Abundance Trends and Nesting Biology of Green Turtles Chelonia mydas (Testudines: Cheloniidae) During Ten Consecutive Breeding Seasons (2012-2021) at Akyatan Beach, Turkey

© 2022 Academia Sinica, Taiwan Open Access Abundance Trends and Nesting Biology of Green Turtles Chelonia mydas (Testudines: Cheloniidae) During Ten Consecutive Breeding Seasons (2012–2021) at Akyatan Beach, Turkey Can Yılmaz1,2,*, Ayşe Oruç2, and Oguz Turkozan3,4 1Vocational School of Health Services, Hakkâri University, Keklikpınar Mahallesi Kampüs Caddesi 30000 Hakkâri, Turkey. *Correspondence: E-mail: [email protected] (Yılmaz) 2WWF Turkey, Buyuk Postane Caddesi No: 19 Kat 5, 34420 Bahcekapi, Eminonu, Istanbul, Turkey. E-mail: [email protected] (Oruç) 3Department of Biology, Faculty of Science and Arts, Aydın Adnan Menderes University, 09010 Aydın, Turkey. E-mail: [email protected] (Turkozan) 4Cyprus Wildlife Research Institute, Kyrenia, Northern Cyprus, Research Associate, Cyprus Received 21 February 2022 / Accepted 25 June 2022 / Published 29 September 2022 Communicated by Yoko Nozawa Data obtained from long-term survey studies are valuable for assessing the population status and trends in critical populations of threatened species, like sea turtles. Akyatan Beach is one of the most important green turtle nesting beaches in the Mediterranean and has been monitored since 2006 without interruption. The beach is 22 km long and more than 100 m wide at some points, and both loggerhead and green turtles nest on the beach. However, loggerhead nesting is very limited compared to green turtles. A total of 3866 C. mydas nests were recorded over ten consecutive years at Akyatan Beach, with a mean of 387 ± 127 nests (range = 201–559). The average nesting density was 17.6 nests km-1 (range = 9.1–25.4 nests km-1). In the 3309 nests, a total of 355,259 eggs were counted. The overall mean clutch size was 112 ± 26.10 eggs. Of these eggs, 50.80% hatched (depredated nests included), and 78.07% of them were able to reach the sea. The overall mean hatching success was 73.07 ± 26.20%. The overall mean incubation duration was 51.4 ± 3.5 days. The clutch sizes and hatching success differed between years, and there was a significant decreasing trend in mean incubation duration over the ten years of the study. A total of 1585 green turtle nests (41.02% of nests) were totally or partially depredated by golden jackals and wild boars, while other predators depredated 20.5% of hatchlings. The nesting data obtained since 2006 showed strong annual fluctuations ranging from 170 (in 2007) to 562 (in 2006) with a slightly increasing but statistically insignificant trend (r = 0.94, p > 0.05). The main threats to the population were depredation by jackals and wild boars. Key words: Endangered species, Conservation, Sea turtles. BACKGROUND Sea turtles are migratory species with a complex life history that spend most of their lives in the sea, and only females come to beaches to reproduce. They have different habitats during these complex life cycles. The creation of practical action plans to protect endangered sea turtles will be possible by obtaining detailed information about these different habitats and revealing the connections between these habitats. In this respect, the Mediterranean is an area containing essential habitats for two species: the green turtle (Chelonia mydas) and the loggerhead turtle (Caretta caretta). It was defined as a regional Citation: Yılmaz C, Oruç A, Türkozan O. 2022. Abundance trend and nesting biology of green turtles Chelonia mydas (Testudines: Cheloniidae) during ten consecutive nesting seasons (2012–2021) at Akyatan Beach, Turkey. Zool Stud 61:53. doi:10.6620/ZS.2022.61-53. Zoological Studies 61:53 (2022) doi:10.6620/ZS.2022.61-53 1 © 2022 Academia Sinica, Taiwan management unit for both species. A total of 17 regional management units (RMUs) were identified for green turtles worldwide, and one of them is the Mediterranean region, which is listed in the “high threats” category (Wallace et al. 2010 2011). Recent studies reveal that the populations of these two species have increased both globally (Mazaris et al. 2017) and regionally (Casale et al. 2018). The loggerhead and green turtles nesting in the Mediterranean prefer the coasts of Turkey, Greece, Libya, Egypt, and Tunisia as