Global Warming and Hurricane Activity and the Impact on U.S. Transportation

forestsaintregisΠετρελαϊκά και Εξόρυξη

8 Νοε 2013 (πριν από 3 χρόνια και 7 μήνες)

62 εμφανίσεις

Global Warming and Hurricane Activity and
the Impact on U.S. Transportation

Dr. Douglas A. Stewart

and

Dr. Lee E. Branscome


Climatological Consulting Corporation

Palm Beach Gardens, Florida

Factors Influencing Hurricane…

Is Global Warming Linked to Hurricane Activity?


Formation frequency


Geographical distribution


Size


Rainfall


Modeling studies


Empirical studies


Natural variability

Impact on Transportation Infrastructure


The linkage debate


Metrics of activity


Intensity

Hurricane intensity

Emanuel’s MPI

(Emanuel 1986,1987,1991)

Holland’s MPI

(Holland 1997)


Underlying SST


Depth of warm water


Wind shear


Upper level disturbances


Eye wall replacement cycles


Lateral moisture flux



Maximum Potential Intensity (MPI) Theory

Hurricane formation frequency


Favorable environment (SST, low wind shear, deep moist layer)


Incipient disturbance of independent origin

Emanuel’s genesis index (
http://wind.mit.edu/~emanuel/gindex.htm
)


G.I.=|10
5
h
|
3/2
(H/50)
3
(V
p
/70)
3
(1+0.1V
s
)
-
2


Distance from equator (Coriolis effect to induce rotation)

h
=Absolute vorticity

H
=600mb R.H.

V
p
=potential wind speed (MPI)

V
s
=wind shear |v
850
-
v
250
|

Geographical distribution of hurricanes


Characteristics/locations of incipient disturbance


(e.g., frontal perturbation, easterly wave)


Quasi
-
stationary large
-
scale circulation features


Spatial distribution of SST anomalies


(http://www.tpc.ncep.noaa.gov/tracks)

1995: 19 named storms

2004: 15 named storms

NAO Effect


NAO Patterns for


July and October


(
+ phase
)


Tropical cyclones during NAO extremes

(monthly NAO data from 1950
-
2005)

10% of storms @ max (
-
) NAO index

10% of storms @ max (+) NAO index

1950

1960

1970

1980

1990

2000


Monthly NAO index averaged
over individual hurricane seasons

ENSO Effect

Tropical cyclones during ENSO extremes

Multivariate ENSO Index data from 1950
-
2005 (NOAA
-
CIRES CDC)

HURDAT best track data from 1950
-
2005


Monthly ENSO index averaged
over individual hurricane seasons

1950

1960

1970

1980

1990

2000

25% of storms @ max (+) ENSO index

25% of storms @ max (
-
) ENSO index

10% of storms @ max (+) ENSO index

10% of storms @ max (
-
) ENSO index

Hurricane size


Radius of maximum winds (RMW); cloud shield

(http://www.tpc.ncep.noaa.gov)

Charley 2004: Cat 4

Wilma 2005: Cat 3


Historical database on RMW is limited in quantity and quality

NCDC image, courtesy Dr. Jack Beven (http://www.mindspring.com/~jbeven/intr0012.htm)


Some damages from small storms

Andrew: cat 5 (35.9B); Charley: cat 4 (15.4B); Camille: cat 5 (7.5B)

(adjusted to 2005 dollars)


Some damages from large storms

Frances: cat 2 (9.1B); Jeanne: cat 3 (7.1B); Floyd: cat 2 (5.1B)

Katrina: cat 3 (e75B); Rita: cat 3 (e10B); Wilma: cat 2 (e12.2B)


Factors governing hurricane size largely unknown


More observations, theory and hi
-
res modeling needed

(adjusted to 2005 dollars)

Impacts from large and small storms

Hurricane rainfall


Large hurricanes produce more rainfall than small


Slow moving storms produce more rainfall locally


10
-
15 inches not uncommon


(Mitch (1998) produced 25 inches in 36 hours in southern Honduras)


Tropical storms may be copious rain producers


(Claudette, 1979 produced 45 inches near Alvin, TX;


Allison, 2001 produced > 30 inches on parts of Houston.TX)


Local topography may result in flash flooding


(often contributing to the majority of fatalities)


City drainage easily overwhelmed in flat coastal regions


Increased hurricane rainfall likely

From modeling studies…as high as 40% increase for intense storms

(Gutowski, 1994) and 18% increase for storms within 100 km of center

(Knutson and Tuleya, 2004) are estimated in a doubled CO2 climate]


Limited historical data on hurricane rainfall amounts


Estimates from radar and satellite are for limited samples only

Changes in a warming climate?

