Content uploaded by Matthew McPherson
Author content
All content in this area was uploaded by Matthew McPherson on Sep 04, 2019
Content may be subject to copyright.
Timeline of severe red tide events on the West Florida Shelf: insights
from oral histories
M. Karnauskas, M. McPherson, S. Sagarese, A. Rios, M. Jepson, A. Stoltz
and S. Blake
SEDAR61-WP-20
21 June 2019
This information is distributed solely for the purpose of pre-dissemination peer review. It does
not represent and should not be construed to represent any agency determination or policy.
Please cite this document as:
Karnauskas, M., M. McPherson, S. Sagarese, A. Rios, M. Jepson, A. Stoltz and S. Blake. 2019.
Timeline of severe red tide events on the West Florida Shelf: insights from oral histories.
SEDAR61-WP-20. SEDAR, North Charleston, SC. 16 pp.
SEDAR61-RWXX
Timeline of severe red tide events on the West Florida Shelf: insights from oral histories
M. Karnauskas1, M. McPherson1, S. Sagarese1, A. Rios1, M. Jepson2, A. Stoltz3 and S. Blake3
1Southeast Fisheries Science Center
75 Virginia Beach Drive
Miami, FL 33149
2NOAA Fisheries Service Southeast Regional Office
263 13th Avenue South St.
Petersburg, FL 33701
3University of Miami
Rosenstiel School of Marine and Atmospheric Science
4600 Rickenbacker Causeway
Miami, FL 33149
*************
Introduction
During a series of summer 2018 workshops led by the Southeast Fisheries Science Center with
stakeholders on the southwest Florida coast, serious concerns were highlighted regarding the
multifaceted impacts of red tide. In addition to the obvious fish kills and water quality issues,
stakeholders have observed extensive habitat damage related to red tide, and have noted that
recovery of fish populations has been increasingly delayed following recent and frequent red
tides. Red tides are impacting not only the fish populations that commercial and for-hire fishing
businesses are dependent upon, but other aspects of the fishing communities such as aquaculture
activities, private recreational fishing, tourism visitation, local seafood markets, and real estate
values. Through these additive and potentially synergistic effects, red tides can have far-reaching
impacts on coastal communities.
In response to these concerns, an initiative was put into place to systematically explore local
ecological knowledge (LEK) regarding red tides with individual fishermen, using oral history
and participatory mapping approaches. Goals of the LEK were to: 1) document red tide
locations, frequency and severity over time and space, 2) document impressions of how red
tides/blooms develop and their impact on different fish populations and habitats in the short and
long-term, 3) identify possible ecological signals and stakeholder-driven hypotheses of red tide
event occurrence and severity, and 4) document the adaptation strategies fishermen have
employed in the face of red tide events over time and any changes to those strategies. Relevant
information was extracted from each of the oral histories and was quantified to compare the
recent 2017-2018 event to previous events in terms of severity, recovery time, temporal extent
and species killed.
Methods
Oral history and participatory mapping process
We identified key fishing communities along the southern Gulf Coast of Florida that had
historically experienced relatively high commercial landings or for-hire fishing activities as well
as red tide events (Table 1). We reached out to community members and other stakeholders to
identify key informants from these communities to participate in red tide oral history interviews.
Key informants are current or retired commercial and for-hire fishermen with extensive time and
experience fishing in state and federal waters along the Florida Gulf Coast. In the oral history
interviews, these fishermen were asked to discuss the major red tide events that they had
experienced during their fishing careers and how these had affected their fishing activities,
livelihoods and the marine environment. During the interviews, fishermen also discussed the
history of their own fishing practices and fishing communities and how these have changed over
time.
An important subcomponent of the oral histories was a participatory mapping activity in which
the fishermen were asked to draw the spatial extent of red tide events they had witnessed on
nautical charts. They were also asked to describe, based on their best recollection, the biological
and socio-economic impacts of each red tide event they identified on the maps including the
following: 1) the exact year (or approximate dates) in which the event occurred; 2) the impacts
on fish and marine life they witnessed; 3) the duration and severity of the event; 4) the
recuperation time of the affected fisheries and habitat; 5) the ways that these events affected their
fishing practices and livelihoods; 6) survival strategies; 7) health impacts and 8) impacts on
overall community well-being.