foraging and wintering grounds (Broderick et al. 2007; Casale and Margaritoulis 2010; Godley et al. 2003; Rees et al. 2008; Snape et al. 2016; Stokes et al. 2015). The nesting of green turtles in the Mediterranean is confined to Turkey, Northern Cyprus, Lebanon, Israel, and Egypt (Türkozan and Kaska 2010; Casale et al. 2018), and the average number of documented nests is 1500 nests/year in the Mediterranean (Casale and Margaritoulis 2010). Of the nesting grounds in the Mediterranean, Turkey, and Cyprus comprise almost 99% of the overall nesting activity (Kasparek et al. 2001; Türkozan and Kaska 2010; Casale et al. 2018). Recent studies using mtSTR described 3 to 4 management units for green turtle stocks in the Mediterranean (Tikochinski et al. 2018; Karaman et al. 2022). Of the 13 major green turtle rookeries in the Mediterranean, six are located on Turkish coasts (Casale et al. 2018), and a total of 452–2051 green turtle nests are estimated annually in Turkey (Türkozan and Kaska 2010). Akyatan Beach has been regularly monitored since 2006, and a -1.2% change was reported in the nest numbers before 1999 and after 2000 monitoring programs (Casale et al. 2018). This decrease is noticeable since other crucial green turtle nesting beaches (Kazanlı and Samandağ, Turkey) provided a 71.4–279.1% increase in nest numbers (Casale et al. 2018). Since Akyatan Beach is a part of a strictly protected area and far from anthropogenic effects, we focused on the long term population parameters to identify possible causes of this decrease. A population size and nest density estimate not based on yearly longterm monitoring studies on nesting beaches may provide misleading results due to biased calculations caused by fluctuations in the number of nests (Sönmez et al. 2021). We, therefore, offered long-term data and ongoing nesting activities on the beach, which hosts about 20% of the total number of green turtle nests recorded in the Mediterranean. MATERIALS AND METHODS Study site Akyatan Beach is located on the eastern Mediterranean coast of Turkey and within the Wildlife Development Area of 15,304 ha, which has held protection status by the Ministry of Forest and Agriculture since 1987. Akyatan Beach is between Akyatan Lagoon and the Mediterranean and is 22 km long (Fig. 1). Both green and loggerhead turtles nest on Akyatan beach, but green turtle nesting is denser. Conservation studies of sea turtles in Akyatan Beach have been regularly carried out since 2006 according to the protocol signed by the Ministry of Agriculture and Forestry, 7th Regional Directorate of Nature Conservation and National Parks, and WWF-Turkey. Nesting data The nesting data for C. mydas were collected Fig. 1. Akyatan Beach on the Turkish coast. N page 2 of 12Zoological Studies 61:53 (2022) © 2022 Academia Sinica, Taiwan over ten nesting seasons (2012–2021) between June 1 and Sept. 15 each year (except for the 2014 and 2015 nesting seasons when the survey started on May 15th). About 14 km of the 22 km beach were monitored daily on foot, and the remaining 8 km were monitored once every three days (Fig. 1). Adult emergence The location of clutches within a nest and nonnesting emergences were determined by carefully using a metal rod. The distance of nests and the apex of nonnesting emergences from the sea were measured using a flexible tape measure in cm. All clutch locations were individually marked and recorded with GPS (Garmin Etrex 20), and non-depredated nests were caged with a wire mesh screen (72 × 72 cm, mesh size 9 cm) to prevent nest depredation by mammals. Depredated eggs were counted in the nests, moved to another location on the beach, and reburied. Clutches laid in areas at risk of flooding were relocated to a site with lower risk and the same dimensions of the original nest. Hatchling emergence The hatchlings reaching the sea were determined daily from hatchling tracks emanating from nests. Counted tracks were obscured by feet to avoid recounting. When the tracks were interrupted by predator tracks, such as golden jackal (Canis aureus) or wild boar (Sus scrofa) tracks, the hatchlings were assumed to be preyed upon before they reached the sea by those predators. Each nest was excavated carefully, by hand, or using a shovel 3 or 5 days after the first emergence of hatchlings. Nest contents