Is global warming influencing

hurricane activity ?

Hybrid Models


Hi
-
res hurricane simulations with CGM BCs


Ignores genesis frequency issues


Tests MPI theory


Rainfall sensitivity

Empirical Studies


Cyclone activity metrics over period of record


Secular trend vs. natural variability


Limited records
-

statistical significance issues


Non
-
uniformity of measurement methods


Tests MPI theory


Stimulates further theoretical inquiry


Hurricane activity in control vs. warm climate simulations


Conflicting results in hurricane frequency


Problems with simulating hurricanes and their genesis


Some robust results

Coupled Global Models (CGM)

Hybrid Model Study

Knutson and Tuleya (2004)
-

J. Climate

Do models show increased hurricane intensity and rainfall in a warmer climate?


Temperature and moisture profiles and SST from 9 CGMs used as background


Simulate control climate and warm climate (1% CO2 increase/yr)


~1300 5
-
day idealized hurricane simulations with the GFDL hurricane model


14% increase in central pressure fall


6% increase in max surface winds


18% increase in precipitation within 100 km of center


2x increase in cat
-
5 storm probability

Cat 3


Cat 5



Cat 4

Hurricane intensity distribution: Aggregate results


Minimum central pressure (mb)

(3 ocean basins (ATL,WPAC,EPAC); 4 convective parameterizations)

Empirical Study

Trenberth (2005)
-

Science

Look for trends in Atlantic SST and hurricane activity [Accumulated Cyclone Energy Index]



(ACE= sum of V
m
2

over cyclone lifetime for all storms in a season)

SST over Atlantic basin

from 1870
-
2004

Annual ACE variability

from 1950
-
2004.

Empirical Study

Emanuel (2005)
-

Nature

Trends in Atlantic and Pacific SST and hurricane activity [Power Dissipation Index]



(PDI= sum of V
m
3

over cyclone lifetime for all storms in a season)

ATL

WPAC

ATL+WPAC

Empirical Study

Webster et al. (2005)
-

Science

Trends in cyclone number, duration and intensity in a warming environment


(satellite era only; 1970
-
2004)


6 ocean basins: NATL, WPAC, EPAC, SPAC, NIO, SIO


No distinct secular trend in storm numbers or duration, globally


Only secular trend in number and duration is in the NATL basin


No trend in maximum wind speed of hurricanes


Distinct secular trend in number of intense hurricanes, globally

ATL Hurricane Activity Metrics
(HURDAT 1870
-
2005)

% of median

# Storms

ACE

PDI

Storm Days

1870

2005

1950

# cat
-
3,4,5

1870

2005

1950

Cat
-
4,5/hur

%

% of median

% of median

% of median

Number

ATL SST variability

ATL SSTA

1870

2005

1950

All (TS + H)

Major H

% of median

% of median

Storm numbers

Storm duration

U.S. Landfalls

Global Warming
-
>Increased Hurricane Activity?


Peer
-
reviewed studies establish a link between increased hurricane

intensity and increased SST
(Trenberth, 2005; Emanuel, 2005; Webster et al., 2005)



Warming SST attributed to increased greenhouse gas emissions
(IPCC,2001)


Skeptics claim:


The skeptics choose the press rather than the scientific peer
-
review process


No trends, only multi
-
decadal fluctuations


SST signal part of the natural variability


Past data are erroneous


Land
-
falling storms do not support linkage


Metrics of hurricane activity are misleading


There are still many unknowns


Why 80
-
90 global hurricanes/yr and not 25 or 200?


Why do storm numbers decrease while intensities increase in most basins?


What is the role of hurricanes in the global heat redistribution?

Impact on Transportation Infrastructure

Increased risk of:


Surge and wave damage to ports, bridges, coastal airports
and highways


Inland wind/rain damage:


Orlando Int. Airport’s damage from Charley, Frances, Jeanne (2004);
Homestead Airport destruction from Andrew (1992)


Traffic signals/sign destruction


Adverse impacts on fuel costs and availability:


Damage to offshore drilling platforms and coastal refineries


Energy price shocks from more frequent threats

Significant impact on infrastructure:


Greater Likelihood of wind/flood/surge damage over the
lifetime of the structure.


Location of new infrastructure will require higher elevations

(Elevation 9 ft)