Activity patterns of Kemp's ridley turtles, <Emphasis Type="Italic">Lepidochelys kempii</Emphasis>, in the coastal waters of the Cedar Keys,Florida

Radio and sonic telemetry were used to investigate the tidal orientation, rate of movement (ROM), and surfacing behavior of nine Kemp's ridley turtles, Lepidochelys kempii, tracked east of the Cedar Keys, Florida. The mean of mean turtle bearings on incoming (48ᇅ °) and falling (232ᆽ °) tides was significantly oriented to the mean directions of tidal flow (37Nj °, P<0.0025, and 234Nj °, P<0.005, respectively). Turtles had a mean ROM of 0.44ǂ.33 km/h (range: 0.004-1.758 km/h), a mean surface duration of 18ᆣ s (range: 1-88 s), and a mean submergence duration of 8.4Lj.4 min (range: 0.2-60.0 min). ROM was negatively correlated with surface and submergence durations and positively correlated with the number of surfacings. Furthermore, ROMs were higher and surface and submergence durations were shorter during the day. Daily activities of turtles were attributed to food acquisition and bioenergetics.     

Satellite telemetry tracking of swordfish, Xiphias gladius, off the eastern United States

Swordfish (Xiphias gladius) were tagged with satellite "pop-off" tags that release from the fish after a preprogrammed time, float to the sea surface, and transmit present position and archived temperature data. Swordfish were tagged on the "Charleston Bump," a topographic feature on the Blake Plateau east of South Carolina and Georgia. This feature is an important swordfishing ground and may be a spawning and nursery area. Swordfish were tagged in spring of 2000 to determine movements in relation to the Charleston Bump, and tags were programmed to pop off the fish at 30 days (n=10 tags), 60 days (n=10), and 90 days (n=9). Although four swordfish were found in the vicinity of the Charleston Bump up to 90 days after tagging, most moved considerable distances to the east and northeast and were subsequently located in association with offshore seamounts, submarine canyons of the Middle Atlantic Bight, and with thermal fronts of the northern wall of the Gulf Stream. The longest minimum (i.e., straight-line) distance tracked was 2,497 km, and maximum speed inferred from tracking was 34 km/day. Seawater temperature data archived by the tags reflected diel vertical migrations in swordfish.     

Intra-Mediterranean migrations of loggerhead sea turtles (Caretta caretta) monitored by satellite telemetry

The movements of four Mediterranean loggerhead sea turtles (Caretta caretta; three females, one male) were tracked via satellite telemetry for between 108 and 457 days. Total length of the routes traveled by the turtles varied between 2554 and 7098 km, and the average travel rate was 1.2 km h^-1. Long-distance movement between the western and eastern Mediterranean basins followed a seasonal pattern and seemed to be triggered by temperature and food availability. In the autumn/winter months turtles moved from west to east in search of warmer waters, and returned to the western basin in spring, where food resources are generally more plentiful. Three (two females, one male) of the four turtles migrated eastward through the Straits of Messina, which is characterized by high fishing pressure and intense boat traffic. Information about turtle migration patterns and routes will serve to plan effective conservation strategies.     

Migration of green turtles <Emphasis Type="Italic">Chelonia mydas</Emphasis> from Tortuguero,Costa Rica

During 1955–2003, flipper tags were attached to 46,983 green turtles and ten turtles were fitted with satellite transmitters at Tortuguero, Costa Rica. Eight satellite-tracked turtles stayed within 135 km of the beach and probably returned to nest after release. The internesting area is more extensive than previously documented. Post-nesting migration routes of satellite-tracked turtles varied. Seven turtles swam close to the coast and three turtles swam through oceanic waters before moving toward nearshore areas. Sea surface height anomaly maps indicate that oceanic movements were consistent with the southwestern Caribbean gyre. Circling and semi-circling turtles could have been disoriented but submergence and surface times suggest they may have been feeding in Sargassum sp. concentrations. Rapid post-nesting migrations (mean 2.2 km hr⁻¹) ended on benthic feeding grounds in shallow waters (<20 m) off Belize (n=1), Honduras (n=1) and Nicaragua (n=8). The spatial distribution of migration end points (n=10) and tag returns (n=4,669) are similar. Fishermen in Nicaragua target green turtles along migratory corridors and on foraging grounds. Management efforts are urgently needed in Nicaragua, particularly in the high-density feeding areas south and east of the Witties (N14°09 W82°45). The proximity of foraging grounds to the nesting beach (mean 512 km) may permit female turtles to invest more energy in reproduction and hence the Tortuguero population may have greater potential for recovery than other green turtle nesting populations. Recovery of the Tortuguero green turtle population will benefit countries and marine ecosystems throughout the Caribbean, especially Nicaragua.     

