Ocean and Coastal Management 248 (2024) 106965

Contents lists available at ScienceDirect

Ocean and Coastal Management

journal homepage: www.elsevier.com/locate/ocecoaman

Humans influence shark behavior: Evidence from the COVID-19 lockdown

André S. Afonso a, b, c
a
MARE – Marine and Environmental Sciences Centre, ARNET – Aquatic Research Network, Department of Life Sciences, University of Coimbra, Calçada Martim de
Freitas, 3000-456, Coimbra, Portugal
b
Departmento de Pesca e Aquicultura, Universidade Federal Rural de Pernambuco, 52171-900, Recife, PE, Brazil
c
CFE – Centre for Functional Ecology, Department of Life Sciences, University of Coimbra, Calçada Martim de Freitas, 3000-456, Coimbra, Portugal

A R T I C L E I N F O A B S T R A C T

Keywords: Achieving the UN SD Goals requires the conservation of keystone marine species. As top predators, many sharks
Acoustic telemetry play a crucial role in balancing marine ecosystems; yet, they have been experiencing severe population declines.
Anthropause Overfishing is indisputably the main threat to shark populations, but little is known about the impact of non-
Fernando de Noronha
extractive human pressure. This study tests the effect of varied human presence on shark behavior at an
Lemon shark
Marine reserve
insular marine reserve. In 2020, humans were virtually absent from Fernando de Noronha (FEN), Brazil, during a
Tiger shark 211-days lockdown period prompted by the COVID-19 pandemics. A local shark tracking program rendered
nearly 280000 acoustic detections in coastal waters from 2016 through 2021. Lemon sharks, Negaprion brevir­
ostris, showed a 43% increase in detection rate during the lockdown which was ascribed to the concomitant
decrease in human presence. In contrast, the detection rate of tiger sharks, Galeocerdo cuvier, decreased by 67%
during the lockdown, but this trend was likely related to seasonality in tiger shark behavior. After the lockdown,
both species’ detection rates tended to return to previous levels. Further, lemon sharks increased diurnal coastal
residency and made more use of areas less restrictive to human use during the lockdown, whilst tiger sharks
increased nocturnal coastal residency during and after the lockdown period. Contrasting trends might relate with
lemon sharks completing their life cycle in FEN and tiger sharks being migrants arriving to FEN at older ages and
being less exposed to local anthropogenic stimuli across ontogeny. This study demonstrates that non-extractive
human disturbance can induce significant changes in the way marine predators explore crucial habitats for
accomplishing trophic, reproductive, and ontogenetic functions. Ascertaining the ecological impacts of marine-
based human development should thus consider cryptic responses by keystone megafauna to the anthropogenic
landscape produced by an ever-growing coastal population. The designation of human exclusion marine areas
could emerge as the single solution to endow shark populations with suitable habitats for optimizing their
resilience to human pressure.

1. Introduction rebound potential is weak (Worm et al., 2013). Fisheries are undis­
putedly the main threat to shark populations. Overfishing is driving over
As high-level predators, sharks play a key role in balancing marine one third of all shark and ray species towards extinction (Dulvy et al.,
ecosystems because they control the abundance and behavior of a 2021) and little refuge from fisheries is available to these resources, even
myriad of mesopredators and other prey (Heithaus et al., 2012; Wcisel in the high seas (Queiroz et al., 2019). However, non-extractive human
et al., 2015), thus preventing unpredictable cascading effects across activities may also inflict considerable damage to marine wildlife
trophic webs (Casini et al., 2012; Myers et al., 2007). Ensuring the health (Halpern et al., 2008), including sharks. Anthropogenic noise, for
of shark populations is utterly important if the services provided by instance, can impair fish behavior and induce negative physiological
marine ecosystems are to be preserved and the UN Sustainable Devel­ responses (Hawkins et al., 2015; Leduc et al., 2021). Hence, noise
opment Goals are to be met. However, the abundance of many shark pollution is developing to become a most threatening process to marine
species has been collapsing during the past few decades (Davidson et al., wildlife globally (Duarte et al., 2021) due to the continuous expansion of
2016). This is problematic because sharks generally grow slowly and the maritime industry and coastal urbanization. Even wildlife tourism,
exhibit low reproductive outputs (Dulvy and Forrest, 2010), hence their which is typically conducted in preserved or legally-protected areas

E-mail address: [email protected].

https://doi.org/10.1016/j.ocecoaman.2023.106965
Received 21 March 2023; Received in revised form 9 November 2023; Accepted 1 December 2023
Available online 8 December 2023
0964-5691/© 2023 The Author. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
A.S. Afonso Ocean and Coastal Management 248 (2024) 106965

dedicated to the conservation of biodiversity, can generate unwanted behaviors and expanding their typical distributions into more urbanized
detrimental impacts on species behavior, condition and fitness (Burgin areas (Corradini et al., 2021; Gordo et al., 2021; Manenti et al., 2020;
and Hardiman, 2015). As a matter of fact, a global review of the related Wilmers et al., 2021). Such responses remain less conspicuous among
scientific literature found that most of the published research reported at marine fauna, though. Positive effects of the anthropause on reef fish
least one negative effect of human recreation with wildlife (Larson et al., populations (Edward et al., 2021; Lecchini et al., 2021) and on the
2016). The current ecotourism pressure on marine habitats is substan­ aquatic environment (Mallik et al., 2022) have been reported, and it is
tial, with a global estimate of >120 million people enrolling in expectable that many marine taxa benefited from e.g., the drop in
ecosystem-based marine activities and generating 47 billion USD maritime traffic and noise pollution (Jones, 2019). Yet, empirical
annually (Cisneros-Montemayor and Sumaila, 2010), of which ~600, research addressing the influence of the COVID-19 lockdown on marine
000 people enroll in activities specifically with sharks (Cisner­ top predators is scarce, with few published data being available (e.g.,
os-Montemayor et al., 2013). Since the global ecotourism market has Huveneers et al., 2021; Séguigne et al., 2022). To address this gap in
been growing sharply and is expected to duplicate its economic value knowledge and contribute to a better understanding of the impacts of
within a 10-year time frame (Himanshu and Deshmukh, 2021 in non-extractive human disturbance on keystone marine megafauna, this
González et al., 2022), it could also become a threat to marine biodi­ study tested the hypothesis that the mandatory confinement during the
versity if sustainable ecotourism practices are not enforced. COVID pandemics, and the subsequent variation-in human coastal ac­
Wildlife responses to anthropogenic disturbance may far exceed the tivities, produced behavioral shifts in large-bodied sharks at a
individual level and prompt changes to community structure and tourism-focused marine reserve.
ecosystem function (Wilson et al., 2020). Understanding the whole
impact of non-extractive human pressure on shark populations is thus 2. Materials and methods
warranted. Human pressure is a complex factor comprising a multitude
of different anthropogenic stimuli and interferences. Therefore, it is 2.1. Study site
more practically addressed by considering a binomial scenario in which
humans are either present or absent. Such a scenario is not so easily This research was conducted in the Archipelago of Fernando de
achievable, though, since the human population is widespread and Noronha (FEN), Brazil, a most prestigious ecotourism destination in the
virtually no marine area in the globe is unaffected by human influence western South Atlantic Ocean. FEN (3◦ 50′S; 32◦ 24′W) is a small, 26 km2
(Halpern et al., 2008). One main approach to quantify non-extractive equatorial insular system located ~345 km off the northeastern edge of
anthropogenic effects on wildlife relies on spatial comparisons of the South American continent. It comprises one main, 17 km2 island and
several biological indicators between areas subject to distinct human 20 other minor islets aligned in the southwest-northeast direction
presence, such as more and less urbanized sites. For instance, French (Fig. 1) and has been acting as a multiple-use marine reserve since 1988.
et al. (2011) related diminished reproductive rates and population de­ About 70% of the archipelago and insular shelf corresponds to a Na­
clines in California sea lions (Zalophus californianus) with the proximity tional Marine Park (i.e., PARNAMAR) and benefits from strict protection
to urbanized areas. In sharks, variability in nutritional quality has been where any extraction activities are prohibited. PARNAMAR occupies the
ascribed to lower-quality prey being consumed by individuals exposed whole windward, southeastern-facing side (hereafter referred to as the
to urban influence compared with conspecifics from less-impacted areas south side) and a portion of the leeward, northwestern-facing side
(Rangel et al., 2021a). Also, sharks were less likely to occur and (hereafter referred to as the north side) of FEN (Fig. 1). In this area,
exhibited more cautious behaviors at impacted areas close to an urban human presence is widely restricted and maritime traffic is limited to a
center than in more pristine areas (Juhel et al., 2019), although such small number of accredited boats involved in SCUBA diving and sea-
effects might not generalize (Hammerschlag et al., 2022). sighting activities, which operate mostly off the north side of FEN.
Another approach to measure human effects on wildlife relies on The north side of FEN harbors most human settlements and the Port
assessing biological responses to temporal shifts in human presence or facilities and includes an area where human activities such as hook-
activity in the same area. If temporary, the duration of the shift should based fishing, boating, and aquatic sports are allowed, i.e., the Envi­
be considered due to potential inertia in biological adjustments to the ronmental Protection Area (EPA). In contrast, the south side is pristine
new environment (Kristensen et al., 2020), which could inhibit the
emergence of observable adaptations in short-term scenarios. However,
longer-term human processes such as seasonality in recreational ocean
usage or alterations in protection status may induce significant effects on
marine communities. For example, Frisch and Rizzari (2019) inferred
recovery trajectories of previously-exploited reef shark populations by
monitoring species abundance and biomass within a network of no-take
or no-entry marine reserves with different ages. Heterogeneity in marine
reserve age and protection strategy enabled a more robust assessment of
shark rebuilding rates under different scenarios after extractive use was
banned. Also, the implementation of shark provisioning activities for
recreational non-extractive use in the cage-diving industry has been
demonstrated to influence shark behavior by increasing residency at
provisioning sites (Huveneers et al., 2013), with shark behavior tending
to resume baseline levels after the amount of pressure produced by this
industry was restricted (Niella et al., 2023). Temporal shifts in human
activity can provide valuable information about the ecological conse­
quences of anthropogenic interference. On that account, the anthro­
Fig. 1. Study area. Map depicting the location of the Archipelago of Fernando
pause triggered by the global COVID-19 lockdown and temporary
de Noronha (FEN) marine reserve, Brazil (inset), and the arrangement of a 15-
reduction in human activity provided an unprecedented opportunity to
unit stationary acoustic receiver array (yellow circles). The polygonal area
address wildlife responses to varied human presence in the environment enclosed by the solid line corresponds to the more-restricted PARNAMAR area,
(Bates et al., 2020). whereas the remainder of the area corresponds to the less-restricted EPA area.
Biological responses to the lockdown became readily noticeable Acoustic stations are numerically coded to enable the spatial interpretation
among terrestrial fauna, with birds and mammals exhibiting bolder of Fig. S4.

