First Characterization of Nesting Behaviors of Leatherback Turtles (Dermochelys coriacea) and Hawksbill Turtles (Eretmochelys imbricata) in Martinique and Inter-Species Comparison

Simple Summary

Sea turtles adopt specific nesting strategies to maximize the survival of their hatchlings. Few studies have examined the nesting behavior of leatherback turtles (Dermochelys coriacea) and hawksbill turtles (Eretmochelys imbricata). The aim of this study was to describe the nesting behavior of these two species, taking into account human and natural impacts, and to compare their behaviors. Nocturnal monitoring was carried out from 2020 to 2024 on three beaches in Martinique: Madiana, Diamant, and Salines. For leatherback turtles, the average nesting time was 101.50 min, with camouflage behavior dominating (27.06% of total time). For hawksbill turtles, nesting duration was 109.49 min, with digging the main activity (31.68% of total time). Human disturbance had no effect on leatherback behavior, whereas hawksbill behavior was strongly influenced by light and the presence of roots. The results showed behavioral differences between these two species and also with other species, such as loggerhead turtles (Caretta caretta) and green turtles (Chelonia mydas). These differences could be explained by morphological variations, particularities of their nesting habitat, beach characteristics, and different evolutionary strategies. Further research is needed to better understand these differences.

Abstract

Sea turtles use specific nesting strategies to maximize the survival of their offspring. Few studies have investigated the nesting behavior of leatherback (Dermochelys coriacea) and hawksbill (Eretmochelys imbricata) turtles. The aim of this study was to characterize the nesting behavior of these two species, taking into account anthropogenic and natural impacts, and to compare behavior between species. Nocturnal monitoring was conducted from 2020 to 2024 on three beaches in Martinique: Madiana, Diamant, and Salines. The average duration of the nesting sequence for leatherback turtles was 101.50 min, with camouflage being the dominant behavior for 27.06% of the total time. For hawksbill turtles, nesting duration was 109.49 min, with digging as the predominant activity (31.68% of total time). No effect of anthropogenic disturbances and obstacles on the behavior of leatherback turtles was observed, while that of hawksbill turtles was significantly influenced by light and the presence of roots. The results revealed behavioral differences between these two species, as well as between two other species: loggerhead (Caretta caretta) and green (Chelonia mydas) turtles. These differences could be explained by morphological differences, nesting habitat peculiarities, beach characteristics, and different evolutionary strategies. Further research is needed to better understand these behaviors and improve conservation efforts.

1. Introduction

Sea turtles have specific nesting strategies designed to maximize the survival of their offspring [1]. The female lays her eggs on the beach and carefully selects a precise location to dig her nest. The choice of nesting site is influenced on several levels [2]. First, turtles return to the same site where they were born: philopatry [3]. Then, they choose a site according to favorable conditions: this is the phenomenon of fixity [4]. Environmental parameters such as lunar activity [5], tides [6,7], and weather conditions [8] influence the turtle’s ascent to the beach. Finally, once on the beach, turtle nesting site selection is influenced by abiotic factors such as temperature, humidity, sand grain size [9], algae and cover [10,11], and biotic factors such as predation [12]. Anthropogenic disturbances associated with artificial lighting [13,14] and human activities on the beach [15] may also be sources of changes in oviposition success.

After laying her eggs, the female covers and disguises the nest before returning to the water. Sea turtles lay multiple eggs in a single nesting season and then return to lay eggs every two to five years, depending on the species [16]. Each sea turtle species shows variation in the behavioral sequence of egg laying, but there are seven main stages: ascent to the beach, body pitting, nest digging, egg laying, nest covering, camouflage, and return to the water [17,18,19,20,21]. Ethograms have been conducted in loggerhead sea turtles (Caretta caretta) [19] and green turtles (Chelonia mydas) [20]. Hailman and Elowson [19] described the timing of various behaviors in loggerhead turtles. In fact, the loggerhead turtle spends an average of 12.90 min digging its nest, 12.30 min laying eggs, and 12.60 min camouflaging. The total average time for the nesting sequence is 63.00 min. In comparison, according to Lindborg et al. [20], the green turtle spends an average of 29.30 min digging its nest, 22.30 min laying eggs and 62.30 min performing its camouflaging. The average total time of the sequence, excluding ascent and descent, is equivalent to 145.60 min. Despite similar behavioral stages, the time spent on each behavior differs between the two species. For the leatherback turtle (Dermochelys coriacea), there are very few recent studies of nesting behavior, and most of the available studies are quite old. Carr and Ogren [22] described the nesting process for one individual in Costa Rica as follows: 3 min to find its nesting site, 17 min for body pitting, 5 min to dig the nest, 15 min to lay their eggs, 5 min to cover the nest, 40 min to camouflage, and 8 min to return to the water, for a total of 93 min. Bacon [23] also reported an approximate nesting time of 90 min for leatherbacks in Trinidad and Tobago. In addition, Pritchard [24] observed a total nesting time of 101.50 min for one turtle. However, no study has described the duration of the different nesting stages for the hawksbill turtle (Eretmochelys imbricata).

The leatherback turtle is the largest species of sea turtle, ranging in size from 1.7 to 2 m [25]. They choose a site with a thick layer of sand to lay their eggs [26]. Each clutch typically consists of approximately 100 eggs [24]. Leatherback females lay an average of 6.17 times per breeding season [27], with an interval of approximately 10 days between clutches [24]. The nesting season in Martinique of leatherbacks extends from February to August [28], and they are less faithful to their nesting site than other species [28]. Nesting occurs with a periodicity of 2.28 years [27].

The hawksbill turtle has a carapace measuring approximately 90 cm (curved carapace length; CCL or straight carapace length; SCL) [29]. It prefers to lay its eggs in vegetated areas such as forest edges or forests [29]. Each clutch contains an average of 150 eggs [30]. Females can lay four to five times in a single laying season [28], with an average interval of about 10 days between each laying [31]. The nesting season typically lasts from June to September [28]. Hawksbill turtles exhibit high fidelity to nesting sites [32]. Hawksbill nesting follows a periodicity of about 3 years [30]. Both species are listed on the IUCN Red List of Threatened Species [33,34]. The hawksbill turtle is listed as critically endangered [33], while the leatherback turtle is listed as vulnerable [34].

Each year, Martinique’s beaches are nesting sites for these two species. The green turtle also frequents the island’s beaches, but very few nests have been recorded [35]. In Martinique, turtles cannot avoid the many physical threats to their nesting sites. Loss of nesting area, loss of vegetation, light pollution, substrate compaction, and disturbance all contribute to the alteration and loss of marine turtle nesting habitat in Martinique [36].

Given their critical status on the IUCN Red List of Threatened Species and the threats to their nesting habitat, an in-depth knowledge of the behavior of these turtles is crucial for their conservation. However, the different stages of the nesting behavioral sequence have been little studied in these two species in Martinique. The development of behavioral knowledge could make it possible to measure behavioral changes in relation to anthropogenic and natural factors [20]. Our study aims to answer the following question: What are the nesting behavioral characteristics of leatherback and hawksbill turtles? The first objective of this study is to better characterize the nesting behavior of leatherback and hawksbill turtles and to highlight the differences between these two species. The second goal is to establish relationships between behavior and other data such as anthropogenic disturbance, natural obstacles, and egg numbers. To achieve these goals, leatherback and hawksbill turtles will be monitored at night on the beaches of Martinique to collect behavioral data.

2. Materials and Methods

2.1. Study Sites

Nesting monitoring was carried out on three different beaches: Madiana beach in Schoelcher, Diamant beach in Diamant, and Salines beach in Sainte-Anne (Figure 1). Each of these beaches has different anthropogenic and environmental characteristics. Salines beach, part of the commune of St-Anne, is a 1.3 km-long beach of fine white sand on the island’s south Atlantic coast. Most of the beach has no slope; only the last 200 m to the west have a slope that varies regularly with the tides and swell. The beach is relatively popular with tourists. Its sandy barrier beach is 1 to 20 m wide, depending on the beach area. The forest is 15 to 35 m wide. A path behind the forest runs along part of the beach for 900 m. Restaurants and snack bars, open only during the day, stretch for around 200 m along the back beach. Diamant beach, in the commune of Le Diamant on the south Caribbean coast, is a 2.5 km-long black sand beach. The sandy zone, with no slope present, is more regular along the beach than that of Plage des Salines. It is around seven meters wide along almost the entire beach. This is a relatively man-made beach, with houses, restaurants and bars stretching for 800 m along the backshore. Some bars and restaurants are also open at night. A road runs along the entire length of the beach, past a forest 60 to 100 m wide in the unmanaged zone. Madiana beach is located in the central Caribbean sector. It belongs to the commune of Schoelcher. It is a gray-black sand beach, with no slope, and much more anthropized than the two previous beaches. It is located directly in a large town and hosts a back-beach restaurant open for lunch and dinner. Beach racquet parties are organized one evening a week. A large spotlight is then switched on. There is also a permanently lit street lamp at the northern end of the beach. The beach is fairly small, measuring 0.2 km in length. Only half the beach is forested. The sand spit measures between 0 and 25 m.

2.2. Nesting Monitoring

Due to COVID-19, the previous monitoring schedule had to be modified. Therefore, the nightly monitoring of sea turtle nesting started in July 2020 and continued until early September. In 2021, it took place from May to early August, and in 2023, from April to late August. In 2024, monitoring began in April and will continue through September. Surveillance is conducted from 7:30 p.m. to 1 a.m., 4 days a week (Monday, Tuesday, Thursday, and Friday). Teams of two to three people patrol the beach looking for tracks or individuals. To avoid disturbing the turtles, all surveys are conducted with red light only. At the beginning of each survey, general information is recorded, such as monitoring information (date, start and end times, distance covered) and environmental conditions (tides, weather, human disturbance).

When an individual is sighted, a distance of 10 m is maintained, and the red light is turned off. General data (species and start time of observation) and behavioral data can then be recorded. The start and end times (HH:MM) of each behavior are recorded. We also quantified and located the presence of obstacles (roots, sargassum, branches), systematically noting their GPS position and measurements (length, width, height). For anthropogenic impacts, light intensity was measured using a luxmeter (Extech Light, Meter LT300, Sigma Aldrich, St Quentin Fallavier, France), and the number of people on the beach was counted every hour. During the nesting period, the state of the turtle [30] allows measurements to be taken on the individual without disturbing it. An observer measures carapace size (length and width) using a phone application (AR Ruler App 2025), performs photo identification (left and right head profiles and pineal spot for leatherbacks), and checks for the presence of tumors, injuries, and flipper tags. The other observer, positioned behind the turtle, counts the number of eggs. GPS position and nest location are also recorded. After the turtle has laid the eggs, the observers remain at a distance and continue to collect behavioral data until the turtle returns to the water. No turtle was touched during the surveys.

2.3. Behavior Description

To facilitate comparison across studies, we decided to utilize the same set of behaviors as previously published by Lindborg et al. [20], where seven different behaviors are described (

Table 1. Description of nesting behavior as described by Lindborg et al. [20].

Behaviors	Description
Ascent	The turtle emerges from the water and moves along the beach using all four limbs simultaneously until it finds the site for its egg-laying. The turtle may take short breaks.
Body pitting	The female moves the sand, mainly using its front flippers in pairs and its rear flippers alternately. The animal may move slightly forward or change the angle of its body.
Digging	The turtle uses its hind flippers alternately to dig its nest.
Egg laying	The female has its two hind flippers on either side of its tail, covering the egg chamber while it lays. Generally, it makes no flipper movements during this phase.
Covering	The turtle uses its hind flippers alternately to move sand and fill the nest. It may use one flipper several times in succession before switching to the other.
Camouflaging	The animal disperses sand, mainly using its front flippers in pairs and its rear flippers alternately. The female may move away from the nesting site and repeat this action several times [21].
Descent	The turtle advances towards the sea using all four limbs simultaneously until it reaches the water. The turtle may take short breaks.

).

Depending on the behavior, there are three types of observation. The search corresponds to the observation of a turtle performing an ascent behavior before returning to the water. Attempting corresponds to a turtle that performs a body pitting and/or starts digging before returning to the water. Laying corresponds to behavior up to the point of egg-laying.

2.4. Data Analyses

All data collected were recorded on field sheets and then entered into an Excel database, including total number of observations of each species, type of observation, number of attempts, number of eggs, presence of anthropogenic disturbances, obstacles and injuries, and duration for each nesting stage. Statistical analyses were performed using R 4.3.1 [37].

First, sampling effort was calculated by calculating the total number of observations, their types, the percentage and average of attempts, and the average number of eggs laid. Then, for each species, the mean duration and standard deviation of each nesting behavior were calculated, along with the range and relative percentages of time. To measure nesting stages, observations that included attempts, anthropogenic disturbances, obstacles, and turtle injuries were excluded. Third, the effects of environmental obstacles, anthropogenic factors and injuries on behavior were tested. For leatherbacks, we were able to test the effects of sargassum, the combination of different anthropogenic factors (light and frequency), injuries, and the combination of anthropogenic factors and obstacles. For hawksbills, we were able to test the effects of sargassum, roots, the combination of obstacles (sargassum, roots, branches), artificial light, the combination of different anthropogenic factors (light and frequency), and the combination of anthropogenic factors and obstacles. Chi-squared tests were used to assess the effect on egg-laying success (yes or no) and presence of an attempt (yes or no). Fisher tests were applied when the data did not follow Cochran’s rule. In addition, the effect of these factors (naturals or anthropogenics) on the number of eggs and total observation time was tested using Student’s t-tests to check the normality of the data and the equality of variance. Mann–Whitney or Welch tests were performed in case of noncompliance. Linear regressions were also used to assess the effect of the number of eggs laid on digging time and laying time. The normality of the data was checked.

