The loggerhead turtle Caretta caretta population of the Eastern Mediterranean faces a host of climate change-induced challenges in the future. The sex of hatchlings is determined by incubation temperature during a genital differentiation phase, producing more females at higher temperatures (generalised here to >29 degrees C). Without behavioural change, either by nesting at a cooler time of year or altering microsite selection, there is a risk that climate change-induced female bias may place the population at risk. Through a combination of in situ dataloggers and biophysical modelling, we estimated offspring sex ratios across 8 Mediterranean countries, including established rookeries (>50 nests yr(-1), n = 17 in 4 countries) and smaller rookeries (n = 5) and beaches that do not currently host nesting (n = 9). Here, we provide for the first time a regional estimate of hatchling sex ratio: 74.5% female, which may rise to 91.2% by 2100 per IPCC SSP3-7.0. Of the well-established rookeries, Patara, T & uuml;rkiye, has the lowest female production (49%), while Alagadi, Northern Cyprus, produces the most females (96%). Several smaller beaches, particularly in North Africa, exhibit male-biased production, including Lake Bardawil, Egypt (19% female), and Kuriat Islands, Tunisia (23% female). Beaches without current nesting activity, such as Qerret, Albania, could provide significant male production (6% female). A phenological shift of 20 d could potentially decrease female hatchling proportion upwards of 60% in areas with rapid mid-summer warming. We present as part of this work an interactive tool to explore these warming and phenological shift scenarios, providing a novel framework to assess the impact of hypothetical phenological and abundance changes on resultant primary sex ratios.

Accurately monitoring sea turtle populations is crucial for informing effective conservation strategies; however, traditional methods for assessing spatial distribution and abundance are time-consuming, labour-intensive, and prone to observer bias. Unoccupied Aerial Vehicles (UAVs) have become valuable tools for collecting high-resolution imagery with minimal disturbance to marine fauna. However, the lack of standardized and user-friendly platforms for processing UAV data limits their broader application. In this paper, we present a lightweight browser-based web application designed to streamline the analysis of video collected by UAVs to monitor sea turtles. The application requires no additional software installations beyond a modern web browser and operates entirely on the client side, preserving data privacy. It supports the integration and execution of artificial intelligence models locally, facilitating the automated detection and classification of turtles in video footage. This tool, freely available, bridges the gap between UAV data collection and effective conservation-oriented decision-making by enabling rapid, standardized, and scalable analysis. Our approach promotes community-driven development and the reuse of AI models, making environmental monitoring practices more accessible and collaborative.

Leatherback sea turtles Dermochelys coriacea are oceanic animals ranging from the equator to sub-polar latitudes. In the Mediterranean Sea, leatherbacks are the third most encountered sea turtle species, albeit at a low frequency. Along the coasts of Greece, 14 records were reported until 1984. Combining these data with 43 additional records collected over the ensuing 40-year period 1985–2024, we infer that most records were reported from the Aegean Sea, north of 39 degrees latitude, in the warmer months of the year (May–November). The size range of measured specimens (123–180 cm curved carapace length) indicate that these were large juveniles and adults. The spatial and seasonal distribution of the recorded leatherbacks, as well as their sizes, suggest that these individuals enter the Mediterranean basin during a foraging migration from their nesting beaches in the western Atlantic. Most of the recorded leatherbacks were captured in fishing gears, primarily gillnets, with only few of them released. It is unknown how many of these animals that enter the Mediterranean return for breeding to their natal beaches. Keywords: Dermochelys coriacea, Mediterranean, fisheries interaction, jellyfish blooms

ARCHELON’s Sea Turtle Rescue Centre (STRC) was established in 1994 in Glyfada, Attica, with the main objectives to rehabilitate injured turtles, assist in determining their threats at sea, and contribute to public awareness. Here, we present the main results of the STRC’s 30-year (1995-2024) operation, evaluating its rehabilitation effectiveness and drawing conclusions on the main threats that turtles encounter in the Greek seas. During this period, 1,539 turtle admissions were registered, with annual numbers showing a significant positive trend. Clinical examination identified the cause of injury for 87% of admissions, with 87.3% of them assigned to fisheries interaction (n=1,170), including those bearing injuries inflicted intentionally following incidental capture (n=595). In total, more than 71% of turtles rehabilitated for more than five days were released to the wild (n=842). Released turtles generally exhibited positive post-release behaviour as indicated by subsequent observations (live sightings or nesting activity) and by satellite telemetry, although more research is needed to confirm this, especially in the case of head traumas. Notwithstanding the relatively small number of released turtles, contributing modestly to sea turtle conservation in terms of increasing population viability and abundance, the role of the STRC is paramount in covering the social need for wildlife welfare, as well as being an important vehicle of public awareness and education.

