Plate 9.1 Green turtle, Chelonia mydas, swimming over reef. 264 Marine Atlas of the Western Arabian Gulf

Transcription

1 Plate 9.1 Green turtle, Chelonia mydas, swimming over reef. 264 Marine Atlas of the Western Arabian Gulf

2 Jeffrey D. Miller University of Montana, USA

3 Plate 9.2 The Gulf short-fingered gecko (Stenodactylus khobarensis) hunts insects at night. Introduction Each of the four main groups of reptiles (lizards, snakes, crocodiles, turtles) has at least one member that is adapted for marine life. The northern Arabian Gulf is host to two of these groups: sea turtles and sea snakes. The coastal mainland provides habitat for lizards, snakes, and a freshwater turtle. Of the seven recognized species of sea turtles, five species live in the Arabian Gulf. Although similar in their general biology, these five species are classified into two families: (Cheloniidae, Dermochelyidae). The Cheloniidae is represented in the Arabian Gulf by four genera, each with a single species (Chelonia mydas, Green turtle; Eretmochelys imbricata, hawksbill turtle; Caretta caretta, loggerhead turtle; Lepidochelys olivacea, Olive Ridley turtle). The family Dermochelyidae contains a single species (Dermochelys coriacea, leatherback turtle). Although there has been no broad-scale assessment of their population biology in the Arabian Gulf, their general biology has been well documented by studies conducted elsewhere (Hirth, 1971; Witzell, 1983; Dodd, 1988; Marquez, et al., 1976; Pritchard, 1969; see also Lutz and Musick, 1997; Lutz, et al., 2003). Figure 9.1 Pictorial key to the identification of marine turtles in the Arabian Gulf (Modified from an Original Key by C.J. Limpus) 266 M a r i n e A t l a s o f t h e W e s t e r n A r a b i a n G u l f CHAPTER 9_FINAL_21APRIL2011_jpk_VERONA.indd /07/12 15:11

4 Plate 9.3 Green turtle, Chelonia mydas, in shallow water as she approaches the beach during the nesting season. The IUCN (2006) Red List gives the current status of the global population of green turtles as Endangered (Seminoff, 2002) and hawksbill turtles as Critically Endangered (Mortimer and Donnelly, 2007). Other species found in the Arabian Gulf are listed as Vulnerable for the olive ridley turtle (Abreu-Grobois and Plotkin, 2006) or as Endangered for the leatherback turtle (Hamann, et al., 2006) and loggerhead turtle (MTSG, 1996). The status of the Arabian Gulf populations is not known. However, it is known that local populations do not exist in isolation; they are linked to and interact with populations at the regional and larger scales (Bowen and Karl, 1997; Plotkin, 2003). Knowledge of local and regional populations is necessary because marine turtles often cross international borders as they move between foraging and nesting areas (Musick and Limpus, 1997; Plotkin, 2003; Bolten, 2003; Hoppell, et al., 2003). Data on habitat use, population distribution, population structure and dynamics, location of nesting sites, growth rates, and reproduction are necessary to support conservation management decisions (Lutcavage, et al., 1997). The number of studies from the Gulf region providing locally and regionally important data on a marine turtles is growing (Brown, 1979; Aspinall and Helleyer, 1995; Baldwin, 1995, 1996; Brown, 1983, 1984, 1985, 1990; Carp, 1976; Al-Ghais, et al., 1998; Health, 1989; Kami, 1997; Kinunen and Walczak, 1971; Loughland, et al., 1998; Loughland and Miller, 2006; Al-Merghani, et al., 1996, 2000; Miller, 1989a, b; Miller, et al., 1989; Miller, et al., 2004; Miller and Loughland, 2006; Miller and Abdulqader, 2009; Al-Mohanna, et al., 1998; Al-Mohanna and Meakins, 2000; Pilcher, 1999, 2000, 2003; Ross and Barwani, 1982; Salm and Salm, 1991; Salm, et al., 1993). Gasperetti, et al. (1993) provide a comprehensive review of marine turtle biology and research around the Arabian Peninsula region (see also Miller 1989a, b; Miller, et al., 1989, 2004). Al-Merghani, et al., (2000) summarize the data collected during studies of nesting green and hawksbill turtles on the Saudi Arabian Gulf islands. Recently, information was published concerning the influence of beach features and the activities of people and predators on nesting by hawksbill turtles along the coast of Qatar (Ficetola 2007, 2008). Plate 9.4 Measuring a green turtle as she returns to the water after nesting. Chapter 9: MARINE AND COASTAL REPTILES 267

5 Kuwait Iran Karan Island Jana Island Kingdon of Saudi Arabia Bahrain Qatar United Arab Emirates Figure 9.2 Satellite view of the Arabian Gulf showing the location of Karan and Jana Islands. 268 Marine Atlas of the Western Arabian Gulf

6 Plate 9.5 Green turtle returning to the water after laying her eggs. Plate 9.6 Hatchling green turtle crawling toward the sea after emerging from the nest. Chapter 9: MARINE AND COASTAL REPTILES 269

7 Methods Data collection and sources of information Published scientific documents, and museum records, as well as unpublished reports and observations, provide the basis for the present assessment of the status of marine turtles and sea snakes in the northwestern portion of the northern Arabian Gulf. The general field methods used to study sea turtles in the Arabian Gulf follow those outlined by Miller (1989b), Eckert, et al., (1999), and Pilcher (2004). Details of the methods used to collect data are provided in the various studies cited herein. Information on the nesting and foraging distribution of marine turtles in the Arabian Gulf territorial waters of Saudi Arabia is based on detailed studies by Miller (1989b), Pilcher (1999, 2000) and Al Merghani, et al., (1996, 2000). Information on the distribution of sea snakes is derived from museum records (Gasperetti, 1988); the biological information concerning sea snakes is based on summarized studies conducted elsewhere (Greer, 1997; Heatwole, 1999; Heatwole and Cogger, 1993) or as part of local assessment (Martens, 1996). Plate 9.7 Close up of a green turtle on the beach during the nesting season. Although some of the materials reviewed have inherent constraints and are limited in their temporal scale (e.g., field reports), taken together the existing data allow the current status of marine turtles and sea snakes in the Arabian Gulf territorial waters of Saudi Arabia to be summarized and assessed with a view for their long-term conservation. Identification Sea turtles can be identified using a combination of characteristics: the number of costal scales, the number of inframarginal scales on the bridge between the carapace and the plastron and the number of scales on the nose between the eyes and the nostrils. Although color and pattern differences occur between adults and hatchlings, the number of scales on the carapace and the bridge to the plastron, as well as between the nostrils and the eyes, remains the same in adults and hatchlings (see Figure 9.1). The configuration and width of the track can be used to identify the species when the turtle is not seen; the size of the remnant nest pit provides an indication of species. Plate 9.8 A turtle tag. Plate 9.9 Green turtle laying eggs. Sea snakes can be identified using a set of standard taxonomic characteristics (e.g., pattern, coloration, mid-body scale row count, number of labial scales, pattern), in many instances, characteristics exhibit a range in expression that often overlaps among species (Gasperetti, 1988). For example, there are similarities in both pattern and coloration among genera, as well as differences between juveniles and adults in some species. Further complicating identification, variation in pattern and coloration occurs among adults, and occasionally between the sexes, in the same species. As a result, multiple characteristics must be used to make a positive identification. Results of Turtle Studies Basson, et al., (1977) identified Karan and Jana Islands as being important breeding sites for green and hawksbill turtles, as well as several species of sea birds (Figure 9.2). Saudi Aramco biologists visited Karan Island in the 1970s and tagged several green turtles (Basson, et al., 1977). 270 Marine Atlas of the Western Arabian Gulf

8 Plate 9.10 Green turtle beginning to dig her nest. Chapter 9: MARINE AND COASTAL REPTILES 271

9 Plate 9.11 Hawksbill turtle, Eretmochelys imbricata, on the beach during the nesting season. Miller (1989b) provided the primary data on the biology of marine turtles in the Saudi Arabian portion of the Arabian Gulf; his reported was based on measurements of and interactions with 1,460 green turtles and 34 hawksbill turtles in 1986 and 1987 on Karan and Jana Islands. Since then, Al-Merghani, et al., (1996, 2000), Pilcher and Al-Merghani (1992, 1994) and Pilcher (1999, 2000) have expanded the existing data. The database assembled from original data sheets and field reports on 4,784 green turtles and 473 hawksbill turtles during the period between 1986 and 1997 forms the basis for the numerical results presented herein (Table 9.2). Plate 9.12 Green Turtle eggs partially exposed during excavation before being counted. Plate 9.13 Green turtle tracks. 272 Marine Atlas of the Western Arabian Gulf

10 Morphometrics Morphometrics of nesting turtles have been reported by each of the studies conducted on the offshore islands (Al-Merghani, et al., 2000). The bulk of the morphometric data comes from nesting female turtles. Only a few male green turtles and one male hawksbill turtle were captured during the years of intensive study. Female green turtles nesting on the Saudi Arabian islands averaged 98.2 cm in curved carapace length (Table 9.3); there was no significant difference for the CCL of the green turtles measured during the various study seasons (Anova: f [7, 2836] = 2.66, NS). The weight of nesting green turtles averaged kg (Table 9.3); although there was a significant difference in weight during the various seasons (Anova: f [6,655] = 5.29, Sig), the average weight and standard deviation are accepted as being representative of the population. Plate 9.14 Weighing green turtle eggs. Female hawksbill turtles nesting on the Saudi Arabian islands averaged 71.5 cm in curved carapace length (Table 9.3); there was no significant difference for the CCL of the hawksbill turtles measured during the various study seasons (Anova: f [8, 490] = 0.89, NS). The weight of nesting hawksbill turtles averaged 38.2 kg (Table 9.3); there was not a significant difference in weight recorded during the various study seasons (Anova: f [8, 236] = 1.76, NS). The pooled mean and standard deviation are accepted as being representative of the population. N Plate 9.15 Candling green turtle eggs. The one on the left does not contain a yolk; the one on the right contains a yolk and will produce a hatchling. The other will not. Plate 9.16 Replacing eggs into the nest chamber after being counted, measured, and weighed. Chapter 9: MARINE AND COASTAL REPTILES 273

11 Plate 9.17 These eggs have been replaced into the nest hole. The smaller eggs are called Yolkless eggs because they are just albumen and shell; they do not Contain a Yolk. Plate 9.18 Green turtle eggs. Plate 9.19 Measuring green turtle eggs. Turtle egg clutch size and hatchling morphometrics Both green and hawksbill turtles produce eggs that are encased in pliable, white, spherical shells (Miller 1987). Green turtles nesting on the Northern Arabian Gulf Islands produce slightly more eggs per clutch than do hawksbill turtles (Al-Merghani et al., 2000). Green turtle clutches contain an average of 87.1 (Std = 16.6, , 91 clutches, 10 eggs measured per clutch) eggs; there was no significant difference in the clutch size is produced by green turtles during the study period (ANOVA: F [3, 87] = 0.52, NS). Hawksbill turtles produce clutches that contain an average of 75.2 (Std = 16.7, , 134 clutches, 10 eggs measured per clutch) eggs; there was no significant difference among the study periods (ANOVA: F [6, 127] = 3.61, NS). The eggs laid by green turtles are larger than the eggs of hawksbill turtles (Table 9.4). The hatchlings produced from eggs laid by green turtles are larger and heavier than the hatchlings produced from hawksbill turtle eggs (Table 9.4). Figure 9.3 Coastal Saudi Arabia showing the distribution of nearshore seagrass beds (dark green) and areas of algal turf (light green) with aerial survey sectors superimposed (dashed lines). Insert Table (a) indicates that more turtles were seen in the most northerly sector and that the numbers of turtles observed decreased progressively into the Gulf of Salwa, coincident with the decrease in the amount of seagrass. Although the occurrence of turtles in sector f was calculated as turtles per square mile, the available data suggest that turtles inhabit the shallow nearshore area. Insert Table (b) provides a range in numbers of turtles nesting on the Offshore Islands. Karan Island hosts the greatest number of nesting green turtles, whereas Jana Island hosts the greatest number of nesting hawksbill turtles (based on Miller, 1989b; Pilcher, 1999, 2000; Al-Merghani et al. 1996, 2000). 274 Marine Atlas of the Western Arabian Gulf

12 Plate 9.20 Normal eggs lined up to be counted; yolkless eggs are placed in a group and counted separately. Plate 9.21 Eggs ready to be replaced back into the nest chamber. Plate 9.22 These eggs have been cleaned of adhering sand before being weighed. Measuring and weighing of eggs is done quickly and the eggs are returned to the nest chamber within one hour of being laid. Chapter 9: MARINE AND COASTAL REPTILES 275

13 Emergent success The average emergent success of hatchling green turtles was about 80% and typically in the range of 60% to 95%. Emergence success for hawksbill turtles was slightly lower; the average was 71%, with a range of 46% to 91%. In the absence of better data, these data are accepted as being indicative for the species in the Arabian Gulf. General species distribution The shallow water near the Saudi Arabian Coast provides large areas of seagrass which support adult, sub-adult, and juvenile green turtles (see Figure 9.2). The seagrass area is interspersed with patchy areas of rocky reef and each of the islands is surrounded by reef; these areas support hawksbill turtles. In addition, adult turtles use mating areas located near the offshore islands and nest on the offshore islands (Miller, 1989b). Foraging areas Green turtles feed on seagrass and algae (Mortimer, 1981; Bonnet, et al., 1985; Lanyon, et al., 1989b). Three foraging areas used by green turtles in Saudi territory were identified by Miller (1989b): the area within Dawhat Abu Ali, the shallow area north of Abu Ali and south of Safaniyah, the reef systems surrounding the offshore islands. Aerial surveys in the Gulf of Salwa showed that turtles occur in the area in low numbers although seagrass and algae appear to be abundant. Other areas along the coast also support seagrass and algal beds that in turn support at least small numbers of foraging green turtles (see Figure 9.3). The distribution of algae and seagrass is restricted to the nearshore, shallow water areas where turbidity is reduced enough for light to penetrate to the substrate. Coverage by seagrass, including Halodule uninervis, Halophila ovalis, Halophila stipulacea, in the shallow coastal and nearshore areas is about 11% of the total area (IUCN, 1987; see also Basson, et al., 1977 and McCain, et al., 1984). Hawksbill turtles feed primarily on sponges and other encrusting organisms that live within the complex environment of coral reefs (Meylan 1985, 1988; Van Dam and Diez, 1997). Their primary foraging habitat incorporates the areas around the offshore coral islands and rocky reef areas that occur on the ocean floor; occasionally the species may be found nearer the mainland where rocky reef areas support encrusting organisms. Mating areas Mating green turtles were observed around Karan Island in June 1986; no mating was observed after the July 10 (Miller, 1989b). Green turtles have been photographed mating in May (2009) near Jana. No mating hawksbill turtles were observed around Jana Island; however, when studies began on Jana, nesting had already begun. It appears that green turtles mate on Karan and Jana reefs, and possibly elsewhere; whereas, no area has been identified where hawksbill turtles mate. Nesting locations Gasperetti (in Litt. to Miller, 1989b) provided the first information based on personal observations of marine turtle nesting. He noted that in the 1940s turtles nested along the coast at places such as on the northern side of Abu Ali and northward along the coast. Burchard (in Litt., Miller, 1989b) stated that in frequent, low density nesting occurred on certain mainland beaches in the vicinity of Ras Tanura; similar information was presented by Basson, et al. (1977). Miller (1989b) conducted air surveys along the Saudi Arabia Coast between the border with Kuwait and Qatar at the bottom of the Gulf of Salwa in May of 1986 and He reported no sign of recent marine turtle nesting, although he did notice one hawksbill track and one green turtle track at Ras Tanura beach. None of the more recent sureys indicate that nesting occurs along the mainland coast of Saudi Arabia. However, this does not preclude the very infrequent use of the coast for nesting by a marine turtle. Both green turtles and hawksbill turtles utilize the offshore islands for nesting (Basson, et al., 1977; Miller, 1989; Al-Merghani, et al. 1996, 2000; Pilcher 1999, 2000), (see Figure 9.3). Karan Island supports the largest number of green turtle nesting attempts; whereas, Jana Island supports the largest number of nesting attempts by hawksbill turtles. Green turtles also nest on Jana Island and hawksbill turtles also nest on Karan Island. More hawksbill turtles than green turtles nest on Jurayd Island; more green turtles than hawksbill turtles nest on Kurayn Island. Very few turtles, mostly green turtles attempt to nest on Harqus Island. There is little information available concerning marine turtle nesting on Al-Arabiyah Island. Basson, et al., (1977) suggested at least some green turtles nest there; however, an aerial survey in May 2009 did not record any sign of nesting. Plate 9.23 Mating green turtles around Karan Island. 276 Marine Atlas of the Western Arabian Gulf

