Saturday, August 29, 2015

Why It's Misleading to Label Katrina as a "Category-3" Hurricane

Hurricane Katrina is often referred to as a "category 3" hurricane. I want to make the case that this label is misleading.

Precisely 10 years ago this morning, Hurricane Katrina made landfall in Louisiana and Mississippi as a category-3 hurricane on the Saffir-Simpson Scale. The first landfall along the Northern Gulf Coast occurred just after 6AM CDT near Buras, Louisiana, where maximum sustained winds were estimated at 110 kts (127 MPH) (Knabb et al. 2011). Katrina then tracked across shallow water and wetlands before making a final landfall near the Louisiana/ Mississippi border, with maximum sustained winds around 105 kts (121 MPH) (Knabb et al. 2011).

 Although Katrina is often referred to as a "category-3" hurricane, referring to its maximum sustained wind speed at landfall, it generated a massive storm surge that flooded approximately 80% of metro New Orleans.  

Both of these intensities are classified as category-3 winds on the Saffir-Simpson Scale.

However, Katrina generated a 27.8 ft (8.47 m) storm surge at Pass Christian, Mississippi (Knabb et al. 2011), which is the highest surge in the history of the Western Hemisphere (Needham et al. 2015). The surge in St. Bernard Parish, Louisiana, topped out at 18.7 ft (Knabb et al. 2011), which is the highest surge level in the modern history of Louisiana (Needham et al. 2013). This storm surge was responsible for flooding approximately 80% of metro New Orleans (Kates et al. 2006) and led to a catastrophe in which more than 1800 lives were lost (McTaggart-Cowan et al. 2008). The storm surge directly caused between 600-700 of these deaths (Boyd 2011). Katrina’s price tag totaled more than $100 billion (Blake et al. 2011), making it the most costly natural disaster in U.S. history (Kessler et al. 2006; Baade et al. 2007).

Katrina's rapidly rising flood waters trapped thousands of people on roofs and attics in the metro New Orleans area. These flood waters directly claimed more than 600 fatalities.  

So how did a “category-3” hurricane generate such a terrible catastrophe?

Fortunately, since Hurricane Katrina, much research has investigated the complex nature of storm surge, helping us understand the complexities of this hazard. The bottom line is that many variables contribute to storm surge height, not the maximum sustained wind speed of a hurricane alone.

Hurricane size, or the area of strong winds, jumped to the forefront, as a variable that was not understood very well preceding Hurricane Katrina. The importance of cyclone size became clear as Katrina generated a higher storm surge than Hurricane Camille of 1969, although Camille produced stronger winds. Katrina’s large size played an important role in this process, as the diameter of hurricane force winds extended over 210 miles of open water and tropical storm force winds extended over 460 miles of open water as the sun set on Katrina the night before landfall (Knabb et al. 2011). This tremendously large wind field displaced a massive amount of water that piled up on the Louisiana and Mississippi coasts.

 As the sun set on Hurricane Katrina on Sunday, August 28, 2005, the storm was centered 130 miles S of the Mouth of the Mississippi River. Maximum sustained winds exceeded 160MPH, the diameter of hurricane force winds extended for 210 miles and tropical storm force winds extended across a diameter of 460 miles. Image: NOAA.

Some notable papers in the past 10 years that investigated the role of hurricane size for generating storm surge include Irish et al. (2008); Nielsen (2009); and Dietrich et al. (2011).

Several hurricanes following Katrina also taught us about the importance of hurricane size for generating storm surge. In 2008, Hurricane Ike made landfall along the Texas Coast as a category-2 hurricane, but generated a storm surge of 17.5 ft (5.33 m) in Chambers County, Texas. (Berg 2010). Ike also inundated much of South Louisiana with a storm surge that exceeded 11.5 ft (3.5 m) and extended inland for more than 33 miles (55 km) (Federal Emergency Management Agency 2008). In 2012, Hurricane Isaac made landfall as a category-1 hurricane in South Louisiana, but generated a storm surge that exceeded 14 ft  (4.3 m) (McCallum et al. 2012). Later that year, Hurricane Sandy approached the coastline of New Jersey as a category-1 hurricane, but generated a massive storm surge that inundated New Jersey and New York with seawater, inflicting a price tag of more than $50 billion and contributing to 147 fatalities (Blake et al. 2013).

In 2012, Hurricane Isaac's large size enabled it to inundate Long Beach, Mississippi (above), although the storm was a category-1 hurricane that made landfall south of New Orleans. 
Hurricanes Ike, Isaac and Sandy were not officially classified as “major” hurricanes, because they made landfall with wind speeds less than the category-3 threshold. However, they collectively generated more than $60 billion in damage (Blake et al. 2011), primarily from large, destructive storm surges. All three of these hurricanes were geographically large, enabling them to generate massive surges.

Hurricane forward speed is another variable that influences surge timing and height. This role of this variable was investigated by (Rego and Li 2009). Hurricane Isaac provides a classic example of the increased surge potential from slow-moving hurricanes. Isaac’s forward motion became nearly stationary along the South Louisiana coast, enabling the storm to push water into Southeast Louisiana, Lake Pontchartrain and Mississippi for an extended period of time.