Scientists from both the Southeast Fisheries Science Center and the Southeast Regional Office
conducted the interviews. The interviews were conducted by a social scientist working in tandem
with a fisheries biologist or ecologist. The interdisciplinary approach ensured the full breadth of
questions and clarifications were asked to obtain relevant information on both biological and
socio-economic impacts, and helped generate observations about the interconnections between
natural and socio-economic systems. Using two interviewers also allowed for a division of labor,
with one asking questions and the other taking notes. All of the interviews were recorded, and
waivers were obtained from fishermen so that the recordings can eventually be added to the
NMFS Voices from the Fisheries oral history collection (https://www.voices.nmfs.noaa.gov).
Researchers did not identify a specific number of interviews to be conducted in each community
during the planning phase of the project. Rather, interviews in each community continued until
researchers felt each subsequent interview was providing largely redundant information
regarding major red tide events in that area. At the time of writing, interviews were still being
completed in some south Florida communities and plans were being made to conduct interviews
in communities north of Clearwater, Florida and in the Panhandle during the summer of 2019.
Derivation of information from oral history recordings
Information relevant to the severity of the red tide events was extracted from the interviews, via
notes taken during the interview process and, when necessary, going back to the interview
recordings to clarify remarks. The information was put into a spreadsheet format, where each
row represents a specific red tide event in a particular year, by a given individual. When
approximate dates were given to describe the timing of an event, a best estimate was used (e.g.,
“about 50 years ago” was assumed to be the year 1969 for the purposes of plotting and analysis).
References to extended periods or cyclical trends (e.g. “every two to four years,” “getting worse
over the last ten years”) were excluded from the analysis. For each individual event, we made
note of commentaries on: the overall scale of the event, how long the event lasted, the spatial
scale of the event, species affected by the event, species and fishing areas not affected by the
event, and recovery time after the event. Descriptions of temporal aspects of the blooms, in
terms of how long the event lasted, and the recovery of the system after the event, were typically
given in months or years and were standardized to a common unit. Species affected by different
red tide events were typically given in extensive lists which were analyzed for mentions of
“groupers” and/or specific grouper species.
Results
At the time of writing, 42 oral history interviews had been transcribed in communities located on
the southwest Florida coast (Table 1). The 42 interviewees identified and described a combined
112 significant red tide events, 97 of which were tied to specific years (or specific time periods,
in the case of blooms spanning multiple years in duration).
Across interviews, three significant recent red tide event periods were consistently identified by
fishermen: years 2004-2005, 2013-2015, and 2017-2018 (Figure 1). There were only two
mentions of significant events occurring outside these three ranges, in the period 2003 to present
(one mention each in 2010 and 2012). Other consistently identified events included the early
1970s, mid-1980s, and the late 1990s. However these latter events were identified by fewer
numbers of participants, and for the purpose of summary and analysis in the present paper,
events prior to 2003 were grouped together into a single bin. Generally these patterns did not
differ among the home residence of the interviewee, and descriptions of the timing of events
were consistent across the different communities; however, interviewees from Charlotte County
did not describe the 2004-2005 or 2013-2015 events.
Overall event intensity was categorized on a 3-part scale (minor, major, devastating) based on
general descriptions or terms used to describe the events, as specified in Table 2. A distinction
between “devastating” and “major” events was deemed important, given that the specific term
“devastating” was used a total of 17 times, while mentions of a “bad event” also occurred 17
times. For the 2018 event, the vast majority of interviewees (94%) described the event as
“devastating” or “major.” This is in contrast to lower percentages (37% - 57%) of “major” or
“devastating” designations for prior events (Figure 2). These results may be partly driven by the
areas that have been covered in the LEK assessment to date (all south of Clearwater), as the 2014
event is known to have occurred generally north of the Clearwater area with particularly severe
effects in the Middle Grounds. However, the overall severity designations for all bloom events
do not differ based on the county of residence of the interviewee (Figure 3), and there do not
appear to be regional trends in the rankings of severity across time (Figure 4). The 2005 and
2018 red tide events have similar spatial footprints based on the FWRI HAB database, with
many high cell counts stretching from Marco Island up to the Tampa Bay area for a significant
period. This spatial domain has been extensively covered by the interviews accomplished to
date, which should allow for robust comparisons between the 2005 and 2018 events as they
would not biased by the locations of the interviews. Comparisons with the 2013-2015 period
should be made with caution, as the communities north of Clearwater thought to be most heavily
impacted by that event have not yet been interviewed.
Of the 112 events described, 99 had accounts of species-specific fish kills, and there were 33
mentions of grouper fish kills. Most frequently, fishermen used the general family name
“grouper” in their indications of fish species killed; where specific species names were used,
goliath grouper was the most commonly cited (14 events), followed by gag and red grouper (6
mentions each) and black grouper (2). The highest proportion of grouper mentions occurred in
the 2014 event (50% of all species-specific fish kill mentions), followed by the 2005 event
(45%); note however these are based on relatively low sample sizes (Figure 5). Of the 34
species-specific mortality descriptions for the 2018 event, just over one-third mentioned seeing
mortalities of grouper species.
Overall, the most recent 2018 event was perceived to have lasted longer than previous events
(Figure 6). On average, the estimated temporal extent of the 2017-2018 event was 9.5 months,
compared to less than 1 month for the 2013-2015 period and 6.6 months for the 2004-2005
period. Additionally, interviewees in some cases described the recovery time of the ecosystem
following a significant red tide event (Figure 7). Of note, some individuals felt that the
ecosystem had still not recovered following the 2004-2005 and 2013-2015 red tide event periods.
At the time the interviews were conducted (November 2018 to May 2019), the vast majority
(92%) of interviewees felt that the ecosystem had not recovered following the 2017-2018 red tide
event.
Discussion
Red tide severity has previously been quantified and used in the assessment framework (Walter
et al. 2013). This index is currently being updated and will provide insights as to the severity of
the red tide events in different time periods. However, the LEK approach lends additional
insights into the impacts of red tides on fish populations that may not be fully represented in the
satellite data. The satellite-derived index is tuned to Karenia brevis count data compiled by
FWRI, which are generally collected onshore and has limited coverage in offshore areas. Thus,
the satellite index may not be fully representative of offshore blooms; also, the satellite imagery
is of limited use in nearshore waters due to confounding with land-based inputs. Also, the
satellite data and the FWRI count data can only give instantaneous snapshots of the bloom status,
but do not inform the larger ecosystem impacts of the bloom, such as species mortality, habitat
loss, or ecosystem recovery. Finally, there are potential side effects of red tide that may have
significant influence on the ecosystem but that would not be reflected in the satellite data or cell
count data. For example, there is emerging evidence that some years of severe red tide are
associated with hypoxia (Diggers et al. 2016, Kelble pers.comm.), which may in itself contribute
to fish mortality.
Besides the question of red tide severity, the LEK approach can yield additional insights useful
for assessment and management. A major information gap from the stock assessment
perspective is understanding the species and age composition of red tide-induced mortality.
Generally, fishermen conveyed the perception that red tide tide effects on grouper populations
are not discriminatory and that all age classes of the stocks are affected equally. These insights
help to validate the current assumption and methods employed for the stock assessment, in which
episodic mortality is applied equally to all age classes. Additionally, to date, red tide mortality
has been modeled in stock assessments as an immediate effect; i.e., mortality is assumed to occur
in the year of the red tide event with no lagged effects. However, local observations point
toward potential habitat loss and slowing recovery rates of populations subsequent to red tide
events, which has potential implications for assessment and rebuilding plans.
The LEK approach also provides information related to how fishermen shift their effort in
response to red tide events. Fishermen generally reported that in the past, red tides were regular
occurrences in the region and that they were able to fish through and around shorter and/or
patchier events. However, to survive the more recent severe events, fishermen often have to shift
their fishing locations to avoid red tide, frequently having to make costly trips to areas that are
distant from their normal fishing grounds. They may try to target different species, or shift into
charter fishing. Many report having to stop fishing to find alternative employment or quit fishing
all together, and these reports were particularly prevalent for the 2018 event -- a number of
interviewees had very recently gone out of business. The interviews suggest that managers may
be able to increase the resilience of the fishing industry by facilitating access to appropriate
substitute species and adopting other strategies that can help fishermen stay in business during
severe red tide events.
Oral histories rely heavily on memory and thus the accuracy of individual interviews may be
affected by factors like subject recall limitation and recency bias. Studies indicate that requesting
information in chronological sequence can help facilitate informants’ memories of events.
Furthermore, the kinds of events that are likely to endure in memory are those that are highly
emotional at the time, perceived as turning points or relatively unique (Hoffman and Hoffman
1994). Asking our informants to recall red tide events in chronological order and associate their
memories with impacts on their fishing businesses and families provided cues to help them
remember important details. For example, interviewees could easily recall years in which they
had to change business practices, shift industries, or seek alternative employment, in response to
major red tide events or other shocks. Furthermore, using multiple interviews in each location
allowed the researchers to identify commonalities and validate the information provided by
different informants. As this research continues, we will continue to refine our history of red
tides in west Florida by cross referencing oral history information with other sources such as
newspaper articles and historical reports and datasets.
Acknowledgements
We greatly appreciate the time and effort of the many fishermen who participated in the red tide
oral history project. We also thank the large number of community members who assisted in this
research for their help in identifying key informants, spreading the word locally about the
project, and providing working space in which to conduct the interviews.
References
Driggers, W.B., M.D. Campbell, A.J. Debose, K.M. Hannan, M.D. Hendon, T.L. Martin, and
C.C. Nichols. 2016. Environmental conditions and catch rates of predatory fishes associated
with a mass mortality on the West Florida Shelf. Estuarine, Coastal and Shelf Science 168: 40-
49.
Hoffman, A.M. and H.S. Hoffman. 1994. Reliability and Validity in Oral History: The Case for
Memory in Memory and History, Essays on Recalling and Interpreting Experience, Ed. J.
Jeffrey and G. Edwall, pp. 107-130. Landham, MD, University Press of America.
Walter, J.F., M.C. Christman, J. Landsberg, B. Linton, K. Steidinger, R. Stumpf, and J. Tustison.
2013. Satellite derived indices of red tide severity for input for Gulf of Mexico Gag grouper
stock assessment. SEDAR33-DW08. SEDAR, North Charleston, SC. 43 pp.
Table 1. Community of residence of interviewees, and number of interviews per community.
Community
Number of interviews
Boca Grande
3
Cape Haze
1
Chokoloskee
2
Clearwater
2
Cortez
6
Everglades City
2
Fort Myers Beach
5
Goodland
1
Madeira Beach
2
Naples
7
Pine Island
6
Placida
2
Plantation Island
1
St. Petersburg
1
Tarpon Springs
1
Table 2. Descriptors included within the three-level scale categories for severity of red tide
events.
Category
minor
major
devastating
Descriptors
3/10
fairly significant
medium/minor
minor
minor - strong
normal
not bad
patchy
significant
small
small events
bad
extensive
intense
major
miserable
pretty bad
really bad
severe
terrible
very bad
worst
9.5/10
10/10
devastating
Figure 1. Number of mentions of significant red tide events, categorized by the home residence
county of the interviewee.
Figure 2. Categorized severity ratings of recent and past red tide events as given by individual
interviewees.
Figure 3. Categorized severity ratings for all described red tide events documented, grouped by
the county of the home residence of the interviewees.
Figure 4. Individual severity ratings for described red tide events, plotted by the identified year
of the event. Each point represents an individual event described by an interviewee; darker
colors indicate overlying points. Shapes denote the county of residence of the interviewee.
Figure 5. Summary of each red tide event in which interviewees described species that were
affected by red tide, plotted to show the proportion in which grouper species were mentioned.
Events are categorized by the major red tide periods.
Figure 6. Temporal extent as described for individual red tide events, plotted by the identified
year of the event. Each point represents an individual event described by an interviewee; darker
colors indicate overlying points. Shapes denote the county of residence of the interviewee.
Figure 7. System recovery time as described for individual red tide events, plotted by the
identified year of the event. Each point represents an individual event described by an
interviewee; darker colors indicate overlying points. Shapes denote the county of residence of
the interviewee.