Finally, the two species were compared using chi-squared test and proportion test to study differences in percentages of observation types and number of attempts. Student’s t-tests were used to compare the number of eggs. Mean comparisons of nesting behavioral stages in both species were tested using Student’s t-tests to check normality of data and equality of variance. If not, Mann–Whitney or Welch tests were performed. Durations and relative percentages of each behavior, as well as differences in sensitivity to anthropogenic and natural factors, were also compared between species.

3. Results

3.1. Leatherback Turtles

3.1.1. Sampling

A total of 47 leatherback turtle observations were recorded during nocturnal monitoring in 2020, 2021, 2023, and 2024. Of these observations, 78.72% (n = 37) were clutches, 14.89% (n = 7) were nest attempts, and 6.38% (n = 3) were nest searches. During the observations, 23.40% (n = 11) of the turtles made at least one attempt, with or without subsequent egg-laying. The mean number of attempts per observation was 1.09 (SD = 0.30), with a maximum of two attempts. On average, leatherbacks laid 89.94 (SD = 25.25) eggs.

3.1.2. Nesting Stages

The complete nesting sequence had an average duration of 101.50 (SD = 6.04) (

Table 2. Duration of different behaviors in leatherback turtles (Dermochelys coriacea) and hawksbill turtles including group size (n), mean and standard deviation ($\overline{\mathrm{X}}$ ± SD), range, and percentage (in %). Times are in minutes. Total time of nesting stages is equal to the sum of the means of all stages. Total time of active nesting stages is equal to the sum of means from body pitting to camouflage.

	Leatherback turtles				Hawksbill turtles
Nesting stages	n	$\overline{\mathrm{X}}$ ± SD (min)	Range (min)	Percent (%)	n	$\overline{\mathrm{X}}$ ± SD (min)	Range (min)	Percent (%)
Ascent	9	10.33 ± 4.03	5.00–18.00	10.18	11	12.73 ± 7.30	5.00–26.00	11.62
Body pitting	11	8.45 ± 4.30	2.00–16.00	8.33	14	5.86 ± 3.46	2.00–15.00	5.35
Digging	16	23.19 ± 8.52	10.00–40.00	22.85	35	34.69 ± 14.22	6.00–60.00	31.68
Egg laying	17	9.94 ± 2.11	5.00–13.00	9.79	46	19.65 ± 6.43	9.00–42.00	17.95
Covering	16	16.19 ± 9.12	8.00–35.00	15.95	42	16.71 ± 6.39	4.00–34.00	15.27
Camouflaging	13	27.46 ± 11.98	11.00–55.00	27.06	39	17.51 ± 11.25	3.00–68.00	15.99
Descent	15	5.93 ± 2.22	3.00–11.00	5.85	44	2.34 ± 4.03	0.00–25.00	2.14
Total active nesting time	17	85.23 ± 7.20	36.00–159.00	83.97	46	94.42 ± 8.35	24.00–219.00	86.24
Total nesting time	17	101.50 ± 6.04	44.00–188.00	100	46	109.49 ± 7.58	29.00–270.00	100

). All seven nesting behaviors were observed (Table 2).

3.1.3. Factors Affecting Nesting

In total, we observed nineteen leatherbacks without disturbance, four with multiple types of anthropogenic factors, eight in the presence of sargassum, four with anthropogenic disturbances and obstacles, and four with injuries.

a.

Anthropogenic factors

Anthropogenic disturbances (light and frequency) had no significant effect on egg-laying (Fisher, p = 1), attempt (Fisher, p = 1), number of eggs (T-test, t = 0.40, df = 17, p = 0.70), and total observation time (Whitney, W = 22, p = 0.21).

b.

Natural factors

Sargassum had no significant effect on egg-laying (Fisher, p = 0.62), attempt (Fisher, p = 0.32), number of eggs (T-test, t = −0.98, df = 19, p = 0.34), and total observation time (T-test, t = 0.99, df = 25, p = 0.34).

The linear regression evaluating the effect of number of eggs laid on digging time shows a coefficient of determination (R-squared) of 0.04, indicating that 4% of the variance in digging time is explained by number of eggs. The overall model is not significant (F = 0.53, p = 0.48). For laying time, the coefficient of determination (R-squared) is 0.02, indicating that 2% of the variance in laying time is explained by the number of eggs. The overall model is not significant (F = 0.33, p = 0.57).

c.

Anthropogenic and natural factors

Observations with a combination of anthropogenic disturbance and obstacles showed no significant effect on laying time (Fisher, p = 0.19), laying attempt (Fisher, p = 1), number of eggs (Whitney, W = 15, p = 1), and total observation time (T-test, t = 1.44, df = 21, p = 0.16).

d.

Injuries

Injuries had no significant effect on egg laying (Fisher, p = 1), attempt (Fisher, p = 1), number of eggs (T-test, t = 1.01, df = 17, p = 0.33), and total observation time (Welch, t = −1.05, df = 2.13, p = 0.40).