2
A.S. Afonso Ocean and Coastal Management 248 (2024) 106965

and nearly empty of human presence. variation in non-extractive human pressure on shark behavior, tagged
FEN’s population has been growing steadily during the last decade sharks in the acoustic database were selected according to the span of
following a striking increase in tourism development, and in 2019 it their detections. Sharks whose detections did not extend throughout the
comprised around 5000 residents (Afonso et al., 2019) and a total of time window of interest (i.e., from at least ~7 months before the lock­
106130 tourists (ADEFN, 2019). An increase in ocean usage by bathers, down until ~7 months after FEN’s reopening to tourism) were not
water-sporters, and transiting boats has naturally followed the incre­ eligible for this analysis. These included migratory sharks with short
ment in human population. Since FEN’s socioeconomic welfare is utterly detection spans, sharks tagged little time before the lockdown, and
dependent on tourism revenue, there is a most direct linkage between sharks tagged a long time before the lockdown whose batteries were
touristic demand and local investment on infrastructure and labor. This depleted during the lockdown. Acoustic data from the remaining in­
has also resulted in a substantial influx of temporary workers and dividuals (hereafter referred to as eligible sharks) were truncated be­
longer-term employees coming from the mainland to work in the con­ tween 211 days before the beginning (i.e., 22 August 2019) and 211
struction sector or in tourism-related companies, boosting up the num­ days after the end of the lockdown period to homogenize the span of
ber of residents. Yet, the outbreak of the COVID-19 pandemics and the individual tracks across three stages of interest (hereafter referred to as
implementation of precautionary control measures prompted the lockdown stages), i.e., the first period before the lockdown, the second
closure of FEN to tourism. From 20 March through October 16, 2020 (i. period corresponding to the lockdown itself, and the third period after
e., 211 days), the archipelago was empty of tourists and most socio­ FEN reopening to tourism. Because the receivers were retrieved from the
economic activities were suspended. Therefore, FEN’s residents tended water for maintenance on 13 April 2021, the third period comprised
to leave the archipelago to spend their lay-off time in the mainland. A only 178 days-worth of data. Then, the number of daily detections
most diminished population coupled with mandatory lockdown regu­ rendered by eligible sharks was calculated, and differences in daily de­
lations resulted in human presence and footprint being virtually absent tections between the three lockdown stages for each species were
in coastal waters off FEN during the referred period. The archipelago examined with boxplots and tested with generalized linear mixed
reopened to tourism on 17 October 2020, but tourist arrivals were low models (GLMM). The number of detections per unit of time provides a
for several months after the lockdown. Indeed, only 12542 tourists were quantitative proxy for the utilization of coastal waters off FEN (i.e.,
registered from October through the end of 2020, which corresponds to coastal residency) by tagged sharks, hence it is useful for assessing
a 3-fold reduction in the number of tourists registered in previous years behavioral shifts derived from lockdown restrictions. GLMMs were fitted
(ADEFN, 2020). using maximum likelihood estimation via Template Model Builder
provided by the ‘glmmTMB’ R package (Brooks et al., 2017), and
2.2. Tagging and tracking procedure included lockdown stage as a categorical predictor and a random effect
for individual shark to account for intraspecific variability in shark
An ongoing shark tracking program using autonomous acoustic behavior. A negative binomial distribution with a log link function was
telemetry was implemented in FEN in October 2016. Focal species were used to address constraints related to data overdispersion and
the tiger shark, Galeocerdo cuvier, and the lemon shark, Negaprion bre­ right-skewness. Also, a zero-inflation parameter was included to account
virostris. These large-bodied species are locally abundant and act as top for the excess of zeros in the dataset. Model diagnosis was conducted
predators. Sharks were caught with bottom longlines and drumlines with the ‘DHARMa’ R package (Hartig, 2022), which tests for issues in
equipped with baited, 17/0 circle hooks, which were deployed for a model performance including overdispersion, goodness of fit, and
maximum of 2 hours at depths ranging from 3 to 70 m. Upon capture, presence of outliers.
sharks were carefully brought on board and placed on the deck with Additionally, to capture the continuous properties of time during the
their eyes covered. A hose with running seawater was inserted in their study span, a time series of the number of acoustic detections per week
mouth for oxygenation. Sharks were sexed, measured for stretched total rendered by eligible sharks was plotted, and the temporal trends in
length (TL) to the nearest centimeter, and fitted with a coded acoustic detection rates were tested with generalized additive models (GAM).
transmitter (V16; Vemco, Canada), which was surgically implanted into GAMs were developed following a Tweedie distribution with identity
their abdominal cavity. After release, tagged sharks were autonomously link and a thin plate-based smooth parameter for week (i.e., an integer
detected by an array of 15 acoustic receivers (VR2W; Vemco, Canada) ranging between 1 and 91, with the latter being the total number of
previously deployed at strategic nearshore locations (hereafter referred weeks comprised by the study span) using the ‘mgcv’ R package (Wood,
to as acoustic stations) around FEN (Fig. 1). Preliminary tests indicated 2017). The family distribution and link function were selected based on
that the detection range of the acoustic signal was ~600 m. Each time a model diagnostics and the Akaike Information Criterion (AIC). Weekly
tagged shark moved at a shorter distance from a receiver, the receiver rather than daily detection rates were used as a response variable to
would record the code of the tagged shark and the date and time of reduce the number of zeros in the dataset. Furthermore, as so to verify
detection. The acoustic array covered most of the areas surrounding whether potential differences between lockdown stages could be
FEN, thus enabling the assessment of shark presence at or absence from ascribed to species-specific seasonality in coastal habitat use rather than
nearshore waters off the archipelago. The detection rate of tagged sharks to variability in human pressure, the mean number of daily detections
across acoustic stations should therefore reflect their spatial distribution per tagged shark in the whole dataset (i.e., including all sharks since
while swimming in nearshore waters off FEN. Likewise, temporal trends their respective tagging dates) was plotted against day of the year (i.e.,
in detection rates should relate with biological responses to intrinsic (e. an integral ranging between 1 and 366 to account for leap years) for
g., trophic behavior) and extrinsic (e.g., environmental forcing) factors each species. The need to standardize the number of detections per
that modulate coastal habitat use. Acoustic receivers were tied to a number of tagged sharks derived from the fact that sharks were tagged at
moored line which was maintained vertically stretched by an under­ different dates, thus the detection rate reflected a varied number of in­
water float, and they were retrieved at least once per year for data off­ dividuals during the initial period of the shark tracking program.
load and battery refurbishment. All procedures were conducted with the Moreover, significant shifts in the proportion of detections per diel stage
approval of the Ethics Committee on Research with Animals of the (i.e., day and night periods) across the three lockdown stages were
Universidade Federal Rural de Pernambuco (license no. examined with Pearson’s χ2 tests.
#23082.025519/2014). To examine spatially-explicit differences in the distribution of
eligible sharks across lockdown stages, Kernel utilization distribution
2.3. Statistical analysis (UD) maps with 50% and 95% contours estimated by the ad hoc method
over a grid with 100 units were generated for each species with the
To examine the effect of the COVID-19 lockdown and subsequent ‘adehabitatHR’ R package (Calenge, 2006) and plotted with QGIS

3
A.S. Afonso Ocean and Coastal Management 248 (2024) 106965

version 3.26.2 (http://qgis.org). Since seasonality in space use could be

also present, the UDs of the first lockdown stage (i.e., the period prior to
lockdown represented by 211-days’ worth of data) were compared with
the UDs generated by the whole data since shark tagging until lockdown.
Such a comparison enabled the detection of potentially confounding
effects derived from seasonal factors which could obscure the interpre­
tation of spatial variability between the three lockdown stages.
All the analyses were conducted with the R statistical software
version 4.0.2 (R Core Team, 2020), with statistical significance being set
at α = 0.05.