Other variables that generate storm surge include bathymetry and coastal shape. The influence of bathymetry, or offshore water depth, is a bit counter-intuitive because storm surges tend to reach higher levels in areas with shallow bathymetry (Needham and Keim 2011). If you’ve visited coastal Mississippi you know that the water is so shallow that you can walk offshore 150 feet (~45 meters) and the water will not even be as deep as your knee in some places. The influence of harbors and bays are also counterintuitive, as they typically provide a safe haven for boats and ships, but during a hurricane-generated storm surge actually enhance surge heights (Needham and Keim 2011).

 A 25.1-ft (7.65-m) rod does not quite reach the height of tree bark removal on East Ship Island, Mississippi, following Hurricane Katrina. The tree bark removal at this site extended to 8.2 m (26.9 ft), indicating an extraordinarily high storm surge at this location. Katrina's surge in Mississippi was the highest surge level in the modern history of the Western Hemisphere. Photo provided by Hermann Fritz.

Two overlooked variables for storm surge generation are the roles of pre-landfall wind speed and size. Jordan and Clayson (2008) first investigated this topic, and several years later I wrote companion papers with Dr. Barry Keim that provided the first data-driven analysis of this phenomenon for the U.S. Gulf Coast region (Needham and Keim 2014a; Needham and Keim 2014b). These analyses quantified the improved correlation of storm surge heights with pre-landfall wind speeds and size, compared to winds and size at landfall.

The graphic below shows the correlation between storm surge height and maximum sustained winds at landfall and at 3-hour increments preceding landfall. The blue bars represent the correlation using an array, or list, of actual wind speeds. The red bars indicate the improved correlation levels when that array of wind speeds is raised to the optimal exponential power. This graphic is adapted from Needham and Keim (2014a).

Needham and Keim (2014a) found that storm surge heights correlate better with wind speeds 18 hours before landfall than any other time period. Graphic adapted from Needham and Keim (2014a).

This graphic shows that storm surge heights and winds have the poorest correlation at landfall, but that correlation dramatically improves as we use the wind speed before landfall. The relationship is optimal at 18 hours before landfall, when the R-squared value approaches 0.70, if the array of wind speeds is raised to the power of 2.2. This analysis reveals that the relationship between surge heights and wind speed is quite non-linear; we found that doubling the strength of pre-landfall winds increases the surge potential by a factor of around 4.60.

Hurricane Katrina was a very large and intense tropical cyclone 18 hours before its final landfall. The maximum sustained winds at that time exceeded 170 MPH (Elsner and Jagger 2013), which would have placed this storm on the threshold of a category-6 hurricane, if the Saffir-Simpson scale were a continuous scale. In fact, Hurricane Katrina’s pre-landfall wind speed was comparable to Camille’s. Katrina’s wind at this time was only about 11 kts, or 13 mph, weaker than Camille’s (at the same time increment) (Elsner and Jagger 2013), but Katrina’s larger size enabled it to generate a higher surge.

 Map of Hurricane Katrina's hourly position and intensity, as well as the height of storm surge and storm tide observations. Katrina was a large, cat-5 hurricane before striking Louisiana and Mississippi. Storm surge and storm tide heights exceeded 16-ft (4.88 m) along the entire Mississippi Coast and portions of SE Louisiana. 
Map archive at:

So we must be careful when we refer to Katrina as a “cat-3” hurricane.  Katrina’s storm surge caused much more damage and loss of life than its strong winds, and levees are built to protect from water, not wind. To call Katrina a “cat-3” storm uses only one metric- the maximum sustained wind speed at landfall, to classify this cyclone. This label is misleading, as the public is aware that the Saffir-Simpson Scale ranges from 1 to 5 and a “cat-3” sounds like an “average” hurricane to the public. This label may be misused to imply that the flood protection around New Orleans could not withstand a strike from an “average” (cat-3) hurricane, which is very misleading.

In reality, Katrina was a large hurricane, with maximum sustained winds that would have approached category-6, if the Saffir-Simpson Scale were a continuous scale. And those wind speeds peaked around 18 hours before final landfall, the time that correlates best for generating catastrophic surges. Taking into account its large size, as well as the shallow bathymetry and presence of bays (like Bay St. Louis) near it’s final landfall, we should not be surprised that Hurricane Katrina generated the largest storm surge in the modern history of the Western Hemisphere.

Perhaps we should say Katrina struck the Northern Gulf Coast with cat-3 winds and a cat-5 surge? It’s a controversial statement, especially after much effort to disassociate Saffir-Simpson wind speed categories from storm surge heights. Yet, in practice, Katrina is regularly called a "cat-3" hurricane, and on the ground, many people still refer to flood protection as "cat-4" or "cat-3" levees. Calling Katrina a cat-3 (at landfall) wind event and cat-5 surge event, would at least communicate the notion that these storms are more complex than just one number on a scale of 1-5. Either that, or perhaps we should just say it was a hurricane that made landfall as a cat-3, but produced the highest surge in the history of our hemisphere.