Artificial light at night (ALAN) is an escalating anthropogenic threat that disrupts coastal and marine ecosystems worldwide, yet its impacts on Mediterranean sea turtles remain poorly quantified. We surveyed 27 experts across 16 Mediterranean countries to evaluate perceived ALAN effects on sea turtle nesting sites and nearshore habitats, identify research gaps, assess monitoring limitations, and prioritize mitigation strategies. Respondents consistently highlighted hatchlings as the most vulnerable life stage, particularly during sea-finding and early dispersal, with ALAN causing disorientation, increased predation risk, and reduced recruitment. Standardized monitoring of ALAN impacts is largely absent, with financial constraints representing the main barrier, alongside gaps in communication, education, and policy. Experts ranked the highest-priority mitigation measures as updating monitoring programs to incorporate ALAN, reducing light exposure on nesting beaches through shielding, downward-directed and long-wavelength lighting, shaded corridors, and nature-based solutions, and integrating advanced technologies into conservation planning. Measures such as policy reform, public awareness campaigns, and scenario-based coastal planning were considered of moderate priority, emphasizing that practical, site-specific interventions may yield the most immediate benefits. Limited expert consensus on the prioritization of mitigation strategies indicates persistent uncertainties and critical regional knowledge gaps that hinder effective conservation planning. ALAN is an established but under-addressed threat to Mediterranean sea turtles. Effective conservation will require integrating evidence-based research, standardized monitoring, and targeted mitigation measures at nesting and nearshore habitats. Implementing such strategies will not only safeguard sea turtle populations in the Mediterranean but also provide a globally relevant framework for managing ALAN in other coastal ecosystems.

Transboundary conservation of migratory marine species is frequently constrained by fragmented and locally isolated biodiversity data, limiting the detection of large-scale ecological change. In the Mediterranean Sea, one of the world's most politically complex and environmentally pressured basins, decades of sea turtle nesting observations have remained dispersed across countries and organizations, preventing a holistic assessment of population trends. Here, we integrate over sixty years of nesting records for Caretta caretta and Chelonia mydas into the first standardized, basin-wide database for the Mediterranean (MedSTN), harmonizing 2290 validated records spanning 1960–2023. By integrating this database into Generalized Additive Models (GAMs), we reveal coherent regional patterns invisible in local assessments. Our results show a fundamental structural disparity in population dynamics: while the Eastern Mediterranean remains the primary reproductive stronghold with stable nesting core areas, for both species, we detect a significant northward range expansion for C. caretta in the Western Mediterranean. This emerging nesting trend is characterized by an exponential increase in unique nesting sites following a distinct temporal breakpoint circa 2010, suggesting a potential northward range expansion that is consistent with climate-mediated thermal suitability and may be influenced by trans-oceanic immigration. Conversely, C. mydas nesting remains almost entirely confined to its Eastern strongholds, exhibiting long-term latitudinal stasis despite localized increases. By transforming a fragmented legacy into a policy-relevant evidence base, MedSTN demonstrates how basin-scale data integration can uncover regional ecological signals and support proactive conservation prioritization. Our framework provides a replicable model for unifying long-term datasets to guide the transboundary management of wide-ranging species under accelerating global change.