3. Results

From 18 October 2016 through 13 April 2021 (i.e., ~4.5 years), a

total of 22 tiger sharks measuring between 166 and 356 cm (mean =
259; standard deviation = 44.1) in total length (TL), and 16 lemon
sharks measuring between 134 and 278 cm TL (mean = 248; sd = 30.3)
were tagged and tracked off FEN, rendering a total of 202461 and 77320
acoustic detections, respectively. The first detection of tiger sharks
occurred on 26 October 2016 and detections spanned over 1629 days
(Fig. S1). For lemon sharks, the first detection occurred on 25 February
2017 and detections spanned over 1508 days. After filtering the dataset
to exclude sharks which were not tracked during the three lockdown
stages, 4 tiger sharks and 3 lemon sharks turned out to be eligible for
analysis. These sharks were all females ranging in size from 134 to 283
cm TL (Table 1). The number of detections rendered by tiger and lemon
sharks after truncating the data to match the span of lockdown stages
totaled 68278 and 40953, respectively. Species-specific temporal and
spatial analyses are presented below.

3.1. Tiger shark

3.1.1. Temporal dynamics in nearshore habitat use

The detection rate of eligible tiger sharks showed considerable
variation across lockdown stages (Fig. 2a), with the number of daily
detections being highest before the lockdown (mean = 42.4, sd = 40.9, Fig. 2. Shark acoustic detection rates. Boxplots depicting the number of daily
median = 35) and lowest during the lockdown (mean = 13.9, sd = 20.3, detections rendered by eligible a) tiger sharks and b) lemon sharks across the
median = 8.5). This corresponds to a 67% reduction in mean detection three lockdown stages, i.e., before the lockdown, during the lockdown, and
rates from the first to the second lockdown stage. During the post- after the lockdown.
lockdown period, the number of daily detections averaged 29.1 (sd =
32.0, median = 19). The GLMM indicated that the observed differences
Table 2
between the first and the remaining lockdown stages were highly sig­
Zero-inflated generalized linear mixed model (GLMM) outputs for the effect of
nificant (p < 0.001) (Table 2), with diagnostics indicating that the fit
lockdown stage (i.e., before, during, and after the lockdown) on tiger shark and
lemon shark daily detection rates off Fernando de Noronha, with reference to
Table 1 regression coefficients (Coef.), standard errors (S.E.), the value of z-statistics (z),
Tagging details of the sharks selected for the acoustic tracking study in Fernando and p-value (p). The stage before the lockdown is the reference level. Individual
de Noronha, with reference to species, acoustic identification number (ID), sex, sharks were included as random effects.
total length in cm (TL), tagging date, date of first and last acoustic detection, and
Species Fixed effect Coef. S.E. z p
number of detections. For these sharks, the overall number of acoustic detections
totaled 122977, but after trimming the data to match the period of time Tiger shark
considered in the analysis (i.e., since 22 August 2019), the number of detections Intercept 3.96819 0.03878 102.34 <0.001
totaled 109231. Lockdown − 1.03191 0.05063 − 20.38 <0.001
After − 0.34180 0.05376 − 6.36 <0.001
Species ID Sex TL Tagging First Last Detections Lemon shark
date date date Intercept 2.74677 0.56138 4.893 <0.001
Lockdown 0.20219 0.06861 2.947 0.00321
Lemon 28030 F 267 16-08- 16-08- 13-04- 22680
After 0.17980 0.07263 2.476 0.01330
2018 2018 2021
Lemon 28031 F 134 25-08- 18-07- 21-04- 2278
2018 2019 2021
Lemon 19599 F 246 20-07- 21-07- 02-04- 21696 was appropriate and that no violations of model assumptions were met
2019 2019 2021 (Fig. S2). A further examination of the variation in the number of tiger
Tiger 28037 F 278 10-08- 10-08- 13-04- 21271 shark weekly detections continuously throughout the three lockdown
2018 2018 2021
stages allowed to identify a peak in coastal habitat use during the first
Tiger 28029 F 273 18-07- 21-07- 12-04- 16038
2019 2019 2021
lockdown stage (i.e., before lockdown), followed by a period of mini­
Tiger 28033 F 283 18-07- 19-07- 12-04- 19363 mum coastal residency coinciding with the lockdown stage, and an
2019 2019 2021 apparent recovery of detection rates during the third lockdown stage
Tiger 28038 F 223 18-07- 21-07- 11-04- 19651 (Fig. 3a). The developed GAM revealed a high approximate significance
2019 2019 2021
of the smooth term (effective degrees of freedom = 7.817, F = 18.87, p

4
A.S. Afonso Ocean and Coastal Management 248 (2024) 106965

Fig. 4. Seasonality in shark acoustic detection rates. Mean number of de­

tections per tagged shark across Julian days rendered by all a) tiger sharks and
b) lemon sharks tagged since October 2016. The vertical dashed red line depicts
the Julian day corresponding to the onset of lockdown, whilst the vertical
Fig. 3. Time-series of shark acoustic detection rates. Number of weekly de­
dotted red line depicts the Julian day corresponding to the end of lockdown.
tections rendered by eligible a) tiger sharks and b) lemon sharks since the
beginning of the first lockdown stage (i.e., 22 August 2019) until the end of the
last lockdown stage (i.e., 13 April 2021). The shaded smooth line depicts the Kernel utilization distributions developed for the first lockdown stage
effect of time on shark detection rates (with a 95% confidence interval) esti­ and for the whole tracking period prior to lockdown since shark tagging
mated with generalized additive models, with the horizontal orange line (Fig. S5). Hence, the spatial data ascribed to the first lockdown stage
depicting the null effect. The vertical dashed red line depicts the onset of should accurately represent the typical distribution of tiger sharks
lockdown, whilst the vertical dotted red line depicts the end of lockdown. before the onset of lockdown, rather than depicting an artifact of e.g.,
seasonality in space use. Before the lockdown, tiger sharks tended to
< 0.001) and an acceptable fit to the data (adjusted R2 = 0.32, explained make indiscriminate use of nearshore areas all around FEN, but resi­
deviance = 29.7%). GAM diagnostics indicated that model fitting was dency was more intense in the north side of the archipelago, inclusively
adequate (Fig. S3). inside the EPA (Fig. 6a). However, a shift in activity core areas from the
Regarding diel variability in nearshore residency across lockdown north side to the south side of FEN was visible during the lockdown
stages by eligible sharks, tiger sharks showed little differences in coastal (Fig. 6b), but core activity tended to shift back to the north side of FEN
habitat use across the diel cycle before the lockdown, but they tended to after the lockdown (Fig. 6c).
significantly decrease (Pearson’s χ2 = 577.87, p < 0.001) their diurnal
coastal residency during the lockdown and after the lockdown stages
(Fig. 5a). Also, seasonality in the mean daily number of detections per 3.2. Lemon shark
tagged shark (i.e., including all tiger sharks tracked in FEN regardless of
their chronology) was evidenced. Tiger shark coastal habitat use 3.2.1. Temporal dynamics in nearshore habitat use
decreased sharply from January through April and then increased In contrast with tiger sharks, lemon shark daily detections were more
smoothly throughout the remainder of the year until peaking in numerous during the lockdown (mean = 26.8, sd = 32.3, median = 18)
December, being lowest during most of the period matching the months and less numerous before the lockdown (mean = 18.8, sd = 32.4, me­
of lockdown (Fig. 4a). dian = 2), translating into a 43% increase from the first to the second
lockdown stage. After the lockdown, an average of 23.1 (sd = 32.1,
3.1.2. Spatial dynamics in nearshore habitat use median = 11) detections per day was observed (Fig. 2b). The GLMM
Eligible tiger sharks tended to make use of the whole area sur­ indicated that the observed differences between the first and the
rounding FEN and their detections were well represented in all acoustic remaining lockdown stages were significant (p < 0.05) (Table 2). Model
stations. Some stations rendered much more detections than others, diagnostics suggested that the fitting was appropriate, with no violations
though (Fig. S4a). Negligible differences were found between spatial of model assumptions being detected (Fig. S2). An examination of the
number of lemon shark weekly detections on a continuous scale

5
A.S. Afonso Ocean and Coastal Management 248 (2024) 106965

Fig. 5. Diel shifts in shark acoustic detection rates. Proportion of acoustic de­
tections rendered by eligible a) tiger sharks and b) lemon sharks within the
acoustic receiver array during the night (dark gray stacked bars) and the day­
time (light gray stacked bars) across the three lockdown stages, i.e., before the
lockdown, during the lockdown, and after the lockdown. Bar width is directly
proportional to the number of acoustic detections. The horizontal dotted line
depicts the 50%-time proportion.

