The longest recorded movement of an inshore common bottlenose dolphin (Tursiops truncatus)

Information on movements and connectivity among populations of animals is important for the delineation of units to conserve, so that demographic parameters, such as abundance, fecundity and mortality, can be placed in an appropriate population and conservation context. Common bottlenose dolphins (Tursiops truncatus) are often considered relatively ‘resident’ and demonstrating strong site fidelity to specific areas. However, this perception may partly be an artefact of the distribution and ‘habitat use’ of cetacean researchers, rather than animals themselves, and bottlenose dolphins have been shown to be capable of substantial movements, often in relatively short periods of time. Here, we report on two long-distance movements of a common bottlenose dolphin within the Mediterranean Sea, across the Tyrrhenian, Ionian and Adriatic Seas, and subsequently back across all three seas to Ligurian Sea, making these the two longest recorded movement for this species in the Mediterranean Sea to date and some of the longest in the world. We also review published records of long-distance movements in this species worldwide. This study highlights the utility of photo-identification and the importance of regional data sharing. We argue that photo-identification comparisons are always worthwhile and the results are informative regardless of the presence or absence of matches, especially with the ongoing advances in automated matching software.

genetic studies (Natoli et al. 2005;Gaspari et al. 2015b). This generally also appears to be true for populations of this species elsewhere in Europe (Cheney et al. 2013;Giménez et al. 2018;Nykänen et al. 2019), as well as globally (Silva et al. 2008;Currey et al. 2009;Rosel et al. 2009). However, this perception may, at least in part, be an artefact of the distribution and 'habitat use' of cetacean researchers, rather than animals themselves. The majority of detailed and longterm studies of bottlenose dolphins via photo-identification (hereafter 'photo-ID') are typically relatively restricted geographically, potentially biasing the overall picture. In some areas, low site fidelity, large proportion of transients or substantial movements have been reported Silva et al. 2008;Defran et al. 2015;Papale et al. 2017;Dinis et al. 2021). In fact, despite the general pattern of some degree of residency in most studied populations in the Mediterranean Sea, bottlenose dolphins in this region have been shown to be capable of substantial movements, often in relatively short periods of time (Bearzi et al. 2011;Gnone et al. 2011;Genov et al. 2016). Moreover, long-term field studies almost inevitably focus on relatively coastal local populations, whereas potentially pelagic populations or individuals, for which there is now also some evidence in the Mediterranean Sea (Gaspari et al. 2015b), may remain overlooked.
Information on movements and connectivity among local populations is important for the understanding of population structure and delineation of units to conserve (Taylor et al. 2010), so that demographic parameters, such as abundance, fecundity and mortality, can be placed in an appropriate population and conservation context. Here, we report on two long-distance movements of a single bottlenose dolphin, first across the Tyrrhenian, Ionian and Adriatic Seas, and subsequently back across the Adriatic, Ionian, Tyrrhenian and Ligurian Seas, making these the longest recorded movements for this species in the Mediterranean Sea to date and among the longest in the world.
A local population of bottlenose dolphins in the Gulf of Trieste and adjacent waters in the northern Adriatic Sea ( Fig. 1) has been studied since 2002 (Genov et al. 2008), with genetic and photo-ID data suggesting it is a distinct unit within the Adriatic Sea (Genov et al. 2009;Gaspari et al. 2015b). Based on mark-recapture models, the total annual abundance during 2013-2018 was estimated to range between 161 (95% CI = 153-170) and 245 (95% CI = 219-273) dolphins (Genov 2021). The population is structured into distinct social clusters, which differ in their temporal use of the study area, as well as their interactions with trawl fisheries (Genov et al. 2019).
A local population of bottlenose dolphins in the Aeolian archipelago in the southern Tyrrhenian Sea (Fig. 1) has been studied since 2005 (Blasi and Boitani 2012) and is thought to consist of about 40 individuals with relatively high site fidelity (Blasi and Boitani 2014;Blasi et al. 2020), although no estimates of absolute abundance are currently available. The population is composed of two main social units that show different association patterns, group sizes and degree of interaction with trammel nets (Blasi and Boitani 2014).
A local population of bottlenose dolphins along the Italian coast of the western Ligurian Sea (Fig. 1) has been studied since 2018 (Ascheri et al. In Press). A total of 123 individuals have been identified, with the majority observed only sporadically and about 20% displaying stronger, yearround site fidelity. Individuals inhabiting the study area are thought to be part of a larger population, with an abundance of the latter estimated at 248 (95% CI = 217-284) individuals (Ascheri et al. In Press).
