Decreasing Trend of Great White Shark Carcharodon carcharias Records in the Mediterranean: A Significant Population Loss or Shifts in Migration Patterns?

Abstract

The Mediterranean subpopulation of great white sharks (Carcharodon carcharias) is elusive and likely in decline, though long-term trends remain uncertain due to opportunistic record-keeping, misidentifications, and changing observation effort. We investigated whether spatial changes in shark occurrences during the 21st century correspond with shifts in Atlantic bluefin tuna (Thunnus thynnus) distribution and habitat conditions. We compiled geographically validated sighting and capture records from 1900 onward, mapped 20th- and 21st-century hotspots, and overlaid these with bluefin tuna potential habitat and long-term sea surface temperature (SST) data. Results reveal a clear redistribution of great white shark hotspots: historic coastal focal areas (e.g., Balearic Islands, Maltese Islands, eastern Adriatic, Sea of Marmara) have diminished or disappeared, while offshore zones (southern Strait of Sicily–Gulf of Gabes) and the Aegean coast of Turkey have emerged as contemporary hotspots. These patterns appear to align closely with shifts in tuna feeding grounds and cooler SST (<18 °C). We highlight limitations in using opportunistic and citizen-reported data due to detection biases and misidentifications, underscoring the need for models that correct detectability. Our findings are consistent with the hypothesis of a link between predator distribution, prey dynamics, and changing ocean conditions, and point toward targeted strategies for future conservation and monitoring of this apex predator in a warming Mediterranean.

1. Introduction

The great white shark Carcharodon carcharias (Linnaeus, 1758) is a cosmopolitan species with one of the broadest habitat and geographic ranges of any fish species tolerating temperatures from 5 to 25 °C and is known for its long-distance seasonal migrations across oceans [1,2]. Being the world’s largest carnivorous fish and broadly distributed in temperate and tropical waters with a high media profile that attracted significant public attention [3] it is not a surprise that the great white shark (GWS) is a species with a long historical track record in the various regions of the Mediterranean Sea. However, despite receiving considerable attention, it is still considered as an elusive species in the Mediterranean due to its low population density and the absence of verified conventional aggregation sites. Most knowledge of the species in the region comes from occurrence records published for different areas [4,5,6,7,8,9,10,11]. Unfortunately, these records provide only limited information. Moreover, the origin and credibility of many records, particularly historical ones, are questionable. Numerous sightings, both historical and contemporary, have been reported without reliable evidence (e.g., photographs or videos), or the available documentation was inconclusive, preventing proper species identification. Nevertheless, such records have often been published. Furthermore, regional records tend to be opportunistic, as they are collected and reported without a standardized field methodology or consideration of various factors that may influence data reliability in a given region. For example, Soldo and Jardas [4,5] documented 61 records of the great white shark in the Eastern Adriatic between 1868 and 2000 but excluded several sightings reported by Fergusson [11] near Šibenik and Lošinj in 1993. Upon tracing the original sources, they discovered that these records were based on misinformation and subsequently omitted them from their list. Nevertheless, as the erroneous records had already been published, some authors [12,13,14] overlooked the findings of Soldo and Jardas [4,5] and continued citing Fergusson [11], thereby perpetuating the error and illustrating how unverified data can persist in the literature and distort conclusions about species distribution and abundance. Additionally, it is well known that an unknown portion of published records suffers from species misidentification, particularly confusion between the great white shark and its close relatives: the shortfin mako, Isurus oxyrinchus Rafinesque, 1810 and the porbeagle, Lamna nasus Bonnaterre, 1788 [15]. This misidentification often leads to erroneous reports of juvenile great white sharks, a mistake that continues to occur even today [16]. In the absence of additional information, many authors have attempted to quantitatively analyze the available diverse set of records, comparing historical and contemporary data in an effort to reconstruct long-term population trends of the GWS in the Mediterranean. Several statistical tools have been developed to analyze opportunistic records to estimate the likelihood of species extinction, an approach also used to assess sub-specific extinctions at regional- and site-specific scales [17]. For example, McPherson and Myers [18] utilized data provided by Soldo and Jardas [4,5] and Soldo and Dulčić [19] for their quantitative analyses and suggested a significant decline in the GWS population in the Eastern Adriatic. Their modeling indicated a decline in abundance ranging from three to twentyfold (median estimate = 8) between 1868 and 1970, with an estimated 84% reduction in abundance over three generations.

Whether considering a reported quantifiable decline or an observed decreasing trend, it is evident that conducting a comprehensive study on the great white shark in the Mediterranean remains highly challenging. Several monitoring studies have attempted to tag GWS in the Mediterranean, but none have been successful so far. Soldo and Peirce [20] spent 23 consecutive days at sea in various areas of the Central Adriatic, aiming to deploy a pop-up archival tag on a GWS. During this period, they employed different chumming techniques, including deploying chum stations at various depths, even below the thermocline, to attract the species. However, although other shark species were observed, no GWS were recorded. Similar studies conducted later on in other regions of the Mediterranean yielded the same results [21,22], further supporting concerns about the significant decline of the GWS population in the region. Several environmental DNA (eDNA) sampling studies have recently been conducted [22,23], but, given the likely low population of GWS and their highly migratory behavior, the effectiveness of such studies remains uncertain, as the result of these studies is only confirmation of GWS presence in a wider area, a fact which is already well known anyway. Additionally, it is important to note that the Mediterranean GWS subpopulation is genetically distinct from the Atlantic subpopulation [24]. This genetic isolation, combined with a documented decline in records since the 1970s and the complete absence of sightings in some areas since the 1990s, led to the species being assessed as regionally Critically Endangered in the Mediterranean by the IUCN [25].

Whether considering the observed decreasing trend or the calculated decline, it is important to note that both are derived from the reported number of sightings recorded per unit of time. This metric is influenced by various factors, including the number of observers (observation effort), their willingness or ability to report sightings, the abundance of individuals available for observation, and their detectability [26]. Moreover, any assumption, whether observational or modeled, relies on the premise that the average likelihood of detecting and reporting a sighting remains unchanged over time, as detectability is considered constant. Analyzing the locations of recorded sightings and captures reveals that most have occurred in coastal areas nearshore. Consequently, the assumption of constant detectability rests on the notion that the factors driving GWS to visit coastal zones have remained stable over time. However, in addition to the well-documented changes in coastal habitat characteristics since the mid-19th century, when many record lists began to be compiled, none of the models account for potential changes in predator–prey interactions, which could also influence sighting rates. Although the great white shark has a broad diet, Atlantic bluefin tuna, Thunnus thynnus, (Linnaeus, 1758) appears to be its primary and preferred prey in the Mediterranean. Soldo and Jardas [4,5] were the first to associate the presence of GWS in the coastal waters of the eastern Adriatic Sea with high bluefin tuna abundance. They suggested that the onset of intensive tuna fishing in the open waters of the Adriatic, particularly during the 1970s, led to the depletion of tuna in coastal areas, which consequently resulted in the disappearance of GWS from records in the final decades of the 20th century. As a result, Soldo and Jardas [4,5] proposed that future records of GWS in the Adriatic Sea would correspond to new migratory routes and areas of tuna abundance. This hypothesis was later supported by the capture of a 5.70 m-long female during tuna purse seining, 15 nautical miles southwest of the isolated Jabuka Island [19]. Afterward, a similar relationship was suggested for other Mediterranean areas [10,13,27]. Additionally, many historical records of GWS in the Mediterranean have been linked to various nearshore tuna fishing methods [16,27,28]. However, since the late 20th century, with the intensification of tuna fisheries, particularly for farming purposes, tuna fishing has shifted farther offshore.

Thus, this study aims to investigate whether there are spatial associations between the decline in great white shark records in coastal areas of the Mediterranean Sea and changes in predator–prey interactions and oceanographic conditions in the same region.