Testing the navigational abilities of ocean migrants: displacement experiments on green sea turtles (Chelonia mydas)

Like many animals migrating through the oceans, sea turtles face difficult navigational tasks when they have to reach distant, specific sites. The paradigmatic case of Brazilian green turtles (Chelonia mydas), which nest on the tiny Ascension Island in the middle of the Atlantic Ocean, has often been the subject of hypotheses concerning their navigational mechanisms. To investigate their nature, we displaced 18 females from Ascension and tracked them by satellite after release from eight different points in the ocean, 60-450 km away from the island. Four turtles moved to Brazil soon after the release, 4 moved in various directions before heading to Brazil, and 10 reached the island. All the successful trips, bar 1, were winding but ended with a final straight segment of variable length, as if the turtles were searching for a sensory contact with the island which they obtained at various distances. The approach to Ascension mostly occurred from the direction opposite to the trade wind, suggesting a navigational role of wind-borne information originating from the island.     

Ecology of neonate loggerhead turtles inhabiting lines of downwelling near a Gulf Stream front

Neonate sea turtles disperse from nesting beaches into the open ocean and develop during a multi-year growth period at sea, but data that characterize their behavior, feeding, and habitat during this developmental period have been few. Limited information has suggested that neonate sea turtles associate with lines of floating debris and biota at areas of surface downwelling. Data from the present study come from measurements of habitat, turtle behavior, and apparent foraging preferences in areas where neonate (post-hatchling) loggerhead turtles (Caretta caretta) were observed and captured. Turtles were observed (n=293) and captured (n=241 of the 293 observed) in downwelling lines that had formed in the slope water near the Gulf Stream front off east-central Florida, USA. Catch-per-unit-effort averaged 12.4 turtles/h from a vessel moving at 2.5 knots. Turtles were largely inactive and were closely associated with floating material, especially pelagic species of Sargassum. Turtles captured along with samples of floating material and given a gastric-esophageal lavage showed a preference for animal material (35.5% of volume in habitat, 70.9% in lavage) over plant material (60.3% of volume in habitat, 22.5% in lavage). Ingested anthropogenic debris included tar (20% of turtles) and plastics (15% of turtles). Ingested animals were principally small (most <1 mm) and were typically slow-moving or non-motile species or stages. Ingested plants were most commonly Sargassum fragments or seagrasses that bore epiphytic animals. Preferred or commonly ingested animals were hydroids, copepods, and pleuston such as Janthina, Creseis, Porpita, and Halobates. Data support a hypothesis describing post-hatchling loggerheads as facultatively active but principally low-energy float-and-wait foragers both within and outside of downwelling lines. Pelagic dispersal of turtles may be best predicted by a "smart" drifter analogy wherein turtle buoyancy, surface advection, and minimal oriented movement determine their distribution at sea. Conservation implications of plastic and tar ingestion are discussed.     

Tracking hawksbill sea turtles (<Emphasis Type="Italic">Eretmochelys imbricata</Emphasis>) during inter-nesting intervals around Barbados

Hawksbill sea turtles (Eretmochelys imbricata) nesting in Barbados (Needham’s Point, 13° 04′ 41.33′′ N, 59° 36′ 32.69′′W) were outfitted with GPS dataloggers over three breeding seasons (2008–2010) to track movement during inter-nesting intervals. Most females established spatially restricted resident areas up current and within 7 km of the nesting beach where they spent the majority of the inter-nesting interval. Females nesting earlier in the season settled on shallower sites. Only experienced remigrant turtles occupied the most distant resident areas. Females tracked for multiple inter-nesting intervals exhibited site fidelity, but the area contracted and the activity of females decreased with each successive interval. Hawksbills may trade off site characteristics with distance from the nesting beach and reduce activity over the course of the breeding season to optimise energy reserves during inter-nesting intervals.     

Diving behavior of immature hawksbills (Eretmochelys imbricata) in a Caribbean cliff-wall habitat

Time-depth recorders were attached to immature hawksbills (Eretmochelys imbricata Linnaeus, 1766) residing at the northwestern cliffs of Mona Island, Puerto Rico. Data on diving profiles obtained from four turtles of 32.0 to 53.4 cm straight carapace length revealed strong similarities in behavioral patterns. During daylight hours, turtles were active an average 8.4 h per day, surfacing 3.6% of the time. Length of foraging dives correlated with turtle size, with mean durations ranging from 8.6 to 14.0 min. Foraging dives, with a mean depth of 4.7 m, were associated with feeding on encrusting sponges. At night, turtles were mostly inactive, surfacing 1.8% of the time and with individual mean submergence intervals of between 30.4 and 37.1 min. From the surfacing behavior of turtles making deep and long dives, it is inferred that dives were aerobic, with the turtles making use of oxygen stores in addition to that of the lung.     

The movements and submergence behaviour of male green turtles at Ascension Island

The movements and submergence behaviour of two male green turtles (Chelonia mydas) on their mating grounds at Ascension Island were investigated by satellite telemetry. During the mating season, males tended to conduct much shorter dives (typically <15 min) than those recorded previously for females during the internesting period at this rookery. This suggests that throughout the mating season males maintained relatively high activity levels, presumably associated with locating and mating with as many females as possible to maximise their reproductive output. At the end of their residence at the mating ground, the two males conducted longer dives (48 min and 21 min, respectively), suggesting that they rested before their migration away from the island. Although very few locations were obtained during this migration, those obtained showed that males migrate to South America, as has been shown previously for females from this population.     

Shark-inflicted injury frequencies, escape ability, and habitat use of green and loggerhead turtles

Interactions between large marine predators and their prey are difficult to observe and little is known about the risk of predation faced by sea turtles. The frequency of predator-inflicted injuries, however, has afforded insights into the predation risk faced by many taxa. We measured the frequency of shark-inflicted injuries on green (Chelonia mydas) and loggerhead (Caretta caretta) sea turtles in Shark Bay, Western Australia with a view to determining differences between species and sex-classes in the risk of predation from tiger sharks (Galeocerdo cuvier). Furthermore, we investigated how escape ability and habitat use might influence the probability of turtles being injured by sharks. Shark-inflicted injuries were more frequent on loggerhead than on green turtles, and most frequent on adult male loggerhead turtles. Species effects could not be attributed to differences in habitat use, since green turtles were found in habitats favored by tiger sharks more often than were loggerhead turtles. Green turtles, however, were faster and maneuvered better than loggerhead turtles, suggesting that escape ability is a factor in interspecific differences in injury frequency. The sex-class difference in injury frequency of loggerhead turtles suggests that males face greater predation risk than females and may take more risks. For green turtles, the lack of a sex difference in injury frequency might be due to greater escape ability lowering overall predation risk or to no differences between sexes in the benefits of risk-taking.     

Horizontal and vertical movements of sailfish (<Emphasis Type="Italic">Istiophorus platypterus</Emphasis>) in the Arabian Gulf,determined by ultrasonic and pop-up satellite tagging

Small-scale movement behavior of sailfish (Istiophorus platypterus) in the Arabian Gulf was evaluated between April 2002 and April 2004 using temperature- and pressure-sensitive ultrasonic telemetry. Of nine sailfish tagged, eight were successfully tracked for periods ranging from 3 h 33 min to 52 h 6 min. Total tracked distances ranged from 5.5 to 78.7 km, while maximum linear displacement from tagging locations ranged from 4.6 to 37.0 km. Average speed based on vessel positioning ranged from 0.29 to 0.75 m s^–1. The cumulative mean vertical distributions showed that 84.3% of time was spent above 10-m depth, even though water temperature altered little with increased depth. Data from two pop-up satellite archival tags deployed in 2002 were used to compare time spent at 5-m depth intervals with data from ultrasonic tags. There was no significant difference (P<0.05) in preferred depths between ultrasonic and pop-up tags for day or night, suggesting that the sailfish in this study recovered from capture stress and returned to normal behavior in relatively short times. Information on vertical and horizontal distribution can reveal preferential habitat that benefits fishery management practices relating to time/area closures, as well as determining optimal gear selection that reduces incidental bycatch and promotes conservation of sailfish in the Arabian Gulf.Communicated by O. Kinne, Oldendorf/Luhe     

Geolocation of free-ranging fish on the European continental shelf as determined from environmental variables