spanning across the three lockdown stages revealed conspicuous dif­

ferences. Before the lockdown, the number of weekly detections was
highly variable and oscillated between the minimum and maximum
values recorded during the study span, but it tended to increase slightly
through time (Fig. 3b). On the other hand, the number of detections
became much more homogenous during the lockdown stage while
exhibiting a most clear increase. In this stage, the detection rate was
consistently higher than those observed in the other stages and peaked
by the end of lockdown. After FEN reopening to tourism, a sudden
Fig. 6. Tiger shark utilization distribution. Spatially-explicit Kernel utilization
decrease in the number of weekly detections was observed, and near­
distribution (UD) of eligible tiger sharks across the three lockdown stages, i.e.,
shore residency through time resumed a highly oscillatory pattern
a) before the lockdown, b) during the lockdown, and c) after the lockdown. Red
similar to the one depicted before the lockdown (Fig. 3b). A GAM of areas and orange areas depict 50% and 95% UDs, respectively. The polygonal
lemon shark weekly detections through time indicated the smooth term area enclosed by the solid line corresponds to the more-restricted PARNAMAR
to be significant (effective degrees of freedom = 2.049, F = 5.461, p = area, whereas the remainder of the area corresponds to the less-restricted
0.003) but the fit was weak (adjusted R2 = 0.03, explained deviance = EPA area.
2.9%) due to the oscillatory behavior of the data. GAM diagnostics
indicated that model fitting was adequate (Fig. S3). (Fig. 4b). Then, detections increased sinuously from July onwards.
Regarding diel variations in lemon shark nearshore habitat use
across lockdown stages, a reduced diurnal coastal residency before and 3.2.2. Spatial dynamics in nearshore habitat use
after the lockdown was noted, with only 29–30% of the detections In contrast with tiger sharks, most lemon shark detections tended to
occurring during the daytime in these stages (Fig. 5b). Yet, this pro­ occur in some acoustic stations from the north side of FEN, with few
portion significantly increased (Pearson’s χ2 = 527.34, p < 0.001) to detections being recorded in the south side (Fig. S4b). Negligible dif­
40% during the lockdown. As for seasonality in the mean daily number ferences were found between lemon shark spatial Kernel utilization
of detections per tagged shark (i.e., including all lemon sharks tracked in distributions during the first lockdown stage and during the whole
FEN regardless of their chronology), a trend comparable to tiger shark period from shark tagging until the onset of lockdown (Fig. S6), sug­
seasonality was evidenced. However, lemon shark detections peaked gesting that the spatial data ascribed to the first lockdown stage accu­
sharply in March before dropping to consistently minimal levels from rately depicts the typical species distribution prior to lockdown. Lemon
April through June, corresponding to the first half of the lockdown shark core activity occurred in the north side of FEN, although a relevant

6
A.S. Afonso Ocean and Coastal Management 248 (2024) 106965

use of some areas from the south side of the island was also noticed 4. Discussion
before the lockdown (Fig. 7a). However, a conspicuous intensification of
lemon shark residency in the north side, including the EPA, was The results obtained indicate that lemon sharks are influenced by
observed during the lockdown stage (Fig. 7b). The magnitude of such a human pressure derived from the tourism industry, whereas tiger sharks
residency decreased after FEN reopening to tourism, but it still may not react to it or react through other unmeasured responses. The
concentrated in the north side of the archipelago (Fig. 7c). expected seasonal dynamics in coastal residency depicted by GAM
modeling did not match the observed trends in lemon shark detection
rate. Hence, the increase in coastal habitat use exhibited by this species
during the lockdown should be ascribed to the concomitantly sharp
reduction in human disturbance. A spatial shift from more-protected
areas into less-protected areas during the lockdown sustains such
interpretation. Lemon sharks are believed to constitute a local popula­
tion because this species uses FEN as a breeding area (Garla et al., 2009)
and all life stages from neonates through adults are abundant in these
waters. Strong site-fidelity in juvenile lemon sharks has been reported in
other Atlantic insular systems (DeAngelis et al., 2008; Murchie et al.,
2010), with residency at natal sites extending for several years after
birth (Chapman et al., 2009), inclusively in a neighboring Brazilian Atoll
(Wetherbee et al., 2007). It is thus plausible that lemon sharks residing
in FEN develop avoidance behaviors towards the increasingly frequent
anthropogenic stimuli produced by a thriving tourism industry. Such
behaviors could be expressed through e.g., a greater use of offshore
deeper waters around the insular shelf and adjacent ocean waters, where
sharks would be undetectable by the acoustic receiver array. While
remaining attached to FEN, large sharks such as the ones in this study
might regulate the time they spend in acoustically-monitored coastal
habitats according to the human landscape. The significant increase in
diurnal coastal residency during the lockdown further corroborates the
relationship between the dynamics in lemon shark detection rates and
human activities. Indeed, human disturbance may produce diel shifts in
wildlife activity and lead to increased species nocturnality (Gaynor
et al., 2018) as a behavioral response to generally diurnal anthropogenic
stimuli. It is thus expectable that previously disturbed sharks would
adopt more diurnal behaviors while humans were virtually absent from
the coastal environment, as observed.
Sharks induce indirect effects on their prey through fear-mediated
behaviors (Wirsing et al., 2007) and it is likely that they are also sub­
ject to an anthropogenic landscape of fear (Bleicher 2017) such as other
large-bodied predators (Lodberg-Holm et al., 2019). These behaviors
could become an energetic burden due to the inherent trade-off between
risk avoidance and some essential activities such as feeding and repro­
ducing, and their prevalence may potentially impair the fitness of in­
dividuals exposed to an unbalanced stimulation of fear. Wildlife
negative responses to human presence include increased flight and
vigilant behaviors, habitat shifts, abundance declines, physiological
stress, and reduced reproductive success (Larson et al., 2016). The
magnitude of such responses can be rather significant, too. For instance,
mammalian movements in wilderness areas were reported to be up to
three times larger than those in more impacted areas (Tucker et al.,
2018). The juvenile-to-female ratio of the ibex (Capra ibex) in a nature
reserve more than doubled after the reserve was closed to visitors
(Zukerman et al., 2021). The density of fishes at a marine reserve
showed a 2.5-fold increase after ecotourism activities were interrupted
due to the COVID-19 lockdown (González et al., 2022). In the present
study, lemon sharks showed a 43% increase in coastal habitat use while
humans were absent, and this value is likely underestimated because the
lockdown period coincided with the season associated with lowest
nearshore activity. Since this species is listed as vulnerable to extinction
(Carlson et al., 2021) and is legally-protected in Brazilian waters (Bra­
zilian Ministry of the Environment, Portaria nº 445 of 17th December
Fig. 7. Lemon shark utilization distribution. Spatially-explicit Kernel utiliza­
2014; Available at https://www.ibama.gov.br/component/legislacao/?
tion distribution (UD) of eligible lemon sharks across the three lockdown stages,
view=legislacao&legislacao=134521), attention should be given to the
i.e., a) before the lockdown, b) during the lockdown, and c) after the lockdown.
Red areas and orange areas depict 50% and 95% UDs, respectively. The impacts of non-extractive human activities on lemon sharks populations.
polygonal area enclosed by the solid line corresponds to the more-restricted On the other hand, tiger sharks did not seem to exhibit any particular
PARNAMAR area, whereas the remainder of the area corresponds to the less- response to the lockdown. While humans were virtually absent, tiger
restricted EPA area. shark detection rates decreased in syntony with the seasonal trends

