The Science Behind Sandy

1
The Science Behind Sandy

• Adam Sobel
• Columbia University

2
Biography of a hurricane
30
29
numbers are dates in October
28
27
26
Yellow tropical storm Red cat 1
hurricane Magenta cat 2
25
24
Blake et al. (2013) NHC report on Sandy
23
22
nytimes.com
3
Sandys grandparentsThe Madden-Julian
oscillationand the North Atlantic oscillation
4
The Madden-Julian oscillation a natural
fluctuation of the tropical climate
Outgoing longwave radiation 15S-15N Blue
rainy Orange clear
Aug 2011
Sep
Oct
Nov
time-gt
Dec
Jan
longitude
Source Matt Wheeler, Centre for Australian
Climate and Weather Research cawcr.gov.au
5
The Madden-Julian oscillation a natural
fluctuation of the tropical climate
Aug 2011
Outgoing longwave radiation 15S-15N Blue
rainy Orange clear
Sep
Oct
Nov
time-gt
Dec
Jan
longitude
Source Matt Wheeler, Centre for Australian
Climate and Weather Research cawcr.gov.au
6
The Madden-Julian oscillation a natural
fluctuation of the tropical climate
Aug 2011
Outgoing longwave radiation 15S-15N Blue
rainy Orange clear
Sep
Oct
Nov
time-gt
Dec
Jan
longitude
Source Matt Wheeler, Centre for Australian
Climate and Weather Research cawcr.gov.au
7
The Madden-Julian oscillation a natural
fluctuation of the tropical climate
Aug 2011
Outgoing longwave radiation 15S-15N Blue
rainy Orange clear
Sep
Oct
Nov
time-gt
Dec
Jan
longitude
Source Matt Wheeler, Centre for Australian
Climate and Weather Research cawcr.gov.au
8
Sandy formed as an active MJO passed through the
north Atlantic basin so it had Pacific ancestry
Blake et al. (2013), NHC report on Sandy
9
The other player was the North Atlantic
Oscillation, natural fluctuation of the
high-latitude jet stream
-
-

-
source ldeo.columbia.edu
10
The other player was the North Atlantic
Oscillation, natural fluctuation of the
high-latitude jet stream
-
-

Jet stream shifted south, cold in eastern US
source ldeo.columbia.edu
11
In mid-October the NAO was heading into strongly
negative territory, conducive to cold-air
outbreaks in eastern US
-

Source NOAA CPC, www.cpc.ncep.noaa.gov
12
And on Oct. 20, 2012, blocking high (typical of
negative NAO) moves into the western north
Atlantic
500 hPa geopotential height
13
And on Oct. 20, 2012, blocking high moves into
the western north Atlantic
500 hPa geopotential height
Jet sream
H
14
Oct. 22, 2012, NHC names Tropical Depression 18,
then a few hours later upgrades to Tropical Storm
Sandy
GOES visible satellite image 10/22 at 15
UTC http//rammb.cira.colostate.edu/
15
Wed. 10/24 landfall in Jamaica (cat 1) Cuba
next day 10/25 (cat 3)
16
Deterministic GFS 6-day forecast for Monday night
10/29, made Wed. 10/24
17
Deterministic ECMWF 6-day forecast for Monday
night 10/29, made Wed. 10/24
18
Interaction with extratropical system, Fujiwhara
effect, landfall
19
Interaction with extratropical system, Fujiwhara
effect, landfall
20
Visible MODIS satellite image from Saturday,
10/28 http//www.wunderground.com/blog/JeffMasters
/
21
Animation from GFS forecast, images by Kyle
Griffin, U. Wisc. PhD student
22
Sandy near landfall was declared post-tropical
note high asymmetry (false color IR)
23
Sandy near landfall was declared post-tropical
note high asymmetry (false color IR)
Because of this, no hurricane warnings were
issued north of NC, causing some confusion
24
NYC storm surge

• Landfall in Jersey at sharp angle moving westward
• Winds are stronger to right of track
• Onshore winds
• Very large wind field
• Worst possible situation

25
Sandy vs. Katrinaclimatecentral.com
26
How rare an event was this?
27
Historic Hurricane landfalls in NY/NJ Figure by
Tim Hall, NASA GISS
1903
Donna (1960)
Sandy
Irene
1893
1938
28
Synthetic track model says NJ landfall at cat 1
at Sandys angle is a 700 yr event. Flood
return time is probably shorter. Not accounting
for climate change.
synthetic storms
1903 vagabond
Sandy
T. Hall and A. H. Sobel, Geophys. Res. Lett., in
press.
29
Storm tide at Battery highest ever recorded by
tide gauge (since 1920) 1821 may have had
larger storm surge
Sandy
Scileppi and Donnelly, 2007, Geochem. Geophys.
Geosys.
30
Climate change influence?
31
Our latest and best models give a mixed
predictions about how the number of hurricanes
will change in a warming climate
Fractional change In number of N. Atlantic
Hurricanes
Different climate models
Zhao et al. (2009) Journal of Climate
32
But we do have some confidence that the most
intense storms will become stronger still
Dan Shaevitz, Suzana Camargo, US CLIVAR
Hurricane Working Group, NOAA/GFDL model
33
But we do have some confidence that the most
intense storms will become stronger still
Dan Shaevitz, Suzana Camargo, US CLIVAR
Hurricane Working Group, NOAA/GFDL model
34
But Sandy was not so intense (cat 1) at landfall
the large surge was due to

• Its huge size, due to extratropical transition
  we know
• nothing about what warming will do to that
• Its track we know little about what warming
  will do to that,
• but

35
Distorted jet, a la Sandy, has been associated
with polar amplified global warming (Francis and
Vavrus 2012)
But climate models give the opposite result (E.
Barnes, L. Polvani, AHS)
36
The really clear and simple link to climate is
via sea level rise.
1 meter
Nicholls and Cazenave, 2010, Science
37
With 1m sea level rise, depending on projections
of TC change, can get increase of 2-20 in flood
frequency. (i.e. 20 means 100-year flood
becomes 5-year)
Lin et al. 2012, Nature Climate Change 2, 462-467.
38
With 1m sea level rise, depending on projections
of TC change, can get increase of 2-20 in flood
frequency. (i.e. 20 means 100-year flood
becomes 5-year)
Lin et al. 2012, Nature Climate Change 2, 462-467.
39
With 1m sea level rise, depending on projections
of TC change, can get increase of 2-20 in flood
frequency. (i.e. 20 means 100-year flood
becomes 5-year)
With this model, 1000-year flood becomes 70-year
Lin et al. 2012, Nature Climate Change 2, 462-467.
40
www.columbia.edu/ahs129/home.html
41
GFS Ensemble run Wednesday 10/24 same model, 20
slightly different initial conditions