A photographic match of an adult male bottlenose dolphin was discovered opportunistically between the three study areas, using multiple identification features, including the shape of the dorsal fin, the nicks and notches on the dorsal fin, as well as scars and tooth rake markings on the body (Figs. 2, 3). The matched dolphin (depicted with ID code PHD18 in Blasi and Boitani 2014 and nicknamed 'Lino') was first photo-identified in the Aeolian archipelago in June 2006. Lino was subsequently observed in all years during 2006-2017, except 2008, in a total of 30 sightings. His number of sightings ranged between 1 and 7 in any given year during this time. The animal was determined to be a male based on photographs of its genital area and was largely solitary or only occasionally associated with a few other males (Blasi and Boitani 2014). He was last observed on 16th July 2017 off the island of Filicudi (Italy, Fig. 1) and has not been observed in the Aeolian archipelago since.
On 8th February 2020 Lino was photographed off the town of Piran (Slovenia) in the Gulf of Trieste, northern Adriatic Sea (Fig. 1), where it had not been seen previously. Throughout the sighting, he was accompanied by a resident male, known since 2008 and sighted in the Gulf of Trieste regularly but mostly on his own (depicted as NEV in Genov et al. 2019). On 10th March 2020, Lino was seen again, this time alone, about 50-100 m from cape Madona, Piran ( Fig. 1), where he engaged in surface feeding activities.
Since then, he has not been seen again in the Gulf of Trieste to date (but NEV continued to be sighted regularly).
On 3rd September 2020, Lino was photographed off Imperia (Italy) in the western Ligurian Sea (Fig. 1), where he had not been recorded previously. He was encountered twice during the same day, first in a group of four individuals and subsequently about 3 h later in a group of 29 individuals, 11 km from the first location. Both groups contained calves and consisted of dolphins considered resident in the area. Lino has not been seen since in any of the three study areas.
Even though the dorsal fin underwent some changes throughout the dolphin's sighting history (some of them already between sightings in the Aeolian archipelago, Fig. 2), the highly distinctive dorsal fin, as well as other subtle markings on the dorsal fin and body, confirmed the match across the three study areas beyond doubt (Fig. 3). The match was also suggested by the finFindR package (Thompson et al. 2022) for program R (R Core Team 2020). To determine if any other matches existed between the three study areas, catalogues (consisting of 396, 42 and 123 individuals for the Gulf of Trieste, the Aeolian archipelago, and the western Ligurian Sea, respectively) were compared. To ensure independence, each research group independently examined the catalogue of the other research groups. In addition, the finFindR package was used to investigate the presence of potential matches, but no other matches were found between the study areas.
To estimate the distance travelled by Lino, we considered several potential scenarios (Fig. 1, Table 1). For the first movement, we primarily assumed (A) the shortest straightline path across water (depicted by the green dashed lines in Fig. 1), with two sub-options: (A-1) Lino moving through the very narrow Strait of Messina between the island of Sicily and the Italian mainland (3.1 km at the narrowest part), or (A-2) moving around Sicily instead. We also considered Fig. 1 The sighting locations, sighting dates and plausible travel routes of the matched individual, with locations cited in the text. The green lines depict the two shortest paths (one through the Strait of Messina and the other around the island of Sicily) of the first movement, from Aeolian archipelago to the Gulf of Trieste; whereas the red lines depict the generic illustration of potential 'coastal' and more shallow-water alternative routes. The second trip from the Gulf of Trieste to the western Ligurian Sea assumes the same potential return trips to the Strait of Messina and the western tip of Sicily, respectively, followed by two potential shortest paths for each (orange lines), and the 'coastal' alternative indicated in purple. The orange star in the western Ligurian Sea represents the second sighting in the same day. Note that the depicted routes represent general possibilities and indicate a hypothetical (perhaps likely) path of movement, as the actual travel path of the dolphin remains unknown 1 3 the possibility that (B) Lino did not travel across very deep areas in the Gulf of Taranto in the Ionian Sea or the southern Adriatic Sea and instead generally followed the coastline, given that the species typically inhabits relatively shallow continental shelf waters or areas around oceanic islands (Bearzi et al. 2009). The same two sub-options were considered here: (B-1) movement through the Strait of Messina or (B-2) around Sicily. Finally, for each of the latter two sub-options, we further considered two alternatives: (B-1-i, B-2-i) the most 'coastal' route along the western Adriatic coast ( Fig. 1, red lines) or (B-1-ii, B-2-ii) Lino initially following the coast, crossing the Otranto Strait and then travelling along the eastern Adriatic coast (Fig. 1, red lines). The latter travel route would be in the direction of the predominant Adriatic current, which moves in an anti-clockwise fashion, moving northward along the eastern shores of the basin (Mauri and Poulain 2001). For the return trip, we considered the same routes back to the (C-1) Strait of Messina or (C-2) the western tip of Sicily, followed by the shortest path from each ( Fig. 1, orange lines), with the addition of a (D) coastal route along the west coast of Italy ( Fig. 1, purple line) as a continuation of previous coastal alternatives. Note that multiple other alternatives exist, including avoiding deep areas but then travelling directly across the relatively shallow central and northern Adriatic Sea, or passing along the islands of Sardinia and Corsica. Not all of these were explored and the presented scenarios are simply a generic representation of potential paths undertaken by the dolphin and are not meant to indicate known or the only viable trajectories. Mapping and distance calculation was carried out using QGIS 3.18.2-Zürich (QGIS Development Team 2021).