Baade, R.A., R. Baumann, and V. Matheson, 2007: Estimating the economic impact of natural and social disasters, with an application to Hurricane Katrina. Urban Studies, 44, 11, 2061-2976.
Berg, R., 2010: Tropical Cyclone Report, Hurricane Ike. The National Hurricane Center, Miami, Florida, United States. Report available on the Web at:
Blake, E.S., C.W. Landsea, and E.J. Gibney, 2011: The Deadliest, Costliest, and Most Intense United States Tropical Cyclones from 1851 to 2010 (And Other Frequently Requested Hurricane Facts). NOAA Technical Memorandum NWS NHC-6. This publication is available on the Web at:
Blake, E.S., T.B. Kimberlain, R.J. Berg, J.P. Cangialosi, and J.L. Beven II, 2013: Tropical Cyclone Report, Hurricane Sandy, (AL182012), 22-29 October 2012. National Hurricane Center, Miami, Florida, USA. Available on the Web at:
Boyd, E., 2011: Fatalities Due to Hurricane Katrina’s Impacts on Louisiana. Dissertation successfully defended April 20, 2011. Louisiana State University Department of Geography and Anthropology, Baton Rouge, Louisiana.
Dietrich, J.C., M. Zijlema, J.J. Westerink, L.H. Holthuijsen, C. Dawson, R. A. Luettich, Jr., R. Jensen, J.M. Smith, G.S. Stelling, and G.W. Stone, 2011: Modeling Hurricane Waves and Storm Surge using Integrally-Coupled, Scalable Computations. Coastal Engineering, 58, 45-65.
Elsner, J.B., and T.H. Jagger, 2013: Hurricane Climatology: A Modern Statistical Guide Using R. Oxford University Press, New York, USA. 430 pp.
Federal Emergency Management Agency, 2008: Hurricane Ike in Texas and Louisiana. Mitigation Assessment Team Report. Available on the web at:
Irish, J.L., D.T. Resio, and J.J. Ratcliff, 2008: The Influence of Storm Size on Hurricane Surge. Journal of Physical Oceanography, 38, 2003-2013.

Jordan, II, M.R., and C.A. Clayson, 2008: A new approach to using wind speed for prediction of tropical cyclone generated storm surge. Geophysical Research Letters, 35, doi: 10.1029/2008GL033564.

Kates, R.W., C.E. Colten, S. Laska, and S.P. Leatherman, 2006: Reconstruction of New Orleans after Hurricane Katrina: A research perspective. Proceedings of the National Academy of Sciences of the United States of America, 103, 14653-14660.

Kessler, R.C., S. Galea, R.T. Jones, and H.A. Parker, 2006: Mental illness and suicidality after Hurricane Katrina. Bulletin of the World Health Organization, 84, 12, 930-939.
Knabb, R.D. J.R. Rhome, and D.P. Brown, 2011: Tropical Cyclone Report, Hurricane Katrina, 23-30 August 2005. Report produced by the National Hurricane Center, Miami, Florida, and published on the Web at:
McCallum, B.E., B.D. McGee, D.R. Kimbrow, M.S. Runner, J.A. Painter, E.R. Frantz, and A.J. Gotvald, 2012: Monitoring Storm Tide Flooding from Hurricane Isaac along the Gulf Coast of the United States, August 2012. Open-File Report 2012-1263. Available on the Web at:

McTaggart-Cowan R., G.D. Deane, L.F. Bosart, C.A. Davis, T.J. Galarneau, Jr., 2008: Climatology of tropical cyclogenesis in the North Atlantic (1948-2004). Monthly Weather Review, 136, 1284-1304.
Needham, H., and B.D. Keim, 2011: Storm Surge: Physical Processes and an Impact Scale. Recent Hurricane Research- Climate, Dynamics, and Societal Impacts. E. Lupo (Ed.). Intech Open Access. Publisher: Croatia.
Needham, H.F., B.D. Keim, D. Sathiaraj, and M. Shafer, 2013: A Global Database of Tropical Storm Surges. EOS, Transactions American Geophysical Union, 94, 24, 213-214.
Needham, H.F., and B.D. Keim, 2014a: Correlating Storm Surge Heights with Tropical Cyclone Winds at and before Landfall. Earth Interactions, 18, 8, 1-26.

Needham, H.F. and B.D. Keim, 2014b: An Empirical Analysis of the Relationship between Tropical Cyclone Size and Storm Surge Heights along the U.S. Gulf Coast. Earth Interactions, 18, 8, 1-15.

Needham, H.F., B.D. Keim, and D. Sathiaraj, 2015: A Review of Tropical Cyclone-Generated Storm Surges: Global Data Sources, Observations and Impacts. Reviews of Geophysics, 53, 2, 545-591.

Nielsen, P., 2009: How storm size matters for surge height. Coastal Engineering, 56, 1002-1004.

Rego, J.L., and C. Li, 2009: On the importance of the forward speed of hurricanes in storm surge forecasting: A numerical study. Geophysical Research Letters, 36, 7.

1 comment: