Source: Papastamatiou et al. faculty paper (FIU assignment)
Tags: white shark, great white, Carcharodon carcharias, predatory behaviour, sex differences, Dyer Island, South Africa, hidden Markov model, movement ecology, marine biology
Difficulty: Introductory | Prerequisites: Basic understanding of marine ecosystems, food webs, and predator-prey relationships.
This topic sits at the intersection of behavioural ecology and marine predator biology. The research examines whether individual white sharks (Carcharodon carcharias) show consistent personal preferences in how and where they hunt, and whether those preferences split along sex lines. If you have covered food webs, trophic cascades, or apex predator roles in OCB 1930, this builds directly on that foundation. Understanding how top predators make foraging decisions matters because it shapes the distribution of prey species and, by extension, entire coastal ecosystems.
Researchers tracked 14 white sharks near Dyer Island, South Africa over four years using accelerometer transmitters and hidden Markov models to classify behavioural states. They found that males and females hunt in different areas and at different times of day, suggesting that individual sharks have consistent, sex-linked foraging strategies rather than all behaving the same way.
Hidden Markov model (HMM)
A statistical model that infers an animal's unobservable internal state (e.g. hunting, resting, travelling) from observable data such as movement speed and turning angle. Think of it as a way to read a shark's "mood" from its GPS track.
Behavioural state
A distinct mode of activity, such as area-restricted searching (hunting), transiting (directed travel), or resting. In this study, HMMs classified each segment of a shark's track into one of these states.
Area-restricted searching (ARS)
A movement pattern where an animal makes frequent turns within a small area, indicating active foraging or hunting. In simple terms, the shark is circling and searching rather than swimming in a straight line.
Chumming
The practice of scattering fish parts, blood, or bait into the water to attract sharks. In this study, both the researchers and external dive-boat operators used chumming, which created a confounding variable.
Confounding variable
An outside factor that was not controlled for and may have skewed the results. In simple terms, it is something that muddied the data and makes it harder to draw clean conclusions.
Transmitter / accelerometer tag
An electronic device attached to the shark that records movement data (speed, depth, acceleration) and transmits it to the researchers. Think of it as a Fitbit for sharks.
Dyer Island
A small island off Gansbaai on the southwestern cape of South Africa, surrounded by reef systems and a large Cape fur seal colony, making it a natural hunting ground for white sharks.
Joubertsdam
A large reef system adjacent to Dyer Island where much of the shark tracking in this study was concentrated.
Sex-based behavioural dimorphism
Differences in behaviour between males and females of the same species. Here, male and female white sharks used different parts of the study area and hunted at different times of day.
Study site: Dyer Island and Joubertsdam reef, near Gansbaai, southwestern cape of South Africa
Chosen because of the high density of both white sharks and Cape fur seals (primary prey)
Sample: 14 white sharks tracked over approximately four years
Sizes ranged from 290 to 450 cm total length
Both males and females included
Tagging method: Sharks were attracted to the research vessel using a bait line and chum, then fitted with accelerometer transmitters via a modified tagging pole
Data collection window: 07:00 to 19:00 only, because nighttime observations were too sparse to produce reliable tracks
Analytical tool: Hidden Markov models (HMMs) were applied to movement data to classify each track segment into a behavioural state (e.g. hunting, transiting, resting)
Females spent more time closer to the bay (nearer to the seal colony and shallow reef areas)
Males spent more time directly off Dyer Island, in more open water
Temporal split:
In one area, shark numbers peaked in the morning and dropped through the day
In the second area, the opposite pattern held: numbers peaked in the evening
This temporal pattern varied between males and females
Individual consistency: The data suggest individual sharks have repeatable foraging preferences rather than switching strategies randomly
External chumming by dive boats: Commercial shark-cage diving operators were chumming in one of the study areas, which likely altered the local distribution of sharks and introduced noise into the spatial data
Restricted observation hours: Data outside the 07:00 to 19:00 window were excluded, so nocturnal or crepuscular hunting behaviour was not captured
Small sample size for habitat conclusions: The researchers noted that they did not have enough data to draw firm conclusions about whether habitat type drove the geographic distribution patterns they observed
Understanding sex-based differences in white shark foraging behaviour has direct implications for marine conservation planning. If males and females use different areas at different times, marine protected areas and beach safety measures need to account for that variation rather than treating all sharks as a single uniform risk. This research also informs the debate around shark-cage diving operations: if commercial chumming measurably changes local shark distribution, regulators can use that evidence when setting rules for dive-boat proximity to seal colonies or swimming beaches.