Given the sighting locations, Lino first had to travel a minimum of 1251 km across the Tyrrhenian, Ionian and Adriatic Seas (Fig. 1, Table 1). This movement (A-1) assumes travelling through the narrow Strait of Messina, across deep waters of the Gulf of Otranto and across deep waters of the southern Adriatic Sea (Fig. 1). On the other hand, if he travelled around Sicily instead (A-2), he would have to travel a minimum of 1743 km (Table 1). Travel distances of other alternative scenarios are shown in Table 1. Lino then had to travel a minimum of 2053 km back across the Adriatic, Ionian and Tyrrhenian Seas, as well as the Ligurian Sea (Fig. 1, Table 1), nearly doubling the distance of the previous movement. This shortest-path movement (C-1) assumes the shortest possible route back to Messina Strait, and then a straight line across the deep part of the Tyrrhenian Sea to the northern tip of Corsica, and across the deep waters of the Ligurian Sea (Fig. 1). The shortest path involving a trip around Sicily (C-2) would amount to 2307 km. Other alternatives to the second movement are provided in Table 1.
Published records of long-distance (≥ 100 km) movements by bottlenose dolphins worldwide are shown in Appendix, Table A1. While it is possible that some were missed, Table A1 likely represents a rather comprehensive overview of relevant published records. This extensive review suggests that both trips reported in our study qualify as the longest recorded movement for this species in the Mediterranean Sea to date, and the second trip is the second longest movement reported worldwide. As inferred from Table A1, most of the recorded 'extremely' long-distance (≥ 1000 km) movements were performed by solitary-sociable dolphins (Nunny and Simmonds 2019), and by the offshore ecotype of this species (Wells et al. 1999). There is no evidence to suggest that Lino is a solitary-sociable dolphin, given his re-sighting history in the Aeolian archipelago, the two sightings in the Gulf of Trieste, as well as the two sightings in the western Ligurian Sea. He was observed interacting with conspecifics in an apparently normal fashion in all three areas and did not display unusual behaviour consistent with that of solitary-sociable dolphins (Nunny and Simmonds 2019). Moreover, it can reasonably be assumed that Lino belongs to the coastal or inshore ecotype. There is currently limited evidence for the presence of a pelagic or offshore ecotype in the Mediterranean Sea (Bearzi et al. 2009;Gaspari et al. 2015b) and there is no reason to believe Lino belonged to such an ecotype.
The only documented movement of free-ranging, nonsolitary-sociable, 'coastal' ecotype bottlenose dolphins comparable to the one reported here, is one by Robinson et al. (2012), of one individual (among eight reported by  the study) re-sighted 1277 km apart, between Scotland and Ireland. All other documented movements in Europe were substantially shorter, with the exception of oceanic movements of up to 1000 km between Macaronesian archipelagos in the eastern North Atlantic Ocean (Dinis et al. 2021). The first movement reported here (1251 km) exceeds the existing longest recorded movements by this species in the Mediterranean Sea by more than twice, whereas the second one (2053 km) exceeds it by more than 4x (Appendix , Table A1). Moreover, the second movement, between the northern Adriatic Sea and the western Ligurian Sea, appears to be the second longest documented movement for any individual of this species worldwide, and the longest reported for any coastal or inshore bottlenose dolphin (Appendix, Table A1). The only documented movement longer than the one reported here was one of a satellite-tagged offshore ecotype bottlenose dolphin in the western North Atlantic Ocean, following a live stranding event and subsequent rehabilitation in captivity lasting 39 days (4200 km, Wells et al. 1999). Another movement reported from the same study, of 2050 km, was of a satellite-tagged 'intermediate' form (intermediate between inshore and offshore ecotype), also following a live stranding event and subsequent rehabilitation of 85 days (Wells et al. 1999). Since both of those individuals were satellite-tagged, their paths were tracked in nearly real time. The live stranding event and the subsequent rehabilitation, handling and tagging may or may not have affected their behaviour and movements (Wells et al. 1999). In the case presented here, however, the distances of 1251 and 2053 km are based on the shortest possible path and it is possible (or even likely) that Lino made substantially longer trips than that (Table 1), whereas its movements can be considered independent of any human care. The causes of the movements reported here can only be speculated upon. Whether Lino left his former home due to decreased quality or carrying capacity of the local habitat (Grant 1978), or due to other reasons, such as competition for mates (Dobson 1982;Lawson Handley and Perrin 2007), remains unknown. Encounter rates in the Aeolian archipelago have declined between 2005 and 2011, suggesting a decline in overall abundance across the study area (Blasi and Boitani 2012). Encounter rates were also found to correlate with the presence of trammel nets (Blasi et al. 2015), which may lead to injuries resulting from retaliation by fishermen or entanglement in this fishing gear (Leone et al. 2019;Bruno et al. 2021).
