Jeffrey A. Seminoff , Wallace J. Nichols, Antonio Resendiz, and Anthony Galvan. 2000. Diet composition of the black sea turtle, Chelonia mydas agassizii, near Baja California, Mexico. In: Abreu-Grobois, F.A., R. Briseno-Duenas, R. Marquez, and L. Sarti, compilers. Proceedings of the Eighteenth International Sea Turtle Symposium. U.S. Dep. Commer. NOAA Tech. Memo. NMFS-SEFSC-436, 293 pp.

INTRODUCTION

The coastal waters of the Baja California peninsula have been considered important to the life history of the black sea turtle, Chelonia mydas agassizii (Townsend, 1916; Carr, 1961; Caldwell, 1962; Felger, 1976, Alvarado and Figueroa, 1992; Cliffton et al., 1982). The Gulf of California on the eastern coast of the peninsula and the eastern Pacific ocean on the west offer a variety of marine habitats. As turtles move into the Gulf, they enter a semi-enclosed body of water that is considered a dynamic and productive ecosystem (Brusca 1980, Pacheco and Zertuche 1996). Supported by seasonal upwelling of nutrient rich waters, coastal areas of this sea host diverse assemblages of fish (Thomson et al.,1979), in- vertebrates (Brusca 1980), marine alga (Norris 1975), and seagrasses (Felger and Moser 1973). Previous data from the Gulf shows that black turtle diet is composed primarily of red algae species (Seminoff et al., In press). In addition, in- vertebrates such as sponge, soft corrals, Sabellid worms, and gastropods are ingested.

Upon dispersal to Pacific coastal waters of the Baja peninsula, turtles enter a region characterized by sandy coasts interspersed with bays and estuaries. These bays are typi- cally soft bottom and host seagrass communities dominated by eelgrass, Zostera marina (Dawson, 1951). Black sea turtles have been documented as historically abundant these areas

(Townsend, 1916; Nelson, 1921; Hodge, 1979). Although to date few data have revealed eelgrass consumption by black turtles in this region (Stinson, 1984), consumption of eel- grass by black sea turtles has been documented at similar latitudes within the Gulf (Felger and Moser, 1973).

While it is apparent coastal ecosystems along the Baja California peninsula may provide a variety of potential food resources for black sea turtles, the diet of this species is poorly understood. Based on three years of data, it appears that black turtles in the Gulf are primarily herbivorous. In the Pacific, it is probable that blacks consume eelgrass, but data are lacking. In this report we present additional data from the Gulf of California and preliminary results from the Pacific coast of the Baja peninsula.

METHODS

Turtle capture was facilitated by the use of two entangle- ment nets (100 m x 8 m, mesh size= 60cm stretched). Nets were regularly monitored during each netting trial and turtles were removed immediately to minimize capture-related stress. Upon capture, straight carapace length (SCL), weight, and other physical data were recorded and diet samples col- lected. Oral examination was used to recover residual food particles and lavage, the esophageal flushing of food components, was performed to recover ingested food samples (Forbes and Limpus, 1993). In-flow and retrieval tubes mea- sured 1.5 m. For small turtles (<55 cm SCL), both tubes were 12 mm inner diameter (I.D.) and 17 mm outer diameter (O.D.). For large turtles (>55 cm SCL) in-flow tubes remained 12 mm I.D. / 17 mm O.D. and retrieval tubes measured 17 mm I.D. and 21 mm O.D. Volume was calculated through water displacement in a graduated cylinder.

Fecal samples were collected from a subset of captures. All fecal samples were collected from the Bahia de Los Angeles study area. For this study turtles were placed into solitary holding tanks (2 m diameter) at the CRIP Sea Turtle Research Station in Bahia de Los Angeles. Turtles were monitored and feces were removed immediately after excre- tion. All turtles were released at the site of initial capture within 24 hours.

RESULTS

Lavage samples were collected from a total of 84 turtles from five sites. Average lavage sample volume was 437 ml. Fecal samples were collected from a total of 34 individuals. All fecal samples were collected at the Bahia de Los Angeles study site. Average sample vol. was 587 ml.

In the Gulf of California, marine algae accounted for 92% of the average lavage sample volume. A total of 20 algae species were recovered, 8 major diet components. Over- all, Rhodophyta dominated with an average of 89% sample volume from the Gulf. The most prevalent was Gracilariopsis laeminoformis (83% of average sample volume, freq. of oc- currence = 71/78 samples). Further, in the 78 lavage samples, red algae were dominant (72 samples with >5% sample vol- ume). Chlorophyta was second most utilized (17 samples with >5% sample volume) and Phaeophyta was the least fre- quent (1 sample with >5% sample volume).

In Bahia de Los Angeles, non-algal diet components included a total of 22 species; 9 spp. recovered from lavage and 20 spp. from feces. The most frequently occurring non-algal ingesta included Sabellid worms (56% of samples), sponges (47%), Stinging Hydroids (41%), small Gastropods (41%), and sea pen tests (23%). Plastic debris was found in 20% of all samples. Substrate particles were found in a total of 61% of lavage and 97% of fecal samples.

Preliminary data from the Pacific show marine algae and seagrass accounted for 99% of the average sample vol- ume. Eelgrass (Zostera marina) was the most prevalent seagrass (44% average sample volume, occurred in 4/6 of samples) and the red algae Gracilaria spp. was the most prevalent marine algae (47%, 4/6). In addition, the seagrass Halodule wrightii (1.6%, 4/6) and marine algae Codium sp. (10%, 1/6), Ulva lactuca (<1%, 1/6), and Sargassum sp. (<1%, 1/6) were found. Non-algal items included soft Gorgonia (<1%, 2/6) and substrate (<1%, 5/6).

DISCUSSION

Black turtles feeding along the shores of the Baja California peninsula exhibit a strong tendency toward herbivory.

Though over 20 algae/seagrass species were recovered in lavage samples from both coasts, there were only eight major diet components. We define major diet components as any food item that comprises >5% of total volume of at least one sample. Other algae with lower consumption levels may have been incidentally taken during foraging. In all cases, the major algae species present in lavage samples were consistent with predominant algae species in the area of capture. For example, the high proportion Gracilariopsis laeminoformis and Gracilaria robusta in lavage samples from Bahia de Los Angeles may reflect the high abundance of these species at this study area (Pacheco-Ruiz, pers. com.).

Fecal sampling yielded the highest number of non-al- gal food items recovered (20 spp.) when compared to lavage samples (9 spp). There are a number of possible explanations for this. First, while lavage samples represent recent feeding, fecal samples may represent feeding over a longer period of time and therefore increase the likelihood of re- covering more species. Second, invertebrate fragments may be too large to pass through the lavage retrieval tube (17 mm O.D.). Nevertheless, this demonstrates the importance of using lavage and fecal sampling in the analysis of diet composition. In our study, the concurrent use of both methods concurrently greatly increased the number of species recovered.

Non-algal species were more prevalent in Gulf of Cali- fornia samples as compared to those from the Pacific. This apparent higher use of non-algal resources may reflect high prevalence of macroinvertebrates in the rocky shoreline habi- tats of the Gulf (Brusca 1980). Of particular interest was the frequent occurrence of Sabellid worms in both lavage and fecal samples. These worms were prevalent in a majority of samples. Considering their patchy distribution, it is possible that these items are actively sought out during foraging activities. A similar scenario may occur with the fleshy sea pen, Ptilosarcus undulatus. This solitary species appears to be uncommon in the study area yet in several fecal samples (N=7) up to 50 tests were recovered.

The frequent occurrence of substrate particles in dietary samples from all sites suggests that feeding turtles may be may be ingesting this material incidentally as they closely crop seagrass and algae. Substrate may also be ingested as turtles feed on Sabellid worms, fleshy sea pens, and other benthic organisms.

CONCLUSION

Although Chelonia mydas agassizii utilize non-algae food resources, their predominantly herbivorous diet is con- sistent with other Chelonia populations. To further elucidate this trend, we must continue to collect dietary samples from a variety of sites along the Baja California peninsula. Addi- tionally, we recommend that analyses of diet composition of other Chelonia populations utilize both lavage and fecal sam- pling in order to maximize the number of food species re- covered.