"White sharks are random, opportunistic hunters." This study provides evidence that individual sharks show consistent foraging preferences tied to sex, location, and time of day. Their hunting is more structured than the popular image suggests.
"All white sharks behave the same way." Males and females in this study used different parts of the study area and showed different temporal activity peaks. Treating white sharks as a homogeneous group overlooks meaningful behavioural variation.
"Chumming only attracts sharks temporarily." The external chumming by dive boats was significant enough to act as a confounding variable in the study, implying that commercial chumming can alter local shark distribution in measurable, lasting ways.
"A small sample size means the study is useless." Fourteen sharks over four years is a meaningful dataset in marine telemetry research, where tagging large predators is expensive and logistically difficult. The limitation is real but does not invalidate the findings; it limits the scope of the conclusions.
⚠️ Be able to explain what a hidden Markov model does in plain language and why it was chosen for this study (inferring unobservable behavioural states from movement data).
⚠️ Know the specific sex-based differences: females closer to the bay, males more offshore, and opposite temporal peaks in the two areas.
⚠️ Be ready to identify the two main confounding variables (dive-boat chumming and restricted observation hours) and explain how each could bias the results.
⚠️ Understand why the researchers chose Dyer Island as a study site (co-occurrence of white sharks and seals, reef habitat, accessibility).
⚠️ If asked about methodology, remember the tagging process: bait line and chum to attract sharks to the vessel, then transmitters inserted via modified tagging pole.
True or False: Male and female white sharks in this study used the same areas at the same times of day. (False)
Fill in the blank: The researchers used ______ models to classify shark behavioural states from movement data. (Hidden Markov)
True or False: The study collected data 24 hours a day. (False, only 07:00 to 19:00)
Fill in the blank: The study site was located near ______ Island, off the coast of Gansbaai, South Africa. (Dyer)
True or False: Commercial dive-boat chumming had no effect on the study's data. (False, it was a confounding variable)
Q: Describe the methods used to tag white sharks in this study.
A: The researchers attracted sharks to their vessel using a bait line and chum. Once a shark was close enough, they attached an accelerometer transmitter using a modified tagging pole. The transmitters recorded movement data that could later be analysed for behavioural states.
Q: What were the main sex-based differences in foraging behaviour?
A: Females tended to stay closer to the bay (nearer to seal colonies and reef), while males spent more time in open water directly off Dyer Island. The two sexes also showed different temporal activity patterns, with peak presence in each study area occurring at different times of day.
Q: Why did the researchers choose hidden Markov models for their analysis?
A: HMMs allow researchers to infer an animal's internal behavioural state (hunting, resting, transiting) from externally observable movement data. Since you cannot directly ask a shark what it is doing, the model uses patterns in speed, turning angle, and acceleration to classify each segment of the track into the most likely state.
Q: Identify two limitations of this study and explain how each could affect the results.
A: First, commercial dive-boat chumming in one study area likely attracted additional sharks or altered their natural distribution, introducing noise into the spatial data. Second, data collection was limited to 07:00 to 19:00, so any hunting behaviour outside daylight hours was missed entirely, potentially skewing conclusions about when and where sharks are most active.
Q: Why was Dyer Island chosen as the study site?
A: Dyer Island offered a high natural density of both white sharks and their primary prey (Cape fur seals), combined with accessible reef habitat (Joubertsdam) for tracking. The co-occurrence of predator and prey in a relatively bounded area made it practical for long-term observational research.
This material connects to optimal foraging theory, which predicts that predators should adopt strategies that maximise energy intake relative to effort. The sex-based spatial and temporal differences here may reflect different optimal strategies for males and females given their different energy budgets (females invest more in reproduction).
It also links to movement ecology more broadly. The use of HMMs to classify behavioural states is a technique you will see across marine biology, from sea turtles to seabirds, so understanding it here gives you a transferable analytical framework.
Finally, the confounding effect of dive-boat chumming ties into human impacts on marine ecosystems and the broader debate about ecotourism, where the economic value of shark tourism must be weighed against its potential to alter natural predator behaviour.
White shark, great white shark, Carcharodon carcharias, shark hunting behaviour, shark foraging strategy, sex differences in sharks, sexual dimorphism behaviour, Dyer Island sharks, Gansbaai white sharks, Joubertsdam reef, hidden Markov model ecology, HMM animal behaviour, movement ecology, accelerometer tagging, shark telemetry, Cape fur seal predation, apex predator foraging, confounding variables marine research, shark-cage diving impact, marine predator conservation, OCB 1930, Intro to Marine Biology FIU, Papastamatiou