were classified as hatched eggs/empty shells (hatchlings emerged), unhatched eggs (unfertilised eggs and eggs without visible embryos or blood formation), and dead embryos (developmentally delayed eggs, early embryo: embryo < 1 cm; middle embryo: embryo 1–2 cm; late embryo: embryo > 2 cm). The hatching success in the nests was calculated as the percentage of hatched empty shells/the total number of eggs in the entire clutch. Predated nests were not considered for the assessment of hatching success. When fragmented eggshells were found, eggshell pieces were reassembled to represent one egg. Also, incubation duration was defined as the number of days from the nesting date to the date of the first hatchling’s emergence onto the surface of the sand. Estimates of population sizes The total number of nesting females was estimated based on clutch frequency (CF) using the clutch frequency of 2.9 (range = 2.0–3.1) for green turtles in the Mediterranean (Broderick et al. 2002), since this information is not available for Akyatan Beach. We considered the remigration interval (RI) as three years for the green turtle populations in the Mediterranean region (Broderick et al. 2002). The total nesting female numbers were calculated with the following formula: Total Nesting Female Numbers = Total Nest Number/ CF * (Total years/RI) (Sönmez et al. 2021) Also, the current female numbers in the last two nesting seasons (2020 and 2021) were calculated with the following formula: Current Female Numbers = (Mean Nest Number/ CF) * RI (Sönmez et al. 2021). Statistical analysis The data were not normally distributed according to Levene and Kolmogorov-Smirnov tests (p < 0.05). However, parametric tests could be used based on the central limit theorem regardless of the shape of our data in large sample sizes (Field 2013). We, therefore, used the One-Way ANOVA test to compare mean clutch size, depredated eggs, hatching success, and incubation duration among nesting seasons. The trend analysis for hatching success, incubation duration, and nest numbers across the years was performed by linear regression. All statistical analyses were performed using the IBM SPSS Statistics 20 software. All means are presented with SD (standard deviation). RESULTS Nests, eggs, and hatchlings A total of 10,767 green turtle emergences were recorded, with 3866 (35.91%) clutches being successfully laid during ten consecutive reproduction seasons (2012–2021) on Akyatan Beach. The annual mean number of nests was 387 ± 127 (range = 201–559 nests), and the mean nest density was 17.57 nests km-1 (range = 9.14–25.41 nests km-1). Due to the risk of inundation of nests by seawater, 42 (1.09%) green turtle nests were relocated. A total of 355,259 eggs were deposited in 3309 nests (excavated nests), and the mean clutch size excluding predated nests was 112 ± 26.10 eggs (range: 4–222) (Table 1). The clutch sizes were significantly different between the nesting seasons (one-way ANOVA, F = 3.105, p < 0.01) (Fig. 2a), and the 2012, 2016, 2020, and 2021 nesting seasons had larger mean clutch sizes (see Table 1 for details). Of the 355,259 eggs, 180,421 (50.80%) produced hatchlings, and page 3 of 12Zoological Studies 61:53 (2022) © 2022 Academia Sinica, Taiwan 140,851 (78.07%) of them were able to reach the sea. When depredated nests were not included, the mean hatching success ranged from 66.53 ± 29.87% to 79.87 ± 21.41%, with an overall mean hatching success of 73.07 ± 26.20% (Table 1). The hatching success was significantly different between the nesting seasons (one-way ANOVA, F = 5.594, p < 0.0001) (Fig. 2b). The nesting seasons of 2019 and 2020 had lower mean hatching success (see Table 1 for details). The mean hatching success in 10 consecutive years showed a significant decreasing trend (r2 = 0.006, d.f. = 2164, p < 0.0001) (Fig. 3a). A total of 54,878 dead embryos were counted in 2227 nests, and of these dead embryos, 69.54% (38.162) were early-stage embryos, 7.84% (4.300) were mid-stage embryos and 22.62% (12.416) were late-stage embryos. Table 1. Biological data about Chelonia mydas at Akyatan Beach. Unhatched eggs = eggs without visible embryos or blood formation; developmentally delayed eggs, early embryo = embryo < 1 cm; middle embryo = 1 < embryo < 2 cm; late embryo = embryo > 2 cm) Nesting Seasons 2012 2013 