7
A.S. Afonso Ocean and Coastal Management 248 (2024) 106965

observed in previous years and nocturnality tended to increase. Further, biased activity in natural conditions, but when provisioning is intro­
tiger sharks increased the use of more protected PARNAMAR areas duced, they become active also during the daytime, when tourism op­
during the lockdown. This species could be less prone to be impacted by erations are typically conducted (Fitzpatrick et al., 2011). In the present
changes in local human pressure than lemon sharks, and the trends study, tiger sharks exhibited small diel differences in coastal habitat use
observed might derive from the intrinsic dynamics in tiger shark regardless of the lockdown stage, but they increased nocturnality with
behavior instead. Tiger sharks are generalist predators with substantial increasing lockdown stage (i.e., they were more diurnal before the
trophic plasticity (Ferreira et al., 2017), but they frequently associate lockdown). Associating this trend directly to a positive taxis towards
with predictable prey sources during food pulses, such as areas with diurnal human influence is cumbersome because nocturnality also
nesting sea turtles (Fitzpatrick et al., 2012; Hammerschlag et al., 2016) increased after FEN reopening to tourism, though. Tiger sharks tagged
or fledgling sea birds (Meyer et al., 2010). Sea turtles abound in FEN off the Brazilian mainland spent more time in shallow waters during the
(Colman et al., 2015; Bellini et al., 2019) and the green turtle nesting night (Afonso and Hazin, 2015), which corroborates species nocturn­
season spans from January through June (Bellini and Sanches, 1996). ality at nearshore habitats. Yet, since movement behaviors are subject to
Such a seasonality in sea turtle nesting behavior could partially explain considerable intraspecific variability (Vaudo et al., 2014) and could be
higher detection rates of tiger sharks before the lockdown. also location-dependent, additional data would be required to clarify
Moreover, unlike lemon sharks, tiger sharks occurring off FEN seem whether human activities at FEN influence tiger shark diel behavior at
to be mostly migrants taking advantage of locally abundant prey all. Other positive taxis exhibited by tiger sharks towards non­
because neonates and early juveniles have not been detected or reported –provisioning human stimuli have been reported. In Hawaii, Meyer et al.
after several years of shark research, even though mating activity was (2009) reported restricted movements by tiger sharks in the vicinity of
noticed in this region (Rangel et al., 2023). The tiger shark is a the Honokohau Harbor potentially indicative of attraction towards
highly-migratory species (Lea et al., 2015), and juveniles likely born on boat-related stimuli. In the Brazilian mainland, the implementation of a
the Brazilian continental shelf have shown to expand their pelagic commercial Port was hypothesized to attract tiger sharks into nearshore
habitat as they grow larger (Afonso and Hazin, 2015) while adopting waters and to trigger a shark hazard outbreak (Hazin et al., 2008). It
wide-ranging oceanic movements which could easily include FEN in should be noted that the tiger shark is listed as one of the most hazardous
their routes (Afonso, 2013). Also, the equatorial Atlantic Ocean seems to shark species either globally (International Shark Attack File, Available
be a most important region for tiger shark transoceanic migrations and at https://www.floridamuseum.ufl.edu/shark-attacks/) as regionally
intercontinental connectivity (Afonso et al., 2017a; Domingo et al., (Afonso et al., 2017b). Hence, any comparison between lemon and tiger
2016), as sustained by the high genetic diversity of the sharks sampled in sharks should consider differences in species personality potentially
FEN indicative of a substantial influx of individuals coming from else­ shaping the decision-making process on the trade-off between foraging
where (Andrade et al., 2021; Carmo et al., 2019). Compared to lemon and risk exposure.
sharks, migratory tiger sharks could become less acquainted with The COVID-19 pandemics and subsequent human confinement
human-derived stimuli across their ontogenetic development and thus measures empowered researchers to directly assess the effects of
respond differently to variations in the anthropic landscape. anthropogenic disturbance on wildlife (Rutz et al., 2020). However, the
Evidence of tiger sharks not responding to human settings as ex­ opportunistic nature of lockdown-derived research generally precluded
pected has been reported. For example, Rangel et al. (2021b) found no proper experimental settings relying on a continuous tracking of wildlife
relevant differences in nutritional quality and metabolic condition be­ behavior to be developed and implemented a priori. In this study,
tween tiger sharks sampled in an urbanized area and in a pristine area, filtering of the acoustic database to crop the data spanning across the
even though another shark species (i.e., the Atlantic nurse shark, Gin­ time window of interest (i.e., the stages before, during, and after the
glymostoma cirratum) sampled in the same locations exhibited clear lockdown) resulted in a small number of sharks being reflected in the
differences in those indicators (Rangel et al., 2021a). In The Bahamas, analysis, raising concerns about the representativeness of the results
where shark-focused diving tourism spans year-round, no effects of achieved. Coincidentally, all these sharks were female besides being
continued provisioning on tiger shark habitat use and diel behavior have unequally distributed across ontogenetic stages. In fact, tiger sharks
been found (Hammerschlag et al., 2017a). Similar results were reported were all below the maturity length of 315 cm TL (Branstetter et al.,
concerning tiger sharks provisioned off French Polynesia (Séguigne 1987), whilst 66% of lemon sharks were mature and 33% were juvenile
et al., 2023). Yet, provisioning proved to largely influence the spatial (size at maturation = 240 cm TL; Brown and Gruber, 1988). Such a
distribution, behavior, and diel activity of other elasmobranchs (Cor­ caveat needs to be considered when interpreting the results herein re­
coran et al., 2013; Barnett et al., 2016; Niella et al., 2023; Séguigne ported because intrinsic factors such as sex and ontogenetic stage may
et al., 2022), although its effects are probably species- and local-specific produce intraspecific variability in shark behavior (Niella et al., 2022).
(Gallagher et al., 2015). This also raises an interesting dichotomy be­ Despite these limitations, a large number of acoustic detections lend
tween shark positive versus negative taxis towards human stimuli. In power to the results obtained and sustain the need to assess whether the
terrestrial mammals, avoidance behaviors frequently lead to increased trends observed in this small dataset would generalize to the whole
species nocturnality due to more intense human activities during the population of the studied species. Since the atypical conditions provided
daytime (Gaynor et al., 2018). Yet, sharks have been usually associated by the lockdown are very unlikely to be replicated in a nearby future,
with more nocturnal behaviors (Campos et al., 2009; Fitzpatrick et al., extracting all possible biological information related with this event to
2011; Papastamatiou et al., 2015) and, even though such trait may not better understand the ecological constraints of the Anthropocene is
be pervasive (Hammerschlag et al., 2017b), they might presumably be highly advisable. This study provides a valuable contribution to bridge
less impacted by diurnal human presence. The observed decrease in gaps in knowledge about how non-extractive human disturbance affects
lemon shark nocturnality during the lockdown suggests however that wildlife, particularly since it addresses vulnerable, understudied taxa
the diurnal period is also important for this species and that human from a region poorly represented in the related scientific literature
processes were somehow hampering coastal habitat use during this diel (Larson et al., 2016).
phase. In the North Atlantic, lemon sharks have been reported to be
more active during the daytime (Bouyoucos et al., 2018), but evidence 5. Conclusion
of increased activity rates during the night is also available (Reyier et al.,
2014), thus diel activity in lemon sharks could be site-dependent. Marine reserves used for tourism may provide safe refugees for
On the other hand, positive taxis towards human stimuli may pro­ species to thrive while promoting pro-conservation values among the
mote more diurnal behaviors in otherwise nocturnal species. For society (Zaradic et al., 2009). However, non-extractive effects of human
example, whitetip reef sharks, Triaenodon obesus, typically exhibit night- activities at these sites could also produce deleterious impacts on

8
A.S. Afonso Ocean and Coastal Management 248 (2024) 106965

wildlife, particularly on more vulnerable, depleted taxa. This study Afonso, A.S., Niella, Y.V., Hazin, F.H.V., 2017b. Inferring trends and linkages between
shark abundance and shark-human interactions for shark hazard mitigation. Mar.
demonstrates that human activities influence lemon shark behavior in
Freshw. Res. 68 (7), 1354–1365. https://doi.org/10.1071/MF16274.
coastal waters off FEN. Such a pressure may potentially increase the Afonso, A.S., Roque, P., Fidelis, L., Veras, L., Conde, A., Maranhão, P., Leandro, S.,
allostatic load of this species and hamper the rebuilding rate of Hazin, F.H.V., 2020. Does lack of knowledge lead to misperceptions? Disentangling
previously-exploited lemon shark populations. Accomplishing the full the factors modulating public knowledge about and perceptions towards sharks.
Front. Mar. Sci. 7, 663. https://doi.org/10.3389/fmars.2020.00663.
rebound potential of these endangered species might depend on Andrade, F.R.S., Afonso, A.S., Hazin, F.H.V., Mendonça, F.F., Torres, R.A., 2021.
excluding humans from shark essential areas. Previous research Population genetics reveals global and regional history of the apex predator
revealed a better performance of no-entry compared to no-take marine Galeocerdo cuvier (Carcharhiniformes) with comments on mitigating shark attacks in
North-eastern Brazil. Mar. Ecol. 42 (2), e12640 https://doi.org/10.1111/
reserves in restoring shark populations (Frisch and Rizzari, 2019), and maec.12640.
this may be the only solution to endow these species with optimal Barnett, A., Payne, N.L., Semmens, J.M., Fitzpatrick, R., 2016. Ecotourism increases the
conditions for enhancing population growth. The impact of human field metabolic rate of whitetip reef sharks. Biol. Conserv. 199, 132–136. https://doi.
org/10.1016/j.biocon.2016.05.009.
disturbance should also be examined in other vulnerable species from Bates, A.E., Primack, R.B., Moraga, P., Duarte, C.M., 2020. COVID-19 pandemic and
FEN such as the Atlantic nurse shark, which reproduces in very shallow, associated lockdown as a "Global Human Confinement Experiment" to investigate
accessible waters (Afonso et al., 2016). More importantly, addressing biodiversity conservation. Biol. Conserv. 248, 108665 https://doi.org/10.1016/j.
biocon.2020.108665.
the human component through dedicated shark literacy programs might Bellini, C., Sanches, T.M., 1996. Reproduction and feeding of marine turtles in the
be crucial to enhance societal perceptions towards sharks within MPA Fernando de Noronha Archipelago, Brazil. Mar. Turt. Newsl. 74, 12–13.
communities and garner support for shark conservation (Afonso et al., Bellini, C., Santos, A.J.B., Patrício, A.R., Bortolon, L.F.W., Godley, B.J., Marcovaldi, M.A.,
Tilley, D., Colman, L.P., 2019. Distribution and growth rates of immature hawksbill
2020).
turtles Eretmochelys imbricata in Fernando de Noronha, Brazil. Endanger. Species
Res. 40, 41–52. https://doi.org/10.3354/esr00979.
Bleicher, S.S., 2017. The landscape of fear conceptual framework: definition and review
Declaration of competing interest of current applications and misuses. PeerJ 5, e3772. https://doi.org/10.7717/
peerj.3772.
The author declares no conflict of interest in the present work. Bouyoucos, I.A., Suski, C.D., Mandelman, J.W., Brooks, E.J., 2018. In situ swimming
behaviors and oxygen consumption rates of juvenile lemon sharks (Negaprion
brevirostris). Environ. Biol. Fish. 101 (5), 761–773. https://doi.org/10.1007/s10641-
Data availability 018-0736-0.
Branstetter, S., Musick, J.A., Colvocoresses, J.A., 1987. A comparison of the age and
growth of the tiger shark, Galeocerdo cuvier, from off Virginia and from the
Data will be made available on request. northwestern Gulf of Mexico. Fish. Bull. 85, 269–279.
Brooks, M.E., Kristensen, K., van Benthem, K.J., Magnusson, A., Berg, C.W., Nielsen, A.,
Acknowledgements Skaug, H.J., Maechler, M., Bolker, B.M., 2017. glmmTMB balances speed and
flexibility among packages for zero-inflated generalized linear mixed modeling. The
R Journal 9 (2), 378–400.
This research is dedicated to the memory of the late Dr. Fábio Hazin, Brown, C.A., Gruber, S.H., 1988. Age assessment of the lemon shark, Negaprion
my mentor and friend who introduced me to the world of shark science brevirostris, using tetracycline validated vertebral centra. Copeia 3, 747–753.
Burgin, S., Hardiman, N., 2015. Effects of non-consumptive wildlife-oriented tourism on
and with whom I would have wished being allowed to collaborate for so marine species and prospects for their sustainable management. J. Environ. Manag.
much more time. Precious field assistance provided by Leonardo Veras, 151, 210–220. https://doi.org/10.1016/j.jenvman.2014.12.018.
Pedro Roque, Leonardo Fidelis, and colleagues from the Universidade Calenge, C., 2006. The package adehabitat for the R software: tool for the analysis of
space and habitat use by animals. Ecol. Model. 197, 1035.
Federal Rural de Pernambuco is deeply acknowledged. Constructive Campos, B.R., Fish, M.A., Jones, G., Riley, R.W., Allen, P.J., Klimley, P.A., Cech, J.J.,
comments provided by anonymous reviewers which helped to improve Kelly, J.T., 2009. Movements of brown smoothhounds, Mustelus henlei, in tomales
the quality of the manuscript are also acknowledged. This study had the bay, California. Environ. Biol. Fish. 85 (1), 3–13. https://doi.org/10.1007/s10641-
009-9462-y.
support of Portuguese funds through Fundação para a Ciência e Tec­
Carlson, J., Charvet, P., Ba, A., Bizzarro, J., Derrick, D., Espinoza, M., Doherty, P.,
nologia, I. P (FCT), under the projects UIDB/04292/2020 and UIDP/ Chartrain, E., Dia, M., Leurs, G.H.L., Metcalfe, K., Pires, J.D., Pacoureau, N.,
04292/2020 (MARE), LA/P/0069/2020 (Associate Laboratory ARNET), Porriños, G., Dulvy, N.K., 2021. Negaprion brevirostris. The IUCN Red List of
Threatened Species. https://doi.org/10.2305/IUCN.UK.2021-1.RLTS.
UIDB/04004/2020 (CFE), and LA/P/0092/2020 (Associate Laboratory
T39380A2915472. T39380A2915472.
TERRA). Carmo, C.B., Ferrette, B.L.S., Camargo, S.M., Roxo, F.F., Coelho, R., Garla, R.C.,
Oliveira, C., Piercy, N.A., Bornatowski, H., Foresti, F., Burgess, G.H., Mendonça, F.F.,
2019. A new map of the tiger shark (Galeocerdo cuvier) genetic population structure
Appendix A. Supplementary data in the western Atlantic Ocean: hypothesis of an equatorial convergence centre.
Aquat. Conserv. Mar. Freshw. Ecosyst. 29 (5), 760–772. https://doi.org/10.1002/
Supplementary data to this article can be found online at https://doi. aqc.3029.
Casini, M., Blenckner, T., Mollmann, C., et al., 2012. Predator transitory spillover induces
org/10.1016/j.ocecoaman.2023.106965.
trophic cascades in ecological sinks. Proc. Natl. Acad. Sci. USA 109, 8185–8189.
https://doi.org/10.1073/pnas.1113286109.
References Chapman, D.D., Babcock, E.A., Gruber, S.H., Dibattista, J.D., Franks, B.R., et al., 2009.
Long-term natal site-fidelity by immature lemon sharks (Negaprion brevirostris) at a
subtropical island. Mol. Ecol. 18, 3500–3507. https://doi.org/10.1111/j.1365-
ADEFN, 2019. Fluxo turístico anual de Fernando de Noronha - 2019. Administração do
294X.2009.04289.x.
Distrito Estadual de Fernando de Noronha. URL: https://www.noronha.pe.gov.br.
Cisneros-Montemayor, A., Sumaila, U.R., 2010. A global estimate of benefits from
ADEFN, 2020. Fluxo turístico anual de Fernando de Noronha - 2020. Administração do
ecosystem-based marine recreation: potential impacts and implications for
Distrito Estadual de Fernando de Noronha. URL: https://www.noronha.pe.gov.br.
management. J. Bioecon. 12, 245–268. https://doi.org/10.1007/s10818-010-9092-
Afonso, A.S., 2013. Bioecology and Movement Patterns of Sharks off Recife, Brazil:
7.
Applications in the Mitigation of Shark Attack Hazard. PhD thesis, University of
Cisneros-Montemayor, A.M., Barnes-Mauthe, M., Al-Abdulrazzak, D., Navarro-Holm, E.,
Algarve, Faro, Portugal. Available at: http://hdl.handle.net/10400.1/2872.
Sumaila, U.R., 2013. Global economic value of shark ecotourism: implications for
Afonso, A.S., Cantareli, C., Levy, R., Veras, L., 2016. Evasive mating behaviour in female
conservation. Oryx 47, 381–388. https://doi.org/10.1017/s0030605312001718.
nurse sharks, Ginglymostoma cirratum, at an insular protected area. Neotrop.
Colman, L.P., Patricio, A.R.C., McGowan, A., Santos, A.J.B., Marcovaldi, M.A., Bellini, C.,
Ichthyol. 14 (4), e160103 https://doi.org/10.1590/1982-0224-20160103.
Godley, B.J., 2015. Long-term growth and survival dynamics of green turtles
Afonso, A.S., Fidelis, L., Roque, P., Dionisio, W., Galindo, R., Veras, L., Hazin, F.H.V.,
(Chelonia mydas) at an isolated tropical archipelago in Brazil. Mar. Biol. 162,
2019. Public support for conservation may decay with residence time in suboptimal
111–122. https://doi.org/10.1007/s00227-014-2585-5.
marine protected areas. Mar. Pol. 108, 103665 https://doi.org/10.1016/j.
Corcoran, M.J., Wetherbee, B.M., Shivji, M.S., Potenski, M.D., Chapman, D.D.,
marpol.2019.103665.
Harvey, G.M., 2013. Supplemental feeding for ecotourism reverses diel activity and
Afonso, A.S., Garla, R., Hazin, F.H.V., 2017a. Tiger sharks can connect equatorial
alters movement patterns and spatial distribution of the southern stingray, Dasyatis
habitats and fisheries across the Atlantic Ocean basin. PLoS One 12 (9), e0184763.
americana. PLoS One 8 (3), e59235.
https://doi.org/10.1371/journal.pone.0184763.
Corradini, A., Peters, W., Pedrotti, L., et al., 2021. Animal movements occurring during
Afonso, A.S., Hazin, F.H.V., 2015. Vertical movement patterns and ontogenetic niche
COVID-19 lockdown were predicted by connectivity models. GECCO 32, e01895.
expansion in the tiger shark, Galeocerdo cuvier. PLoS One 10 (1), e0116720. https://
https://doi.org/10.1016/j.gecco.2021.e01895.
doi.org/10.1371/journal.pone.0116720.

9
A.S. Afonso Ocean and Coastal Management 248 (2024) 106965

Davidson, L.N.K., Krawchuk, M.A., Dulvy, N.K., 2016. Why have global shark and ray Hazin, F.H.V., Burgess, G., Carvalho, F.C., 2008. A shark attack outbreak off Recife,
landings declined: improved management or overfishing? Fish. Fisherman (Syd.) 17, Pernambuco, Brazil: 1992–2006. Bull. Mar. Sci. 82, 199–212.
438–458. https://doi.org/10.1111/faf.12119. Heithaus, M.R., Wirsing, A.J., Dill, L.M., 2012. The ecological importance of intact top-
DeAngelis, B.M., McCandless, C.T., Kohler, N.E., Recksiek, C.W., Skomal, G.B., 2008. predator populations: a synthesis of 15 Years of research in a seagrass ecosystem.
First characterization of shark nursery habitat in the United States Virgin Islands: Mar. Freshw. Res. 63, 1039–1050. https://doi.org/10.1071/MF12024.
evidence of habitat partitioning by two shark species. Mar. Ecol. Prog. Ser. 358, Huveneers, C., Jaine, F.R., Barnett, A., Butcher, P.A., Clarke, T.M., Currey-Randall, L.M.,
257–271. et al., 2021. The power of national acoustic tracking networks to assess the impacts
Domingo, A., Coelho, R., Cortes, E., Garcia-Cortes, B., Mas, F., Mejuto, J., et al., 2016. Is of human activity on marine organisms during the COVID-19 pandemic. Biol.
the tiger shark Galeocerdo cuvier a coastal species? Expanding its distribution range Conserv. 256, 108995.
in the Atlantic Ocean using at-sea observer data. J. Fish. Biol. 88, 1123–1128. Huveneers, C., Rogers, P.J., Beckmann, C., Semmens, J.M., Bruce, B.D., Seuront, L.,
Duarte, C.M., Chapuis, L., Collin, S.P., Costa, D.P., Devassy, R.P., Eguiluz, V.M., Erbe, C., 2013. The effects of cage-diving activities on the fine-scale swimming behaviour and
Gordon, T.A.C., Halpern, B.S., Harding, H.R., Havlik, M.N., Meekan, M., space use of white sharks. Mar. Biol. 160, 2863–2875.
Merchant, N.D., Miksis-Olds, J.L., Parsons, M., Predragovic, M., Radford, A.N., Juhel, J.B., Vigliola, L., Wantiez, L., Letessier, T.B., Meeuwig, J.J., Mouillot, D., 2019.
Radford, C.A., Simpson, S.D., Slabbekoorn, H., Staaterman, E., van Opzeeland, I.C., Isolation and no-entry marine reserves mitigate anthropogenic impacts on grey reef
Winderen, J., Zhang, X., Juanes, F., 2021. The soundscape of the Anthropocene shark behavior. Sci. Rep. 9, 2897. https://doi.org/10.1038/s41598-018-37145-x.
ocean. Science 1979, 371. Jones, N., 2019. The quest for quieter seas. Nature 568, 158–161.
Dulvy, N.K., Forrest, R.E., 2010. Life histories, population dynamics, and extinction risks Kristensen, T.N., Ketola, T., Kronholm, I., 2020. Adaptation to environmental stress at
in chondrichthyans. In: Carrier, J., Musick, J., Heithaus, M. (Eds.), Sharks and Their different timescales. Ann. N. Y. Acad. Sci. 1476 (1), 5–22. https://doi.org/10.1111/
Relatives II: Biodiversity, Adaptive Physiology, and Conservation. CRC Press), nyas.13974.
Florida, USA, pp. 655–696. Larson, C.L., Reed, S.E., Merenlender, A.M., Crooks, K.R., 2016. Effects of recreation on
Dulvy, N.K., Pacoureau, N., Rigby, C.L., Pollom, R.A., Jabado, R.W., Ebert, D.A., animals revealed as widespread through a global systematic review. PLoS One 11
Finucci, B., Pollock, C.M., Cheok, J., Derrick, D.H., et al., 2021. Overfishing drives (12), e0167259. https://doi.org/10.1371/journal.pone.0167259.
over one-third of all sharks and rays toward a global extinction crisis. Curr. Biol. 31, Lea, J., Wetherbee, B., Queiroz, N., et al., 2015. Repeated, long-distance migrations by a
4773–4787.e8. philopatric predator targeting highly contrasting ecosystems. Sci. Rep. 5, 11202
Edward, J.P., Jayanthi, M., Malleshappa, H., Jeyasanta, K.I., Laju, R.L., Patterson, J., https://doi.org/10.1038/srep11202.
Raj, K.D., Mathews, G., Marimuthu, A.S., Grimsditch, G., 2021. COVID-19 lockdown Lecchini, D., Brooker, R.M., Waqalevu, V., Gairin, E., Minier, L., Berthe, C., Besineau, R.,
improved the health of coastal environment and enhanced the population of reeffish. Blay, G., Maueau, T., Sturny, V., Bambridge, T., 2021. Effects of COVID-19 pandemic
Mar. Pollut. Bull. 165, 112124. restrictions on coral reef fishes at eco-tourism sites in Bora-Bora, French Polynesia.
Ferreira, L.C., Thums, M., Heithaus, M.R., Barnett, A., Abrantes, K.G., Holmes, B.J., Mar. Environ. Res. 170, 105451.
Zamora, L.M., Frisch, A.J., Pepperell, J.G., Burkholder, D., Vaudo, J., Nowicki, R., Leduc, A.O.H.C., Nunes, J.A.C.C., Araújo, C.B., Quadros, A.L.S., Barros, F., Oliveira, H.H.
Meeuwig, J., Meekan, M.G., 2017. The trophic role of a large marine predator, the Q., Simões, C.R.M.A., Winandy, G.S.M., Slabbekoorn, H., 2021. Land-based noise
tiger shark Galeocerdo cuvier. Sci. Rep. 7, art7641. pollution impairs reef fish behavior: a case study with a Brazilian carnival. Biol.
Fitzpatrick, R., Abrantes, K.G., Seymour, J., Barnett, A., 2011. Variation in depth of Conserv. 253, 108910.
whitetip reef sharks: does provisioning ecotourism change their behaviour? Coral Lodberg-Holm, H.K., Gelink, H.W., Hertel, A.G., Swenson, E., Domevscik, M.,
Reefs 30 (3), 569–577. Steyaert, S., 2019. A human-induced landscape of fear influences foraging behavior
Fitzpatrick, R., Thums, M., Bell, I., Meekan, M.G., Stevens, J.D., et al., 2012. of brown bears. Basic Appl. Ecol. 35, 18–27.
A comparison of the seasonal movements of tiger sharks and green turtles provides Mallik, A., Chakraborty, P., Bhushan, S., Nayak, B.B., 2022. Impact of COVID-19
insight into their predator-prey relationship. PLoS One 7 (12), e51927. https://doi. lockdown on aquatic environment and fishing community: boon or bane? Mar. Pol.
org/10.1371/journal.pone.0051927. 141, 105088 https://doi.org/10.1016/j.marpol.2022.105088.
French, S.S., González-Suárez, M., Young, J.K., Durham, S., Gerber, L.R., 2011. Human Manenti, R., Mori, E., Di Canio, V., Mercurio, S., Picone, M., Caffi, M., Brambilla, M.,
disturbance influences reproductive success and growth rate in California sea lions Ficetola, G.F., Rubolini, D., 2020. The good, the bad and the ugly of COVID-19
(Zalophus californianus). PLoS One 6 (3), e17686. https://doi.org/10.1371/journal. lockdown effects on wildlife conservation: insights from the first European locked
pone.0017686. down country. Biol. Conserv. 249, 108728.
Frisch, A.J., Rizzari, J.R., 2019. Parks for sharks: human exclusion areas outperform no- Meyer, C.G., Clark, T.B., Papastamatiou, Y.P., Whitney, N.M., Holland, K.N., 2009. Long-
take marine reserves. Front. Ecol. Environ. 17 (3), 145–150. https://doi.org/ term movement patterns of tiger sharks Galeocerdo cuvier in Hawaii. Mar. Ecol. Prog.
10.1002/fee.2003. Ser. 391, 223–235.
Gallagher, A.J., Vianna, G.M.S., Papastamatiou, Y.P., Macdonald, C., Guttridge, T., Meyer, C.G., Papastamatiou, Y.P., Holland, K.N., 2010. A multiple instrument approach
Hammerschlag, N., 2015. Biological effects, conservation potential, and research to quantifying the movement patterns and habitat use of tiger (Galeocerdo cuvier) and
priorities of shark diving tourism. Biol. Conserv. 184, 365–379. Galapagos sharks (Carcharhinus galapagensis) at French Frigate Shoals, Hawaii. Mar.
Garla, R.C., Garcia, J., Veras, L.B., Lopes, N.P., 2009. Fernando de Noronha as an insular Biol. 157 (8), 1857–1868.
nursery area for lemon sharks, Negaprion brevirostris, and nurse sharks, Murchie, K.J., Schwager, E., Cooke, S.J., Danylchuk, A.J., Danylchuk, S.E., et al., 2010.
Ginglymostoma cirratum, in the equatorial western Atlantic Ocean. Marine Spatial ecology of juvenile lemon sharks (Negaprion brevirostris) in tidal creeks and
Biodiversity Records 2, e109. https://doi.org/10.1017/S1755267209000670. coastal waters of Eleuthera, the Bahamas. Environ. Biol. Fish. 89, 95–104.
Gaynor, K.M., Hojnowski, C.E., Carter, N.H., Brashares, J.S., 2018. The influence of Myers, R.A., Baum, J.K., Shepherd, T.D., Powers, S.P., Peterson, C.H., 2007. Cascading
human disturbance on wildlife nocturnality. Science 360, 1232–1235. effects of the loss of apex predatory sharks from a coastal. Ocean 315 (5820),
González, M.O., Reyes-Bonilla, Héctor, Arreola-Alarcon, Isabel Montserrat, Valdovinos 1846–1850. https://doi.org/10.1126/science.1138657.
Uribe, Regina, Olivier, Damien, 2022. COVID-19 lockdown reveals fish density may Niella, Y., Butcher, P., Holmes, B., et al., 2022. Forecasting intraspecific changes in
be much higher in marine reserves. bioRxiv 2022 5 (17), 492376. https://doi.org/ distribution of a wide-ranging marine predator under climate change. Oecologia
10.1101/2022.05.17.492376. 198, 111–124. https://doi.org/10.1007/s00442-021-05075-7.
Gordo, O., Brotons, L., Herrando, S., Gargallo, G., 2021. Rapid behavioural response of Niella, Y., Udyawer, V., Drew, M., Simes, B., Pederson, H., Huveneers, C., 2023. Multi-
urban birds to COVID-19 lockdown. Proc. R. Soc. A B 288, 20202513. https://doi. year effects of wildlife tourism on shark residency and implications for management.
org/10.1098/rspb.2020.2513. Mar. Pol. 147, 105362 https://doi.org/10.1016/j.marpol.2022.105362.
Halpern, B.S., Walbridge, S., Selkoe, K.A., et al., 2008. A global map of human impact on Papastamatiou, Y.P., Watanabe, Y.Y., Bradley, D., Dee, L.E., Weng, K., Lowe, C.G.,
marine ecosystems. Science 319, 948–952. Caselle, J.E., 2015. Drivers of daily routines in an ectothermic marine predator: hunt
Hammerschlag, N., Bell, I., Fitzpatrick, R., Gallagher, A.J., Hawkes, L.A., Meekan, M.G., warm, rest warmer? PLoS One 10 (6), e0127807. https://doi.org/10.1371/journal.
Stevens, J.D., Thums, M., Witt, M.J., Barnett, A., 2016. Behavioral evidence suggests pone.0127807.
facultative scavenging by a marine apex predator during a food pulse. Behav. Ecol. Queiroz, N., Humphries, N.E., Couto, A., et al., 2019. Global spatial risk assessment of
Sociobiol. 70 (10), 1777–1788. sharks under the footprint of fisheries. Nature 572 (7770), 461–466. https://doi.org/
Hammerschlag, N., Gutowsky, L.F.G., Rider, M.J., Roemer, R., Gallagher, A.J., 2022. 10.1038/s41586-019-1444-4.
Urban sharks: residency patterns of marine top predators in relation to a coastal Rangel, B.S., Afonso, A.S., Bettcher, V., Bucair, N., Amndres, N., Veras, L.B., Hazin, F.H.
metropolis. Mar. Ecol. Prog. Ser. 691, 1–17. https://doi.org/10.3354/meps14086. V., Garla, R., 2023. Evidence of mating scars in female tiger sharks (Galeocerdo
Hammerschlag, N., Gutowsky, L., Gallagher, A., Matich, P., Cooke, S., 2017a. Diel cuvier) at the Fernando de Noronha Archipelago, Brazilian Equatorial Atlantic.
habitat use patterns of a marine apex predator tiger shark, Galeocerdo cuvier at a high Environ. Biol. Fish. 106, 107–115. https://doi.org/10.1007/s10641-022-01380-z.
use area exposed to dive tourism. J. Exp. Mar. Biol. Ecol. 495, 24–34. Rangel, B.S., Hammerschlag, N., Moreira, R.G., 2021a. Urban living influences the
Hammerschlag, N., Skubel, R., Calich, H., Nelson, E.R., Shiffman, D.S., Wester, J., nutritional quality of a juvenile shark species. Sci. Total Environ. 776, 146025
Macdonald, C., Cain, S., Jennings, L., Enchaelmaier, A., Gallagher, A.J., 2017b. https://doi.org/10.1016/j.scitotenv.2021.146025.
Nocturnal and crepuscular behavior in elasmobranchs: a review of movement, Rangel, B.S., Moreira, R.G., Neilla, Y., Sulikowski, J., Hammerschlag, N., 2021b.
habitat use, foraging, and reproduction in the dark. Bull. Mar. Sci. 93 (2), 355–374. Metabolic and nutritional condition of juvenile tiger sharks exposed to regional
https://doi.org/10.5343/bms.2016.1046. differences in coastal urbanization. Sci. Total Environ. 780, 146548 https://doi.org/
Hartig, F., 2022. DHARMa: Residual Diagnostics for Hierarchical (Multi-Level/Mixed) 10.1016/j.scitotenv.2021.146548.
Regression Models. R package version 0.4.6. https://CRAN.R-project.org/packa R Core Team, 2020. R: A Language and Environment for Statistical Computing. R
ge=DHARMa. Foundation for Statistical Computing, Vienna, Austria. URL. https://www.R-project.
Hawkins, A.D., Pembroke, A.E., Popper, A.N., 2015. Information gaps in understanding org/.
the effects of noise on fishes and invertebrates. Rev. Fish Biol. Fish. 25, 39–64. Reyier, E.A., Franks, B.R., Chapman, D.D., Scheidt, D.M., Stolen, E.D., et al., 2014.
https://doi.org/10.1007/s11160-014-9369-3. Regional-scale migrations and habitat use of juvenile lemon sharks (Negaprion