Tag recoveries from a more than 40-year tagging programme of loggerhead turtles nesting at the three major nesting areas of Greece (Zakynthos, Kyparissia Bay, Rethymno) revealed their post-nesting whereabouts in the Mediterranean. Most tag recoveries, signalling potential foraging areas, occurred in the Central Mediterranean, with major concentrations in the Adriatic Sea and the Tunisian Plateau (both characterized by shallow waters and rich benthic fauna), whereas fewer recoveries were recorded in the Aegean Sea. While turtles from Zakynthos and Kyparissia Bay shared the Adriatic Sea and the Tunisian Plateau, turtles from Rethymno (Crete) preferred the Aegean Sea and the Levantine Sea. Tag recoveries showed a much wider dispersal of post-nesting turtles than the dispersal recorded from satellite telemetry, highlighting the benefits of long-term tagging programmes. More than 60% of the recovered turtles were reported dead, indicating the substantial threats that reproductively active individuals, the most biologically valuable group in the loggerhead population, face in the Mediterranean Sea.

Fisheries interaction is one of the most significant threats to sea turtles in their marine habitats. Sea turtles are reptiles and therefore need to surface intermittently to breathe. This dependence on the surface means they inhabit much of the vertical space in coastal and neritic habitats, exposing them to encounter fisheries in a wide range of depths. Turtle dive profiles may vary greatly depending on their species, behaviour and the ambient water temperature, for example they may undertake longer dives in the winter, exhibiting a type of dormancy. The dive behaviour of Loggerhead sea turtles (Caretta caretta) in the Mediterranean is relatively well known, at least for females where nesting allows for easy access, but knowledge is lacking for adult males of the species. To address this lack, we tracked three adult male turtles from their breeding area in Kyparissia Bay, Greece to their foraging areas in the eastern Mediterranean using devices that transmitted depth and temperature time series data. Tracking duration ranged from 98 to 547 d. The turtles demonstrated different behaviours that were influenced by time and ambient temperature. The turtles showed diel changes in dive behaviour, split-level dives in the presence of thermoclines, and extended dive durations when waters approached 15 °C. All turtles spent over 50

Sea turtles are long-lived wide-ranging animals undertaking long-distance breeding migrations showing fidelity to both foraging and breeding sites. This fidelity has led to population structuring with regional management units defined for globally distributed species. Loggerhead sea turtles (Caretta caretta) have established such a management unit in the Mediterranean, with the greatest number of turtles breeding in Greece. Kyparissia Bay, Greece, has been identified as the location with largest nesting aggregation of loggerhead turtles in the Mediterranean. Determining where this aggregation of turtle migrates to outside the breeding season is important for its continued conservation. Long-distance flipper tag recaptures have identified certain high use areas for these turtles but lack route data and certainty that recaptured turtles are resident in the area of capture. Satellite tracking has been used to validate such tag return data and generate insights into turtle migrations patterns and their relationship with phenological and fecundity metrics. Here we recorded turtles' curved carapace lengths as a metric for general body size and deployed Argos satellite transmitters on 11 adult loggerhead sea turtles (8 females, 3 males) during their breeding period in Kyparissia Bay, to examine the implications of their selected migratory routes and foraging strategies. Average carapace length (± SD) was 82.0 cm (± 4.6 cm) for the 8 female turtles and 86.0 cm (± 5.5 cm) for the 3 males. Turtles were tracked for an average of 326.9 days. End points for the tracked turtles were grouped into four distinct regions The Aegean Sea was the end point for 3 turtles, and the Adriatic Sea was the end point for another 3 turtles with one turtle spending an extended period in the northern Ionian. The northern Ionian Sea was the end point for a further 3 turtles and the Tunisian plateau was the end point for final 2 turtles. Six turtles migrated to restricted area foraging sites, 1 turtle remained nomadic for the entirety of its tracking duration, 2 turtles were semi-nomadic incorporating both restricted area foraging and large-scale movements during their tracking period, and the tracks of the final 2 turtles ceased before their behaviour type could be determined, but sedentary behaviour type was inferred. Notably the one turtle that remained nomadic was 6 cm shorter than the next shortest turtles, suggesting carry-over effects of the nomadic life-history. The turtles generally moved to locations that had been identified by flipper tag recaptures, however the tracking identified routes taken to get there, which were often highly convoluted. These foraging sites, also identified through other tracking studies of loggerhead turtles nesting elsewhere in Greece and from Cyprus, were often sites of high fishing activity and cause for concern for turtles present there. Given that up to around 2,000 turtles may nest in Kyparissia Bay during a single summer, we conclude that a larger number of turtles, over multiple years, should be tracked to obtain a more robust assessment of post-breeding migrations, foraging strategy adoption and their carryover effects, which can then be used to better inform conservation and management actions.

