MARINE MAMMAL SCIENCE, 21(2):327-335 (April 2005)

0 2005 by the Society for Marine Mammalogy

ENVIRONMENTAL CORRELATES OF CETACEAN MASS

STRANDING SITES I N FLORIDA
J. WALKER'
REBEKAH
EDWARD0. KEITH
E. YANKOVSKY
ALEXANDER
Nova Southeastern University,
8000 North Ocean Drive
Dania Beach, Florida 33004, U.S.A.

K. ODELL
DANIEL
Hubbs-SeaWorld Research Institute,
6295 Sea Harbor Drive
Orlando, Florida 32821, U.S.A.

The cause(s) of cetacean strandings remain uncertain, and though numerous

theories abound, relatively few are supported by substantial evidence. Mass
strandings (ie., more than two animals of the same species) can span one or more
days and range over miles of shoreline (Walsh et al. 2001). Recent explanations for
such strandings include bottom topography, coastal configuration, or geomagnetic
topography; meteorological or oceanographic events; extreme conditions in the
environment; auditory trauma; toxicity of pollutants in the environment; and
parasitism (Perrin and Geraci 2002). Contributory factors may also include unusual
tides, sea state, nature of the adjacent seafloor, and meteorological events such as
electrical storms (Warneke 1983). It is important not to make generalizations about
the causes of mass strandings (Ode11 1987). Although many strandings may be due to
a common cause, others may simply be a result of experiencing conditions for which
a group of animals was not prepared. A stranding may also reflect prior events while
the animals were some distance out to sea (Best 1982; Walsh et al. 1991, 2001).
One factor that has not been examined in depth is the effect of seasonal
environmental parameters on the frequency of mass strandings. The effect of wind
on nearshore oceanic circulation is highly variable and also changes between and
within seasons. The effects of wind-forcing on across-shelf ocean circulation have
been noted as an important aspect of the ocean environment and shelf dynamics
(Lentz 2001, Fiedler 2002). Although it is unlikely that wind alone could cause
a mass stranding, the wind does influence across-shelf exchange and water mass
characteristics. This may affect prey assemblages, and in turn influence cetaceans.
Wind-driven transport in the near-surface water layer is deflected by the Coriolis
force at 90" to the right in the northern hemisphere (Ekman transport). Since the
Address all correspondence to: 317 Tall Oak Avenue, Charleston, South Carolina 29407, U.S.A.;
e-mail: walkerr@nova.edu.

327
328 MARINE MAMMAL SCIENCE, VOL. 21, NO. 2,2005

water moves perpendicular to the wind, alongshore wind produces across-shore

Ekman transport (Barber and Smith 1981).
Coastal upwelling occurs in the northern hemisphere when wind blows with the
coast on its left, driving surface water offshore and bringing deeper water to the
surface near the coast. Downwelling is the opposite; it occurs when wind blows
with the coast on its right, thus piling up surface water and deepening the surface
layer at the coast (Brink 1991). The average prevailing winds during the winter
are southward, that is, upwelling-favorable on the west coast of Florida and
downwelling-favorable on the east coast (Fernald and Purdum 1981). In contrast,
during the summer average prevailing winds are northward, which is downwelling-
favorable on the west coast and upwelling-favorable on the east coast.
Cetacean distribution varies with the physical, chemical, and biological
characteristics of water masses (Forcada 2002). Cetaceans respond to changes in
water properties, and thus upwelling and downwelling conditions may influence
their movements. The interface between water masses of different properties,
termed thermal fronts, is associated with high biological productivity (Mann and
Lazier 1991). These fronts are among the abiotic factors that have been shown to
influence the distribution of cetaceans (Polacheck 1987, Evans 2002). Interactions
between bathymetry and frontal structure similarly increase primary and secondary
productivity by affecting circulation patterns, increasing the availability of prey and
generating areas able to sustain cetaceans (Owen 1981, Tynan 1997). The objective
of this study was to investigate seasonal and other abiotic factors that may be
correlated to cetacean mass strandings in Florida.
Data collected by the Southeastern United States Marine Mammal Stranding
Network contained 7 6 mass stranding occurrences between 1977 and 2001 in
Florida. Restranding events were discarded to ensure specific events were not over-
represented in the data set. The total number of mass strandings reported during
each month was determined. In addition, a seasonal index of the number of strand-
ings each month (Spiegel and Stephens 1999) was calculated with December-
February representing winter; March-May representing spring; June-August
representing summer; and September-November representing fall. Figure 1 shows
the location of the 76 reported mass strandings in Florida between 1977 and 2001.
The total number of mass strandings reported during each month in Figure 2 shows
a bimodal trend with more strandings occurring in the winter and summer seasons.
Seasonal indices indicated the same bimodal tendency.
Short-finned pilot whales (Globicepbala macrorbyncbas) mass stranded more
frequently than all other species (32 of 76 events) and, therefore, stranding events
for this species were examined separately to determine if seasonal trends existed.
A two-sample t-test did not show a significant difference between the seasonality of
G. macrarbyncbas mass strandings and the mass stranding of other species ( P > 0.05).
June and December had frequent strandings by species other than G. macrorhyncbzls,
but none of that species, perhaps due to seasonal differences in proximity to shore.
Due to the physical and oceanographic differences of the two coasts of Florida,
data were divided by the month and the location of stranding. This analysis was
separate from the previous analysis and all species were considered. The orientation
 NOTES 329