Demersal fish cannot readily be tracked using satellite-based or light-based geolocation techniques. As an alternative, we describe the tidal location method, which uses tidal data recorded by electronic data storage tags (DSTs), to determine geoposition. Times of high water (H) and tidal ranges (R) recorded by DSTs moored at known locations, and from free-swimming tagged plaice, Pleuronectes platessa, were compared with a North Sea tidal database to identify all positions with matching values of H and R. Within the recording precision of the tag (ǂ.2 m, ᆞ min) and the predicted accuracy of the model generated tidal data (ǂ.15 m, ᆨ min), geolocations over much of the North Sea and eastern English Channel were predicted to be accurate to within 40 km, sometimes to within 10 km. Positional estimates of the moored tags were within 15.7Dž.5 km of the actual locations. Geolocations made from tagged plaice within 5 days of release and 5 days pre-recapture were within 35ᆬ km and 37ᆫ km of release and recapture positions respectively. Our results demonstrate the ability of this method to accurately describe the migrations of North Sea plaice throughout their geographical range with a level of accuracy unattainable using light-based geolocation. The method could equally be applied to any shelf-dwelling demersal fish that periodically rests on the sea-bed for the duration of a tidal cycle. In fisheries management, the method has clear potential application in defining the movements and migrations of other commercial species.     

Interactions among endocrinology,seasonal reproductive cycles and the nesting biology of the female green sea turtle

We collected data on plasma levels of testosterone+5a-dihydrotestosterone (T+DHT) and corticosterone (CORT) from adult female green sea turtles (Chelonia mydas) from southern Queensland during distinct stages of their reproductive cycle. Those females capable of breeding in a given year had elevated plasma steroid levels (T+DHT 0.91ǂ.08; CORT 1.05ǂ.29 ng/ml), associated with follicular development, until courtship began in October. At the beginning of the nesting season in November plasma levels of CORT were related to when the female first nested (r²=0.06; F=10.45; P=0.01). However, they were not correlated with the number of clutches a female laid in that season (F=3.65; P=0.08). We repeatedly sampled 23 turtles over the nesting season and profiled changes in steroids immediately following oviposition of each clutch. Levels of T+DHT (range 0.41-0.58 ng/ml) and CORT (range 2.13-2.81 ng/ml) were similar through the early stages of the nesting season and inter-nesting period, and declined to near basal levels (T+DHT 0.37ǂ.03 and CORT 1.85-ng/ml) following the last clutch for the season. Steroid hormone levels were also low (T+DHT 0.38ǂ.16; CORT 0.46ǂ.21 ng/ml) in four independent post-breeding (atretic) females; samples for these females were taken at a time when body condition was presumably at the lowest for the season. Subtle changes in the nesting environment, such as variation in nesting habitat or the time of night that nesting occurred, were associated with a small and slow CORT increase. We suggest CORT is increased in nesting females to assist in lipid transfer to prepare the ovarian follicles and/or the reproductive organs for ovulation.     

Ecology of subtropical hermit crabs in SW Madagascar: short-range migrations

Many mobile animals migrate because of the different benefits provided by different localities in time and space. For hermit crabs, such benefits include resource (shell, water, food) acquisition and gamete release. One of the more successful crustacean land-invaders, Coenobita hermit crabs, undertake complex short-range migrations in SW Madagascar. Number of active hermit crabs was inversely related to wind strength and positively related to tidal range, emphasising that movement would conserve water. A circadian component was also recorded in the locomotory activity of Coenobita pseudorugosus and C. rugosus. Path linearity varied with many of the same parameters, but also with beach slope. Movement was primarily perpendicular to shore in small individuals, but the parallel proportion increased with hermit crab size and tidal range, probably driven mostly by shell and food searching. Despite the costs of movement and shell carriage in the terrestrial environment, C. pseudorugosus and C. rugosus were as fast as their marine counterparts. Their speeds varied principally with individual size and were approximately 20% faster without shells and about 20% slower when climbing up a 20° slope, compared to horizontal or downhill travel. Hermit crabs, which are highly numerous and speciose in SW Madagascar, do not seem to partition niches by differential movement patterns. Aside from provision of shells in middens and capturing large adults for bait or pets, human activity may have a profound effect on hermit crab movement: observations at rare uninhabited marine reserves like Nosy Ve show that considerable diurnal activity may take place despite the apparent hostility of the environment to an essentially marine animal.     