14 Nesting season The definition of the nesting season is based on the pattern of nesting recorded during several extended research periods spent on the islands (Table 9.2, Miller, 1989b). Hawksbill turtles begin nesting on the offshore islands in early May (sometimes late April); the numbers reach maximum (greatest number of turtles attempting to nest) in early June. The numbers quickly reduce so that very few hawksbill turtles are nesting on any of the offshore islands by the first week in July. In contrast, green turtles begin to nest in early June and continue nesting through September; the peak nesting time occurs in early July. Nesting attempts and re-nesting interval Miller (1989b) reported that the majority of green turtles attempted to nest three or more times at Karan Island during both the 1986 and 1987 nesting surveys, with an average of four nesting attempts; Pilcher (1992) stated that the average number of clutches laid by green turtles was 1.9. The majority of green turtles re-nested between nine and 18 days (average 14.3, Pilcher, 1992) following the previous sighting. The average number of clutches produced by hawksbill turtles was 1.7 during 1991 with an average re-nesting interval of 18.2 days (Pilcher, 1992). These values are within the normal range of the renesting interval for the species (Hirth, 1997; Witzell, 1983) Plate 9.24 Whenever eggs are being handled (during excavation, being weighed and measured, and placing them back into the nest chamber) care is taken to not invert or rotate them. Change of a rookery For unknown reasons, a small number of turtles change their nesting location during the nesting season or between nesting seasons. Miller (1989b) reported four turtles that changed their nesting site from Jana Island (June 24, 1986) from among the 37 turtles tagged that night. The time intervals to recapture were not equal (12, 28, 41, 17 days). No hawksbill turtles were recorded changing nesting sites. Remigration interval More recent surveys on both Karan and Jana Islands recaptured turtles that had been tagged in previous seasons (Table 9.5). The remigration interval between tagging and recapture range between two and nine years. The average interval between nesting seasons for green turtles in the Arabian Gulf is 3.8 years (SD = 1.458, range = 1-7, n = 90). One green turtle was recorded nesting in the next year after it was tagged (T8062, July 19, 1992 to June 20, 1993). A remigration interval of one year is highly unusual; more typically the remigration interval is two or more years (Hirth, 1997). One green turtle (K1255) was recorded nesting after an interval of four years (May 26, 87 - May 27, 91) and again after an interval of two years (May 20, 1991 to June 19, 1993) (see Miller, 1989; Pilcher, 1992). Plates 9.25 Green turtle laying eggs; the nest chamber is nearly full and she will begin to cover the eggs in a few minutes. The average interval between nesting seasons for hawksbill turtles in the Arabian Gulf is 3.9 years (SD = 1.859, range = 2-8, n = 14). No hawksbill turtles tagged in one year were recaptured attempting nesting in the next season; however, three (K1255, K1911, K6008) were recorded at two-year intervals. The two longest internesting intervals recorded were seven (K1265) and eight (K1002) years (see Miller, 1989; Pilcher, 1992). During a visit to Jana and Karan Islands in June 2008, five previously tagged turtles were encountered in a sample of 61 turtles examined (T13350, T13741, T13823, all green female adults found among 54 green turtles at Karan Island; T8933 green female adult; T13757 hawksbill female adult found among 7 turtles examined at Jana Island) and measured (no tagging was done during this short visit). The records for only one green turtle (T8933) were contained in the database available. It was originally tagged on June 24, 1993 at Jana Island and was recaptured on June 21, 08 at Jana Island. This turtle was originally measured to be 98.5 cm CCL and had a CCL of was cm when recaptured. These records indicate this turtle grew 12.7 cm in 14 years (0.9 cm/y). Chapter 9: MARINE AND COASTAL REPTILES 277

15 Figure 9.4 Diagram of marine turtle life cycle. Based on Lanyon, et al., Nesting site disturbance The accumulation of debris on the islands is a continuing problem for nesting turtles. Miller (1989b) noted that approximately 18% (15% to 22% counted on six nights during 1986) of turtles that attempted to nest on the northern side of Karan Island were prevented from doing so on their first attempt by accumulated debris. He also noted that dhows anchored offshore of Karan Island did not apparently impact on the number of turtles nesting on that side of the island; however, when fishing nets were set parallel to and within 50 m of the beach, the number of turtles nesting in that area was reduced. Even though the beaches were cleaned following the 1991 Gulf War, debris has accumulated (Pilcher, 1999) and debris, including discarded fishing nets, entangles and/or traps hatchlings as they try to cross the beach (Al-Merghani, et al., 1996, 2000; Pilcher 1999, 2000). Plate 9.26 Early morning view of the nesting area on Karan Island with green turtle tracks leading back toward the ocean. 278 Marine Atlas of the Western Arabian Gulf