..._..-.::
,.__..-+
......,._...-

Panhandle Coast

\ "\
_..7 . Keys
' Florida
0
-..-
L.."'

Figure 1. Map of locations of reported cetacean mass stranding events in Florida from
1977 to 2001. Boxed sections represent defined areas used for analysis by stranding location.

of the west coast of Florida is such that the Panhandle area and the southwest coast
may be influenced by different factors, so four categories resulted: east coast, Keys,
southwest coast, and Panhandle coast. Analysis of strandings by location suggested
an influence of wind forcing on stranding patterns. Of the 76 events, 21 occurred on
the east coast, 23 occurred in the Keys, and 32 occurred on the west coast (12 on the
Panhandle and 20 on the southwest coast). Figure 3 shows the number of strandings
for all four coastal divisions. The east coast and Panhandle had more strandings in
the winter, while the southwest coast and Florida Keys had more strandings in the
summer. The seasons during which strandings more likely occurred on the east and
southwest coasts coincided with the occurrence of downwelling-favorable winds as
mentioned above. Again, seasonal indices agreed with the trends.
Analysis of the importance of wind-forcing focused on the east coast of Florida.
When coastlines are straight, alongshore winds create upwelling or downwelling
conditions (Brooks and Mooers 1977). Coastline irregularities complicate water
flow, making shelf responses to wind forcing difficult to determine (Li and
Weisberg 1999). Due to the complexity of the coastline and the difficulty obtaining
in situ historical data, the effects of wind on water circulation in the Panhandle,
west coast, and the Florida Keys were not investigated.
330 MARINE MAMMAL SCIENCE, VOL. 21, NO. 2, 2005

l2 1
10 -

$ 8

-2z
B

6 -
L

n
J
2
4 -

2-

o i I,

Figure 2. Number of cetacean mass strandings in Florida each month from 1977 to
2001.

Hourly wind data from NOAA National Data Buoy Center (http:ii
www.ndbc.noaa.govirmd.shtm1)buoys were examined with respect to 15 of the
mass stranding events occurring on the east coast of Florida between 1977 and
2001. The orientation of the coastline relative to true north was measured for each
stranding point for which wind data were available, and these angles were then used
to determine alongshore and across-shore wind components. Alongshore wind
speeds were smoothed with a low-pass Gaussian filter to remove fluctuations with
periods less than 24 h.
Winds influence the overall movement of the water, which in turn changes local
sea levels. However, the sea level can be affected by both local and remote winds.

east coast Panhandle southwest coast Keys

Figure 3. Number of cetacean mass strandings in Florida from 1977 to 2001, separated
by coastal location of stranding.
 NOTES 33 1