Diel movement patterns of the Hawaiian stingray,<Emphasis Type="Italic"> Dasyatis lata</Emphasis>: implications for ecological interactions between sympatric elasmobranch species

The Hawaiian stingray, Dasyatis lata, is a common benthic elasmobranch in nearshore Hawaiian waters. Acoustic telemetry was used to track the movements of seven rays in Kaneohe Bay, Oahu, Hawaii. Rays were tracked continuously over 31-74 h periods. Geographical movements were analyzed to determine space utilization and rate of movement. Rays were found to utilize significantly larger activity spaces at night (0.83ǂ.70 km²) (mean-SD) than during the day (0.12ǂ.15 km²). Mean total activity space for rays tracked was 1.32ǂ.75 km². Rates of movement were also significantly higher at night (0.34ǂ.30 km h^-1) than during the day (0.15ǂ.22 km h^-1). Average straight-line swimming speed was 0.64ǂ.16 km h^-1, with a maximum observed swimming speed of 1.9 km h^-1. Tidal stage had no effect on rate of movement. Comparison with previously published data on juvenile scalloped hammerhead sharks, Sphyrna lewini, in Kaneohe Bay revealed a high degree of overlap in habitat use and time of activity, suggesting possible ecological interactions between these two species.     

Site fidelity and homing behavior in juvenile loggerhead sea turtles (<Emphasis Type="Italic">Caretta caretta</Emphasis>)

To investigate site fidelity and homing behavior in juvenile loggerheads ( Caretta caretta, L.), a mark-recapture study spanning four years (1998–2001) was conducted in Core Sound, N.C., USA. Each year of the study, approximately half of the turtles captured were tagged and released near the capture sites ( n=207), while the remaining turtles were displaced 15–20 km and released ( n=198). Loggerheads in both groups were recaptured in equal proportions near the original capture sites and many individuals were also recaptured in subsequent years. These data imply that juvenile loggerheads often returned to their capture sites following displacement, because if turtles dispersed randomly or remained near their release sites, then fewer displaced turtles should have been caught again. Moreover, because turtles migrate out of North Carolina sounds each winter, turtles recaptured at the same locations in different years evidently returned to specific sites following long migrations. To further investigate homing behavior, a small number of displaced turtles ( n=28) were tracked using radio telemetry following their release. Although transmitters detached from most turtles within a few days, analyses of initial headings showed strong orientation in the direction of the capture site. In addition, four turtles successfully tracked for longer periods of time all returned rapidly to the vicinity of the capture location and remained in the area. Taken together, the results of this study indicate that juvenile loggerheads exhibit fidelity to specific areas during summer months and possess the navigational abilities to home to these areas following forced displacements and long-distance migrations.     

Spatial structure, sampling design and abundance estimates in sandy beach macroinfauna: some warnings and new perspectives

We discuss methodological aspects directed to quantify the across-shore population structure and abundance of sandy beach macroinfauna. The reliability of estimates derived from design-based (stratified random sampling) and model-based (geostatistics, kriging) approaches is discussed. Our analysis also addresses potential biases arising from environmentally driven designs that consider a priori fixed strata for sampling macroinfauna, as opposed to species-driven sampling designs, in which the entire range of across-shore distribution is covered. Model-based approaches showed, spatially, highly autocorrelated and persistent structures in two intertidal populations of the Uruguayan coast: the isopod Excirolana armata and the yellow clam Mesodesma mactroides. Both populations presented zonation patterns that ranged from the base of the dunes to upper levels of the subtidal. The Gaussian model consistently explained the spatial distribution of species and population components (clam recruits and adults), with a minor contribution (Е%) of unresolved, small-scale variability. The consistent structure of spatial dependence in annual data strongly suggests an across-shore-structured process covering close to 35 m. Kriging predictions through cross-validation corroborated the appropriateness of the models fitted through variographic analysis, and the derived abundance estimates were very similar (maximum difference=7%) to those obtained from linear interpolation. Monthly analysis of E. armata data showed marked variations in its zonation and an unstable spatial structure according to the Gaussian model. The clear spatial structure resulting from species-driven sampling was not observed when data was truncated to simulate an environmentally driven sampling design. In this case, the linear semivariogram indicated a spatial gradient, suggesting that sampling was not performed at the appropriate spatial scale. Further, the cross-validation procedure was not significant, and both density and total abundance were underestimated. We conclude that: (1) geostatistics provides useful additional information about population structure and aids in direct abundance estimation; thus we suggest it as a powerful tool for further applications in the study of sandy beach macroinfauna; and that (2) environmentally driven sampling strategies fail to provide conclusive results about population structure and abundance, and should be avoided in studies of sandy beach populations. This is especially true for microtidal beaches, where unpredictable swash strength precludes a priori stratification through environmental reference points. The need to use adaptive sampling designs and avoid snapshot sampling is also stressed. Methodological implications for the detection of macroecological patterns in sandy beach macroinfauna are also discussed.     

