Hurricane Threat and Risk Analysis in Rhode Island

Graduate School of Oceanography

Isaac Ginis, Richard Yablonsky,

and Tracy McCormick

The Hurricane Threat and Risk

Analysis in Rhode Island

Hurricane Gloria on Sept. 27, 1985 (NOAA)

Beach SAMP Stakeholder Meeting:

Hurricanes and Storm Recovery in Rhode Island

July 24, 2014

• Frequency and severity

– How often do hurricanes make landfall?

– Where?

– How strong?

• Physical hazards

– What is the spatial pattern of the wind?

– How high is the storm surge?

– How much rain falls?

• Vulnerability

How much damage is caused by physical hazards?

RI Hurricane Threat

Tropical Cyclones Impacting RI since 1851

Total Number: 56

Tropical Cyclones Impacting RI since 1851

Great Colonial Hurricane of 1635

−10−

Figure 2.1 Track of the Great Colonial hurricane of 26 August 1635, with hourly positions in local standard time

and central pressure in millibars. Possible 80°F SST isotherm also shown.

Jarvinen (2006)

Storm Tides

−11−

Figure 2.2 Track of the Great Colonial hurricane of 26 August 1635, with hourly positions in local standard time,

pressure in millibars and SLOSH model maximum over water 1-minute wind speed in miles per hour. Circles

represent location of maximum wind with radius given in statute miles. Wind vectors show where maximum wind

is occurring at that time. Wind barbs in mph.

Great September Gale of 1815

−18−

Figure 3.1 Track of the Great September Gale on September 23, 1815, with hourly positions in local standard time

and central pressure in millibars.

Figure 3.2 Track of the September 23, 1815 hurricane with hourly positions in local standard time, pressure in

millibars and SLOSH model maximum over water 1-minute wind speed in miles per hour. Circles represent

location of maximum wind with radius given in statute miles. Wind vectors show where maximum wind is

occurring at that time. Wind barbs in mph.

−19−

Figure 3.3 Graphical computation of the storm tide hydrograph from the addition of the SLOSH and tide

hydrographs at two locations. The peak of the storm tide hydrograph is compared to the observed height. SLOSH

model over water 1-minute wind speeds in miles per hour are plotted with wind barbs indicating direction.

Jarvinen (2006)

Storm Tides

Storm Surge: New England Hurricane (1938)

Hurricane Sandy, 2012 1938 Hurricane

Sept 20

Sept 21

Oct 26

Oct 29

A tale of Two Hurricanes

• 8:30 am – hurricane centered near Cape Hatteras

• 2:30 pm – made landfall in Long Island

• 4:00 pm – made landfall in CT, RI

• 6:00 pm – reached Vermont

• 10:00 pm – crossed into Quebec

September 21, 1938

Forward speed reached 70 mph, the highest recorded!

Sandy: Maximum Sustained Winds (kt)

Sandy: Maximum Wind Gusts (kt)

Hurricane Wind Measurements – WeatherFlow Mesonet

GSO Anemometer: Monday, Oct 29, 2012

100 anemometers designed to survive hurricane winds

Invaluable source of high quality hurricane wind measurements

Inland Flooding: Connie & Diane (1955)

CT

State

Library,

State

Archives,

File

Name

55flood17

NOAA/WPC

NOAA/WPC

Naugatuck, CT: August 19, 1955

Inland Flooding: Esther (1961)

NOAA/WPC

Inland Flooding: Irene (2011)

NOAA/WPC

Margaretville, NY VT Route 100

Windham, NY

Windham, NY Wilbur’s Pt., Fairhaven, MA

• Most hurricanes approaching RI undergo a

transition from pure Tropical to “Extratropical”.

This transition implies significant changes in the

storm size, wind structure and rainfall pattern.

• The area of high winds and rain often expands

significantly. As a result, a wider area is affected

and storm’s total energy increases in many

cases.

Common Characteristics of RI Hurricanes

Cyclones: Tropical vs. Extratropical (Nor’easters) Hurricane Katrina: August 28, 2005 “Superstorm”: March 13, 1993

• Derives energy from ocean surface via release of latent heat in convective clouds

• Derives energy from horizontal temperature gradients (baroclinic instability)

• Develops best in a barotropic environment, far from jet stream disturbances

• Develops best in a highly baroclinic environment, close to jet stream disturbances

“Late” extratropical transition

of Hurricane Arthur (July 1-7, 2014) Escuminac, N.B.