2014 2015 2016 2017 Total number of nests 444 249 433 201 559 349 No. of nests depredated 110 82 159 97 270 159 No. of nests with eggs uncountable due to depredation 64 39 80 29 70 50 No. of nests left on the beach 15 14 15 10 28 12 No. of nests excavated 319 153 259 94 261 178 No. of nests excavated and eggs countable 365 196 338 162 461 287 Total number of eggs 40758 20401 35559 16448 48474 29396 • Unhatched eggs 671 589 884 396 1702 869 • Early embryo 5430 1807 4016 1327 4608 3221 • Middle embryo 479 193 612 113 589 289 • Late embryo 2052 431 999 247 1485 1215 • Depredated eggs 5150 4104 7438 6435 18914 10401 • Total number of hatchlings (hatched eggs) 26976 13277 21610 7930 21176 13402 • Hatchlings depredated 7526 3594 3300 2067 3538 2180 • Dead hatchlings in nests 305 91 178 103 193 146 • Hatchlings reached the sea 19145 9592 18132 5760 17445 11076 Mean clutch size (no. of eggs/nest) * 116.04 ± 26.06 108.42 ± 24.71 109.05 ± 25.42 107.64 ± 22.56 113.74 ± 22.92 107.60 ± 22.47 Mean Hatching success (%)* 74.09 ± 27.26 79.87 ± 21.41 76.00 ± 25.50 79.41 ± 21.50 72.48 ± 26.21 71.34 ± 25.65 Mean incubation duration (days) 51.39 ± 2.92 51.25 ± 2.38 52.19 ± 3.12 53.27 ± 2.09 50.39 ± 2.75 52.38 ± 3.54 Nesting Seasons 2018 2019 2020 2021 Overall Total number of nests 341 348 532 410 3866 No. of nests depredated 130 210 198 131 1546 No. of nests with eggs uncountable due to depredation 3 27 90 2 454 No. of nests left on the beach 9 - - - 103 No. of nests excavated 202 148 334 279 2227 No. of nests excavated and eggs countable 329 321 442 408 3309 % Total number of eggs 34925 34234 49982 45082 355259 100 Unhatched eggs 1191 754 1516 1312 9884 2.78 Early embryo 2908 2715 7862 4268 38162 10.74 Middle embryo 568 541 514 402 4300 1.21 Late embryo 1664 1119 1786 1418 12416 3.49 Depredated eggs 12581 18188 12974 13890 110075 30.98 Total number of hatchlings (hatched eggs) 16011 10917 25330 23792 180421 50.80 Hatchlings depredated 2169 1292 7026 4357 37049 20.53 Dead hatchlings in nests 132 98 916 359 2521 1.40 Hatchlings reached the sea 13710 9527 17388 19076 140851 78.07 Mean clutch size (no. of eggs/nest) * 110.13 ± 27.50 110.63 ± 25.19 114.38 ± 32.46 113.98 ± 22.83 111.93 ± 26.10 Mean Hatching success (%)* 71.34 ± 28.73 66.53 ± 29.87 67.39 ± 28.28 76.36 ± 20.60 73.07 ± 26.20 Mean incubation duration (days) 51.41 ± 3.98 52.48 ± 3.70 51.52 ± 4.64 49.61 ± 2.77 51.39 ± 3.46 * Predated nests were not considered for assessment of mean clutch size and hatching success. page 4 of 12 Zoological Studies 61:53 (2022) © 2022 Academia Sinica, Taiwan The mean incubation durations ranged from 49.6 ± 2.77 to 53.3 ± 2.09 days, with an overall mean of 51.4 ± 3.46 days (Table 1). The incubation duration was significantly different during the breeding seasons (one-way ANOVA, F = 12.960, p < 0.0001) (Fig. 2c). The mean incubation durations in 10 consecutive years showed a significant decreasing trend (regression analysis, r2 = 0.007, d.f. = 1528, p < 0.001) (Fig. 3b). Three peak nesting periods were recorded according to the temporal distribution of green turtle nests between nesting seasons (Table 2). The first nest was recorded on the 19th of May, while the last nest was recorded on the 23rd of Aug. (Table 2). However, based on ten nesting seasons, the mean peak-nesting season on Akyatan Beach was between the 29th of June and the 8th of July (Fig. 4). Hatching starts during the third week of July (except in 2018, when it started in the first week of July), and there were two different periods for the peak-hatching season among nesting seasons (Table 2). The overall peak-hatching season was in the period of the 10th–19th Aug. (Fig. 5). The average distance of green turtle nests from the tide line was 43.17 ± 18.00 m (n = 3882, range 5.60–125.4; Fig. 6). A total of 1585 green turtle nests (41.02% of nests) were completely or partially depredated by golden jackals (Canis aureus) and wild boars (Sus scrofa). Of the 1585 nests, 1204 (75.96%) were depredated by golden jackals and 381 (24.04%) by wild boars. While the number of eggs depredated by golden jackals was counted in most cases, the number of eggs depredated by wild boars could not be counted. A total of 110,075 eggs were destroyed by golden jackals (Table 1). The number of depredated eggs significantly differed between the nesting