In most wildlife populations, sex and age have profound effects on the chances of an individual dying, or producing offspring (Neal, 2004), thus these factors are routinely considered when studying population structure and dynamics (Ojasti and Dallmeier, 2000). Other factors with the potential to influence population structure are site fidelity (SF) and philopatry (Bose et al., 2017). SF has been defined as the “tendency to return to a previously occupied location” (Greenwood, 1980; Switzer, 1993), whereas philopatry is the tendency to stay in a familiar environment, mainly related to the natal habitat (Begon et al., 1995). Several environmental factors have been suggested to influence species’ site fidelity and/or philopatry, such as habitat stability, predictability of reproductive failure, variability in territory quality within a habitat, and population pressure; individual characteristics may also include an individual’s previous reproductive success, age, and knowledge of other sites (Switzer, 1993).

In cetacean studies, SF and residency patterns are evaluated using information on the number of recaptures, the duration of stay, and/or the average recurrence of individuals (see Ballance, 1990 and Morteo et al., 2012b). In this regard, Ballance (1990), stated that “a high number of resightings, a long period time between the first sighting and the last resighting, and a short time interval between adjacent sightings describes an animal with a high degree of residence.” Thus, individual residency patterns can be considered as a manifestation of SF and philopatry.

For cetacean social species such as delphinids, other factors that are thought to shape population and social structure are kinship and reproductive condition (Bigg, 1982; Wells et al., 1987). Sex and age also influence the way that delphinids use their habitats. In general, females try to ensure the survival of their offspring, thus their distribution is often closely related to the quality and/or accessibility of food resources and habitats that have a lower risk of predation; conversely, the distribution of males is often related more to female distribution than to the availability of food (Gowans et al., 2007). Delphinids routinely travel throughout the day and can be exposed to large areas, their entire home range, or even multiple habitats on a daily basis (Gowans et al., 2007). Consequently, individual home ranges -that may depend on the age and sex of the individual- result in structured societies that are usually stable over time (Wells et al., 1980, 1987), in which females tend to be more resident to specific areas than males (Gowans et al., 2007).

Since Ballance’s (1990) work, several methods have been used to evaluate cetacean residence and this lack of standardization hinders proper comparison among SF studies (Tschopp et al., 2018, see below). Tschopp et al. (2018) found that, in general, the methods to quantify site fidelity include three approaches: (1) proportions (a ratio between the number of sightings or resightings, and a measure of effort), (2) categories (e.g., high, moderate, low), and (3) models (maximum likelihood methods). Recently, Tschopp et al. (2018) developed and compared the performance of a series of standardized indexes -based on Ballance’s parameters- to be able to quantify the site fidelity degree and ensure accurate comparability across related investigations. Regarding delphinids, several authors (e.g., Gowans et al., 2007; Wells and Scott, 2018) have predicted that in offshore environments, where food availability is patchy and unpredictable and predation risk might be high, dolphins range more widely and form larger groups to forage on sparsely distributed prey schools and to reduce predation. In contrast, in complex inshore environments with predictable resource availability and potentially lower predation, dolphins are predicted to remain resident in relatively small areas and to form smaller groups to avoid feeding competition. This seems to be the case for the bottlenose dolphin (Tursiops truncatus) populations inhabiting semi-enclosed bays or sounds, or inshore habitats in the Gulf of Mexico (GoM). This species has been widely studied in the northern GoM on the basis of photo-ID methods and capture-mark-recapture (CMR) models (see review in Vollmer and Rosel, 2013). In the northern GoM, where long-term, year-round residents are best documented, there is also evidence for seasonal changes in abundance; most often these are thought to result from movements of seasonal, short-term residents and/or transients (very short-term visitors) to the area. It is generally thought that these shorter-term residents and visitors come from the adjacent coastal population, although comprehensive and directed studies to address this question have yet to be performed (Vollmer and Rosel, 2013).