Eastern Canada

Fredericton, N.B.

Frequency: Some computer models indicate either

reduction or increase in TC frequency. However, most

models show reduction of 0-20%. We have very low

confidence in projected changes in individual basins.

Tropical Cyclone Projections

Due to Global Warming

Intensity: More intense tropical cyclones (2-11% for an

IPCC A1B scenario). The frequency of the most intense

(rare/high-impact) storms will likely increase by a

substantially larger percentage in some basins.

Rainfall: Rates are likely to increase. The projected

magnitude is on the order of +20%.

“Downscaling” Method to Model the Impact of

Global Warming on Frequencies and Intensities

of Atlantic hurricanes

Source: Bender et al., Science, 2010.

21st Century Climate Warming Projected

Changes in Atlantic Hurricane Frequency

Colored bars show changes for the 18 model CMIP3 ensemble (27 seasons); dots

show range of changes across 4 individual CMIP models (13 seasons).

Cat 4+5 frequency:

81% increase, or

10% per decade

Source: Bender et al., Science, 2010.

Estimated net impact

of these changes on

damage potential:

+28%

Hurricane Risk Analysis in RI

• How well do we understand the hurricane risk?

• Do we have necessary modeling tools?

…Not really…

• The primary tool is FEMA’s HAZUS based on

NOAA’s SLOSH model developed in the mid

1960s

Parametric Wind Model Used in SLOSH

Radial distance

Win

d S

peed

Simplified Axisymmetric Wind Profile

Typical parameters used: Vm, Rm, Pc, Po

Asymmetries

are added by

including

translation

velocity

vector.

Not suitable for RI

hurricanes

undergoing extra-

tropical transition!

A Need for Actionable Science

for Risk-informed Decisions

• RI needs robust, scientifically defensible

modeling tools to quantify the combined

coastal and inland hazards from hurricanes.

• Advanced modeling tools will help to more

accurately and clearly communicate

hurricane risks to RI stakeholders, including

threats to life, property and existing or

planned infrastructure.

Hurricane Risk Modeling Strategy for RI

To use advanced

hurricane-ocean coupled models at

open-ocean scales (a),

multiple coastal ocean circulation,

surge, and wave models from

the shelf (b) to estuaries (c), to

urbanized estuary-tributary interface (d),

combined with watershed rainfall runoff

and river flood models and environmental

biogeochemical/ecological models

Numerical Weather Prediction Hurricane Models

•The 3-D physical

processes governing the

evolution of the

hurricane are simulated,

resulting in realistic

estimates of the surface

wind and rainfall.

•This approach must be

adopted for hurricane

risk analysis in RI

U.S. Operational Hurricane Models

• GFDL/GFDN – used by National Hurricane Center (NHC)

in the Atlantic Ocean and East Pacific since 1995 and

Joint Typhoon Warning Center (JTWC) in all ocean basins

since 1998

• HWRF – used by NHC in the Atlantic since and East

Pacific in 2007 and JTWC in the West Pacific since 2013

The GSO hurricane research group has been involved in the

development and improvements of GFDL/GFDN and HWRF

models in collaborations with NOAA and Navy scientists.

Nested Movable Grid Configuration

in Hurricane Models

GFDL Hurricane-Ocean Coupled Model

Forecast of Hurricane Katrina (2005)

Applying NOAA’s GFDL model for wind field

simulation during hurricane landfall

Hurricane

Isabel (2003)

Numerical weather prediction vs.

parametric wind fields

Landcover Variability in RI and Hurricane Wind

•Different landcovers (or

surface roughness) have

different “frictional”

characteristics which

dramatically affect the winds

near the ground.

• Variations in landcover are

responsible for most of the

local variability in the wind

fields over relatively small

distances.

(Landcover refers to the homes, forests, fields, and rivers)

Hypothetical “Hurricane Rhody”

High-impact, physically realistic scenario based

on historical hurricanes that affected RI

Summary

• Robust hurricane risk analysis is required for

better preparation and mitigation of hurricane

hazards in RI.

• Advanced wind, coastal and inland flood models

will improve accuracy of storm-induced and

post-storm environmental impacts.

• Climate change-induced increases in hurricane

intensity, rainfall, and sea level rise will

exacerbate wind, inland and coastal flooding

impacts.