This analysis of diet composition is the first step towards understanding the feeding ecology of the black sea turtle in Baja Californian waters. Future study will include the analysis of local movement and food availability within individual homeranges. When this information is coupled with diet composition, a more thorough understanding of black sea turtle behavior and feeding ecology will be gained. Furthermore, by learning what resources they are most commonly using and where they are moving on a daily basis, we can begin to make educated management decisions regard- ing gill netting activities and commercial algae harvest.

ACKNOWLEDGEMENTS

We are indebted to the following individuals for their generousassistanceduringthisstudy: AnaBarragan,Marcos Blanco, Kim Cliffton, Richard Felger, Jennifer Gilmore, Amanda Jaksha, Isai Pacheco-Ruiz, Martin Pepper, Bety Resendiz, Mauro Rosini, Cecil Schwalbe, Francisco Smith, Yoshio Suzuki, Donald Thomson, and all the Earthwatch team members. Further, we thank Earthwatch Institute, Wallace Research Foundation, University of Arizona, and National Geographic Television for financial and logistical support. This research was performed under permits issued by Secretaria de Medio Ambiente, Recursos Naturales, y Pesca (SEMARNAP); Permiso Pesca de Fomento No. 150496-213- 03; No. 269507-213-03, and No. DOO 750-07637. We would like to give our thanks to the Government of Mexico.

LITERATURE CITED

Alvarado, J. and A. Figueroa. 1992. Recapturas post- anidatorias de hembras de tortuga marina negra (Chelonia agassizii) marcadas en Mich, Mex. Biotropica 24(4):560- 566.

Brusca, R.C. 1980. Common intertidal invertebrates of the Gulf of California. University of Arizona Press, Tucson, AZ. 513pp.

Carr, A. 1961. Pacific turtle problem. Natural History 70:64- 71.

Cliffton, K., D.O. Cornejo, and R.S. Felger. 1982. Sea turtles of the Pacific coast of Mexico. In: K. Bjorndal (Ed.), Biology and Conservation of Sea Turtles. Smithsonian Inst. Press, Wash., D.C. pp. 199-209.

Dawson, E.Y. 1951. A further study of upweling and associated vegetation along Pacific Baja California, Mexico. J.Mar.Res. 10:39-58.

Felger, R.S., and M.B. Moser. 1973. Eelgrass (Zostera marina L.) in the Gulf of California: Discovery of its nutritional value by the Seri indians. Science 181:355- 356.

Forbes, G. and C. Limpus. 1993. A non-lethal method for retrieving stomach contents from sea turtles. Wildl. Res. 20:339-343.

Hodge, R.P. 1979. Geographic distribution: Chelonia mydas agassizii. Herp. Rev. 12(3):83-84.

Nelson, E.W. 1921. Lower California and its natural resources. Mem. Natl. Acad. Sci. 16(1):194pp.

Norris, J.N. 1975. Marine Algae of the Northern Gulf of California. Ph.D. Dissertation. University of California, Santa Barbara. 575pp.

Pacheco-Ruiz, I. and J.A. Zertuche-Gonzalez. 1996b. Green algae (Chlorophyta) from Bahia de Los Angeles, Gulf of California, Mexico. Botanica Marina 39: 431-433.

Seminoff, J.A., W.J. Nichols, and A. Resendiz. In Press. Diet composition of the black sea turtle, Chelonia mydas agassizii, near Bahia de Los Angeles, Gulf of California, Mexico. Proceedings of the 17th annual Symposium on Sea Turtle Biology and Conservation.

Stinson, M.L. 1984. Biology of the sea turtles of San Diego Bay, California, and in the northeastern Pacific Ocean. Unpublished MS Thesis, S.D. State U. 285pp.

Thomson, D.A., L. Findley, and A. Kerstitch. 1979. Common reef fishes from the Sea of Cortez. John Wiley and Sons, New York. 302pp.

Townsend, C.H. 1916. Voyage of the Albatross to the Gulf of California in 1911. Bull. Amer. Mus. Nat. Hist. 35(24):399-476.