The bottlenose dolphin has been widely documented in some localities in the southern GoM (Ortega Ortiz, 2002; Ortega Ortiz et al., 2004; Ramírez-León et al., 2020), including waters off the Tamiahua Lagoon System (Galindo et al., 2009; Martínez-Serrano et al., 2011; Valdés-Arellanes et al., 2011); Nautla (Morteo et al., 2019); and the Alvarado Lagoon System (ALS) in Veracruz state (Morteo, 2011; Morteo et al., 2012b, 2014, 2017, 2019; Morales-Rincón et al., 2019), and some coastal areas in the states of Tabasco, Campeche and Yucatán (Delgado-Estrella, 2015).

Off the ALS, Morteo et al. (2014) found that this population is sexually segregated, where females were more resident, had higher site fidelity, and had weaker associations with a higher number of partners than males. On the other side, males were assumed to be primarily responsible for gene flow among adjacent locations (Morteo et al., 2014).

Capture-mark-recapture (CMR) studies have been used as a general sampling and analysis method to assess population status and trends in many biological populations (White and Burnham, 1999). The photographic documentation of long-lasting natural marks on the dorsal fin or flukes (i.e., photo-identification, photo-ID) has been used since the early 1970’s to study aspects such as group structure, site fidelity, movement patterns, and abundance using CMR models in cetaceans (Würsig and Jefferson, 1990). Models usually require a set of restrictive assumptions about the properties of the population under study, such as individual homogeneity in capture and survival probabilities, among others (Krebs, 1972). Failure to comply with these assumptions implies that the model does not adequately fit the data, and thus may introduce severe bias in parameter estimates (Krebs, 1972). In structured delphinid populations, heterogeneity in capture or survival probabilities is often produced by age, sex, or size of the individuals (Williams et al., 1993), as well as inconspicuous marks, low photographic quality, and social bonds (Morteo et al., 2012a).

In recent studies, researchers started to include site fidelity and residency pattern as factors to account for heterogeneity in the data when studying demographics or population ecology of several cetacean species, including the southern Australian bottlenose dolphin (Tursiops cf australis, Zanardo et al., 2016; Passadore et al., 2017), the Australian humpback dolphin (Sousa sahulensis, Hunt et al., 2017), the Risso’s dolphin (Grampus griseus, Carlucci et al., 2020) and the fin whale (Balaenoptera physalus, Schleimer et al., 2019). These researchers calculated sighting rates and site fidelity indexes for each individual and then used agglomerative clustering methods (Legendre and Legendre, 1998) to identify clusters of individuals with similar degrees of site fidelity, before running CMR analysis. In particular, Haughey et al. (2020) used for the first time Tschopp et al. (2018) standardized indexes to stratify the sample and estimate population parameters of Indo-Pacific bottlenose dolphin (T. aduncus) off western Australia by residency pattern.

In this paper, we used the recently developed standardized site fidelity index (SSFI) IH₄ (Tschopp et al., 2018) to study potential differences in population parameters between resident and non-resident common bottlenose dolphins (Tursiops truncatus) that use the open, coastal waters off Alvarado Lagoon System, south-western Gulf of Mexico. On the basis of our knowledge of the study area, we hypothesized that bottlenose dolphins using the marine coastal waters adjacent to the ALS are part of an open population within which exists a core community (sensu Wells et al., 1987), that behaves as a closed population, in the sense that it is composed of all the groups and individuals “… that share large portions of their ranges and interact with each other to a much greater extent than with members of similar units in adjacent waters, but genetics exchange occurrs between communities.”

For this study, we used the photo-ID catalog and database from the Marine Mammal Lab at Universidad Veracruzana (LabMMar-IIB-ICIMAP). This database includes the sighting histories of 231 bottlenose dolphins from 75 survey trips conducted between May 2006 and August 2010 (Morteo, 2011; Morteo et al., 2012b, 2014, 2017). Previously, Morteo et al. (2014, 2017) used a partial subset of these databases, covering the period May 2006–April 2008, to explore the abundance and social ecology of this dolphin population under open population modeling by considering the population as a homogeneous unit, and without considering temporal variations or population structure in the parameters. Detailed definitions of methodological terms are presented in Supplementary Material 1.

