Transcript AOSS480_2015_Sandy_Attribution_Presentationx

Hurricane Sandy
and Climate Change
AOSS 480 - 4/20/2015
Bukowski, Frey, Loeffler, Slevin
http://eoimages.gsfc.nasa.gov/images/imagerecords/79000/79553/sandy_goe_2012302_1745_lrg.jpg
Presentation Outline
1) Analysis of Hurricane
Sandy
2) Attribution to Climate
Change
3) Discussion of Forecast
Models
4) Conclusions
From hurricanewarningcenter.com
Analysis of
Hurricane Sandy
Timeline
●
●
Tropical wave left
Western Africa
●
Storm classified
as Tropical
Depression 18,
south of Jamaica
6 hours later:
Named Tropical
Storm Sandy
●
●
Began to turn north,
reached secondary peak
strength of 100 mph
Reclassified as
extratropical
Center made landfall in
New Jersey with 80 mph
winds and 945 mb central
pressure
10/24 10/25
10/18
Center
became ill
defined over
Ohio as storm
rapidly
weakened
10/27
2012
10/11
10/22
Wave enters
Caribbean Sea
and begins to
strengthen
Classified Hurricane
Sandy with 85+ mph
winds, passed over
Jamaica
10/29
Classified Major
Hurricane Sandy
with 115+ mph
winds, made
landfall in Cuba
10/31
Lost, then regained
hurricane status after
passing through the
Bahamas and winds
had doubled in radius
H
Records and Significance
Winds/Pressure
Size
●
●
●
Diameter of Tropical Storm force winds extended
1000 miles near time of landfall in New Jersey
Due in part to partial transition to extratropical
system, then back to tropical, then finally fully
extratropical
Largest Atlantic hurricane since records began in
1988
Rainfall
●
●
●
●
Peak winds of 115 mph at landfall in Cuba
(Category 3 strength)
Secondary peak of 100 mph of eastern US coast
Winds of 80 mph at landfall in New Jersey
Low pressure of 945 mb made it the strongest
storm to strike north of Cape Hatteras, NC since
the 1938 New England Hurricane
Storm Surge
● Maximum of
28 inches of rain in
Jamaica
● Maximum US
rainfall of 13 inches
in Maryland
●
●
NWS Advanced Hydrologic Prediction Service
Impacted water levels from
Florida to Maine
Battery Park (tip of
Manhattan) saw 14 foot
storm tide - 4 feet
higher than previous
record from 1992
Impacts
•
•
•
•
Damage estimated at $50 billion - second only to Hurricane Katrina (2005)
72 direct deaths in US, most due to storm surge
• Deadliest non-southern hurricane since Agnes (1972)
• 87 indirect deaths - most due to extended power outages
650,000 homes damaged, 8.5 million people lost power
Nearly 20,000 flights canceled
http://darkroom.baltimoresun.com/wpcontent/uploads/2012/10/REU-STORMSANDYHURRICAN-12.jpg
http://blogs.agu.org/geospace/files/2014/10/Hurricane_Sandy_New_Jersey_Pier_cropped.jpg
Attribution to
Climate Change
Extreme Events
•
•
•
Non-extreme events can
have extreme impacts
• Occurring
simultaneously with
other events
• Location
Not all extreme events
lead to serious impacts
Sandy as an extreme event
http://en.wikipedia.org/wiki/Hurricane_Sandy
Climate Attribution
•
•
•
•
•
•
Analyze observations and
climate relationships while
experimenting with climate
models for comparison
Separate signal from noise
Attribution important for
decision making
Fingerprinting
Joint attribution
Event attribution
Santer et al., Nature (1996)
Extreme Event Attribution
•
•
•
Extreme events are rare
Probabilistic event
attribution
• Find fraction of risk only
from anthropogenic
drivers
• Event could happen
naturally by chance
Event recurrence interval
• How often an event will
occur
http://www.southwestclimatechange.org
Tropical Cyclones
• Tropical cyclone (TC): rotating,
organized system of clouds and
thunderstorms that originates
over tropical waters
• Historical data on TCs is
unreliable, and link to climate
change uncertain
• Low confidence in observed
increase in TC activity
• Incomplete understanding of
physical mechanisms linking
TCs to climate change
http://sos.noaa.gov/Education/forecast.html
Tropical Cyclone Attribution
• Increase in TC damage
• Difficult to attribute due
to quality of data and
internal variability
• Signal has not emerged
• No individual TC can be
directly attributed to
climate change
• Weather and climate are
not independent
Vecchi and Knutson (2007)
Tropical Cyclone Predictions
•
•
•
•
•
•
Increase in average intensity
Decrease in overall frequency
Increase in frequency of most
intense storms, max wind
speeds, and precipitation rates
Storm track will shift
Sea level rise will increase storm
surge
Population increase in risk-prone
areas
Bender et al. 2010
Sandy Impact Attribution
• 1950 Impact Probability
Occurrence:
• 435 years - New Jersey
• 2330 years - NYC
• 2013 Impact Probability
Occurrence:
• 295 years - New Jersey
• 1570 years - NYC
• Sea level rise will increase
risks
Bulletin of the American Meteorological Society - 2013
Attribution Summary
•
•
•
•
•
•
Not all extreme events have extreme effects, and nonextreme events can have severe impacts
Extreme events are a natural part of climate variability
Climate change is a contributor to extreme events, not a
cause
Climate change alters frequency, intensity, extent, and
duration of extreme events
Use Probabilistic Event Attribution Framework and Event
Recurrence Intervals to quantify anthropogenic forcing
Treat climate and weather together, not separately
Discussion of
Forecast Models
Model Problems
Not this model
problem...