16 Discussion Plate 9.27 Green Turtle in a body pit using her hind flippers to Cover the Nest Site; Soon she will throw sand with her front flippers to Cover the Site Before she Returns to the Sea. Marine turtle biology The following summary emphasizes the biological information on the green and hawksbill turtles that nest and forage in the Arabian Gulf; because marine turtles share a general life cycle (Figure 9.4) and a general set of biological characteristics it contains general information on all species (Table 9.6; see also Miller 1989b, 1995, 1997; Dodd, 1988; Hirth, 1997; Witzell, 1983; Marquez, et al., 1976; Pritchard 1969). It should be remembered that most research and conservation efforts occur on beaches (Lutz and Musick, 1997; Lutz, et al., 2003). As a result, it is imperative that conservation of marine turtle populations must take into account their temporal and spatial requirements, as well as their habitat use. For example, marine turtles nesting on an island do not forage in the same area; they migrate from several, widely separated foraging areas. Conversely, all the reproductively active turtles in a foraging area will not migrate to the same nesting site, but will utilize multiple nesting beaches that maybe located across different geopolitical boundaries. Further, it must be remembered that the ability of marine turtles to respond to rapid environmental change is constrained by their reproductive strategies, and habitat requirements, as well as their behavioral and morphological characteristics. During their life cycle marine turtles use different habitats, including shallow water foraging areas, nesting beaches, and the open ocean (Bjorndal, 1997) (Figure 9.4; Table 9.6). Juvenile, sub-adult and adult marine turtles forage in shallow areas (usually < 10 m). In these areas green turtles feed on seagrass and algae (Mortimer, 1981; Bonnet, et al., 1985; Lanyon, et al., 1989) and hawksbill turtles forage for sponges and other encrusting organisms in complex reef habitats (Meylan 1988, 1985; Van Dam and Diez, 1997). In foraging areas younger turtles grow to maturity while mature turtles forage in preparation for reproduction. When they are ready to reproduce, turtles migrate to mating areas, usually located closer to the nesting beach than to the foraging area. Turtles migrate between their foraging area and their nesting area repeatedly during their reproductive lifetime (i.e., nest site and forging area fidelity). After mating, females leave the water to oviposite their eggs at about 2 weekly intervals. Eggs incubate in beaches without maternal care. Developing embryos are subject to the environmental conditions affecting that particular beach (see below). After emerging onto the beach, hatchlings descend the beach and disperse to the open ocean where they remain for a decade or more. A reproductively active marine turtle will produce between 50 and 100+ eggs per clutch and produce several clutches (3-8) during a nesting season. Between nesting events within one season, female turtles rest and prepare the next clutch of eggs in a nearby inter-nesting habitat. At the end of its reproductive season the female marine turtle returns to her foraging area, where she recovers and prepares for the next reproductive season several (2 to 5, or more) years into the future. Most species of marine turtles show iteroporus reproduction (i.e., typically they do not nest every year). In Eastern Australia it appears that males breed on a 1, 2 or 3 year reproductive cycle (Hamann, et al., 2003); whereas, the female turtles breed on a much longer cycle (three to five years or more) (Hamann, et al., 2003). Both sexes must migrate to the mating area for reproduction. Both male and female turtles may mate with multiple partners. As a result the eggs in a single clutch have the possibility of being fertilize by sperm from several different males; the contribution by each male changes in successive clutches. Females are receptive to males for about 3 weeks, after which she stores sperm for the remainder of that reproductive season. Each female arrives at the nesting beach with ovarian follicles that were formed while in her foraging area. After insemination, the process of egg formation (i.e., ovulation of the follicles, mobilization and deposition of the components of the albumin, and secretion of the egg shell) requires approximately 2 weeks. This time requirement determines the interval between successive clutches. Chapter 9: MARINE AND COASTAL REPTILES 279

17 Beaches used for nesting must be accessible from the sea and contain few obstacles that restrict access or digging. The sand must be moist enough to retain stability and not collapse as the turtle digs the egg chamber. Also, the nesting beach must be stable through time because marine turtles return after several years to the same beach or one nearby. Further, the sand of a nesting beach must be low in salt content and high in moisture, as well as being well-drained and well ventilated to be suitable for incubation. A nesting beach must also be adjacent to currents that help the hatchlings disperse to the open ocean. Not all beaches that have appropriate characteristics are utilized by marine turtles for nesting (e.g., Johannes and Rimmer, 1984). The beach must provide a range of temperature between about 25 C and 33 C to support embryonic development. The temperature experienced by the embryo during the middle third of incubation determines the sex of the hatchling (i.e., temperature dependent sex determination; Georges, et al., 1994; Wibbels, 2003). Female hatchlings emerge from eggs incubated at temperatures above 29 C and male turtles emerge from eggs incubated at temperatures below 29 C. The pivotal temperature (theoretical point at which a 50:50 sex ratio is produced, sensu Wibbels, 2003) has not been determined for any nesting population in the Arabian Peninsula region. Data from elsewhere in the world indicate that the pivotal temperature is approximately 29 C; it may be higher. After emerging onto the beach, hatchlings disperse from their natal beach to the open ocean for a number of years. The dispersal pattern of hatchlings produced from the Arabian Gulf islands has not been determined (i.e., only in the Arabian Gulf or if they exit through the Straits of Hormuz into the Indian Ocean). Small turtles experience high mortality from a variety of fishes (e.g. grouper, sharks) as they grow slowly toward maturity; mortality is highest among smaller turtles and reduces as the turtles become larger. Marine turtles require several decades to reach sexual maturity; studies elsewhere in the world suggest that 30 to 50 years may be required for a hatchling to become a member of the breeding population. The long time to sexual maturity supports a long breeding life. How long a marine turtle may live, and whether it may reach reproductive senescence are not known. Plate 9.28 Measuring a green turtle. Plate 9.29 Filming a green turtle as she is laying eggs. Plate 9.30 After carefully digging out the sand behind a laying green turtle, the eggs dropping into the nest chamber was filmed without disturbing the turtle. 280 Marine Atlas of the Western Arabian Gulf

18 Plate 9.31 Green turtle tracks. Marine turtle nesting Little sign of recent marine turtle nesting has been reported along the Coast of Saudi Arabia in several decades; the only substantial nesting activity recorded since 1986 has occurred on the offshore islands. However, because marine turtles occasionally use nesting sites in the general vicinity of the primary nesting beach, a turtle may infrequently nest along the mainland coast. This can only be interpreted as an irregular nesting. An annual aerial survey along the coast and around the offshore islands should allow identification of any areas utilized for nesting. Because remnant nest pits may remain on the beach for many years and because turtles utilize traditional nesting sites, nesting areas are relatively easy to distinguish by the presence of numerous nest pits. The presence of nest pits provides an indication of where turtles choose to nest but the number of pits does not correlate with the number of nesting turtles beyond a very general level because the evidence of pits remains visible over several years (Miller, 1989b). Tag returns The recapture of a turtle that was tagged previously in the current nesting season provides insight into the repetitive nesting behavior of the turtles and provides data on the internesting interval. Many of the turtles tagged between 1986 and 1994 re-nested multiple times within the nesting season (Table 9.2). Because coverage of each island was not continuous throughout the nine years of intensive study, it is likely that some nesting attempts were missed; the actual number of clutches laid by green and hawksbill turtles may be slightly higher than recorded. The data clearly indicates that these turtles, like other marine turtles around the world, nest more than one time in a reproductive season. A small proportion of the animals tagged in the early years were recaptured nesting in subsequent years. The number of turtles returning to lay eggs in a subsequent nesting season is probably higher than recorded because tags applied improperly fall off, as do tags of animals that exhibit long remigration intervals. More turtles may well have returned to nest at re-migration intervals that fall outside of the period of monitoring. Only long-term studies that cover the nesting period are necessary to better define the remigration patterns. Interchange among nesting sites Although the exact reasons why a turtle would shift from one nesting site to another within a nesting season are difficult to identify, the shift in nesting site is costly to the turtle and most likely results from some disturbance. The interchange among the islands within and between nesting seasons clearly indicates that the turtles perceive these islands as being part of a single nesting area, although individuals show a strong preference to one island, the fact that they will shift nesting from Jana to Karan or vice versa indicates that they know where these islands are and the repetitive nesting within and between seasons identifies these islands as being extremely important to both green and hawksbill turtles. No other species of marine turtles has been recorded nesting on the Saudi Arabian Gulf islands. Marine turtle foraging areas Marine turtles spend the majority of their lifetime in their foraging areas, growing to maturity and preparing for reproduction, or recovering from a breeding episode. Once a turtle has reached sexual maturity, the foraging area provides the food (energy) to support growth and reproduction. The quality of the general foraging habitat, as well as the food it contains, provides the resource base for the population and its reproductive output. Chapter 9: MARINE AND COASTAL REPTILES 281