Remote winds generate coastal-trapped, or shelf, waves that spread the ocean's
response in the alongshore direction, propagating southward along the US east coast
(Yankovsky and Garvine 1998). These waves do not produce upwelling fronts at the
surface. Sea level data were, therefore, compared with local winds to determine local
response that was likely associated with formation of upwelling fronts as opposed
to coastal-trapped waves. Hourly sea level data obtained from NOAA's National
Oceanographic Data Center (http:l/www.nodc.noaa.gov) were corrected for at-
mospheric pressure changes, filtered, and examined relative to wind changes.
Only one of the 15 events examined occurred outside of winter, which is the
downwelling season on the east coast, as described above. Although this single
event occurred during the upwelling season, it was preceded by two weeks of
prolonged downwelling-favorable winds. No events occurred during extended
upwelling-favorable winds (extended events considered to be 2 2 d). In all of the
east coast records examined, a change in wind direction from upwelling-favorable
to downwelling-favorable conditions occurred during the week prior to the event.
In addition, simultaneous oscillations in sea level were apparent and support the
hypothesis that changes in wind direction influenced water circulation prior to the
stranding. Figure 4 gives an example of plots analyzed for wind and sea level
conditions during the two weeks prior to each stranding event. An increase in wind
speed and a decrease in sea level indicate an upwelling regime. Alternatively,
a reversal of wind speed towards negative values and an increase in sea level indicate
a downwelling regime. Figure 4 illustrates the two weeks before 3 January 1998,
when seven short-finned pilot whales (Globicephala macrorhyncbus) came ashore. The
shift in wind from upwelling to downwelling can be seen 4 d prior to the stranding
event. Each event was analyzed in this way, and the number of days prior to each
stranding event when the wind changed from upwelling favorable to downwelling
favorable is shown in Figure 5. With the exception of three events, this change
occurred within 4 d prior to the stranding, suggesting an importance of winds in
the week prior to the stranding.
Distances from all stranding sites to the 10 m, 20 m, and 50 m isobaths were
measured. For stranding sites located in the Florida Keys, a straight-line distance
from the site to each isobath was measured, bisecting an island if necessary, to obtain
the shortest distance to a set isobath. To confirm that Florida shores themselves are
not skewed, 76 random points were chosen along the coast and measured to the lO-m,
20-m, and 50-rnisobath to compare with stranding data. A Wilcox signed rank test
was performed using SPSS (SPSS Inc., Chicago, IL) on each set of isobath data.
The mean distance from shore to all isobaths was significantly shorter for
stranding sites than for the random sites (10-m isobath P = 0.005; 20-m isobath
P = 0.002; 50-m isobath P = 0.01 1). In addition, skewness statistics for distance to
stranding sites (10-m isobath 1.596; 20-rn isobath 0.790; 50-rn isobath 0.773) and
bottom slope (10-m isobath 2.314; 20-rn isobath 1.319; 50-m isobath 1.243) are
all positive denoting a tendency towards the shorter distances and steeper slopes.
Strandings in Florida are relatively evenly distributed along the coastline, with
the exception of southeastern Florida and the Big Bend region on the west coast
(Fig. 1). Increased across-shelf water exchange is one physical condition common
332 MARINE MAMMAL SCIENCE, VOL. 21, NO. 2, 2005

V
$ 0

-10
354 356 358 360
Year day
362
- 364 36E
:hange in wind
I
1
b)
,1800-
E

-
a
m

1400 ~

I I
J
- 364 36E

Figure 4. Change in wind speed and sea level during the two weeks prior to a stranding
of seven Globicephala macrorhynchus in Brevard county, FL (27"54.9', 80'28.7') on 3 January
1998. Note the change in wind direction and sea level from upwelling-favorable to
downwelling-favorable conditions 4 d prior to the stranding, which is shown as the last day
on the chart. a) Positive wind speed values indicate upwelling-favorablewinds and negative
values indicate downwelling-favorable winds. b) A rise i n sea level values indicates
downwelling-favorable winds and a fall in values indicates upwelling-favorable winds.

to these two areas. The Gulf Stream passes very close to southeastern Florida,
ventilating the inshore water and dominating the physical oceanography of the area.
Across-shelf transport is increased in the Big Bend area because of recirculation
causing frequent upwelling events (Yang et a/. 1999).
Mass strandings in Florida show a definite seasonal trend with more occurrences
in the winter and summer seasons, and fewer in the spring and fall. There was no
significant difference between species, although seasonal changes in proximity to
the shoreline may account for the timing of a particular species stranding. When
strandings were analyzed relative to the four coastal divisions, strandings were more
likely to occur in seasons with downwelling-favorable conditions. The Florida Keys
did not show as clear a seasonal change, perhaps due to the convoluted nature of the
shoreline that leads to less direct influence by wind-forcing. The tendency for
changes in wind direction from upwelling-favorable to downwelling-favorable in
the week prior to the event is of interest because this change results in frontal
system formation. An upwelling front is formed when cooler upwelled waters meet
warmer surface waters (Shanks eta/. 2000). This front forms close to shore, and then
moves offshore, eventually reaching a quasi-stationary regime while wind continues
(Mann and Lazier 1991). Wind reversal to downwelling-favorable winds allows the
front to move back toward the shore (Shanks et al. 2000).
 4

3
i

0
1 2 3 4 5 6 7
Number of days before stranding wind change occured

Fzgwe 5. Number of days before each mass stranding event analyzed when wind
changed from upwelling to downwelling favorable.