2. Materials and Methods

2.1. Study Area

The Mediterranean Sea covers approximately 0.81% (2.514 million km²) of the Earth’s total water surface. Its coastline is shared by 23 countries, stretching about 4000 km from the Strait of Gibraltar to the Bosporus, which connects to the Black Sea. The sea reaches its greatest depth of 5267 m at the Calypso Pit in the Ionian Sea, off the southern coast of Greece. The Mediterranean is divided into two main sub-basins—eastern and western—separated by the Sicily–Tunisia ridge. The eastern sub-basin experiences significant oceanographic variability, with surface temperatures ranging from approximately 15 °C in winter to 30–32 °C in summer. In contrast, the western sub-basin sees temperatures of about 11 °C in winter and up to 23 °C in summer. Salinity levels also differ, with 39‰ in the east compared to 36‰ in the west [29].

2.2. Great White Shark Database

Records of great white sharks in the Mediterranean were gathered from multiple sources. The majority were obtained from scientific papers published in peer-reviewed journals [10,13,14,16]. Additional records were sourced from various media outlets, including newspaper articles, news websites, and online portals, as well as unpublished data provided by researchers to institutional databases such as MEDLEM (Mediterranean Large Elasmobranchs Monitoring), which operates under the auspices of the General Fisheries Commission for the Mediterranean Sea [30].

For this study, records with uncertain locations, those deemed false or highly uncertain, and duplicate entries of the same record were excluded. Additionally, known false records were also excluded; however, this was only from the Adriatic Sea, as more local knowledge is needed for such a decision in other areas. Consequently, only records documented from the year 1900 onward were included in the analysis.

2.3. Data Analysis

Previous studies have attempted to account for several variables when comparing historical and recent sighting data. McPherson and Myers [18] assumed that detectability is directly proportional to abundance and that the number of observers and their reporting propensity constitute a joint process. They also assumed that the rate of change in abundance remained constant over the study period, that is, abundance increased or decreased by a fixed percentage from one time unit to the next. Moro et al. [14] followed McPherson and Myers’s [18] approach, assuming that the expected number of sightings per statistical unit depended on the number of potential observers (i.e., observation effort), their propensity to report sightings, white shark population abundance, and shark detectability. However, such approaches have several drawbacks as they rely on the assumptions that the Mediterranean Sea has not undergone major environmental changes and that predator–prey relationships have remained stable. In reality, the Mediterranean Sea has experienced significant long-term changes in oceanographic properties, particularly since the 1990s [31]. Taking into account that broadscale oceanographic conditions are known to influence the occurrence and habitat use of great white sharks [32], it can be presumed that changes in habitat characteristics can substantially affect detectability. In addition, it is well established that predators follow migratory prey, leading to shifts in their own distribution patterns. This behavior can create short- or long-term hotspots of predator activity in regions where migratory prey congregate. Thus, shifts in prey migration patterns influence predator distributions and, consequently, detectability, an aspect overlooked by the aforementioned studies. Given the risks of drawing conclusions from overly complex models based on many limited or uncertain assumptions, we adopted a simpler and more direct approach: comparing the spatial overlap of the bluefin tuna and great white shark distributions in the Mediterranean Sea during the 21st century to explore potential patterns of co-occurrence, while recognizing that these are illustrative rather than conclusive.

Geographical coordinates representing GWS occurrences in the Mediterranean Sea before and after 2000 were obtained. When publications provided only locality names, Google Earth was used to extract the corresponding coordinates, while records with clear geocoding errors were excluded. QGIS was employed to verify the accuracy of all occurrence records prior to analysis. For graphical visualization, QGIS heat maps were used to depict species records along with their densities [33]. Additionally, a long-term daily average sea surface temperature (°C) map of the Mediterranean Sea for the period 1991–2020 was generated using data from the National Oceanic and Atmospheric Administration (NOAA) Physical Sciences Laboratory, with minor modifications to enhance its appearance.

3. Results and Discussion

The spatial distribution of great white sharks in the Mediterranean during the 20th century (Figure 1) reveals several key hotspots. From west to east, these include the Balearic Islands; the western and northwestern Italian coast, spanning from the Gulf of Genoa to the northern Tyrrhenian Sea; the Strait of Sicily, from the Gulf of Tunis to the Aegadian Islands; the Strait of Messina; the eastern Adriatic coast, particularly the Kvarner Bay area; and the Sea of Marmara. Additionally, other significant sites of GWS occurrence were the Gulf of Lyon and the Maltese Islands.