Animal photo-identification (photo-ID), the process of identifying individual animals from images, has proven to be a valuable tool for various studies on sea turtles, increasing the knowledge of their ecology and informing conservation efforts. Photo-ID in sea turtles is predominantly based on the geometric patterns of the scales of their two head sides, which are unique to every individual and different from side to side. As such, both manual and automated photo-ID techniques are traditionally performed under a side-specific setting. There, an image showing a single profile of an unknown individual is compared only to images showing the same side of previously identified individuals. In this paper, we show for the first time an inherent visual similarity between left and right facial profiles of the same individuals in three sea turtle species. We do so by employing two state-of-the-art automated neural network-based photo-ID methods, one local feature-based and one deep embedding-based, designed to rank profiles based on their similarities. Both methods rank the similarity of the left and right profiles of the same individual higher than those of different individuals. These similarities are detectable even when images are taken years apart under diverse conditions. We further show that the exploitation of this similarity results in improved accuracies when compared to the traditional side-specific photo-ID setting. Our results indicate two concrete guidelines for improving automated sea turtle photo-ID workflows. When trying to match a photo of a given profile, searches should not be restricted only to photos of the same profile. As the first method of choice, a deep embedding model finely-trained using a photo-database of the focal sea turtle population should be used. In the absence of such training database, a neural network-based local feature method is preferable, but in that case searches should be performed with both the original query image and its horizontally flipped version.

Predation of nests by a multitude of terrestrial predators is a major threat to sea turtle populations worldwide. Destruction of eggs decreases hatching success and reduces hatchling recruitment. The 43.5-km beach of Kyparissia Bay in Greece, hosting a large nesting population of loggerhead sea turtles (Caretta caretta), is subject to heavy nest predation by canids. Over 50% of nests were annually depredated in the 1980s and this triggered in 1992 an intensive long-term nest protection program through which the predation rate was reduced to about 13%. Continuous beach monitoring over decades revealed that nest numbers started to increase after 17-20 years from the onset of nest protection and in recent years reached exponential dimensions. Similarly, yearly tagging of nesting turtles showed, in the last decade, a significant increase in the percentage of neophyte turtles, i.e., those considered to be nesting for the first time. We attribute this extraordinary increase of nests largely to the maturing of hatchlings saved due to the intensive nest protection, since the time lag of 17-20 years falls within the boundaries of the maturation time of Mediterranean loggerheads. Our conclusion is further discussed in relation to the evolution of nest numbers at the nearby predator-free Zakynthos rookery that over the same time frame shows no significant increase of nests, although both populations share the same foraging habitats, and exhibit low nesting interchange, similar temperature regimes and female mortalities.

Sporadic nesting events (i.e., any evidence of nesting or attempted nesting) of loggerhead sea turtles Caretta caretta, outside of the systematically monitored nesting areas, were recorded over 40 years (1985–2024) in the Ionian Sea, Greece. Most nesting events were reported by the public. Accumulated records show a wide distribution of nesting events along the coasts, without a considerable north–south gradient—contrasting with a similar study in the Aegean Sea, where nesting was rather uncommon in northern latitudes. This discrepancy is attributed to the different temperature regimes between the two marine areas. Beaches with frequently reported nesting events should be surveyed systematically to assess their nesting magnitude. The consequences of a possible northward expansion of loggerhead nesting in the Ionian are discussed.