seasons (one-way ANOVA, F = 32.438, p < 0.0001). The depredated eggs in 10 consecutive years showed a significant increasing trend Fig. 2. Interannual variation of green turtle (a) clutch size, (b) hatching success, and (c) incubation duration on Akyatan Beach for 10 years. Shown are medians (horizontal line), interquartile ranges (upper and lower box limits), range (vertical lines), and outliers (circles). Fig. 3. Temporal trend in hatching success (a) and incubation durations (b) of green turtle nests over 10 consecutive years on Akyatan Beach. page 5 of 12 Zoological Studies 61:53 (2022) © 2022 Academia Sinica, Taiwan (r2 = 0.009, d.f. = 1238, p < 0.0001). A total of 37,049 (20.53%) hatchlings were predated on their journey from the nest to the seawater edge (Table 1). Of the predated hatchlings, 36,242 (97.82%) were predated by jackals, 345 (0.93%) by crabs, 45 (0.12%) by seagulls, and 60 (0.16%) by wild boars. The fate of 357 (0.96%) hatchlings that were disoriented and entered the forest is unknown. Furthermore, 2521 (1.40%) hatchlings were found dead in the nests without predation (Table 1). Nesting trend and abundance A total of 6034 green turtle nests were recorded during 16 consecutive years (2006–2021), with a mean of 377 ± 127.2 (range = 170–562) nests per year on Akyatan Beach. The number of nests showed a strong annual fluctuation ranging from 170 (in 2007) to 562 nests (in 2006), with a difference of 231% (Table 3). The yearly number of nests across 16 consecutive years showed a insignificant but increasing trend (regression analysis, r2 = 0.012, d.f. = 14, p > 0.05) (Fig. 7). The total number of females nesting each year was estimated as 2081 (range = 1947–3018) (Table 3). The total mean number of females nesting on Akyatan Beach was 390 (range = 365–566). The number of females on Akyatan Beach in 2020–2021 was 487 (range = 456– 707). The mean estimated population size (390 females) of nesting females for all years (2006–2021) was lower than for the current year (487 females) (2020–2021). Threats on the beach During the research period, 12 green turtles were found stranded on the beach. Furthermore, seven adult green turtles were killed by golden jackals while nesting. Human disturbance on the beach at night was Fig. 4. Temporal distribution of green turtle nests on Akyatan Beach, Turkey (Data are grouped into 10 day bins). Table 2. Dates for peak-nesting/hatching seasons, nesting and hatching season during 10 nesting seasons for green turtle in Akyatan Beach Season Nesting Peak-nesting season Hatching Peak-hatching season 2012 May – 5th Aug 29th June – 8th July 1st July – After 15th Sept 10th – 19th Aug 2013 May – 23rd Aug 19th – 28th June 19th June – After 15th Sept 20th – 29th Aug 2014 19th May – 5th Aug 29th June – 8th July 25th June – After 15th Sept 10th – 19th Aug 2015 28th May – 13th Aug 9th – 18th July 29th June – After 15th Sept 20th – 29th Aug 2016 May – 20th Aug 29th June – 8th July 18th June – After 15th Sept 10th – 19th Aug 2017 May – 17th Aug 29th June – 8th July 23rd June – After 15th Sept 20th – 29th Aug 2018 May – 8th Aug 19th – 28th June 19th June – After 15th Sept 10th – 19th Aug 2019 May – 11th Aug 19th – 28th June 24th June – 27th Sept 10th – 19th Aug 2020 May – 18th Aug 29th June – 8th July 30th June – 5th Oct 20th – 29th Aug 2021 May – 26th July 19th – 28th June 20th July – 9th Sept 10th – 19th Aug page 6 of 12Zoological Studies 61:53 (2022) © 2022 Academia Sinica, Taiwan low due to its entry restriction its protected status as a Wildlife Development Area. The main anthropogenic threat was caused by several tractors illegally entering the beach, which compacted the sand, affecting hatchling emergence and leaving tracks that trapped hatchlings. DISCUSSION Nests, eggs, and hatchlings Of the 13 major green turtle rookeries in the Mediterranean, six are located in Turkey, and these beaches comprise almost 80% of the overall nesting activity in the Mediterranean (Türkozan and Kaska 2010). The annual mean number of nests on Akyatan Beach (387 nests) constitutes almost 17.51–23.39% of the overall nesting in the Mediterranean, while it comprises 27.4–37.6% of the broad green turtle nesting activity along the Turkish coasts. The annual mean nest number of Akyatan is higher than other nesting colonies in the Mediterranean (see