19 Summary of marine turtle reproductive characteristics in the Arabian Gulf Green and hawksbill turtles show distinct seasonality in nesting in the Arabian Gulf (Miller, 1989b). Hawksbill turtles nest earlier in the summer than green turtles, although some overlap occurs at both Karan and Jana Islands in June (Miller, 1989b). This timing of nesting probably results from the populations being constrained by differing environmental conditions. At Tortuguero in Costa Rica green turtles nest before hawksbill turtles (Bjorndal, et al., 1985) and both species nest in concert in the northern Great Barrier Reef, Australia (Limpus, unpublished data; Limpus and Miller, 1994). Green turtles are larger than hawksbill turtles and they produce larger, as well as more, eggs (Table 9.7). In the Arabian Gulf nesting green turtles are about the same size as those found nesting in Oman (Ross and Barwani, 1982); they are slightly smaller than nesting green turtles from around the world (Hirth, 1997; Miller, 1989b; Van Buskirk and Crowder, 1994). In contrast, Arabian Gulf nesting hawksbill turtles are among the smallest in the world (Witzell, 1983; Miller, 1989b; Van Bushkirk and Crowder, 1994). Both species produce clutches that contain fewer eggs than green and hawksbill turtle populations nesting elsewhere in the world (Hirth, 1997; Witzell, 1983; Miller, 1989b; Van Buskirk and Crowder, 1994; Pilcher and Al-Merghani, 1994). Unfortunately, the number of hatchlings produced (emergence success) has not been determined. The number of hatchlings produced is a better indicator of the health of the incubation system and of the population using the beach than is the number of eggs laid or the size of nesting turtles. Plate 9.32 Plate 9.33 Plate 9.34 Plates Hatchling green turtle emerging from the sand and starting to crawl toward the ocean. 282 Marine Atlas of the Western Arabian Gulf

20 Plate 9.36 Green turtle searching for a place to dig her nest chamber. Plate 9.37 Hatchling green turtle beginning to swim in the shallow water next to the beach. Chapter 9: MARINE AND COASTAL REPTILES 283

21 LWM 2 metres Plate 9.38 All marine turtles come ashore to lay their eggs and, as a result, share their nesting habitat with ghost crabs, birds, and many other animals. The beaches on which they nest are a boundary zone between the ocean and the land, with vegetation on one side and ocean on the other. In the water, different species of turtles forage for different foods; green turtles eat mostly seagrass and hawksbill turtles consume sponges and other encrusting organisms. Turtles share their foraging habitat with a mryiad of sea creatures, including fish, sharks, and sea snakes (drawing by Kym Wissemann). Plate 9.40 Different species of sea snakes forage for different species of small, elongated fish in different parts of the marine environment. Some species prefer the shallow water of the lagoon; whereas, most forage among the complex structures of the coral living on the reef slope. A few species forage along the bottom in the deeper water or near the surface of the open water. Although sea snakes in the Arabian Gulf do not come ashore to lay eggs, they all must come to the surface to breath (drawing by Kym Wissemann). 26 Significance of Saudi Arabia for marine turtles Sea Snakes The shallow water located offshore of Saudi Arabia provides important foraging areas for green and hawksbill turtles and the offshore islands provide essential nesting sites in the Arabian Gulf. Because the area between the coast and the islands includes rocky reef embedded in expanses of seagrass, both species occur together in some areas. Saudi Arabian islands in the Northern Arabian Gulf support the largest populations of nesting green and hawksbill turtles in the Arabian Gulf. Virtually no nesting occurs along the Saudi Arabian Coast (Miller, 1989b); however, nesting occurs in Iran (Kinunen and Walczak, 1971; Kami, 1997). A small amount of nesting turtles occurs along the Coast of Qatar (Ficetola 2007, 2008) and occasionally in Kuwait (Al-Mohanna, et al., 1998; Al-Mohanna and Meakins, 2000). Future directions The basic reproductive characteristics of the northern Arabian Gulf populations of green and hawksbill turtles have been defined by the nine years of intensive study (Miller, 1989b; Al-Merghani et al., 1996, 2000; Pilcher 1999, 2000). The general lack of even intermittent intensive study of the nesting populations of the offshore islands means that in order to measure change in the populations as a result of the increasing utilization of the area surrounding the offshore islands (principally for oil and gas extraction), intensive studies need to be conducted for at least a few more consecutive years to collect new data on the populations. In addition, two other issues need to be addressed during the reevaluation of the nesting populations. First an intensive tagging program should be conducted over two or more years coupled with regional cooperation to find and link the foraging areas used by these turtles within the Arabian Gulf. This will require a concerted effort by all nations that border the Arabian Gulf. Second, the dispersal of hatchlings needs to be defined. The questions of (1) whether hatchlings remain in the Arabian Gulf, or (2) whether they enter the Gulf of Oman and the range into the Indian Ocean need to be answered. While the hatchlings remain in the Arabian Gulf, they are subject to the conditions of the Gulf, including seasonal changes in salinity, seasonal changes in temperature, and the impact of oil spills and debris. Broad-scale genetic sampling at nesting and foraging areas throughout the Arabian Gulf and the Gulf of Oman would address these issues and provide for the focusing of more detailed research efforts at specific locations. Sampling nesting turtles and emerging hatchlings is relatively easy; however, sampling both males and females (including juvenile and some adults) in the foraging areas is more difficult. Methods exist for doing this type of work. Local biologists could be trained to collect samples, which could be sent to a single laboratory for analysis to ensure uniform application of analytical methods; a data rich report that defines the interrelationships among the various nesting sites could then be returned to the individual research units and member states for consideration and implementation of conservation management action. Plate 9.39 During hatching, the baby turtles often stop trying to get out of the egg shell to rest. Hatchlings have a small tuberical on the top of the nose that helps them to tear through the egg shell (drawing by Kym Wissemann). Classification Sea snakes belong to the family Elapidae (Greer, 1997). In recent years the family and contained genera have gone through a number of revisions. For example, the generic name for Hydrophis (Microcephalophis) gracilis has been reconsidered a few times (see McDowell, 1972; Cogger, et al., 1983; Gasperetti, 1988; Greer, 1997; Heatwole, 1999; Heatwole and Cogger, 1993 for a history of the taxonomy); this species has a very small head and looks quite different from the other members of Hydrophis. There will be future revisions of the taxonomy of the sea snakes as more is learned about their genetics, morphology, and biology. All species are highly venomous. Data concerning the distribution of sea snakes in the Arabian Gulf, as in many other parts of the world, has been collected ancillary to other work. Most records of sea snakes in the Northern Arabian Gulf come from incidental sightings from ships, aerial surveys, or from beach washed specimens found along the coast (Gasperetti, 1988; Martens, 1996). Nine species representing five genera have been recorded in the Arabian Gulf (Table 9.8); whereas, only seven species have been recorded in the northern portion. This reduction in number may result from a physiological limitation (i.e., temperature or salinity tolerance), or the lack of surveys. General distribution With the exception of Pelamis platurus, sea snakes are restricted to tropical and subtropical waters, including the semitropical waters of the Arabian Gulf (Gasperetti, 1988; Greer, 1997; Heatwole, 1999; Heatwole and Cogger, 1993). The one exception (Pelamis platura) is tolerant to a wide range of environmental conditions (i.e., temperature, salinity) and it is able to tolerate cooler waters than other species. The Arabian Gulf appears to be suitable habitat for nine species, with a 10th species recorded in the Gulf of Oman (Table 9.8). Distribution in Arabian Gulf Gasperetti (1988) provided a review of the distribution of sea snakes in the Arabian Gulf based on museum records; his maps show gaps in the known distribution that may result from the paucity of records or some environmental or physiological limitation. Martens (1996) confirmed the presence of Pelamis platura, Hydrophis cyanocinctus and Hydrophis lapemoides from the vicinity of Abu Ali and Jubail on the Saudi Arabian coastline; he also reported seeing 21 Hydrophis sp. Drifting on the surface of the water or while they were swimming slowly above the sandy sea bottom. The addition of these specimen records underscores the need for systematic surveys to clarify the occurrence and distribution of all species of sea snakes in the Arabian Gulf. For example, the museum records presented by Gasperetti (1988) indicate that Hydrophis lapemoides does not occur in the northern Gulf. However Hydrophis lapemoides was reported by Farmer (1983) in Kuwait and by Martens (1996) near Abu Ali Island. Astrotia stokesii may venture into the lower portion of the Arabian Gulf but it has not been recorded from inside the Arabian Gulf (Gasperetti, 1988). The record of the distribution of another species in the Arabian Gulf is also questionable; Enhydrina schistose has been recorded from the Straits of Hormuz and may well enter the Gulf; however, its distribution with in the Arabian Gulf needs to be confirmed. Gasperetti (1988) listed several museum records that had been labeled only as Persian Gulf; unfortunately, the locality Persian Gulf could mean specimens were collected in the Arabian Gulf or in the Straits of Hormuz or Gulf of Oman. The distribution of Praescutata viperina in the Arabian Gulf is less problematic. There is a museum record of from the territorial waters of U.A.E. and several other records from the Persian Gulf (location(s) not recorded, possibly the Gulf of Oman). Current records indicate that this sea snake does not occur in the northern portion of the Arabian Gulf; however, this may reflect the lack of surveys in the northern Gulf. The other seven species (Table 9.8) have been collected in the Northern Arabian Gulf. 284 Marine Atlas of the Western Arabian Gulf Chapter 9: MARINE AND COASTAL REPTILES 285