Between May 2006 and August 2010, we conducted 75 surveys, totaling 6,328.8 km surveyed, 288.1 h of search effort, and 110.4 h photographing dolphin groups (Table 1). Search effort ranged from 3.8 to 8.15 h day⁻¹ (x̄ = 3.6 ± 1.1 SD) and the full study area was covered in each and all survey trips. In general, the effort was greater during the first two seasons, but we found no evidence of a difference in search effort among seasons (Kruskal-Wallis, H = 0.91, p = 0.63). Photo-ID effort (h day^–1) ranged from 0.0 to 3.6 (mean 1.4 ± 0.8 SD), and no seasonal differences were found (Kruskal-Wallis, H = 10.55, p = 0.10).

Two hundred and sixty-three groups were encountered totaling 2,320 dolphins sampled, of which 231 distinct individuals were identified. We processed 30,402 pictures of which 10,958 (36%) were useful for photo-ID. Group size averaged 9.0 individuals (± 11.2 SD, range: 1–100), and remained similar among seasons (Kruskal-Wallis, H = 2.62, p = 0.27). The number of recaptures for marked dolphins ranged from 0 to 12 occasions, where 36.8 % (n = 85) were transients (Figure 2).

Population structure of highly mobile marine organisms can be complex and difficult to study, but it is important to understand how individuals within a population partition their environment, in order to better address conservation challenges (Vollmer et al., 2021). In this regard, recent advancements in the implementation of standardized site fidelity indexes (Tschopp et al., 2018) and considerations on residency patterns when studying cetacean population dynamics (e.g., Zanardo et al., 2016; Hunt et al., 2017; Passadore et al., 2017; Schleimer et al., 2019; Carlucci et al., 2020; Haughey et al., 2020) provide improved methodological tools for dealing with heterogeneity in capture or survival probabilities inherent to the dynamics of social species with fission-fusion societies such as bottlenose dolphins (Connor et al., 2000). This is the first study to explore the population dynamics of the common bottlenose dolphin (T. truncatus) in the Gulf of Mexico based on its residency patterns, and to provide quantitative evidence of the existence of a “core community” (sensu Wells et al., 1987) in an open, coastal habitat. As this study is an extension of previous work by Morteo (2011) and Morteo et al. (2014, 2017), we included new data on previously unknown parameters, improving the temporal coverage as well as a more comprehensive analytical approach that yielded more refined estimates, and a more robust assessment of the population structure.

Due to the longevity of bottlenose dolphins, we acknowledge that short-term studies such as this one may not encapsulate the population processes related to factors such as mortality or emigration (Haughey et al., 2020). Despite the high number of transient individuals in the sample (Figure 2), the asymptotic trend in the discovery curves (Figure 4) indicated that most of the marked individuals in the population had been photographed by the end of the study period. For the full sample, the existence of two plateaus and decrease in the discovery rate of new individuals after the rainy season of 2009 suggest that the addition of new members is low, with sporadic immigration pulses as occurs in this, and other areas within the southern Gulf of Mexico (e.g., Martínez-Serrano et al., 2011; Valdés-Arellanes et al., 2011; Morteo et al., 2014, 2017, 2019; Delgado-Estrella, 2015). On the other side, the similarity between the discovery curves of the whole sample, occasional visitors, and transients showed that the dynamics of this population is mainly regulated by the influx of non-resident individuals.

Catchability of RR (mean 0.67 ± 0.27 SD) individuals was greater than for OR (0.35 ± 0.20) and OV (0.32 ± 0.27 SD) individuals, and thus abundance estimates for this cluster will be more precise (Williams et al., 2002). In general, catchability should be directly related to research effort alone but, in this case, it probably reflects that regular residents are more catchable by unit of effort than individuals in any other clusters (Williams et al., 2002).

High values of annual survival for the core community (cluster RR + OR) (1.00, 95% CI: 1.00–1.00) contrast sharply to those of occasional visitors: (0.75, 95% CI: 0.59–0.87). As transients were not included in the analysis, and RR and OR clusters behave as a closed population, these highest values for the core community are most probably the result of both a high survival rate and a very low emigration rate.