The fact that Lino has been determined to be a male is consistent with the general notion that males tend to be the dispersing sex, both in this genus (Krützen et al. 2004;Möller and Beheregaray 2004;Bearzi et al. 2011) and in mammals generally (Greenwood 1980;Dobson 1982;Lawson Handley and Perrin 2007), although genetic evidence from the Mediterranean Sea and the North Atlantic Ocean has suggested that female bottlenose dolphins may be equally important (Natoli et al. 2005;Rosel et al. 2009) or potentially even more important (Gaspari et al. 2015a) in promoting gene flow. Kin cooperation and familiarity with the natal area, particularly in relation to potential foraging specialisation, may oppose sex-biased dispersal and promote natal philopatry (Lawson Handley and Perrin 2007;Rosel et al. 2009). On the other hand, the potentially diminished quality of local habitat in the Aeolian archipelago and the declining dolphin abundance there (Blasi and Boitani 2012) due to fishing pressures or other reasons (Bruno et al. 2021) may have promoted dispersal, particularly if the individual was largely solitary. The tendency of being largely solitary or only interacting with males for short periods of time in the Aeolian archipelago (Blasi and Boitani 2014) is interesting to note. During his first sighting in the Gulf of Trieste Lino was also observed with a local male bottlenose dolphin, who tends to be rather solitary himself (Genov et al. 2019), and subsequently observed alone. In the western Ligurian Sea, however, Lino was seen in a rather large group including females with calves.
It is unknown how long Lino stayed in the Gulf of Trieste, but he was not seen again after his last sighting in March 2020. Roughly 200 bottlenose dolphin sightings were recorded in the Gulf of Trieste since then, with group sizes up to 45 individuals (Morigenos, unpublished data), but Lino was not among them. Likewise, he has not been observed in any of the three study areas since his last sighting in the western Ligurian Sea.
The existing evidence from photo-ID, genetics and ecological markers such as stable isotopes largely suggests that most local populations or subpopulations of bottlenose dolphins in the Mediterranean Sea are largely discrete or at least somewhat well-defined units (Natoli et al. 2005;Genov et al. 2009;Gnone et al. 2011;Gaspari et al. 2015aGaspari et al. , 2015bGonzalvo et al. 2016;Giménez et al. 2018;Pleslić et al. 2019). However, bottlenose dolphins in the region have also been documented undertaking movements over substantial distances (Bearzi et al. 2011;Gnone et al. 2011;Genov et al. 2016). Such movements, and particularly the two extreme cases presented in this paper, challenge the traditional view of such relatively discrete units, especially considering the short time frame of the second movement. This may partly explain some of the patterns observed through population genetics, such as apparent close genetic relatedness among some individuals from distant locations within the Mediterranean Sea (Gaspari et al., unpublished), and may help place the results of genetic analyses into proper context. While the natal philopatry and site fidelity appear to be a marked feature of most populations, it may be that bottlenose dolphins make substantial movements more often than currently recognised, and such movements are likely to promote gene flow. Their perceived absence, or rather the paucity of reports indicating long-distance movements, may be due to the lack of consistent sharing of photo-ID catalogues. While this type of collaboration is generally encouraged, it seldom happens in practice. There may be reluctance because the process can be tedious or due to concerns about confidentiality and data ownership. Arguably, exchange may not happen simply due to the expectation that matches among distant areas are unlikely. However, such attempts are worthwhile. For instance, a sighting of a common dolphin (Delphinus delphis) in the northern Adriatic Sea, a region where the species had become extremely rare (Bearzi et al. 2004;Genov et al. 2021), prompted image comparison with the Ionian Sea (Bearzi et al. 2008), resulting in the longest movement reported for this species worldwide (Genov et al. 2012). Other examples include matches of humpback whales (Megaptera novaeangliae) between ocean basins (Robbins et al. 2011;Stevick et al. 2011Stevick et al. , 2013Acevedo et al. 2022). Ideally, various approaches, including photo-ID and population genetics, should be combined as much as possible to infer movement rates and connectivity of bottlenose dolphins and other species in the Mediterranean Sea and elsewhere. The existing and ongoing advances in automated matching software (such as the finFindR package used here) may facilitate the exchange and matching of catalogues in the future. Photo-ID matching is a labour-intensive and time-consuming endeavour, but automated approaches (see Karczmarski et al. 2022 for several examples) can potentially eliminate some of the common obstacles to efficient data sharing by making the process easier and more time-efficient.