22 General biology and ecology Although the habitat preferences of most sea snakes are not known, they typically live in water less than 30 m deep that contains a complex habitat (Gasperetti, 1988; Heatwole and Cogger, 1993; Greer, 1997; Heatwole, 1999, Guinea, 2007). The habitats used by sea snakes range from turbid water, to seagrass areas, and three-dimensional rocky reef areas (Heatwole, 1999; Guinea, 2007). For example L. curtus and H. ornatus hunt in nearshore turbid water over muddy and sandy substrates. Other species hunt in seagrass areas; whereas other species hunt around rocky reefs. Most species feed on several species of small, elongate, bottom-dwelling fish, such as gobies. Some specialize in catching eels (H. spiralis, H. gracilis) or consuming fish eggs (e.g., H. gracilis). Other species feed on a narrow range of prey; for example, Enhydrina schistosa feeds mostly on puffer fish and small catfish. In contrast, the pelagic Pelamis platurus is a generalist that feeds on a variety of small fish at the surface (19 families of fish, Kropach, 1975). The majority of dietary items have an elongate body shape; presumably selection of these prey species facilitates swallowing. Habitat use by each species may simply reflect the occurrence of prey species rather than selection for the structure of the habitat by the snake. As a result of habitat used by prey species, sea snakes often occur in mixed populations (Heatwole and Cogger, 1993). The assemblage of sea snakes in the Arabian Gulf is probably separated by diet and, therefore, by differences in the habitats in which the prey species live, as much as by physiological tolerances for salinity and temperature. More research needs to be conducted in the Arabian Gulf to determine the habitat use and prey species utilized by sea snakes. Reproduction Sea snakes of the Arabian Gulf give birth at sea to young that must immediately fend for themselves: getting to the surface to breathe, finding food, and avoiding predators. There are only a few observations on the reproduction of sea snakes in the Arabian Gulf. Martins (1996) reported that a Pelamis platurus (500 mm SVL) contained three well-developed embryos in January 1992; his report also noted that the ovaries of a female Hydrophis cyanocinctus contained 12 eggs (large follicles) in May and that the ovaries of a female Hydrophis lapemoides collected in the Gulf of Salwa contained nine eggs. As with all aspects of the biology of sea snakes in the Arabian Gulf, their reproduction, including the seasonality and reproductive output requires more data before the viability of the populations can be determined. Places where sea snakes give birth have not been identified; it is presumed that they give birth in the complex habitats in which they forage rather than moving to other areas. Plate 9.41 The capsian turtle (Mauremys caspica) can be found in freshwater or tidal brackish water along the coast. Coastal reptiles and amphibians Over the years, the terrestrial herpetofauna of the Eastern Province has been described as part of broad-scale assessments of the Arabian Peninsula (Joger, 1984; Balletto, et al., 1985; Arnold 1986, 1987; Gasperetti, 1988; Leviton, et al., 1992; Gasperetti, et al., 1993); these studies deal more with taxonomy than with autecology but include ecological information. The few small-scale studies better define the habitat(s) used by the resident species and provide detailed information about their ecology (e.g., Martens, 1996; Ross 1988, 1989, 1993). Recently, there has been a renewed interest using modern mapping methods (e.g., see Gardner, 2009) and re-evaluation of taxonomy (e.g., Wilms and Böhme 2000, 2007; Wilms, et al., 2009) that is contributing to a better understanding of the herpetology of the region. Currently, the terrestrial herpetofauna of the Arabian Gulf Coast includes 27 species of lizards, and 10 species of snakes, as well as one amphisbaenian, one freshwater turtle and one amphibian. 286 Marine Atlas of the Western Arabian Gulf

23 Plate 9.42 The toad-headed agama (Phrynocephalus arabicus) observing its habitat from a rocky outcrop. The only amphibian known from the coast of the Eastern Province is the marsh frog, Rana ridibunda, which occurs in the oasis areas at Al Qatif and Al Ahasa; it has not been reported from other areas along the coastal margin, or any of the other nearshore or offshore islands (Balletto, et al., 1985). The Marsh frog lives in freshwater areas, such as the irrigated oasis of Al Qatif and wetlands at Al Ahasa, where it forages and breeds. This frog has a wide distribution in central Europe, extending into Iran to the Straits of Hormuz. Its occurrence in Al Qatif may well be the result of an introduction during trade across the Arabian Gulf. The freshwater turtle, Mauremys caspica, inhabits the freshwater channels and wetlands of the Al Qatif, Al Ahasa and Al Uqair oasis. Although it prefers freshwater, this turtle may be found in drainage ditches that lead into tidal areas and it is able to tolerate excursions into salt water. It forages for a wide range of prey along the bottom, consuming both animal and plant material. In Bahrain, the recently introduced freshwater turtle, Trachemys scripta, appears to be an ecological competitor of Mauremys caspica and is becoming the dominate species in several oasis (Gasperetti, et al., 1993). Because the main distribution of the species is to the north of the Arabian Peninsula, including western Iran to the Caspian Sea, the populations in these oasis, and in Bahrain, may have resulted from human transport or they may be relics of the Würm glaciations and intervening warmer periods (Gasperetti, et al., 1993). The single species of amphisbaenian, Diplometopon zarudnyi, appears snake-like, but it is actually a legless lizard. This nocturnal lizard spends the majority of its life burrowing through soft, mesic sandy soil foraging for food; infrequently, it may be found under pieces of wood or other debris (Leviton, et al., 1992). Its habitat is restricted to mesic soils; for the most part, such areas do not occur along the coastal margin or on the islands. Terrestrial lizards and snakes are the common reptiles found along the coast and on the islands. Martens (1996) reported 14 species of lizards representing five families (Agamidae, Gekkonidae, Lacertidae, Scincidae, Varanidae) and four species of snakes from three families (Boidae, Colubridae, Viperidae) from the Jubail Wildlife Marine Sanctuary. Because these species are able to cross short distances of salt water, their distribution is not limited to the coastal mainland and several species may occur on the nearshore islands. However, because systematic surveys of the coastal islands of Saudi Arabia have not been done, the occurrence of these species on nearshore islands is not documented; there is probably a reduced number of species on the nearshore islands resulting from the reduced vegetation and habitat diversity. Only two species of lizards have extended their range to include the offshore islands. No snakes have been recorded from the offshore islands. In 1987, both Joger and Arnold summarized the zoogeography of reptiles and amphibians of the Arabian Peninsula based on their known distribution and taxonomic affinities in the context of the general physiographic characteristics (such as Central and Eastern arid regions and the mesic Southwestern and Southern highlands) of the region. The herpetofauna of the Arabian Peninsula has strong affinities with the large desert areas to the east and west. Although about 55% of the amphibian and reptile species are endemic to the Peninsula, most belong to genera that are spread over the entire Saharo-Sindian subregion (Arnold, 1987; Joger, 1987; Balletto, et al., 1985; Gasperetti, 1988; Baha El Din, 1996). Chapter 9: MARINE AND COASTAL REPTILES 287