The loggerhead sea turtle ( Caretta caretta ), has the widest distribution among sea turtle species in the Mediterranean, requiring regional and international collaborations in addition to local efforts to better inform conservation actions. Molecular techniques are powerful tools for the assessment of population dynamics at large scales, especially in determining the connectivity among different nesting and foraging sites, and genetic diversity. In this study, a large sample was collected synchronously in the nesting areas located in the north, south and eastern Mediterranean. Recently confirmed nesting sites from Albania and other nesting sites represented by lower sample sizes were also included in order to better assess the genetic composition of the region’s rookeries. Samples from 698 individuals were collected and the longer (815 bp) mtDNA D-loop fragment of these samples was sequenced. We recorded 15 haplotypes, three of which were novel. In addition, our results show that some haplotypes considered of Atlantic origin, have a wider dispersal in the Mediterranean than previously thought, albeit with low levels of representation. Our results contribute to determining the likely origin of haplotypes that were previously only recorded from foraging sites. They highlight the utility of broad-scale sampling, with increased sample sizes and the longer mtDNA sequence to determine genetic diversity and connectivity. This study also demonstrates the importance of continued monitoring for the contribution of Atlantic-origin haplotypes to the Mediterranean population and the effects of climate change on the resident Mediterranean population, which is expected to expand its geographical range for reproduction. This work is important for, mixed stock analyses (MSA) that seek to determine the natal regions of stranded or accidentally caught sea turtles and those purposefully obtained from foraging sites. In doing so biogeographic linkages between areas of the Mediterranean can be elucidated for conservation purposes.

female Loggerhead Turtles (Caretta caretta) migrate from foraging to breeding areas to lay multiple clutches over 1-2 mo. As sea turtles may travel long distances between nests, efforts are needed to determine their at-sea movements and habitat preferences within breeding areas. In turn, the predictable presence of turtles in these habitats can facilitate the use of straightforward management measures. We used satellite telemetry, novel GSM telemetry, and drone surveys to determine the distribution of turtles at-sea during the breeding period at the largest nesting site for Loggerhead Turtles in the Mediterranean Kyparissia Bay, Greece. We found that turtles largely remain within the 50 m isobath and close to where they nest, although specific sites (i.e., protected waters of a nearby harbor) farther away from the nesting beach were also used. We also combined our satellite telemetry data with previous data to update the estimated mean clutch frequency (the number of clutches laid by turtles during a single breeding period) for this population. The new estimated mean clutch frequency ranged from 3.3-3.5, which were slightly lower than the previous estimate (3.8) for the same breeding population even though they remain higher than values published for other locations in the Mediterranean. Our findings confirm that conservation measures can be adequately defined within spatially local and temporally restricted boundaries and that population sizes based on clutch frequencies lower than 3.3 are likely to be overestimated in Kyparissia Bay.

Assessing sea turtle movements and connectivity among different areas is pivotal to understanding their biology and implementing efficient conservation actions. In the Adriatic Sea, one of the most important sea turtle foraging areas in the Mediterranean, a total of 311 capture-mark-recapture (CMR) records (mostly bycatch) from 294 loggerhead turtles (Caretta caretta) in the period 1984-2021 were analysed. A general fidelity pattern to Adriatic subareas was indicated by a significantly shorter CMR distance than the potential dispersal distance and by the significantly higher proportion of re-encounters in the same area of release than expected. No seasonal pattern was detected between subareas, and shorter re-encounter distances were observed in turtles released and re-encountered in the same season, suggesting different winter and summer residential areas. Results suggest that turtles frequenting the Adriatic can go anywhere in the Mediterranean basin and may exhibit a wandering behaviour regardless of their size. A substantial connectivity with nesting sites in Greece was observed, confirming with empirical evidence that this is the most important breeding area for turtles foraging in the Adriatic Sea. This study highlights the value of cooperation among different groups and shows a main behavioural pattern of fidelity to neritic foraging grounds.

Long-term monitoring programs are valuable in assessing population trends and evaluating conservation status especially for threatened species exhibiting delayed maturity such as marine turtles. The loggerhead sea turtle Caretta caretta is a globally distributed species with a regional population within the Mediterranean Sea. Loggerhead nesting in the Mediterranean occurs mainly in the eastern basin, with nesting areas classified as per their magnitude and density. A "moderate-dense" nesting area in Greece is the 2.7 km Koroni beach which has been monitored by ARCHELON since 1995 with the aim to collect reproductive data and to protect nests. Data collected over 25 years showed an average annual number of 55.8 nests, a nesting success (percentage of emergences resulting in egg-laying) of 38.0% and a nesting density of 20.7 nests/km. Nest numbers exhibited a significantly increasing trend in recent years, while clutch size showed a significant downward trend. Incubation durations, considered to be an indicator of incubation temperature and subsequently hatchling sex ratio, have been significantly decreasing over the years-a possible sign of global warming. A major threat is nest predation by foxes and dogs, which has been effectively controlled through the fencing of nests. This nesting population, despite its moderate size, may contribute to the genetic homogeneity of the larger western and eastern nesting aggregations of loggerhead turtles in Greece. The nesting beach has been recently included in the European Union's NATURA 2000 network of protected areas. Continuation of this long-term monitoring program is expected to provide further insights into the reproductive traits of this important loggerhead population.