table S11; Casale et al. 2018) and therefore, still the most critical site for green turtles. Long-time surveys indicated a 47% increase in the number of nests laid by Mediterranean green turtle populations (Casale et al. 2018). However, the comparison of average nest numbers obtained due to the monitoring studies carried out before 1999 and the average number of nests in the monitoring studies carried out after 2000 provided a -1.2% change on Akyatan Beach (Casale et al. 2018). In contrast, Kazanlı Fig. 6. Spatial distribution of green turtle nests on Akyatan Beach. Fig. 5. Temporal distribution of green turtle hatching on Akyatan Beach, Turkey (Data are grouped into 10-day bins). page 7 of 12 Zoological Studies 61:53 (2022) © 2022 Academia Sinica, Taiwan and Samandağ nesting beaches (both in Turkey), two other essential nesting grounds in the Mediterranean, showed a remarkable increase (71.4–279.1%) during the same period. The decrease is surprising since Akyatan Beach is a strictly protected area, far from adverse anthropogenic effects. We believe that this could result from the shifting of adult females since the nest site fidelity of green turtles seems to be region specific rather than beach specific (Karaman et al. 2022). The interchange of nesting individuals between Akyatan-Sugözü and Samandağ-Syria was also reported with mark-recapture studies previously (Sönmez et al. 2017). On the other hand, nest numbers showed an increasing trend which is possibly the result of long term conservation efforts in the region. The overall mean clutch size recorded for Akyatan Beach during ten consecutive reproduction seasons (112 eggs) is within the range for green turtle populations in the Mediterranean (Alagadi/North Cyprus: 116 eggs, Egypt: 101 eggs, Latakia/Syria: 108 eggs, Kazanlı/ Turkey: 111 eggs, Akyatan/Turkey: 114 eggs, see table S15; Casale et al. 2018). The clutch size in green turtles increases with the body size of the female (Broderick et al. 2003). Bjorndal and Carr (1989) indicated that Table 3. Number of nests recorded through the years and estimated number of females based on clutch frequency (CF) study in the Mediterranean region according to Broderick et al. (2002) (± SD) Years Nest number Estimated number of nesting females (Number nest/CF) Max (Number nest/CF) Min (Number nest/CF) References 2006 562 193.8 281.0 181.3 Yılmaz et al. (2015) 2007 170 58.6 85.0 54.8 2008 542 186.9 271.0 174.8 2009 261 90.0 130.5 84.2 2010 289 99.7 144.5 93.2 2011 346 119.3 173.0 111.6 2012 444 153.1 222.0 143.2 This study 2013 249 85.9 124.5 80.3 2014 433 149.3 216.5 139.7 2015 201 69.3 100.5 64.8 2016 559 192.8 279.5 180.3 2017 349 120.3 174.5 112.6 2018 341 117.6 170.5 110.0 2019 348 120.0 174.0 112.3 2020 532 183.4 266.0 171.6 2021 410 141.4 205.0 132.3 Overall 6036 2081.4 3018 1947 Fig. 7. Temporal trend in the number of green turtle nests over 16 consecutive years on Akyatan Beach. page 8 of 12 Zoological Studies 61:53 (2022) © 2022 Academia Sinica, Taiwan carapace lengths of females with a small mean clutch size in the 1976 nesting season were significantly shorter than all other years. The current study’s significantly different mean clutch sizes between the nesting seasons may be due to size differences of nesting females during those breeding seasons. However, since we have no measure of female size, we cannot precisely comment on this topic. Hatching success in our study (73.07%) was higher than Egypt (53.7%) and lower than Alagadi/ North Cyprus (84.2%), Latakia/Syria (80.0%), Kazanlı/Turkey (78.3%), Akyatan/Turkey (75.6%) and Samandağ/Turkey (81.1%) (see table S15; Casale et al. 2018). Also, the mean hatching success among the nesting seasons was determined to have significant differences, and showed a crucial decreasing trend over ten consecutive years. Hatching success can be affected by the nest’s location and its microhabitat (Zárate et al. 2013), temperature (Godley et al. 2001; Pike 2013; Hays et al. 2017; Türkozan et al. 2021), fungal infections (Limpus et al. 1983) and moisture (Wood and Bjorndal 2000). The hatching success is reduced when the temperature is higher (Godley et al. 2001; Pike 2013; Hays et al. 2017). Furthermore, Türkozan et al. (2021) reported that hatching success declines after conditions exceed the 33°C thermal thresholds