3.2. Hawksbill Turtles

3.2.1. Sampling

During the four years of nocturnal monitoring, 271 observations of hawksbill turtles were recorded. Of these, 73.43% (n = 199) were clutches, 11.81% (n = 32) were nest attempts, and 14.76% (n = 40) were nest searches. During the observations, 23.25% (n = 63) of the turtles made an attempt, with or without subsequent egg-laying. The mean number of attempts per observation was 2.48 (SD = 1.77), with a maximum of seven attempts. On average, hawksbills laid 135.29 (SD = 34.60) eggs.

3.2.2. Nesting Stages

The complete nesting sequence had a mean duration of 109.49 (SD = 7.58) (Table 2). All seven nesting behaviors were observed (Table 2).

3.2.3. Factors Affecting Nesting

In total, we observed one hundred forty-one hawksbills without disturbance, fourteen with light pollution, fifteen combining at least two anthropogenic factors, thirteen in the presence of sargassum, thirty in the presence of roots, eight with multiple types of obstacles, and thirty-one with anthropogenic disturbances and obstacles.

a.

Anthropogenic Factors

Light significantly reduced oviposition success (chi-squared, χ² = 6.26, df = 1, p = 0.01). However, light had no significant effect on attempt (Chi-squared, χ² = 2.60, df = 1, p = 0.10), number of eggs (Wilcoxon, W = 242, p = 0.13), and total observation time (Whitney, W = 660.5, p = 0.53). Anthropogenic disturbance (light and frequency) had no significant effect on oviposition (Fisher, p = 0.73), attempt (Fisher, p = 1), number of eggs (T-test, t = −1.52, df = 112, p = 0.13), and total observation time (Whitney, W = 681.5, p = 0.51).

b.

Natural Factors

Sargassum had no significant effect on egg laying success (Fisher, p = 0.71), attempt (Fisher, p = 1), number of eggs (T test, t = 0.92, df = 113 p = 0.36), and total observation time (Whitney, W = 761, p = 0.97).

Roots significantly decreased egg-laying success (chi-squared, χ² = 4.26, df = 1, p = 0.04) and significantly increased the number of attempts (chi-squared, χ² = 16.29, df = 1, p < 0.001). However, light had no significant effect on the number of eggs (T-test, t = 0.36, df = 23.87, p = 0.72) and on the total observation time (Whitney, W = 1659.5, p = 0.80).

The combination of obstacles (trunk, roots, sargassum) significantly reduced egg-laying success (Fisher, p < 0.01) and increased the number of attempts (Fisher, p = 0.03). However, the combination of obstacles had no significant effect on the number of eggs (Whitney, W = 13, p = 0.20) and the total observation time (Whitney, W = 428.5, p = 0.88).

The linear regression evaluating the effect of the number of eggs laid on digging time shows a coefficient of determination (R-squared) of 0.005, indicating that about 0.54% of the variance in digging time is explained by the number of eggs. The overall model is not significant (F = 0.17, p = 0.68). For laying time, the coefficient of determination (R-squared) is 0.38, indicating that 37.97% of the variance in laying time is explained by the number of eggs. The overall model is significant (F = 26.32, p < 0.01).

c.

Anthropogenic and natural factors

Observations with a combination of anthropogenic disturbance and obstacles significantly increased attempts (Chi-squared, χ² = 5.82, df = 1, p = 0.02). On the other hand, egg laying success (Chi-squared, χ² = 0.14, df = 1, p = 0.70), number of eggs (T-test, t = −0.57, df = 122, p = 0.57), and total observation time (Whitney, W = 1669, p = 0.63) were not significantly affected by the combination of anthropogenic disturbance and obstacles.

3.3. Comparison Between Species

3.3.1. Proportion of Observation Types and Attempts and Number of Eggs

There was no significant difference in the distribution of observation types (search, attempt, laying) between the two species (chi-squared, χ² = 3.86, df = 2, p = 0.15). Similarly, there was no difference in the proportion of attempts (with or without subsequent nesting) between leatherbacks and hawksbills (proportion test, χ² = 0.06, df = 1, p = 0.80). Hawksbill turtles laid significantly more eggs than Leatherback turtles (Student’s t-test, t = 5.56, df = 59, p < 0.001).

3.3.2. Means and Percentages of Nesting Behavioral Stages

The means of the seven nesting behavioral stages were compared for leatherback and hawksbill turtles (Figure 2).