24 On a much smaller scale, Martens (1996) divided the reptiles found at 14 sites in the hinterland of Jubail into three general assemblages, each associated with different substrates and vegetation (Table 9.9). Among the stenoecious species, four species were found only in the Aeolian sand areas (3 lizards; 1 snake); whereas, five species (3 lizards; 2 snakes) and three species (3 lizards; 0 snakes) were found in the Sabkha and consolidated substrate areas, respectively. Among the euryoecious species that utilize the wide range of habitats around Jubail, two lizards and the amphisbaenian were found in all three habitats; whereas, three lizards and one snake were found in two of the habitat areas. Three species (Hemidactylus flaviviridi; Hemidactylus persicus; Mabuya aurata septemtaeniata) were collected outside the boundaries of the Jubail study area and were not associated with the habitats being surveyed (Martens, 1996). Another group of four lizards and one snake that may occur along the coastal margin were not found by Martens (1996) (Table 9.10); the lizards are typically less than 25 cm in total length, diurnal in habit, and utilize a wide range of habitats. The snake is medium sized (40-50 cm total length) and nocturnal. The species that comprise these loose assemblages of desert reptiles differ in many aspects of their ecology. The lizards utilize different habitats with the Jubail hinterland; some of the species are stenoecious, whereas others are euryoecious (ratio 9:6). The stenoecious species utilize microhabitats within the more general habitat; for example, each of the three diurnal species using the Aeolian sand areas uses the microhabitat differently (Baha El Din, 1996). Scincus mitranus forages on the surface, often near vegetation, but burrows into loose sand to hide. Acanthodactylus gongrorhynchatus actively forages in the open areas situated among hummocks of vegetation; when threatened, it hides in the debris at the base of the plants. Phrynocephalus arabicus wiggles itself into the loose surface sand to watch for prey and avoid predators. In contrast, the euryoecious species use the entire habitat area. Lizards use their preferred habitats (and microhabitats) at different times. At the general level, 10 of the species in the study in the Jubail hinterland were diurnal and five were nocturnal. Among the snakes, two are considered nocturnal and two diurnal. Although useful in thinking about the general ecology of these reptiles, it must be remembered that these are somewhat artificial distinctions. For example, some lizards are active at different times of the day. Three species of Acanthodactylus living in the same area in Eastern Saudi Arabia began foraging at successively later times of the day (Ross, 1989); a similar pattern was found among three unrelated species in the U.A.E. (S. mitranus was active before A. schmidti, which was active before P. arabicus) (Arnold, 1984). In addition, the seasonal change in temperature may affect the foraging times of lizards; for example, many diurnal species have a single activity period during cooler months, but as the daytime temperatures increase, they shift their activity pattern to being active in the morning and later in the afternoon, rather than during the hottest time of day. Like the lizards, the diurnal snakes may become more crepuscular as daytime temperatures increase. Key: D = Diurnal, N = Nocturnal, W = Wide distribution, R = Restricted distribution, B = Burrowing Besides habitat use and activity time, the coastal reptiles can be separated into loose groups based on size and food preference. The two largest lizards (Varnus griseus, known as monitor; Uromastyx aegyptius, known locally as dhub) grow to be more than 75 cm long; whereas, the others are much smaller (usually less than 25 cm in total length). The dhub is primarily herbivorous and the monitor is mostly carnivorous, in contrast to the smaller lizards which are insectivorous. There is much more ecological research to be done to elucidate the details of the ecology of these desert reptiles. 288 Marine Atlas of the Western Arabian Gulf

25 Threats to Reptiles and Amphibians Plate 9.43 Stenodactylus khobarensis. Saudi Arabia is home to a wide variety of reptiles (20 Families; 54 Genera; 103 Species) and a few amphibians (3 Families; 4 Genera; 7 Species) ( 2010). The complex terrestrial and marine habitats of the coastal zone of the Eastern Province host a subset of this diversity. These areas are used extensively by people for a variety of activities, including camping, grazing, oil facilities, agriculture and fishing. As the coastal zone habitats are modified for human use, the complex habitats used by reptile and amphibian species are destroyed or degraded to a level where the populations cannot survive. Habitat loss and degradation are the largest threats to reptile and amphibian populations. On the terrestrial side, for example, grazing of livestock in densities that alter the vegetation or the increase the compaction of the substrate can disrupt foraging and breeding. The eastern coastal plain has a series of off-road tracks that parallel the beach. Such roads compact the substrate, damage the drainage, and destroy vegetation. In addition reptiles may be killed while crossing or basking on the off-road tracks, such as lizards, e.g., Dhub (Uromastyx aegyptius). Stream and water body pollution is another cause of habitat loss. The single frog species found in the Eastern Province is particularly susceptible to changes in water quality and pollution from agriculture and houses. The freshwater turtle is more tolerant of these changes but is still affected by the changes in the environment in which it lives and forages. The lizards and snakes require areas of suitable habitat in which to live. Dhubs live in loose colonies and need large areas of undisturbed habitat in which to forage; monitors are solitary and roam over large areas as they forage. As the human population increases in an area, so does the number of reptiles killed (particularly snakes) because the people do not understand the role these animals have in controlling the population of other species (e.g., mice, insects). Also, an increase in the number of people in an area leads to an increase in the number of cats and dogs; in turn, this increases the incidence of reptile predation, particularly near human settlements. The greatest risk to terrestrial reptiles and amphibians in the Kingdom of Saudi Arabia comes from the uncoordinated, localized development along the coast. Development in one area may not appear to cause much change in the hinterland; however, as more and more areas are developed (sabbkas filled, hummocks flattened, vegetation over grazed) the overall habitat required by the herpetofauna will degrade. Marine reptiles are subject to the threat of habitat loss/destruction; they are also subject to direct harvest and fisheries by-catch. Foraging and reproductive habitats are absolutely essential to support the life cycle of marine reptiles (Dobbs, et al., 2007), just as they are for terrestrial herpetofauna. Rocky reefs areas are necessary for hawksbill turtles where they feed on encrusting organisms; shallow areas of seagrass meadows are required to support green turtles. Sea snakes also forage in these areas. In addition, because marine turtles come ashore to lay eggs, human harvest of adults on nesting beaches, human harvest of eggs, and predation of eggs/hatchlings by other species further impact the population. Sea snakes do not come ashore to give birth; rather they use complex habitats of the reef and seagrass areas when giving birth. Both marine turtles and sea snakes are caught in fishing nets (both commercial and artisanal); although some turtles and sea snakes may be released, many are drowned. Chapter 9: MARINE AND COASTAL REPTILES 289