10
A.S. Afonso Ocean and Coastal Management 248 (2024) 106965

brevirostris) in the US South Atlantic. PLoS One 9 (2), e88470. https://doi.org/ Wetherbee, B.M., Gruber, S.H., Rosa, R.S., 2007. Movement patterns of juvenile lemon
10.1371/journal.pone.0088470. sharks Negaprion brevirostris within Atol das Rocas, Brazil: a nursery characterized by
Rutz, C., Loretto, M.C., Bates, A.E., Davidson, S.C., Duarte, C.M., Jetz, W., Johnson, M., tidal extremes. Mar. Ecol. Prog. Ser. 343, 283–293.
Kato, A., Kays, R., Mueller, T., Primack, R.B., 2020. COVID-19 lockdown allows Wilmers, C.C., Nisi, A.C., Ranc, N., 2021. COVID-19 suppression of human mobility
researchers to quantify the effects of human activity on wildlife. Nat. Ecol. Evol. 4 releases mountain lions from a landscape of fear. Curr. Biol. 31, 3952–3955. https://
(9), 1156–1159. doi.org/10.1016/j.cub.2021.06.050.
Séguigne, C., Mourier, J., Vignaud, T., Buray, N., Clua, É., 2022. Effects of a COVID-19 Wilson, M.W., Ridlon, A.D., Gaynor, K.M., Gaines, S.D., Stier, A.C., Halpern, B.S., 2020.
lockdown-induced pause and resumption of artificial provisioning on blacktip reef Ecological impacts of human-induced animal behaviour change. Ecol. Lett. 23,
sharks (Carcharhinus melanopterus) and pink whiprays (Pateobatis fai) in French 1522–1536.
Polynesia (East-Pacific). Ethology 128 (2), 119–130. Wirsing, A.J., Heithaus, M.R., Dill, L.M., 2007. Fear factor: do dugongs (Dugong dugon)
Séguigne, C., Bègue, M., Meyer, C., Mourier, J., Clua, É., 2023. Provisioning ecotourism trade food for safety from tiger sharks (Galeocerdo cuvier)? Oecologia 153 (4),
does not increase tiger shark tiger fidelity. Sci. Rep. 13, 7785. https://doi.org/ 1031–1040. https://doi.org/10.1007/s00442-007-0802-3.
10.1038/s41598-023-34446-8. Wood, S., 2017. Generalized Additive Models: an Introduction with R, second ed.
Tucker, M.A., Böhning-Gaese, K., Fagan, W.F., et al., 2018. Moving in the Anthropocene: Chapman and Hall/CRC, p. 496.
global reductions in terrestrial mammalian movements. Science 359 (6374), Worm, B., Davis, B., Kettemer, L., et al., 2013. Global catches, exploitation rates, and
466–469. rebuilding options for sharks. Mar. Pol. 40, 194–204.
Vaudo, J.J., Wetherbee, B.M., Harvey, G., Nemeth, R.S., Aming, C., et al., 2014. Zaradic, P.A., Pergams, O.R.W., Kareiva, P., 2009. The impact of nature experience on
Intraspecific variation in vertical habitat use by tiger sharks (Galeocerdo cuvier) in the willingness to support conservation. PLoS One 4, e7367. https://doi.org/10.1371/
western North Atlantic. Ecol. Evol. 4 (10), 1768–1786. https://doi.org/10.1002/ journal.pone.0007367.
ece3.1053. Zukerman, Y., Sigal, Z., Berger-Tal, O., 2021. COVID-19 restrictions in a nature reserve
Wcisel, M., O’Riain, M.J., de Vos, A., Chivell, W., 2015. The role of refugia in reducing reveal the cost of human presence for the threatened nubian ibex (Capra nubiana).
predation risk for Cape Fur seals by white sharks. Behav. Ecol. Sociobiol. 69, Frontiers in Ecology and Evolution 9. https://doi.org/10.3389/fevo.2021.751515.
127–138. https://doi.org/10.1007/s00265-014-1825-5.

11