Small-scale spatial and seasonal differences in the distribution of beach arthropods on the northwestern Tunisian coast. Are species evenly distributed along the shore?

Spatial distributions of sand beach arthropods were studied at Zouara (Nefza), a beach dune system along the northwestern coast of Tunisia. Two transects with pitfall traps, perpendicular to the shoreline and placed from sea towards land, captured spontaneously crawling arthropods. The pitfalls were used to obtain data on horizontal zonations for 2 days during different seasons (April, October). In April, in order to assess local, long-shore distributions of surface-active beach invertebrates, ten transects with pitfall traps were placed every 100 m, for 2 days. Core samples were also taken in correspondence with each of the ten transects to obtain the distribution of burrowing individuals. Sand samples were taken for successive laboratory analyses (salinity, grain size and organic contents). Local climatic conditions were also recorded. One isopod (Tylos europaeus, subdivided into three different size classes), two amphipods (Talitrus saltator, Talorchestia brito and indistinguishable juveniles) and four coleopteran species (Eurynebria complanata, Scarites laevigatus, Phytosus nigriventris, Phaleria acuminata) were considered. Mean zonations perpendicular to the shoreline showed that crustaceans were found closer to the sea-line limits than coleopterans and that between seasons some species changed their mean zonation more than others. Local, long-shore distributions of active and burrowing individuals showed that most species were unevenly distributed along the shore. These discontinuous distributions differed according to the species and age class and were mainly related to specific microclimatic conditions and to nutrition.     

Effects of Illegal Harvest of Eggs on the Population Decline of Leatherback Turtles in Las Baulas Marine National Park,Costa Rica

Abstract: Within 19 years the nesting population of leatherback turtles (Dermochelys coriacea) at Parque Nacional Marino Las Baulas declined from 1500 turtles nesting per year to about 100. We analyzed the effects of fishery bycatch and illegal harvesting (poaching) of eggs on this population. We modeled the population response to different levels of egg harvest (90, 75, 50, and 25%) and the effect of eradicating poaching at different times during the population decline. We compared effects of 90% poaching with those of 20% adult mortality because both of these processes were present in the population at Las Baulas. There was a stepwise decline in number of nesting turtles at all levels of egg harvest. Extirpation times for different levels of poaching ranged from 45 to 282 years. The nesting population declined more slowly and survived longer with 20% adult mortality (146 years) than it did with 90% poaching (45 years). Time that elapsed until poaching stopped determined the average population size at which the population stabilized, ranging from 90 to 420 nesting turtles. Our model predicted that saving clutches lost naturally would restore the population when adult mortality rates were low and would contribute more to population recovery when there were short remigration intervals between nesting seasons and a large proportion of natural loss of clutches. Because the model indicated that poaching was the most important cause of the leatherback decline at Las Baulas, protecting nests on the beach and protecting the beach from development are critical for survival of this population. Nevertheless, the model predicted that current high mortality rates of adults will prevent population recovery. Therefore, protection of the beach habitat and nests must be continued and fishery bycatch must be reduced to save this population.     

Post-nesting migrations of green turtles (Chelonia mydas) at Wan-An Island, Penghu Archipelago, Taiwan

 During the summers of 1994 to 1997, eight green turtles (Chelonia mydas) nesting at Wan-An Island, PengHu Archipelago, Taiwan, were equipped with satellite-telemetry transmitters. Using the Argos-linked satellite system, the turtles' migration routes were tracked until the transmissions stopped. The turtles migrated widely on the continental shelf to the east of Mainland China. The migration distances ranged from 193 to 1909 km, and the migration speeds from 1.2 to 2.8 km h⁻¹. The turtles apparently utilize several coastal areas as temporal residential foraging sites, and their migrations consist of both trans-oceanic and coastal legs. The wide distribution of the foraging sites of the turtles comprising this rookery reflects the extent to which the green turtle migrates in northeast Asia; regional and international cooperation will therefore be needed to conserve this declining population. Received: 14 December 1998 / Accepted: 8 May 2000