26 Further, climate change looms as a threat to benthic and island organisms; the projected rise in sea level will limit the amount of light reaching the benthic communities through the turbid water of the Gulf, which in turn, will alter the ecological interactions and food availability. The projected increase in storms will likely produce tidal surges that will inundate the low island beaches, reducing nesting habitat for turtles and sea birds. Another important threat to sea snakes and sea turtles in the Arabian Gulf is from oil spills because they must come to the surface to breathe. Although oil spills occur infrequently and are usually cleaned up rapidly, a localized oil spill may foul their nostrils and skin when they come to the surface to breathe. Such fouling may cause slow starvation by impairing their ability to find food. The greatest risk to marine turtles and sea snakes, as well as other marine species and their habitats, in the Kingdom of Saudi Arabia comes from the localized development in the marine environment. Development in one area offshore may not appear to cause much of a change in the sea grass or the light visible from the islands; however, as more and more areas are developed, the overall habitat required by marine reptiles will degrade. For example, in 1986 virtually no lights could be seen from Karan Island although there was a glow on the horizon from the port area of Jubail; however, in June 2008, multiple onshore and offshore facility lights were visible in virtually every direction from the island. Bright lights disorient hatchling sea turtles and may cause them to linger in areas where there is an increase in predation; similarly, adult turtles can be disoriented during their nesting. The solution lies in finding a balance between the need for the lights for safety of people working on the facility and the need for sea turtles to view a horizon without artificially bright areas. Plate 9.44 Mesalina brevirostris Recommendations for Further Research and Action Given the rate of development of along the coast and near the offshore islands, there will be greater pressure for commercial development including oil and gas extraction, as well as an increase in commercial and artisanal fishing, and recreational activities. In addition, long-term conservation can only be achieved through a sea-change in human behavior, and the best way to affect such a change is through education which helps people to understand and revalue the native species and their habitat. In order to make appropriate decisions that will benefit all parties, the stakeholders should be involved in defining the areas and acceptable use, as well as the making of regulations. This style of decision making must be supported by research to assess specific impacts, such as camel and sheep grazing on reptile numbers and diversity in the area and the impact of fishing on sea snakes. Another aspect of moving toward an integrated management of the coastal margin, the nearshore islands, and the offshore islands is the development of a public education program about the importance of conservation of natural habitats, and their flora and fauna. One shorter term action is to preserve large areas of habitat as nature reserves; this does not preclude human use of an area but does restrict the activities allowed in the area so that the habitat is not degraded. For example, grazing can be allowed in nature reserves as long as the number of livestock is restricted to avoid damage to the substrate and vegetation. Another example would be to restrict human recreational activity to certain sites to ensure other areas remain undisturbed. Conservation management planning must be based on a set of clear objectives and clear definition of those areas where the animals live. Because many species of reptiles, amphibians, mammals, birds, and invertebrates utilize natural areas, it is better to think in terms of conservation efforts in broad areas of habitat than to focus efforts on a single species. A conservation program that deals with management of a region incorporating several habitats will benefit all species that live in the area. The definition of the special management areas where human activity is regulated will benefit the other species that share the habitats (e.g., sabbkas, hummock-dunes, seagrass meadows, rocky reef areas). In the marine environment, this is also good management for fisheries because larval and juvenile fishes live in seagrass and or reef areas. Protecting areas does not mean that there can be no human activity; however, it does mean that the human activity must be regulated. For example, commercial fishing boats could continue to use an area but with appropriate fishing practices (e.g., Turtle Exclusion Devices, TEDs) and the regular checking of set fishing nets so that turtles and other marine species do not become entangled and drown. Recreational fishermen can also use areas inhabited by marine turtles and sea snakes, as long as they respect areas designated as go slow zone wherein boat speed should not to exceed 10 knots. Terrestrial areas could be designated for camping or off-road driving, while other areas are protected. The first step in achieving such a broad goal as conserving habitat on a large scale is to assemble the existing data and identify what additional data must be collected to support informed decision making. Unfortunately, conservation management intended to protect reptile and amphibian species is constrained by insufficient data regarding the status of their populations and their habitat requirements. 290 Marine Atlas of the Western Arabian Gulf

27 Plate 9.45 Trapelus persicus using a shrub as a basking and observation site. (Photograph by Michael R. Reilly) Plate 9.47 The Arabian toad-headed agama (Phrynocephalus arabicus) will bury itself in the fine sand to hide when chased. (Photograph by Michael R. Reilly) Plate 9.46 A dhub (Uromastyx aegyptius) displaying mock aggression. (Photograph by Michael R. Reilly) Plate 9.48 Acanthodactylus schmidti forages widely among coastal habitats. (Photograph by Michael R. Reilly) Chapter 9: MARINE AND COASTAL REPTILES 291

28 Detailed methods are available for assessing terrestrial reptile and amphibian populations on large and small scales (e.g., Heyer, et al., 1994; Ryan et al., 2002; Campbell and Christman, 1982; Manley, et al., 2006; Szaro, et al., 1988; Margules and Austin, 1991). Assessment of the herpetofauna and threatening processes along the coastal margin requires sampling at multiple sites over several seasons; although labor intensive and logistically complex, such sampling would provide much needed data. Besides surveying the coastal margin, habitats contained the in secure areas inside the fences of airports, industrial facilities and military bases should be surveyed because they are de facto preserves and should provide insight into the composition and function of relatively undisturbed populations. The methods necessary to collect the required data for marine turtles are well established (Miller 1989b, 1999; Eckert, et al., 1999; Pilcher, 2004). Working in the high density nesting period would mean that a research team could spend approximately 2 1/2 weeks on each of Jana and Karan Islands and maintain a critical dataset on the populations through which changes can be detected. Data concerning important population characteristics such as the curve carapace length, weight, number of eggs laid, number of hatchlings produced, and re-nesting interval are required to detect change in the status of the populations. Because the threat to marine reptiles, and other marine organisms, continues from the slow accumulation of changes in their environment good conservation based management must start with monitoring the populations to develop a database against which change can be measured. Similarly, the populations of sea snakes should also be assessed systematically to determine their distribution, population structure, habitat use, reproductive characteristics, and food. In the marine environment, Saudi Arabia has already established baseline data on the distribution of nesting areas and several foraging areas used by marine turtles; it has baseline data on the reproductive characteristics of the populations that breed on the offshore islands. Saudi Aramco has mapped the benthic structure of much of the Arabian Gulf Coastline using GIS. This information can be used to provide an initial definition of foraging areas. In support of these efforts, the nearshore benthic communities and habitats should be mapped and ground truthed, such as is under way in Saudi Arabia (e.g., Khan, et al., 1992; Farrand, et al., 2002; Purkis and Riegl, 2005) and in Bahrain (GEOMATIC, 2006). Further, fishing activity in these areas needs to be monitored to assess (with a view to reducing) the impact on the bycatch, as well as the target species (e.g., Abdulqader, 2006). When the existing data on the areas used by commercial fishermen are integrated with both the habitat and the distribution of marine turtles, areas of potential conflict can be identified. As Saudi Arabia is doing, the other countries that surround the Arabian Gulf should manage sea turtle and sea snake populations and their habitats within their national borders. However, given that marine species do not recognize international borders, the national efforts would be more effective, if they were coordinated through a region-wide organization. Plate 9.49 The toad-headed agama. (Phrynocephalus arabicus) Viable populations of coastal reptiles and amphibians as well as marine turtles and sea snakes will continue to thrive in Saudi Arabia as long as the coastal terrestrial habitats, and marine areas (seagrass meadows, rocky reef areas, and the beaches of the islands) are protected. 292 Marine Atlas of the Western Arabian Gulf Chapter 9: MARINE AND COASTAL REPTILES 293

29 Map 9.1 Jana Island indicating Turtle Nesting Sites.

30 Map 9.2 Karan Island indicating Turtle Nesting Sites.