In contrast, records of GWS in the 21st century (Figure 2) indicate a notable shift in distribution, particularly further from the coast and mainly offshore. Several previously important hotspots from the 20th century have diminished in importance, with only isolated or no recent records reported. The Balearic Islands and Maltese Islands are no longer significant areas for GWS occurrences, as only a single contemporary record has been documented for each, while no records have been reported from the Gulf of Lyon, the Gulf of Genoa, the northern Strait of Sicily, or the Sea of Marmara. Furthermore, sightings in the Adriatic Sea have shifted toward the central offshore area, far from the mainland.

Currently, the primary hotspots are the southern areas of the Strait of Sicily, extending to the Gulf of Gabes, and along the Aegean coast of Turkey, where the first confirmed juvenile GWS individuals were recorded [10]. Considering that only the first quarter of the 21st century has passed, the Northern Tyrrhenian Sea may also be considered as an important region for GWS occurrences in the Mediterranean. These results highlight a significant regional shift in the distribution of GWS in the Mediterranean during the 21st century. While a population decline is the most straightforward explanation for this shift, other ecological and anthropogenic factors may also play a role. Soldo and Jardas [4,5] demonstrated that the decline in GWS records along the eastern Adriatic mainland coast coincided with the shift in tuna fishing from coastal waters, where coastal seines were used, to offshore areas, where purse seines became a predominant fishing gear. This trend is supported by new GWS records from the central offshore Adriatic (Figure 2). Similarly, Kabasakal et al. [10] suggested that the disappearance of GWS from the Sea of Marmara aligned with the collapse of the bluefin tuna fishery that occurred in the region [34]. Our updated dataset appears consistent with this hypothesis, although we cannot directly quantify predator–prey linkages with the available data.

Historically, bluefin tuna in the Mediterranean were primarily targeted using coastal seines and traps near the mainland, with fishermen relying on their knowledge of tuna migration routes. This use of coastal and passive gear remained largely unchanged until the late 20th century, when fishing intensified with the development of purse seine fleets that started to target tuna far offshore. Today, the tuna fishery is strictly regulated through seasonal closures, quotas, minimum catch-size restrictions, and a ban on fish-spotting by aircraft. Purse seines now account for 60–80% of the total bluefin tuna catch in the Mediterranean, with fishing operations occurring almost exclusively in offshore waters [35].

In addition to fisheries, broader ecological shifts, including climate change, predator–prey dynamics, and pollution, have altered bluefin tuna migration patterns in the 21st century [36]. Druon [37] produced a map (Figure 3) illustrating the potential feeding habitat of Atlantic bluefin tuna in the Mediterranean Sea for the 21st century, based on data collected between 2003 and 2012. It should be noted that this bluefin tuna habitat model does not fully overlap temporally with the shark dataset from the 21st century. However, given the documented shift in tuna fishing from coastal to offshore areas [35], it can be considered indicative. This map, which also aligns with key tuna fishing areas identified by Cermeño et al. [38], indicates that several historically important tuna fishing hotspots are no longer favorable feeding grounds for bluefin tuna. At the same time, these areas correspond to regions where the great white shark is no longer observed. These include the Balearic Islands, the Maltese Islands, the eastern Adriatic coastline, and the Sea of Marmara, all of which were previously identified as significant hotspots for great white shark occurrences during the 20th century. Conversely, several of the most prominent contemporary bluefin tuna feeding areas now correspond to the newly established primary hotspots for GWS, such as the Aegean coast of Turkey, particularly Erdemit Bay and its surrounding region. Given the likely low population of GWS in the Mediterranean and species’ apparent dependency on tuna as prey, future field studies requiring direct interaction with these sharks, such as tagging research, could benefit from targeting known seasonal bluefin tuna aggregation sites to increase the likelihood of shark encounters. Such an approach would enhance the probability of intentionally locating GWS, a task that has so far proven unsuccessful in the Mediterranean. Interestingly, the importance of this study can already be verified by comparing the main contemporary hotspots of GWS (Figure 2) with recent unsuccessful attempts to locate this species in the Mediterranean [21,22,23], which were based on areas identified as suitable for detecting GWS from older historical records. None of the areas where these studies were conducted were indicated as the contemporary hotspot for GWS: Micarelli et al. [21] concentrated their main efforts along the Italian coast of the Adriatic and central Tyrrhenian sea, and also, as well as Ferreti et al. [22] and Jenrette et al. [23], more eastern in the Gulf of Gabes, than it is indicated by this study (Figure 2).