Leatherback and hawksbill turtles show differences in the time distribution of each nesting stage. For Leatherback turtles, camouflage is the dominant behavior, accounting for 27.06% of the total time, while for Hawksbill turtles, digging is the dominant behavior, accounting for 31.68% of the total time. The percentage of time spent on ascent and camouflage is relatively similar between the two species, differing by only 1.44% and 0.68%, respectively. Hawksbill turtles are less likely to spend time on their bodies, but these differences are only 2.98% and 3.71%, respectively. However, a more significant difference is found in the time spent digging, which is 8.83% higher in hawksbills than in leatherbacks (31.68% vs. 22.85%). Similarly, the time spent laying eggs is 8.16% less for leatherbacks than for hawksbills (9.79% vs. 17.95%). Conversely, Leatherback turtles spend relatively more time camouflaging, with a percentage 11.07% higher than that of Hawkbill turtles (27.06% vs. 15.99%).

3.3.3. Factors Affecting Nesting

Anthropogenic disturbance had no effect on nesting for leatherback turtles (Fisher, p = 1). In hawksbill turtles, however, light had a significant effect on nesting success (χ² = 6.26, df = 1, p = 0.01).

Sargassum had no significant effect on nesting behavior in either species. However, roots and the combination of obstacles significantly affected egg-laying (roots: χ² = 4.26, df = 1, p = 0.04; combination: Fisher, p < 0.01) and attempt (roots: χ² = 16.29, df = 1, p < 0.001; combination: Fisher, p = 0.03) in hawkbill turtles.

The association of anthropogenic disturbance and obstacles showed significant effects on attempt (Chi-squared, χ² = 5.82, df = 1, p = 0.02) only for hawksbill turtles.

Finally, the number of eggs had a significant effect on the incubation period only in hawksbill turtles (F = 26.32, p < 0.01). In contrast, there was no effect of egg number on digging duration in either species (Leatherback: F = 0.53, p = 0.48; Hawksbill: F = 0.17, p = 0.68).

4. Discussion

The aim of this study was to improve our understanding of the behavior of leatherback and hawksbill turtles in relation to anthropogenic and natural factors. Using nest monitoring, we were able to observe the complete nesting behavioral sequence for both species.

The average duration of the nesting behavioral sequence of leatherback turtles is 101.50 min. The nesting behavior is characterized by a predominance of camouflaging, which represents 27.06% of their total nesting time, followed by digging (22.85%) and covering (15.95%). Our results are consistent with the observations of Pritchard [24], who also found an average nesting sequence of 101.50 min. Although the studies of Carr and Ogren [22] and Bacon [23] showed variations of 8.50 min and 11.50 min, respectively, with our results, these differences are relatively similar to our data. Carr and Ogren [22] observed longer times for body pitting (17 min and 8.45 min), laying (15 min and 9.94 min), and camouflaging (40 min and 27.06 min), but shorter times for digging (5 min and 23.19 min) and covering (5 min and 16.19 min). These variations in duration are important because they may be indicators of environmental constraints leading to changes in turtle behavior (e.g., the duration of egg-laying phases). This may highlight differences in adaptation and evolutionary responses to environmental pressures (e.g., more camouflage possibly because of more predators). Then, environmental constraints can directly influence the number of attempts or the risk of egg-laying abandonment. For example, the presence of roots in the sand can increase the number of attempts and the risk of egg-laying abandonment (as there is no suitable place to lay eggs, or the turtle expends too much energy to complete the egg-laying process).

However, these results must be interpreted with caution because the study by Carr and Ogren [22] was based on a single individual. Differences may be due to different environmental conditions or different methodologies between studies. For example, Carr and Ogren’s [22] study was conducted in Costa Rica, while Bacon’s [23] study was conducted in Trinidad and Tobago. Beach characteristics such as slope, tide, and obstacles may influence the duration of nesting behavior. In our case, however, tidal variations appear to be minimal across the beaches as they are all bordered by the Caribbean Sea.

In comparison, hawksbill turtles had an average nesting sequence duration of 109.49 min. Digging dominated with 31.68% of the total nesting time, followed by laying (17.95%) and camouflaging (15.99%).

Although the proportion of observation types and attempts did not show significant differences between the two species, there were marked differences in several behavioral stages. Hawksbills spent significantly more time digging than leatherbacks (34.69 min (31.68%) and 23.19 min (22.85%), respectively). Leatherbacks dig deeper nests than hawksbills, but the latter still spend more time digging [38,39]. This variation is primarily explained by morphological differences. Leatherbacks have larger flippers, which may allow them to dig faster than hawksbills [24,29]. In addition, the nesting habitat of hawksbill turtles, which is often along forest edges, may slow their digging due to harder sand and obstacles such as roots [40,41]. Nest recovery time did not show a significant difference between the two species (Leatherbacks: 16.19 min (15.95%); hawksbills: 16.71 min (15.27%). As a consequence of the aforementioned circumstances, the differences in digging time in our study are mainly due to the characteristics of the nest habitat.

