East Atlantic Tropical Cyclone Track Variability and the U'S' Mainland

1
East Atlantic Tropical Cyclone Track Variability
and the U.S. Mainland

• John Sullivan

2
Motivation

• The majority (and often strongest) of Atlantic
  storms form far from the United States but how
  do they get here and what are the preferred
  paths?
• Are there general environmental patterns that
  lead to a U.S. landfall?
• Could patterns be sub-divided into preferred U.S.
  coastal impact zones?

3
Data Methodology

• NOAA Coastal Services Center
• Best Track Dataset (GIS shapefile format)
• Used Years 1968 2004
• http//hurricane.csc.noaa.gov/hurricanes/download.
  html
• Analysis of tracks and densities created using
  ArcGIS
• manipulated using queries and database joins to
  extract specific parts of the dataset
• raster calculations to normalize densities
• NCEP/NCAR Reanalysis Data
• NOAA-CIRES Climate Diagnostics Center
• Used Daily and Summed Daily Composites
• http//www.cdc.noaa.gov/Composites/Day/
• Data was extracted using CDC website, converted
  using FORTRAN program, and manipulated using
  ArcGIS

4
Atlantic Track Density

• Climatology is known to be a useful baseline
  (used in skill assessment) for tropical systems
• People, corporations and governments would like
  to know of risks well in advance of storms if
  possible

• Only used data after 1968
• more reliable 60s began usage of satellite data
  for tracking storms out at sea (Elsberry 1987)
• reanalysis data availability and means
• strong hurricanes were over-reported in wind
  strength for the years 1944 to 1969 (and possibly
  even before 1944)

East Atlantic Box basically includes all storms
south and east of 20N, 60W
5
Density Variation
129

• density is normalized according to the number of
  storms
• shows the variability among the differing amount
  of storms

101
28
A lot of variation, but a general tendency for
storms to recurve and miss the U.S.
2 preferred paths
6
Across the Decades
Storm totals 70s 25 80s 32 90s 47 00s 29
(through 04)
7
Variability of Track According to Impact Zone

• All storms occurred in July, August, and
  September
• 20N, 60W significant especially for Gulf
  Coast Florida

More southern, often follow up the East Coast as
well
Generally start south in box
Widest swath most variable tracks
Often graze Florida, head for Carolinas
8
Contributions to Track

• Large Scale Environment (TEJ, ridges, troughs,
  strength of trades)
• Attempt to locate a pattern to the subtropical
  ridge and/or troughs for U.S. Impact Zones from
  the East Atlantic
• Analyzed 925mb and 500mb Geopotential Heights
• Daily Composite Mean (average from day storm(s)
  crossed out of East Atlantic box)
• Anomaly (mean - total mean)
• Total Mean (1968-2004)
• Seasonal Variability (alter environment)
• Intraseasonal Variability (alter environment)
• Internal Oscillations (last test)

9
925mb Gulf Coast vs. East Coast
500mb Gulf Coast vs. East Coast
Gulf Coast anomaly north of 20N, 60W extends
further south, low over Hudson Bay region East
Coast higher heights over Midwestern U.S.
Gulf Coast west side of subtropical high more
pointed, broader low over Baja Cali East Coast
subtropical high more north-south on western edge
10
Comparing a Mean Member to the Mean Composite

• 500mb Claudette - Texas
• some similarities in the anomalies nearest the
  storm
• other patterns dont match up well

11
Why not landfall in the U.S.?

• Less intense anomalies over U.S. especially to
  recurve is the natural tendency?

• Generally ridge is weaker, but significance?

12
Noteworthy Points

• Less storms impact Florida as opposed to the
  other coasts, but tracks seem to come from most
  multiple angles, esp. to the east
• Variability across the decades is great
• Able to identify patterns across storm impact
  zones, but unable to significantly satisfy a
  trend for individual storms (deviation)
• Seems pattern usually more favorable for
  recurving takes something special to bring to
  the U.S. mainland
• 20N, 60W significant for U.S.

13
Other Possible Research Paths

• Statistical significance calculation of anomalies
• Data variability before 1968
• Other variables (SST, steering column winds,
  thickness)
• Invoke a radius of storm effect, more data
• Paul Heberts Box

14
References

• Elseberry, R. L., 1987 Tropical Cyclone Motion.
  Chap. 4, A Global View of Tropical Cyclones.
  Office of Naval Research, Arlington, VA 22217,
  91-171.
• Harr, P.A., R.L. Elsberry, 1991 Tropical Cyclone
  Track Characteristics as a Function of
  Large-Scale Circulation Anomalies. Mon. Wea.
  Rev., 119, 1448-1468.
• Hart, R. E., 2006 class notes.
• Kalnay, E. and Coauthors, 1996 The NCEP/NCAR
  Reanalysis 40-year Project. Bull. Amer. Meteor.
  Soc., 77, 437-471.

15
Questions
Thanks to Dr. Hart for knowledge, Clark Evans for
valuable input, and the Fuelberg Lab for putting
up with my antics.