For long-lived species with complex life history processes, such as cetaceans, adult survival is expected to be high and variable with age, sex, and individual fitness (Ralls et al., 1980). Higher survival rates for resident individuals were indirectly assumed by Morteo et al. (2012b), who showed that these animals interact less frequently with fisheries, when compared to non-residents, thus facing lower exposure to entanglements in fishing gear and retaliation measures by fishers. Therefore, although much lower survival estimates for the visitor fraction is largely based on a high proportion of dolphins permanently leaving the study area shortly after the sampling, it is likely that higher mortality for non-residents is also due to a greater risk of predation by sharks and higher risk of entanglement in fishing gear (Morteo et al., 2012b, 2014, 2017; Rechimont et al., 2018; Morales-Rincón et al., 2019). The latter is supported by reports from recently stranded animals – recorded between 2002 and 2019 (decomposition code 1–3, according to Geraci and Lounsbury, 1993) – showing that fisheries-related injuries only occurred in individuals that had not been previously photographed in the study area (Fuentes Del Muro and Morteo, unpublished data, LabMMar-UV).

Survival values recorded for regular and occasional residents in this study are higher than those reported elsewhere (e.g., Wells and Scott, 1990; Speakman et al., 2010; Daura-Jorge et al., 2013; Tezanos-Pinto et al., 2013; Fruet et al., 2015; Vermeulen and Bräger, 2015; Vermeulen et al., 2017; Methion and Díaz López, 2018), probably because of differences in the way we stratified the sample by residency pattern. It is important to notice that the null standard error in the point estimate of survival for both RR and OR is commonly attributed to problems with the data (Cooch and White, 2019, p. 6–24); nevertheless, because of the stratification by residency type, it probably reflects true high values and precision of the estimates due to effective population closure within the study timeframe (i.e., 5-y).

In general, coastal bottlenose dolphins are known to exhibit a wide spectrum of residency patterns, which include transients, seasonal migrants, year-round residents, and a combination of occasional long-range movements, and repeated local residency (Morteo et al., 2014; Wells and Scott, 2018). This seems to be the case for the population in our study area, where Ruíz-Hernández (2014) and Morteo et al. (2019) recorded a limited exchange of individuals between coastal waters off ALS and two northern locations (the Veracruz Reef System and Nautla, see Figure 1), for which dolphins need to travel at least 100 and 230 km, respectively. Our results are similar to other contributions made to the south of our study area, where it was documented that some individuals traveled 270 km on average in a few months, including a dolphin that traveled more than 800 km between the states of Quintana Roo and Tabasco (Delgado-Estrella, 2015, see Figure 1).

Differences in recruitment rate and timing between the core community (RR + OR) and visitors, as found in this study, could correspond to sex-related movement patterns, as reported previously in the Gulf of Mexico, the Caribbean (e.g., Urian et al., 2009; Caballero et al., 2012; Wells, 2014) and also in the study area, through greater residency in females and larger dispersion by males (Morteo et al., 2017, 2019). In this sense, natal site philopatry of both sexes is common and associated to age, as it occurs in the Sarasota population (Wells, 2003, 2014; Sellas et al., 2005). In the northern Gulf of Mexico, molecular data also suggest that some females may move and breed among different communities (Duffield and Wells, 2002), but implications at a local level remain to be investigated.

The existence of two plateaus in the discovery curves of the full sample, OV and OR individuals, indicates at least one occasional pulse of individuals entering from the superpopulation to the study area. These results are consistent with previous findings (Morteo, 2011; Morteo et al., 2014, 2017), suggesting that it is an open population, but the rate of incorporation of new individuals is low in the medium-long term.

Comparable absolute abundances of several estuarine bottlenose dolphin communities around the world are often estimated between 60 and 150 individuals (e.g., Williams et al., 1993; Wilson et al., 1999; Wells, 2003; Balmer et al., 2008; Vermeulen and Cammareri, 2009; Fruet et al., 2011, 2015; Félix et al., 2017). Monthly averages of daily abundances by Morteo et al. (2017) in the study area were about 125 dolphins under the Jolly-Seber model. This number looks consistent with our estimates both within and between years, but results are not directly comparable because of differences in the treatment of the samples.