Cetaceans are known to track frontal convergences, and thus might, hypo-
thetically, follow an upwelling front. A change from upwelling-favorable winds
to downwelling-favorable winds would cause the front to move inshore, and if
cetaceans were following the front, this may explain their movement towards
shore. At some point the frontal structure would disappear, perhaps “confusing”the
animals. Because every switch from upwelling to downwelling conditions does not
cause a stranding it is obvious that this alone does not cause the stranding. This
analysis suggests that prevailing winds and their influence on frontal structures near
shore may be important during the week prior to an event.
Bathymetry was also be an important factor. A number of studies have concluded
that cetaceans are most likely to strand on gently sloping beaches (Mazzuca et al.
1999). Our results suggest that cetaceans are more likely to strand on Florida
beaches having a gentle slope with a sudden drop in depth close to shore. Some
authors attribute the correlation between gently sloping beaches and stranding sites
to the possibility that sonar is not reflected properly in these areas. Florida data
suggest that the gently sloping beach may be located near an area of rapid depth
change, and that this change in sea-floor relief may be the important factor.
Most species that mass strand in Florida occupy offshore, deepwater habitats, and
are not coastal inhabitants (Reeves et al. 2002). Cetaceans may occur outside their
usual habitat because of fluctuations in food availability and oceanographic
conditions and they are more likely to be found near regions of high sea-floor relief
(Selzer and Payne 1988). Some species will also move farther inshore following
migrations of their prey (Reilly 1990), or frontal structure movements, rather than
prey directly (Tynan 1997, Bjorge 2001). Analysis of Florida mass strandings found
both bathymetry and wind-induced water circulation to be important factors. Both
influence oceanic frontal structures, and thus suggest the importance of abiotic
factors in the locations of mass stranding events.
334 MARINE MAMMAL SCIENCE, VOL. 21, NO. 2, 2005

ACKNOWLEDGMENTS

We would like to gratefully acknowledge all volunteers and coordinators of the

Southeastern United States Marine Mammal Stranding Network for their invaluable time
and effort collecting the data used for this analysis, and for giving permission for its use in
this study. Alexander Yankovsky was supported by NSF grant OCE-0002885. The authors
would also like to thank two anonymous reviewers for their helpful comments.