While this study supports a link between bluefin tuna abundance and GWS distribution, other potential factors may also contribute to these changes. Notably, changes in GWS distribution during the late 20th and early 21st centuries coincided temporally with abrupt and significant changes in the Mediterranean Sea oceanographic characteristics [31]. Environmental shifts, particularly those driven by recent climate change in the Mediterranean, such as rising sea temperatures, may influence both GWS and bluefin tuna distribution. Cermeño et al. [38] demonstrated that temperature plays a crucial role in bluefin tuna behavior and migration. Regarding GWS, the species is known to tolerate temperatures ranging from 5 to 25 °C. At first, it may seem surprising that the only region where GWS occurrences do not align with bluefin tuna feeding areas [37] or tuna fishing grounds [38] is the eastern Mediterranean, where only a few historical GWS records have been documented since the early 20th century, with none in recent decades. However, an analysis of the daily average long-term sea surface temperature (SST) in the Mediterranean from 1991 to 2020 (Figure 4) reveals that SST perfectly delineates the species’ preferred habitat. A comparison of average SST and GWS occurrences indicates that GWS in the Mediterranean generally inhabit only yellow and green areas (Figure 4), suggesting a possible preference for waters where the mean SST is below approximately 18 °C, although this requires confirmation with in situ data. This thermal preference aligns with findings from other regions of the world [32,39]. Given the ongoing rise in SST across the Mediterranean [40], particularly in the southern and eastern Mediterranean, it is reasonable to conclude that the distribution of GWS is likely to shift northward in the future.

Also, what needs to be further investigated is the suggested magnitude of GWS population decline in the Mediterranean. Although it is highly likely that anthropogenic pressures, primarily fishing and pollution, have contributed to a population reduction, as similar cases have been observed in other regions and with other species based on more robust data [41,42], it is important to understand that all previous population decline estimates have been based on reported sightings. This study highlights that the likelihood of detecting and reporting GWS occurrences has significantly changed and decreased due to the species’ shift toward offshore areas, where observational effort is substantially lower. Consequently, historical comparisons of occurrence data may underestimate the current population due to decreased detectability rather than actual abundance. In addition, while citizen science initiatives have contributed valuable occurrence records of great white sharks in the Mediterranean, these data are subject to several limitations. Reports are typically opportunistic, lack standardized methodology, and are heavily biased toward nearshore, high-traffic areas, leading to an underrepresentation of offshore habitats where GWS may now be more prevalent. The absence of systematic effort and verification protocols raises concerns about both false negatives (areas assumed to be devoid of sharks due to low reporting) and false positives (misidentifications, especially of other large shark species). As such, while citizen-reported data can offer preliminary insights into spatiotemporal trends, they should be treated with caution and complemented by scientifically rigorous methods for robust population assessments. Future estimates of population status must therefore incorporate models that correct for detection bias. Ideally, future research should build upon the findings of this study to design and implement field surveys, targeting likely aggregation areas, that generate more reliable data, ultimately providing a clearer understanding of the status of this apex predator in the Mediterranean.

4. Conclusions

This study provides important insights into the distribution shifts in great white sharks in the Mediterranean, highlighting potential associations with bluefin tuna abundance and climate-driven environmental changes. However, given the complexities of predator–prey interactions and the ongoing changes in marine ecosystems, further interdisciplinary research is crucial to better understand the factors influencing great white shark distribution. Thus, future studies, particularly those incorporating more advanced tracking and monitoring techniques, will be essential for accurately assessing the species’ status and predicting its future distribution patterns in a rapidly changing Mediterranean environment.