Hawksbill nesting is slower than Leatherback nesting (19.65 min (17.95%) and 9.94 min (9.79%)). On average, hawksbills lay significantly more eggs than leatherbacks (135.29 eggs and 89.94 eggs, respectively), which may explain the difference in laying time. In fact, we observed that the number of eggs had a positive effect on the laying time of hawksbill turtles.

Leatherbacks spend more time camouflaging their nests than hawksbills (27.46 min (27.06%) and 17.51 min (15.99%)). Burns et al. [21] have shown that leatherbacks spend more time and effort in this phase than hawksbills. While most studies suggested that nest camouflage served to create a lure for predators, the work of Burns et al. [21] shows greater complexity in this phase. In fact, this behavior first involves covering the eggs and the nest, in order to physically protect the eggs and create an environment conducive to successful incubation, then creating lures for potential predators around the nest but also at a greater distance. This phase, which requires extra time and effort, seems to be specific to each individual, regardless of species. The difference observed in our study may be due to the fact that leatherback nests are positioned on the beach in the open, whereas hawksbill nests are partially protected by vegetation and therefore require less camouflage.

Then, there is no significant difference in ascent time between the two species (leatherbacks: 10.33 min (10.18%); hawksbills: 12.73 min (11.62%)). Hawksbills lay their eggs at a greater distance from the sea than leatherbacks [41,42], suggesting that hawksbills move faster on the nesting beach. This could be explained by morphological differences between the two species, as hawksbills are smaller and lighter [25,29], allowing them to move more quickly across the sand. However, leatherbacks take longer than hawksbills to return to the water (5.93 min (5.85%) and 2.34 min (2.14%)). There is a noticeable decrease in the time between ascent and descent. The ascent would be a time of recognition and searching for a place to lay eggs [19], a wandering behavior according to Hendrickson [17]. The descent would be a simple return to the water where the turtle may have memorized the route. It would be interesting to study the turtle’s return journey. To explain the significant difference in descent time, we can hypothesize that due to the longer camouflage time and its peculiarities [21], the energy depletion generated by this stage in leatherback turtles [21,43] could be greater than in hawksbill turtles, resulting in a longer return to the water. Finally, total nesting time was not significantly different between the two species (leatherbacks: 101.50 min; hawksbills: 109.49 min).

In terms of factors affecting nesting, the results show differences between the two turtle species. Anthropogenic disturbance had no significant effect on leatherback nesting success. On the other hand, artificial light significantly affected the nesting success of hawksbill turtles. This may indicate that hawksbills are more sensitive to human disturbance than leatherbacks. The study by Attum and Nagy [14] supports our findings and shows a negative effect of light pollution on the nesting success of hawksbill turtles. However, the observations by Hu et al. [44] contradict these results by reporting a negative impact of light pollution on leatherback nesting. It is important to treat these results with caution due to the low number of observations of leatherback turtles exposed to anthropogenic disturbance. This low number of observations could also indicate that these turtles are not coming out of the water to lay eggs due to the disturbance. It would be necessary to establish a specific protocol to study the effects of anthropogenic disturbance.

Sargassum has not been shown to have a significant effect on the nesting behavior of either species. The study by Maurer et al. [11] suggests that sargassum may act as a barrier and alter habitat availability for hawksbill turtles. However, like us, they show no effect of sargassum on nesting success. Again, our results should be treated with caution due to the limited number of observations and lack of information on the amount of sargassum present. A specific protocol would also be required to further investigate the effect of sargassum on nesting behavior. Roots and the combination of obstacles significantly reduce nesting success and increase the number of attempts for hawksbill turtles. Due to their nesting habitat, hawksbill turtles are highly exposed to the presence of tree roots and branches. Kamel and Mrosovsky [41] have shown that these obstacles make the nesting site less favorable and increase the risk of abandonment of the nesting process. This effect could not be tested in leatherback turtles because these types of obstacles were not present in their nesting habitat. Obstacles (roots and a combination of several obstacles) have been shown to reduce nesting success and increase the number of nesting attempts in hawksbills. Indeed, obstacles can make nest digging difficult, forcing the turtle to change location or dig longer. This requires energy (and more possibly stress from predation), which may lead the turtle to abandon egg-laying. Hawksbill turtles are particularly affected because they lay their eggs close to vegetation, where obstacles are more frequent.

For hawksbill turtles, the number of eggs had a positive effect on the incubation period. No significant effect of egg number on digging time was observed in either species. It would be interesting to study the relationship between nest depth and egg number.

The association of anthropogenic disturbance and obstacles confirms the significant impact of these elements on nesting only in hawksbill turtles. These results underscore the importance of continuing to improve our understanding of the effects of these features on sea turtle nesting. Such efforts would contribute to the conservation and reduction in anthropogenic impacts on sea turtles.