for twofifths of the incubation period in Akyatan Beach. But the hatching success was increased despite the reduced incubation duration caused by the higher temperature in the 2021 nesting season. The low mean incubation duration in the 2021 nesting season probably resulted from more depredated nests (65.65%) laid until May and June 15th, possibly with a high incubation duration. Of the recorded embryonic deaths in nests, 69.54% were early-stage, 7.84% were mid-stage, and 22.62% were late-stage embryos. Compared to Yılmaz et al. (2015), while the early-stage embryos in our study were lower, mid-stage and late-stage were higher (earlystage embryos: 76.4%, mid-stage embryos 4.4%, and late-stage embryos: 19.2%). Environmental factors such as precipitation and temperature influence embryonic deaths (Rafferty et al. 2011). Booth and Dunstan (2018) found that the proportion of early embryo death was remarkable in two green turtle nests that experienced the highest nest temperature, lowest oxygen (PO2), and highest carbon dioxide (PCO2) during the first week of incubation. Also, the high temperatures experienced in the early stages of incubation might cause malformations, leading to increased mortality rates (Booth et al. 2020). The overall mean incubation duration (51.39 ± 3.46) was in the range of values reported for green turtle populations in the Mediterranean (Alagadi/North Cyprus: 51.1 days, Egypt: 46.5 days, Kazanlı/Turkey: 52.2 days, Akyatan/Turkey: 52.9 days, and Samandağ/ Turkey: 52.9 days, see table S15; Casale et al. 2018). Also, the incubation duration was determined to be significantly different between the nesting seasons. While variation in incubation durations is related to temperature profiles and sand characteristics on other nesting beaches (Yılmaz et al. 2015), the same nesting beach is affected by the temperature profile in different nesting seasons. Türkozan et al. (2021) found that incubation temperatures will significantly increase in future years in the Mediterranean region. This situation will further reduce the incubation durations of green turtle nests in the Mediterranean. The significant decrease in mean incubation durations over ten consecutive years found in our study supports this conclusion. The overall peak-nesting season is between 29th June–8th July on Akyatan Beach (Fig. 3), with the first nesting event on 19th May and the last nesting event on 23rd Aug (Table 2). Yılmaz et al. (2015) reported the latest nest records as 7th Aug. 2008 for green turtles in Akyatan Beach. Broderick and Godley (1996) reported that the first nest was recorded on 31st May 1994, and the last nest on 25th Aug. 1993 for green turtles in Cyprus. For Caretta caretta, the earliest and the latest nest records were reported on 19th May 2001 and 14th Sept. 1984, respectively, in Greece (Margaritoulis 2005). Furthermore, Weishampel et al. (2005) reported that earlier nesting of C. caretta was correlated with sea surface warming. The peak nesting season for green turtle nests in the Mediterranean is between 19–28 June in Akyatan Beach (Yılmaz et al. 2015) and 12–19 June in Northern Cyprus (Broderick and Godley 1996). The first, last and peak-nesting season of sea turtle nests in the same or different populations may respond to climate change between nesting seasons. Patel et al. (2016) reported that the nesting season of the loggerhead turtle could shift to an earlier date by as much as 50 to 74 days in the Mediterranean by the year 2100 because of the rise in air and ocean temperatures. Although non-depredated nests were caged with wire mesh in the morning survey, the predation rate (41.02% of nests) was high on Akyatan Beach. This predation rate is higher than the predation rate (33.4%) from the other study done on Akyatan Beach (Yılmaz et al. 2015). Also, the depredated eggs showed a significant increasing trend for ten consecutive years. This is because the nests were predated by wild boar (S. scrofa) despite being caged with wire mesh, especially in the 2012, 2014, 2016, 2017, and 2020 nesting seasons. Although the cages are placed at a depth of about 20 cm from the surface of the nests, the nests were predated by wild boar (S. scrofa). Engeman et al. (2016) demonstrated that cages and screens offer little page 9 of 12Zoological Studies 61:53 (2022)