Model Problems
...this model
problem
Discussion of Forecast Models
• In this section we will discuss the
differences in the GFS and the ECMWF
models, as well as each of their
performances during Hurricane Sandy
GFS and ECMWF
• Both are global domain, spectral models
• Different microphysical schemes
• GFS - Simple Cloud Scheme
• ECMWF - Predicted cloud liquid and ice,
rain, snow, and cloud fraction scheme
Model Performances
Model Performance
• The ECMWF forecasted a turn to the
coast two days before the GFS
• Why was it able to do forecast so much
better?
Computing Power
• At the time of Hurricane Sandy, the
ECMWF had superior computing power.
• Process a higher resolution, resolve
smaller phenomenon
GFS Improvements
• Recently in January 2015, the GFS upgraded
their system as a part of the Disaster Relief
Appropriations Act of 2013.
• Authorized $60 billion in disaster relief
• $300 million to NWS
• Of $300 million, $23.7 million was portioned
to improve the American models
GFS Improvements
• The GFS now has superior computing
power for the first time since the early
90s.
• GFS - 2,600 teraflops
• ECMWF - 2,217 teraflops
Is the GFS Problem Solved?
• Shortly after the update, the GFS performed
better than the ECMWF in forecasting the
New England blizzard event.
• Due to a small sample size and a wide variety
of extreme weather events, it is far too early
to say that the GFS is now “better” than the
ECMWF.
ECMWF Website (Before Blizzard)
ECMWF Website (After Blizzard)
Why choose GFS?
• Easy to judge after the event
• One model vs another, not
overwhelming
Conclusions
●
●
●
●
●
Hurricane Sandy was an extreme event
Climate change does not cause extreme events
but contributes to them
As sea level rises, so does the chance for
Sandy-like damage
GFS model failed to predict Sandy’s storm track
Updates to GFS may not be obvious in the near
future
Questions?
References
http://www.nhc.noaa.gov/data/tcr/AL182012_Sandy.pdf
http://www.wunderground.com/blog/JeffMasters/comment.html?entrynum=2283
http://www.iowastatedaily.com/news/article_e9c808dc-22df-11e2-bed4-0019bb2963f4.html
http://www.usatoday.com/story/todayinthesky/2012/11/01/airline-cancellation-tally/1673823/
http://www.wmo.int/pages/prog/arep/press_releases/2006/pdf/iwtc_summary.pdf
http://earthzine.org/2011/04/17/changing-the-media-discussion-on-climate-and-extreme-weather/
http://www.ametsoc.org/2012extremeeventsclimate.pdf
Vecchi, G., & Knutson, T. (2007). On Estimates Of Historical North Atlantic Tropical Cyclone Activity. Journal of Climate, 21, 1-1.
Knutson, T., et al. (2010). Tropical Cyclones And Climate Change. Nature Geoscience, 3, 157-163.
Knutson, T., et al. (2013). Dynamical Downscaling Projections of Twenty-First-Century Atlantic Hurricane Activity: CMIP3 and
CMIP5 Model-Based Scenarios. Journal of Climate, 26, 6591-6617.
Larow, T., L. Stefanova, and C. Seitz, (2014). Dynamical Simulations of North Atlantic Tropical Cyclone Activity Using Observed LowFrequency SST Oscillation Imposed on CMIP5 Model RCP4.5 SST Projections. Journal of Climate, 27, 8055-8069.
Hegerl, G., Von Storch, H., Hasselmann, K., Santer, B., Cubasch, U., & Jones, P. (1996). Detecting Greenhouse-Gas-Induced Climate
Change With An Optimal Fingerprint Method. Journal of Climate, 2281-2306.
Santer, B., et al. (1996). A search for human influences on the thermal structure of the atmosphere. Nature, 382, 39-46.
Rosenzweig, C., et al. (2008). Attributing Physical And Biological Impacts To Anthropogenic Climate Change. Nature, 353-357.
Bindoff, N.L. et al., (2013) Detection and Attribution of Climate Change from Global to Regional: Climate Change 2013: The
Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel
on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.
Nicholls, N., et al. (2013) Changes in climate extremes and their impacts on the natural physical environment: IPCC Special Report
on Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation. Cambridge University
Press, Cambridge, United Kingdom and New York, NY, USA.
Bender, M., Knutson, T., Tuleya, R., Sirutis, J., Vecchi, G., Garner, S., & Held, I. (2010). Modeled Impact of Anthropogenic Warming
on the Frequency of Intense Atlantic Hurricanes. Science, 327, 454-458.