1

FEMA REGION II Alan Springett, October 18, 2012 Senior Engineer

Introduction to Risk MAP NYC Coastal Study

2

Components of FEMA Region II Coastal Surge AnalysisOverview of entire process at a glance

ADCIRC model development – Comparison to SLOSHModel inputs

Topography/bathymetry Land use

Field reconnaissanceStorm characterization & selectionJPM Statistics and Synthetic TracksWind pressure and field modeling

Surge analysisADCIRC model validationWave setup and surge modelingReturn period analysis

WHAFIS and overland wave heightWave runupCoastal hazard modeling

Agenda

3

Why Are We Here?

Empirical Track + JPM-OS

EST

Tide Gage

JPM-OS-Q

4

Why Are We Here?

5

What Was Done?

November 1983 – First detailed FIS of NYC (new 2’ topography up to elevation 20)

1990’s – Small revision to delete unnumbered A Zones in Staten Island & Queens

1990’s – Two revisions to study several small streams in Staten Island by detailed methods

2007 – Updated NYC DFIRM for purpose of digitizing

6

5/31/13 – New preliminary DFIRM – new 2’ LIDAR & new coastal hydraulics

19 coastal counties in NY and NJ

9 additional counties up the Hudson Valley

What Are We Doing?

7

Numerical ModelsSLOSHOWI PBL Hurricane ModelADCIRC-UNSWAN Surge & Wave Model

Significant Wave Heights and wind vectorspredicted during Hurricane Dennis

on the Basin Scale grid at 50 hours into the simulation.

UNSWAN Model

Models

8

Sea, Lake, and Overland Surges from Hurricanes (SLOSH)

9

ADCIRC Dynamic atmospheric model. Requires five input

parameters: Central Pressure, po Background Pressure, pb Radius of Maximum Winds,

RMW Maximum Wind Speed, Vmax

@10m Forward Velocity, Vf Multiple number of dynamic

surface friction coefficients.

SLOSH Simplified parametric model. Requires three input

parameters: Central Pressure, po Background Pressure, pb Radius of Maximum Winds,

RMW Limited number of static surface

friction coefficients. No tidal interaction Not currently run for extra-

tropical events

Wind Model Comparison ADCIRC & SLOSH*

*From Bailey, James R., PhD.,P.E. FSAR 2.4.5, “Probable Maximum Surge and Seiche Flooding”

10

ADvanced CIRCulation Model (ADCIRC)

A (PARALLEL) ADVANCED CIRCULATION MODEL FOR OCEANIC,COASTAL AND ESTUARINE WATERS

The ADCIRC source code is copyrighted, 1994-2006 by R.A. Luettich, Jr. and J.J. Westerink

A system of programs solving time dependent, free surface circulation and transport problems in two and three dimensions using finite element methods and unstructured grids.

11

Simulating WAves Nearshore (SWAN)

‘Tight’ Coupling of SWAN + ADCIRC

Communication is optimized for high-performance computing:

Solves the action balance equation:

Models use same unstructured mesh; Information passed dynamically

SWAN is as accurate as WAM and STWAVE

Coupled model is efficient to 1000s of computational cores

12

Coastal Surge Analysis Components

Acquire map data and field reconnaissance

Develop model inputs for topo/bathy grid, land use, etc.

Characterize the local storm climate (tropical & extra-tropical)

Develop method of forward projection

Create surge heights with numerical models

Analyze recurrence statistics

Develop 0.2%,1.0%,2.0% & 10.0% surge heights with wave set-up

WHAFIS overland wave conditions & BFEs

13

Coastal Study Technical Advisory Panel (TAP)

Technical Advisory Panel (TAP) was established to engage coastal study subject matter experts

Meetings were held on: 4/14/2011, 5/10/2011 and 11/10/2011

Some of the Members: Academics and Non-Profit Agencies:

Jacques Cousteau National Estuarine Research Reserve (under Rutgers University) Richard Stockton College Coastal Research Center Monmouth University Urban Coast Institute Barnegat Bay National Estuary Program Sustainable Jersey, Climate Adaption Task Force

State and Local Governments: New Jersey Department of Environmental Protection New York State Department of Environmental Conservation New York City (multiple departments: Office of Long Term Planning and Sustainability,

Buildings and Planning) Other Federal Agencies:

NOAA; USGS; USACE;