Here we present 38 yrs of nesting data of the loggerhead population in Laganas Bay, Zakynthos, Greece, adding data for 2010-2021 to existing published data for 1984-2009. This rookery, encompassing 6 discrete beaches of 5.5 km total length, features the highest nesting density in the Mediterranean. The annual nest numbers, ranging from 667 to 2018, showed no significant long-term trend. Beach nesting contributions increased on some public-accessed beaches (as did nesting success), but decreased on a remote beach that previously held > 50% of all nests. These changes might be attributed to management actions by the National Marine Park of Zakynthos that improved conditions for sea turtles (e.g., restricting nighttime access to the public), indicating that the previously recorded high nesting density on the remote beach may have been atypical. Although no significant phenological changes were detected with the available data in the 20-yr period 2002-2021, the date of the first hatched nest shifted significantly toward earlier dates in the 38-yr data set. Clutch size and body size of nesting turtles exhibited significantly decreasing trends. Hatching success, hatchling emergence success, and in-nest hatchling mortality showed significantly increasing trends. The number of viable hatchlings decreased over the seasons, albeit not significantly, possibly a result of the decreasing clutch size. Incubation durations were revealed as shortening on all beaches, a potential sign of global warming, with a consequent suggested increase of female primary sex ratio. Continuation of this long-term monitoring program is expected to provide further insights in the reproductive traits of this regionally important loggerhead population.

Delineating spatial boundaries that accurately encompass complex, often cryptic, life histories of highly migratory marine megafauna can be a significant conservation challenge. For example, marine turtles range across vast ocean basins and coastal areas, thus complicating the evaluation of relative impacts of multiple overlapping threats and the creation of coherent conservation strategies. To address these challenges, spatially explicit ‘regional management units’ (RMUs) were developed in 2010 for all marine turtle species, globally. RMUs were intended to provide a consistent framework that organizes conspecific assemblages into units above the level of nesting rookeries and genetic stocks, but below the species level, within regional entities that may share demographic trajectories because they experience similar environmental conditions and other factors. From their initial conception, RMUs were intended to be periodically revised using new information about marine turtle distributions, life history, habitat use patterns, and population structure. Here, we describe the process used to update the 2010 RMU framework by incorporating newly published information and inputs from global marine turtle experts who are members of the IUCN Marine Turtle Specialist Group. A total of 48 RMUs for 6 of 7 marine turtle species and 166 distinct genetic stocks for all 7 species are presented herein. The updated RMU framework reflects a significant advance in knowledge of marine turtle biology and biogeography, and it provides improved clarity about the RMU concept and its potential applications. All RMU products have been made open access to support research and conservation initiatives worldwide.

: The northwestern Indian Ocean hosts globally crucial marine biodiversity, which is relatively understudied. There has, however, been significant research and conservation effort focused on marine turtles across the region in recent decades. We undertook an exhaustive review of 251 relevant publications to extract and collate the current information concerning marine turtles. To ensure completeness, we received input on a preliminary draft from a large group of experts who have worked extensively in the region. Data presented here are from all 13 countries with significant coastlines in the region (Bahrain, Djibouti, Egypt, Eritrea, Iran, Kuwait, Oman, Qatar, Saudi Arabia, Somalia, Sudan, United Arab Emirates, and Yemen) on occurrence, nesting biology, morphology, foraging areas, population status, threats, and relevant national legislation and international conservation strategies. The experts independently provided details on what they felt were the (1) most significant threats to marine turtles, (2) most critical knowledge gaps for marine turtle ecology, (3) impediments to practical conservation, and (4) essential strengths and opportunities in the region.