Our results support the assumption that the dolphin population that uses the coastal waters off ALS is open (Morteo et al., 2017), but emphasizes the existence of a core community of resident individuals that, occasionally, receives an influx of individuals from neighboring waters, with no apparent seasonal trend. It is noteworthy that, unlike other study areas, this core community -living in an open habitat- largely behaves as a closed population. This situation may be more common than previously thought for the species across the Gulf of Mexico, where many individuals remain in relatively small but well provisioned areas, whereas short-term residents and visitors come from the adjacent coastal populations, or have nomadic habits in their constant pursuit of food and mates (Shane, 1980; Irvine et al., 1981; Wells et al., 1987; Ruíz-Hernández, 2014; Delgado-Estrella, 2015; Morteo et al., 2019, among others). The existence of such core communities in the northern Gulf of Mexico has been generally established based on qualitative criteria, mainly the overlap in the presence of transients with year-round and seasonal residents (i.e., Shane, 1980, 1990; Wells et al., 1987; Balmer et al., 2008; Tyson et al., 2011, among others) and, in most cases, refer to dolphin populations living in inshore, estuarine habitats. However, in a coastal habitat adjacent to the Sarasota Bay, Fazioli et al. (2006) found that the dolphin community that prefer the Gulf of Mexico is primarily composed of transients, seasonal residents an individuals with a home range greater than the study area, with fewer year-round residents; nevertheless, details on the process to compute individual residency, and quantitative approaches on the classification of such individuals based on the dynamics of the population is generally lacking.

The permanent presence of dolphin groups and the existence of a core community in the coastal waters off ALS are probably related to a predictable supply of prey and a sheltered environment. Interestingly the dynamic flux of individuals, abundance at a local level (i.e., the core community) seems stable over time, which suggests that a sort of carrying capacity effect is in place. High site fidelity and/or restricted ranging patterns are likely driving population parameters for the core community of bottlenose dolphins in the coastal waters of Alvarado; as this could increase their chances of living in a provisioned habitat, it could also make them more prone to detrimental effects by both documented and currently unknown local threats. Future work should focus on exploring at a finer scale the relationship between these population parameters, the structure of the population (sex/age), and relevant environmental variables for the species.

Differences in residency patterns for the dolphins in our study area could be explained by the socioecological model of Gowans et al. (2007) such that, as resources are spatially and temporally predictable, dolphins remain resident in relatively small areas (Morteo et al., 2014, 2017). Conversely, competition with local dolphins may lead non-resident individuals (mostly males) to range widely to find sufficient resources and mating opportunities (Gowans et al., 2007). This strategy helps to prevent inbreeding (Caballero et al., 2012), and would cause male dolphins of this population to prey on other species that may be available off river mouths and estuaries along coastal waters off the southwestern Gulf of Mexico (Martínez-Serrano et al., 2011; Morteo et al., 2014, 2017, 2019). As the existence of many resident communities has evidenced limited genetic exchange among adjacent sites all along the Gulf of Mexico (Sellas et al., 2005; Caballero et al., 2012; Vollmer and Rosel, 2013), this raises the question of a probable structure at metapopulation level, which should also be investigated.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. The reduced m-array produced by the MARK output, that contains summary information on the numbers of individuals released at each occasion, and is the basis for the estimation of parameters, is presented in Supplementary Material 3 for reproducibility and replicability.

This animal study was accomplished through federal permits SGPA/DGVS/00351/06, 01407/08 (EM) and SGPA/DGVS/00870/07, 02788/07, 01344/08, and 01649/08 (M. C. Bazúa) issued by the Subsecretaria de Gestion para la Proteccion Ambiental.

JB-J: conceptualization, formal analysis, investigation, methodology, validation, visualization, and writing – original draft. EM: conceptualization, data curation, funding acquiring, methodology, project administration, resources, supervision, validation, and writing – review and editing. CD-A and JB-P: methodology, supervision, validation, and writing – review and editing. PF and ES: conceptualization, methodology, supervision, validation, and writing – review and editing. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.