14

Surge Field Reconnaissance

15

Hurricane Measurements

Period of RecordSearch & Capture Areas

16

Storm Selection for Synoptic Climatology

1938_04 NOTNAMED*1944_07 NOTNAMED*1948_03 NOTNAMED1952_03 BAKER1953_02 BARBARA1953_04 CAROL1954_03 CAROL1954_05 EDNA1954_09 HAZEL1955_02 CONNIE1958_04 DAISY1960_05 DONNA*1961_05 ESTHER1967_04 DORIA1969_07 GERDA

1972_02 AGNES1976_03 BELLE1978_06 ELLA1985_07 GLORIA*1990_02 BERTHA1991_02 BOB1993_05 EMILY1996_05 EDOUARD1996_08 HORTENSE1999_06 FLOYD2002_08 GUSTAV2003_09 ISABEL2004_01 ALEX2007_16 NOEL2009_03 BILL

Thirty storms selected for synoptic climatology

* Used as “AdCIRC/UNSWAN Verification Storm “

All candidate storms (1900-2009; 236 storms, ~ 8,400 snapshots)

17

Nor’easter Storm Analysis

1950 11/251961 04/131962 03/061964 01/121966 01/231968 11/121970 12/171971 02/081971 11/251972 02/191974 12/021979 01/251984 03/29*1987 01/231991 10/31*

1992 12/11*1993 03/141994 03/031994 12/241995 02/041995 11/151996 01/081996 10/201996 12/061998 01/281998 02/052005 10/252007 04/162008 05/122009 11/13

Storm Dates (Y M/D)

* Used as AdCIRC/UNSWAN Verification Storm

18

ADCIRC/SWAN Mesh

Nodes: 604,790Elements: 1,188,640Min mesh size: 70 meters1 second time step goal

19

ADCIRC/SWAN Mesh

20

Topography Data Sources

NYNYC DOITT 3 ft DEM

Westchester 2’ FEMA compliant contours

Hudson Valley USGS NED 1/3 arc second DEM

21

Bathymetric Data Sources

NOAA Hydrographic SurveysNational Geophysical Data CenterOffice of Coast Survey

NOAA Navigation ChartsFrequently provides data in areas

where digital survey data is not available (fills gaps)

USACE SHOALS/CHARTSHydrographic LiDAR Limited coverage due to light penetration in the water column

USACE District Surveys (Retrieved from NAN and NAP) Limited to Federal navigation projects or project specific study

areas

22

Shoreline and Boundary Extraction

Accurate shoreline essential for seamless DEM Other uses:

Boundary (ADCIRC), WHAFIS, DFIRMS Extracted and corrected for each

county Upper limits of flooding generated

and extracted.

23

Probabilistic Model for Storm Characteristics

*

**Eq. for median:

*** truncated at 26 deg for NJ

Parameter Distribution Type∆P (mb) Truncated Weibull* U=41.2 K=2.05Rp (km) Lognormal median=Vickery-Wadhera

(2008)**σlnRp=0.44

Vf (kt) Normal mean=6+0.4∆P σ=7θ (deg) Normal*** mean=22 (23 in LI) σ=10B Normal mean=1.1 σ=0.2

Distribution Parameters

0[ ] exp[ ( / ) ]; (33mb)kP P x c x u x P∆ > = − > ∆

24

JPM-OS1 storms (final* model)

*Latitude-dependent ∆P and Rp

25

Nor’easter Storm Analysis

1950 11/251961 04/131962 03/061964 01/121966 01/231968 11/121970 12/171971 02/081971 11/251972 02/191974 12/021979 01/251984 03/29*1987 01/231991 10/31*

1992 12/11*1993 03/141994 03/031994 12/241995 02/041995 11/151996 01/081996 10/201996 12/061998 01/281998 02/052005 10/252007 04/162008 05/122009 11/13

Storm Dates (Y M/D)

* Used as AdCIRC/UNSWAN Verification Storm

26

Summary of Storm Selection

Hurricanes JPM-OS with 35 Synthetic Storms each on a set of

approximately 6 to 9 tracks (depending on storm size) Consider only storms with DP> 33 mb (Cat > 2)

Extra-tropical Storms 30 Historic storms 12 are “significant”

27

Next Steps

Forward Modeling,Return Period Analyses

and Wave Parameters for WHAFIS

28

Example Envelope of Maximum Compute Surge Heights

159 STORM SIMULATIONS(## STORM-TYPES)

PARALLEL TRACK SETS (W/LANDFALLS BEYOND AREA)