Our study complements previous behavioral research by Hailman and Elowson [19] on loggerhead sea turtles (Caretta caretta) and Lindborg et al. [20] on green sea turtles (Chelonia mydas), allowing for comparisons among four species of sea turtles. Analysis of nesting behavior revealed differences in the duration of different stages of the process (

Table 3. Mean duration (in minutes) of nesting stages of four species of sea turtles: loggerhead (Caretta, Caretta), green (Chelonia mydas), hawksbill (Eretmochelys imbricata), and leatherback (Dermochelys coriacea), including group size (n). Total time of nesting stages is equal to the sum of the means of all stages. Total time of active nesting stages is equal to the sum of the means of body pitting to camouflage. Loggerhead and green turtle times are from observations by Hailman and Elowson [19] and Lindborg et al. [20].

Nesting Stage	Caretta caretta (n = 16)	Chelonia mydas (n = 14)	Dermochelys coriacea (n = 17)	Eretmochelys imbricata (n = 46)
Ascent	7.30	9.90	10.33	12.73
Body pitting	5.50	30.20	8.45	5.86
Digging	12.90	29.30	23.19	34.69
Egg laying	12.30	22.30	9.94	19.65
Covering	10.50	10.50	16.19	16.71
Camouflaging	12.60	62.30	27.46	17.51
Descent	5.80	4.10	5.93	2.34
Total active	54.80	145.60	85.23	94.42
Total	63.00	X	101.50	109.49

).

Ascent is faster for loggerheads (7.30 min), although the difference between the four species is relatively small. Body pitting is relatively similar for loggerheads, hawksbills, and leatherbacks, but higher for green turtles (30.20 min). Nest digging takes the longest for hawksbills (34.69 min), closely followed by green turtles (29.30 min). Leatherback turtles lay their eggs the fastest (9.94 min), while green turtles take an average of 22.3 min. Camouflage time is similar between leatherbacks (16.19 min) and hawksbills (16.71 min), and between green turtles (10.50 min) and loggerheads (10.50 min). Camouflage time is much longer for green turtles (62.30 min), followed by leatherbacks (27.46 min). Finally, descent is more than twice as fast for hawksbills (2.34 min) as for other species. Differences between species can be due to several factors. Firstly, morphological: leatherbacks are much heavier and therefore slower than other species. Nest depth (greater in leatherbacks and green turtles) has a direct impact on the time required for digging. The factor(s) influencing egg-laying speed (egg ejection) merits further investigation. The duration of nest camouflage also needs to be investigated, as it may be due to the number of predators present on site.

These differences in nesting behavior among the four species of sea turtles can be explained by adaptations specific to each species and its nesting environment. First, the morphology of different turtle species can play a critical role in nesting behaviors such as digging, covering, ascending, and descending [24,29,45,46]. The shape of the flippers, size, and weight of the turtle are determining factors. Second, beach characteristics such as slope, width, and tide also influence ascent and descent behavior. Studies by Hailman and Elowson [19] and Lindborg et al. [20] conducted on Florida beaches may show significant differences in slope, width, sand granulometry, tides, and predation compared to Martinique beaches. For example, Florida beaches, located on the Atlantic coast, have higher tides than Martinique beaches, which may affect the ascent and descent behavior of sea turtles.

In addition, the four species do not share the same nesting habitat [1,10,26,29], which may directly influence digging. For example, hawksbill turtles often dig at forest edges where the sand is hard and roots are present [40]. This may explain their longer digging times than other species. Body pitting, covering, and camouflage strategies vary among species. Miller [27] suggests that these strategies are evolutionary adaptations to maximize nesting success in response to local threats and environmental conditions. For example, body pitting is particularly long in green turtles. One hypothesis is that this behavior allows the turtle to properly slope according to the slope of the beach [17]. Camouflage, which would have the function of creating decoys [21], is used differently by different species, with longer camouflage durations in leatherback and green turtles. The number of eggs laid affects the incubation period, as we observed in hawksbill turtles. Since each species has a different number of eggs [4,24,30,47], this may explain the variation in laying time. Finally, sequences of action movements may vary. Hailman and Elowson [19] have shown differences between leatherback, hawksbill, green, and loggerhead turtles, particularly during digging and laying [17,24,48].

In conclusion, our study has allowed us to increase our knowledge of the nesting behavior of leatherback and hawksbill turtles in Martinique and to make detailed comparisons with other species. However, it is essential to continue research on nesting behavior and the factors that influence it. To this end, the use of video, as used in the article by Lindborg et al. [20], could provide a more accurate analysis of movements and action sequences. This would complement the existing ethograms of green and loggerhead turtles [19,20]. Ethograms would provide a new way to measure behavioral changes in response to anthropogenic disturbance. They would complement specialized protocols for sea turtle conservation goals.