It remains to be seen if Lino is sighted again in any of the three study areas or elsewhere in the future. Regional photo-ID comparisons may provide further insights into the movements of this and other dolphins and will be helpful in improving our understanding of movements of bottlenose dolphins and other cetaceans across the region and worldwide (Carpinelli et al. 2014). This in turn may be relevant to the Conservation Management Plan for the Mediterranean Sea bottlenose dolphin and other species, currently being developed in the framework of the Agreement on the Conservation of Cetaceans of the Black Sea, Mediterranean Sea and contiguous Atlantic Area (ACCOBAMS).
This study further demonstrates the utility of photo-ID to study marine mammal movements. It also shows that our understanding of the behaviour and movement patterns of these animals is still limited, and highlights the importance of regional data sharing in understanding cetacean ranging patterns, connectivity and population structure. We hope it will serve as a simple yet valuable example and encouragement for a more frequent collaborative effort in field studies, especially if they take place in relative geographic proximity.

Appendix
See Appendix Table A1. Only values larger than 100 km are listed, even if some animals from the study populations performed shorter movements. When distances were not provided directly in the published papers, they were derived from Google Earth based on geographical locations and descriptions in the text and are therefore approximate. In cases of solitary-sociable dolphins where multiple movements were performed, only maximum distances are reported. 'Photo-ID' is listed as the method for both 'standard' photo-ID studies as well as cases of solitary-sociable dolphins, as it was assumed that some form of photo-ID or visual identification had to be carried out for the latter.
Note that the study by Mate et al. 1995 has been cited in some of the existing literature as an example of long-distance movements. However, that study reported cumulative distance of multi-directional movements of a tagged dolphin within Tampa Bay (Florida), which therefore does not constitute a long-distance movement.  Camphuysen and Peet, (1996) One confirmed (135 km) and two possible (325 km) long-distance movements Gladilina et al. (2018)  Kotnjek and Natalija Žlavs. We also thank the local fisherman Rok Domnik for reporting his dolphin sighting to Morigenos, which allowed an immediate response and led to the second sighting of this individual in the area. We thank all volunteers of Filicudi WildLife Conservation that assisted with the field data collection and analyses that led to the available information about this individual in the Aeolian archipelago. We thank all the Delfini del Ponente team members and volunteers who participated in the field work and sorting of the photographs involved in this matching, in particular Enrico Carta. MFB's participation at the meeting of the ACCOBAMS-ASCOBANS Joint Bycatch Working Group was in the framework of the Life DELFI project, co-financed by the European Community under the LIFE programme. We thank Laetitia Nunny for providing a difficult to find piece of literature. The authors would also like to thank ACCOBAMS and ASCOBANS Secretariats, in particular Jenny Renell, as well as the co-chairs of the ACCOBAMS-ASCOBANS Joint Bycatch Working Group, Ayaka Amaha Öztürk and Peter Evans, for organising the online meeting that eventually led to the findings reported here. Finally, we thank the Editors Leszek Karczmarski and Stephen C.Y. Chan, and one anonymous reviewer, for constructive comments that helped improve the manuscript.
Author contributions TG: made the initial match and designed the study. TG and JŽ: carried out detailed matching. DA: made the subsequent match. All authors confirmed the match and the lack of other matches among catalogues. JŽ: carried out matching in finFindR. TG, MFB and EF: collected photographs in the field. JŽ: carried out the mapping and distance calculations in GIS with input from TG. TG: coordinated the manuscript and was responsible for writing the draft manuscript. All authors carefully reviewed and approved the final manuscript.

Conflict of interest Authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.