LITERATURE
CITED
BARBER, R. T., AND R. L. SMITH.1981. Coastal upwelling ecosystems. Pages 31-68 in A. R.
Longhurst, ed. Analysis of marine ecosystems. Academic Press Limited, London, UK.
BEST,P. B. 1982. Whales: Why do they strand? African Wildlife 36(3):96-101.
BJ0RGE, A. 2001. How persistent are marine mammal habitats in an ocean of variability?:
Habitat use, home range and site fidelity in marine mammals. Pages 63-92 in P. G. H.
Evans and J. A. Raga, eds. Marine mammals: Biology and conservation. Kluwer
Academic, New York, NY.
BRINK,K. H. 1991. Coastal-trapped waves and wind-driven currents over the continental
shelf. Annual Review of Fluid Mechanics 23:389412.
BROOKS,D. A., AND C. N. K. MOOERS.1977. Wind-forced continental shelf waves in the
Florida Current. Journal of Geophysical Research 82:2569-2576.
EVANS,P. G. H. 2002. Habitat pressures. Pages 545-548 in W. E. Perrin, B. Wursig
and J. G. M. Thewissen, eds. Encyclopedia of marine mammals. Academic Press,
San Diego, CA.
FERNALD, E. A., AND E. D. PURDUM, EDS. 1981. Atlas of Florida. University Press of Florida,
Tallahassee, FL.
FIEDLER, P. C. 2002. Ocean environment. Pages 824-830 in W. E. Perrin, B. Wiirsig
and J. G. M. Thewissen, eds. Encyclopedia of marine mammals. Academic Press,
San Diego, CA.
FORCADA, J. 2002. Distribution. Pages 327-333 in W. E. Perrin, B. Wursig and J. G. M.
Thewissen, eds. Encyclopedia of marine mammals. Academic Press, San Diego, CA.
LENTZ,S. J. 2001. The influence of stratification on the wind-driven cross-shelf circulation
over the North Carolina shelf. Journal of Physical Oceanography 3 1:2749-2760.
LI, Z., AND R. H . WEISBERG. 1999. West Florida shelf response to upwelling favorable wind
forcing: Kinematics. Journal of Geophysical Research 104(C6):13,507-13,527.
MANN,K. H., AND J. R. N. LAZIER.1991. Dynamics of marine ecosystems: Biological-
physical interactions in the ocean. Blackwell Scientific Publications, Boston, MA.
MAZZUCA, L., S. ATKINSON, B. KEATING AND E. NITTA.1999. Cetacean mass strandings in
the Hawaiian Archipelago, 1957-1998. Aquatic Mammals 25:105-114.
ODELL,D. K. 1987. The mystery of marine mammal strandings. Cetus 7(2):2-6.
OWEN,R. W. 1981. Fronts and eddies in the sea: Mechanisms, interactions and biological
effects. Pages 197-234 in A. R. Longhurst, ed. Analysis of marine ecosystems.
Academic Press Limited, London, UK.
PERRIN,W. F., AND J. R. GERACI.2002. Stranding. Pages 1192-1197 in W. E. Perrin,
B. Wursig and J. G. M. Thewissen, eds. Encyclopedia of marine mammals. Academic
Press, San Diego, CA.
POLACHECK, T. 1987. Relative abundance, distribution and inter-specific relationship of
cetacean schools in the eastern tropical Pacific. Marine Mammal Science 3:54-77.
REEVES, R. R., B. S. STEWART, P. J. CLAPHAM AND J. A. POWELL. 2002. Guide to marine
mammals of the world. Alfred A. Knopf Inc., New York, NY.
REILLY, S. B. 1990. Seasonal changes in distribution and habitat differences among dolphins
in the eastern tropical Pacific. Marine Ecology Progress Series 66: 1-1 1.
SELZER, L. A., AND P. M. PAYNE.1988. The distribution of white-sided (Lagenorbynchus
acutus) and common dolphins (Delpbinus delphis) us. environmental features of the
 NOTES 335

continental shelf of the northeastern United States. Marine Mammal Science 4:

141-153.
SHANKS, A. L., J. LARGIER AND L. BRINK. 2000. Demonstration of the onshore transport of
larval invertebrates by the shoreward movement o f an upwelling front. Limnology and
Oceanography 45 :230-2 36.
SPIEGEL,M. R., AND L. J. STEPHENS. 1999. Analysis of time series. Pages 434-349 in
Schaum’s outlines statistics. Third Edition. McGraw Hill, New York, NY.
TYNAN,C. T. 1997. Cetacean distribution and oceanographic features near the Kerguelen
Plateau. Geophysical Research Letters 242793-2796.
WALSH, M. T., D. K. ODELL,G. YOUNG,E. D. ASPERAND G. BOSSART. 1991. Mass strandings
of cetaceans. Pages 75-83 in J. E. Reynolds 111 and D. K. Odell, eds. Marine mammal
strandings in the United States. Proceedings of the Second Marine Mammal Stranding
Workshop, Miami, Florida, December 3-5, 1987. NOAA Technical Report NMFS 98.
U S . Department of Commerce, NOAA, NMFS, Silver Spring, MD.
WALSH, M. T., R. Y. EWING,D. K. ODELLAND G. D. BOSSART.2001. Mass strandings
of cetaceans. Pages 83-96 in L. A. Dierauf and F. M. D. Gulland, eds. CRC hand-
book of marine mammal medicine: Health, disease, and rehabilitation. CRC Press,
Boca Raton, FL.
WARNEKE, R. 1983. Whale stranding-accident or design? Australian Natural History
21A3-47.
YANG,H., R. H. WEISBERG, P. P. NIILER,W. STURGES AND W. JOHNSON. 1999. Lagrangian
circulation and forbidden zone on the West Florida Shelf. Continental Shelf Research
19:1221-1245.
YANKOVSKY, A. E., AND R. W. GARVINE.1998. Subinertial dynamics on the inner
New Jersey shelf during the upwelling season. Journal of Physical Oceanography 28:
2444-245 8.
Received: 4 April 2004
Accepted: 19 October 2004