RANDOM SEMI-DIURNALTIDE PHASES

29

General Flow Chart for NY/NJ Production Runs

30

General Flow Chart for NY/NJ Production Runs

Input Checks-• File correct• Scanning

31

General Flow Chart for NY/NJ Production Runs

Input Checks-• File correct• Scanning

Local Checks –• Files completion• Initial MEOWs• Initial time series

32

General Flow Chart for NY/NJ Production Runs Full QA/QC

• MEOWs plots• Time-series plots• Animations• Automated scripts

Input Checks-• File correct• Scanning

Local Checks –• Files completion• Initial MEOWs• Initial time series

33

Inclusion of Astronomical Tide

New York-New Jersey area; (a) Water elevation difference (meters) of ST minus S at time of maximum ST surge; (b) water elevation deviation (meters) from linear superposition with extrapolated tides. Black line is shoreline; brown line is the limit of the model mesh.

34

Sum Exceedance Frequencies for Each Value of Surge Height

0

0.05

0.1

0.15

0.2

0.25

1 1.5 2 2.5 3 3.5 4

Annu

al E

xcee

danc

e Fr

eque

ncy

Surge Height (m)

Combined

Hurricane

Extra-Tropical

0.046(H'cane)+0.038(N'easter)=0.084(total)

35

Composite Surge/Frequency Ranking

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0 20 40 60 80 100 120 140 160 180 200

Wav

e He

ight

(m),

Perio

d (s

)

Surg

e El

evat

ion

(m)

Storm Rank (based on surge)

Surge Wave Height Wave Period

36

Transect Layout and Model Setup

37

Overland Feature CharacterizationCoastal hazard analysis and overland wave modeling require information on land cover and obstructions (buildings) within the study area. Dunes, buildings, plant types and their density control overland wave dissipation.

38

Coastal Structure Evaluation

The presence of coastal structures along a section of shoreline can affect how erosion analyses are conducted. FEMA has specific guidelines on how to evaluate and treat coastal structures in coastal hazard analysis.

39

Erosion Methodology and Analysis

FEMA guidelines and specifications define a standard erosion methodology for dunes (i.e., the “540 rule”). A review of the geology and shoreline types will be made to determine the applicability of standard erosion methods and determine the necessity of non-standard approaches.

40

Wave Height Analysis for Flood Insurance Studies (WHAFIS)

Wave transformation modeling will be conducted from the shoreline and overland using FEMA’s WHAFIS model. The WHAFIS model has been used in Flood Insurance Studies since 1980 to incorporate the effects of wave action on FIRMs for communities along the Atlantic and Gulf Coasts. WHAFIS is a one-dimensional model that will be applied to each transect in the Region II study area. The model uses a specified Stillwater Elevation (SWEL), and the starting wave conditions as input. Simulations of wave transformations are then conducted with WHAFIS taking into account the storm-induced erosion and overland features of each transect. Output from the model includes the combined SWEL and wave height along each cross-shore transect allowing for the establishment of BFEs and flood zones from the shoreline to points inland within the study area.

41

Wave Runup and Over-topping

1938 Long Island Express waves hitting seawall -Photo courtesy of NASA

Wave runup is defined as the maximum vertical extent of wave uprush on a beach or structure. FEMA’s 2007 Guidelines and Specifications require the 2% wave runup level be computed for the feature being evaluated (cliff, coastal bluff, dune, or structure). Each transect defined within the Region II study area will be evaluated for the applicability of wave runup.

42

Coastal Flood Hazard Mapping Coastal flood hazard mapping is the process where the overland wave modeling results are assimilated with the topography data to delineate the boundary of the Special Flood Hazard Area (SFHA) for the 1% annual chance stillwater elevation (100-year conditions), along with mapping the location and extent of Zones VE, AE, and X.

43

FIRM Production

44

Coastal Outreach Advisory Team (COAT)

The Coastal Outreach Advisory Team was established to support the New Jersey and New York City Coastal Flooding Outreach and Education Program

Technical Advisory Panel (TAP) is focused on technical aspects of the flood risk program

45

Introduction to Risk MAP NYC Coastal Study

Questions?-

ContactsAlan Springett, Senior Engineer, FEMA Region II Mitigation, Risk Analysis26 Federal Plaza, Rm 1337New York, NY 10278212-680-8557 – Desk347-633-4342 – MobileAlan.Springett@fema.dhs.gov

J. Andrew Martin, CFMRSC II Lead Coordinator Dewberry Management & Consulting Services15 East 26th St, 7th FloorNew York, NY 10010-1505646.216.3677 amartin@dewberry.com