model and examine test cases
• Examine uni-coupling model of short- and
  long-wave models
• Examine two-way coupling of short- and long-wave
  models

5
Storm Tides
Hydrodynamic Model (Long waves)
Wave model (short waves)
Storm surges
Storm tides
(Schematic showing of Storm Tides, Graber 2006)
Coupling hydrodynamic and wave models to describe
the storm tides
6
Storm Surge vs. Storm Tide
Source NHC / NOAA (http//www.nhc.noaa.gov/HAW2/en
glish/storm_surge.shtml)
7
Domain Area 8.347106 km2
Computational Nodes 52,774
Triangular Elements 98,365
Minimum Node Spacing 0.5 km
Maximum Node Spacing 160 km
Boundary Spacing 6.0 km
Boundary Nodes 7,111
Atlantic Ocean
Gulf of Mexico
Caribbean Sea
8
Western North Atlantic Tidal Model Domain
Continental Shelf Break (183 m)
Gulf of Mexico
Edge of Blakes Escarpment (1200 m)
Atlantic Ocean
Caribbean Sea
Open-Ocean Boundary, 60W Meridian
9
Western North Atlantic Tidal Model Domain
Gulf of Mexico
Atlantic Ocean
Caribbean Sea
Open-Ocean Boundary, 60W Meridian
10
St. Johns River with Major Basins (Sucsy and
Morris 2002)
11
St. Johns River
Longest River (500 km) Contained Wholly Within
Florida Slow-Moving River With Low Slope (2.2
cm/km)
12
Large-Scale Approach (WNAT-SJR Mesh)
75,436 Nodes 138,622 Elements Maximum Element
Size 160 km (Deep Ocean) Minimum Element
Size 50 m (St. Johns River)
13
Pseudo-Operational Mesh
26,543 nodes 47,763 elements Max node space
40km Min node space 50 m
Florida Coast
14
St. Johns River Inlet to Lake George (maps and
photos courtesy of USGS with graphics by
Funakoshi 2006)
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
122-Day Simulation Tides Full Meteorology
122-Day Run Length (June 1, 2005 September 30,
2005) 3-Day Forcing Ramp 2-Second Time
Step Large-Scale Approach (i.e., WNAT-SJR Mesh
Application) Local-Scale Approach (i.e.,
SJR-Inlet Mesh Application) Open-Ocean Boundary
Elevation Forcings K1 O1 M2 S2 N2 K2
Q1 Advective Terms Enabled River Inflows USGS
Gage Data Surface Forcings 1-Hour Pressures
Winds Provided by Oceanweather Inc.
20
Mayport
I-295 Bridge West End
Buffalo Bluff
21
Bar Pilots Dock, FL
22
Domain Extent Sensitivity I-295 Bridge, West
End, FL
23
Domain Extent Sensitivity Red Bay Point, FL
24
Buffalo Bluff, FL
25
Root Mean Square Error (cm)
Station Index 122 Days (6/1/05 9/30/05) 122 Days (6/1/05 9/30/05) 122 Days (6/1/05 9/30/05)
Station Index WNAT-SJR Tidal BC Hydrograph BC
1 16.8 17.9 16.2
2 15.7 16.4 16.0
3 11.1 11.4 11.4
Station Index Ophelia (9/6/05 9/15/05) Ophelia (9/6/05 9/15/05) Ophelia (9/6/05 9/15/05)
Station Index WNAT-SJR Tidal BC Hydrograph BC
1 26.4 35.5 26.1
2 20.2 29.5 20.3
3 12.3 21.1 12.4
4 10.3 16.8 10.2
26
(No Transcript)
27
Uni-Coupling and Coupling
Uni-Coupling
Coupling
28
Nested Wave Model Grids
0.02 Degree SWAN Grid
Fernandina Beach
Mayport
0.005 Degree SWAN Grid
St. Augustine Beach
0.1 Degree WAM Grid
29
Fernandina Beach
30
Mayport
31
St. Augustine Beach
32
Conclusions

• A faithful representation of the physical system,
  forcing processes (wind, pressure, tides,
  riverine flows, waves) and of the flow itself
  (through grid resolution and accurate algorithms)
  is critical to a truly predictive astronomic and
  storm tide model.
• Localized high resolution is critical to capture
  the physics of storm tide generation and
  propagation in any spatially complex system.
• Physics of the storm tide are complicated.
• Wind forcing for the St. Johns River is equal to
  or greater than that of astronomic tides and
  generally supersedes the impact of inflows.
• Pressure variations have minimal impact.
• Water levels inside the St. Johns River depend on
  the wind forcings in the deep ocean however, if
  one applies an elevation hydrograph boundary
  condition from a large-scale domain model to a
  local-scale domain model the results are highly
  accurate.

33
Conclusions

• Regardless of whether one uses uni-coupling or
  two-way coupling, wind-induced waves result in an
  approximately 10 15 higher peak storm tide
  level than without any coupling.
• The wave-current interaction described by a
  two-way coupling model results in decreasing
  peaks and increasing troughs in the storm tide
  hydrograph.
• Wind drag formulations that are presently
  employed are spatially and temporally dependent
  and for the purposes of recreating the entire
  storm tide the present formulations are
  inadequate.

34
Future Work

• Examine the impact of inundation areas and tidal
  marshes for the 122-day hindcast period.

• Calibrate the model with Mannings coefficients.

35
Questions? shagen_at_mail.ucf.edu